Megajoules & Gigawatts

Polarity, trigger, & bias scheme

2012-01-05T18:56:00.001-07:00

This will be uncharacteristically concise as my new job + domestic duties = dead-tired Prof. Gomez.

So! It occurred to me I hadn't posted how the switch was to be biased, triggered, etc. here it is, hopefully you can make it out:

Note that power dissipation in the biasing resistors can be non-trivial - we'll need high voltage, high value, reasonably precise resistors for this purpose. Naturally none of those I have lying around are even close to the correct values...

good news

2012-01-02T22:28:00.001-07:00

My air drier -- which is a necessary component for operating the triggered switch -- is in perfect working order, near as I can tell, aside from the pilot lamp in the power switch being burned out.

Here's a miserable picture of the thing:

It has a little humidity indicator window which contains an "indicating" silica gel, and through which the output air passes, to give a general indication of whether the supplied air is reasonably dry. The silica gel is impregnated with cobalt chloride (which by the way is both a heavy metal poison AND a carcinogen, so treat with enormous respect) which is all-blue at humidity levels below 5% RH. Unfortunately, that corresponds to a dew point of only about -2ºF. The clay zeolite itself is capable of delivering dew points as low as -20ºF. Since we want our air to be as dry as reasonably practicable, this calls for a down stream air drier of silica gel, which will get us down to -40ºF. Achieving a dew point lower than that would require cryogenic temperatures, or bottled gas. If I used silica gel to do the brute force drying of our lovely unpredictable Colorado air, I would be regenerating it constantly. This two-stage setup with the first stage being self-regenerating strikes me as a pretty good way to get relatively free ultra-dry gas, considering I arrived at the hard part on the cheap. I picked up the air drier at an amateur radio swap meet for, I think, $20 USD.

Now then, the humidity indicator was fine the last time I mucked about with the thing, indicating the supply of dry(ish) air. But when I hauled it out again a few years later (last month) it showed all orange and pink for the first hour after I ran it. Oh dear.

According to the preventive maintenance manual I found floating on the internet, when this happens, the zeolite has become saturated and must be replaced. And oh by the way, attempting to service the drying towers in the field voids the warranty, no user serviceable parts inside, refer servicing to qualified service personnel (ie; factory) and blahblahblah.

Now it's been my (admittedly limited) experience that zeolites can be regenerated (ie, their absorbed molecules desorbed) by heating or application of high vacuum or both. In fact, this is done all the time as a matter of course. There are vacuum pumps that work via the principle. Well and so. Simply running the unit wasn't solving the problem, and the unit was supposed to be self-regenerating.

The PM manual mentioned that the towers would not be properly regenerated if the units "short cycled" (the unit turning on and off too often) or if it was operated at a pressure it outside the range it was designed for. Aha. The last time I'd been using it, I had been trying to get it to run continuously, cycling between the two towers as designed, but delivering dry air at a low pressure and flow. I'd been fiddling about with the pressure regulator (which is used in a strange way in this design) and the compressor cutoff switch. I'd managed to get the operating interval and pressure so far out of whack that the towers weren't getting completely regenerated, resulting in a slow build-up of water vapor.

Once I understood better how the thing needs to operate, I was able to set the regulator and the compressor cutoff switch such that the towers were regenerated (assisted by a heat lamp) in just a few hours.

I still need a downstream silica gel drying cartridge to get the final dew point down in the -30 or -40F range, which I'll either pick up on eBay (I'm not liking the prices I'm finding for new ones) or I'll fabricate, since it looks easy and the bulk agent is cheap.

Dew point monitors/sensors for process gas streams exist, but BOY HOWDY are they expensive. With one, the silica gel cartridge might or might not be necessary. With a (far less expensive) silica gel cartridge, the dew point monitor isn't necessary.

The gas system will probably look something like this:

...except that I can already see that I need a small accumulator and a one-way valve between the air drier and the rest of the system. Its output pressure fluctuates too much as is. It's intended to work into a big reservoir, ie; waveguides.

I'm still figuring this out, because I'm trying to make use of as much on-hand parts as possible. I seem to have nearly everything (certainly the big expensive things, like the vacuum pump) but I'm still short a few odds and ends. I need to be able to control pressure very accurately and repeatably, so a capacitance manometer (Baratron) is probably on the shopping list. I've got thermocouple gauge tubes around, but they aren't useful at higher pressures. The Baratron will cover from above atmosphere to 1x10^-3 with almost single-Torr precision. (absolute accuracy I care less about than I do precision for this application, assuming the repeatability is there)

The idea is to only pull fresh dry air (or other working gas mix) through the switch when purging it after a shot. We don't want flowing gas during a shot, as it may introduce instabilities we don't want. It would almost certainly affect jitter although I don't happen to care about that yet. Yet.

So that's what the purge valve is for. We open that after a shot. Then close it and pull the switch down to operating pressure. Depending on how hard it turns out to be to maintain the switch at a constant pressure, I might either add some valves to isolate the switch completely, or I might be forced to bleed a small amount of gas through it all the time to maintain the (very low) pressure, in which case I'll plumb the purge valve around a needle valve upstream of the switch.

Speaking of the switch, here's another really excellent phonecam pic of same, attempting to reveal the geometry or relative position of the trigger plane electrode and the adjacent main electrode:

If I had been paying attention when I took this photo, two o-rings in the insulator portion, which are uncompressed, would have been removed, allowing everything to sit as it would when compressed, and the two electrodes would look about .040 closer than they do. But you get the idea.

Now, I haven't yet got a way to model fields, but based on what I've seen in many papers which depict calculated fields through switches, and the few which lead me to think this design with these shapes of electrodes could work -- based on all that I say, I think the adjacent electrode needs to be a hell of a lot more adjacent. Which is to say, I want the round bit to be closer to the plane of the trigger electrode, but still not protruding through the hole.

Right now, when the switch is fully assembled, the main electrodes are separated by .978". The trigger plane opening is 1.017". We want the hole larger than the main electrode spacing, but that's cutting it a bit finely. I think I may -- eventually -- cut new insulator housing pieces from acrylic, with shorter dimensions. I may only have to replace one piece actually. The one which is the short side now looks about right for the new long side, and a new piece cut from acrylic will be the new short side. I only used this horribly yellow bubbly urethane crap because I had it lying around and it was already exactly the the right ID and OD. It would have made better rollers for some mechanical application...

Anyway, shorter spacing is okay only up to a point. I can't let the main electrode separation (the 'D' in the Paschen Pressure-Distance curve) get too low or my operating pressure will be impractically high and pseudochannel operation may not be possible. And since I originally figured all of this out based around one case: the main electrodes I had on hand (specifically, their radius and height) and the right spacing for operating at 10kV in the pseudospark pressure regime... I should probably figure out the limits before I waste time machining new housing bits.

But as it is, I do not believe the field will distort far enough fast enough (through the trigger hole) to ensure not involving the trigger electrode, not to mention achieve the fast current rise time and commutation time I am hoping for. Besides, shorter is lower inductance. There's a reason many commercial switches look the way they do. And there's a reason my switch resembles a scaled down version of the classic T-508 switch originally developed by Physics International and now sold by Titan Pulsed Systems Division of L-3. Yes, this thing is long and huge (tho half the size of a T-508) but if the circuit geometry is right, that will be an advantage, not a disadvantage. If I'd started any smaller I'd have driven myself mad, not to mention would have had to fabricate more parts from scratch.

Having said all that, I think I will give this thing a shot or three as it is before I rebuild the housing. Parts for the gas circuit and some minor tooling to help me get the holes drilled in the end caps. MAYBE, possibly, even using the drill press, which means much sooner rather than waiting until the mill is up again - no promises on that however. It remains to be seen how clever I can be. I'd rather do good work slowly than sloppy work quickly, so if I think I really need the mill to do this without ruining my pretty end caps (which I would be REALLY upset over if I had to make them again) then it'll just have to wait until I get a phase converter. It's mainly an issue of stability and stiffness and whether the rotary table will fit on the drill press table, and how exactly I'm going to mount my end cap fixture tool to the rotary...

This work was supported by the Joss Research Institute.

Step 1 of {mumble} complete!

2011-12-29T23:08:00.000-07:00

As you can hopefully see from the no-doubt indistinct phone-cam shot accompanying this post, the reconfiguration of the trigger electrode is complete.

Next step is to add the extra holes to the switch end-caps that I mentioned in my previous post. To do that, I need to be able to mount the lathe mount I made for the switch end caps into the mill vise. And to do that, I need a big pair of vee blocks than the pair I have. Right then, one more item added to the shopping list. Oi. Been meaning to get a pair for a very long time. I scored a small precision pair on eBay some time ago, but they are too small for this job.

After the mods to the switch are complete, I have to focus on getting the gas handling system and some kind of very fast trigger ready.

For the gas handling part, I have an air drier the media of which needs drying out, a vacuum system which lacks a precision manometer, and so on - about 3/4 of what I need is ready to go or can be. My vacuum system is technically down, but I only need the roughing pump, there's an additional port for that.

As for triggering this thing in field distortion mode, I do not believe the EG&G TM-12 which I have on hand, or indeed any trigger source EG&G made -commercially- is fast enough (gasp! Yes, I said it, tho I'll be delighted to be proved wrong, it just wasn't their usual bailiwick) to multichannel this sucker without going trigatron on me. Trigatron operation means a spark develops between the adjacent electrode and the trigger electrode, and that spark provides the ionization which fires the gap. Some switches are designed to work that way. Unfortunately, in this switch it would probably result in very bad erosion of the trigger electrode, so it's an operating mode to be avoided. I also don't believe that the Pacific-Atlantic Vector Inversion Generator (trigger pulser) is in working condition, although I haven't nailed that status down with certainty yet. I have another miniature VIG in a trigger panel for a Pulsatron I scored from eBay, but I don't know yet whether I can get that unit working either. There are one or two other options, and of course there is always the Marx, except that I think its output might be too high. I know, right? There are worse problems to have. OTOH, it isn't built yet.

Odds are good that I'll try running the switch with just the TM-12 for a trigger, at least once, because I'm dying to run the thing at all. Hopefully running it at relatively low power levels will keep electrode erosion down. The first test shots will be done with a 100Ω load at 10kV, resulting in 1MW dissipated in the load resistor.

And one way or another, I have to get the Bridgeport some 3-ph power again because so many parts and features I need to make require a mill. I am looking at a rotary phase converter from American Rotary, most likely. I'm done mucking about with inverter VFDs, I think. And if any reader is wondering why one doesn't simply swap the motor for a single-phase version, there are various very good reasons which I have arrived at through my own homework and which I feel certain the reader will find far more convincing if she arrives at them himself.

I think I shall shop for a capacitance manometer (Baratron, MKS Instruments) and a simple display for same (Duniway's TeraNova displays are nice) on eBay. I'll need to be able to read the gas pressure in the switch accurately and repeatably. It will take time to find what I need for a price I'm willing to pay, so I'd best start searching now.

This work was supported by The Joss Research Institute, Laurel, MD, USA.

why so serious?

2011-12-27T20:22:00.001-07:00

For some time now, and as previously reported, I have been unable to make much headway on any of my Mad Science projects. There are various reasons for this, some good, some not-so-good.

Here are some good ones:

* I've been unemployed since May, and every single project needs parts which must be purchased. Many of said parts are relatively inexpensive, but there has been NO disposable income until recently, and that due solely to the generosity of friends. More on this down below.

* my wife just went through total hip replacement surgery. If you're not familiar with that, please take my word for it that it is brutal and very challenging to recover from. This will be the third such surgery she has endured. Recover takes months, happens at home, and I am all of the nursing staff (poor girl).

* I've got health issues of my own which are severely hampering my ability to do what I'd like to be doing. I'll spare you the details.

And now the good news: I start a new job in January and should have paychecks coming in mid-month.

I've also been honored to receive an honorarium / stipend / fellowship from The Joss Research Institute, a private R&D firm in Maryland. It came with some funds which for now went into paying bills, but which will be pulled back out of my future paychecks and earmarked exclusively for expenditure on my science projects. I think it is entirely appropriate that said money be used exclusively for that, especially given the spirit in which it was given, which was to help remove obstacles to my getting shit done. It's also easy to do this since our credit union gives us several free accounts to stick money into for various purposes.

I have to say, I am quite flattered given the smarts of the guys who gave it to me. We all communicate with several other folks with similar and branching interests along the lines of pulsed power, high voltage, high vacuum, physics experiments, lasers,

To make this post a little better than a place-holder, I offer the following:

My Triggered Spark Gap Switch requires some rework before I strap it to one of the 3kJ pulse caps and a pulse resistor and transfer a 1 kA / 1MW pulse through it.

It turns out I can do at least part of said re-work without spending money. I just need to get a different project off of the lathe so I can work on the trigger plane electrode. Two things need doing:

Thing One
I'll be making the ID about twice as large, all the better to pass a large diffuse plasma discharge, my dear, preferably without actually involving the trigger electrode - remember, this is a field-distortion switch, we hope)

-and-

Thing Two
I'll be removing the radius on the ID, leaving instead an acute angle with a sharp edge. This angle will be tapered on the large-gap side and in-plane on the small-gap side. The better to distort the field with, my dear. However, a sharp-edged trigger plane must be PREZACKLY biased with respect to the ratio of the two gap distances. I have the vague recollection that the ratio of diameter to main electrode distance was to be in the range of 1.0 - 1.6, but my memory is notoriously unreliable, so we'll just leave that a hypothetical until I find the collection of photons and spins states that constitutes a "white paper" these days...

Looky here:

That brass disk with the hole in it, bottom-center, is the trigger plane electrode. The hole in the middle needs to be about twice as big or a bit bigger. Someone mentioned (in some switch white paper I have somewhere) an empirically-arrived-at rule of thumb for main electrode distance vs. hole size for this style of switch, but finding it again will take some time...

As is probably obvious, this switch has not had a single shot through it. I realized after I finished it that the changes were needed. Just now got around to diddling with it in between thirteen other things.

I lied unintentionally: THREE things need doing. A second set of holes in each switch end plate needs to be made, identical to the ones that are there now, but countersunk in the opposite direction, for bolting the switch into a transmission line or to coaxial test loads. Having a bolt circle lets me fashion large-diameter connections and current paths which translate in the general case to low distributed inductance, which informs the minimum attainable pulse rise time. Readers who have read my earlier descriptive material here and on Flickr will have gleaned that a big part of what I am hoping to do with this switch is not merely transfer big lumps of energy around, but to achieve very fast switching speeds and commutation times -- fast, that is, for a rank amateur with not formal schooling in pulsed power!

Oh yes, I will be studying calculus in 2012, or that's the plan anyway.

I'm just not in any shape to finish the project currently occupying the lathe and which I don't want to re-center later. Humph. But I am not idle. I'll get some of the above done soonish I think. Things ought to be a LOT better in 2012 than they have in a very long time.

Stay tuned.

even yet more further additional progress on my "Last Tesla Coil", book two, part twelve, chapter 27...

2011-10-11T21:55:00.001-06:00

You might think this is the nearly-finished primary of a Tesla coil. And you would be reasonable to think so. But you would be mistaken. This is a laceration engine.

Well, I couldn't wear gloves because I needed the dexterity and the clearance between turns. I tried. I tried rubber gloves, but they didn't protect, just as I'd expected.

After I get done putting Band-Aids™ on all my fingertips, I'll have one more go at straightening and evening-up the turns, leaving them a bit tighter (to keep them from shifting -- as much), which is to say less of a perfect spiral, and more of an octagon. Then I will cut the outer end to its final length and screw it down. Wee! All this should be done a few hours after this post goes up.

It will be badly tempting to lash up a primary circuit for it just so I can try it out, but although I have most of the parts, nothing is put together. Also, I don't have quite enough capacitance, methinks. Gotta go through the calcs one more time after I measure the as-built primary inductance and plug it into the formulae. Furthermore, the HV transformer for this coil is practically unreplaceable, and I'll be damned if I will blow it up, which means I really need to finish the "Terry Filter" for it before running it. *sigh* Did I ever mention this? This coil will be fed by DC, so no resonant charging. I wish to be very kind to my precious 120mA NST, and I am effectively isolating it from the primary circuit entirely.

I wish there was a way to make that primary stay this pretty pink color forever, but there isn't. If I spray it with acrylic now, the acrylic will get all over the finished wood and mess up its finish. It wasn't practical to spray it before installing it - I thought about it for all of ten seconds. There's over 90 feet of 1" copper strip here, so I'm not gonna be dabbing Futura on it with a cotton swab, or laying it out on the sidewalk to spray it.

I'll spray or coat (Futura floor wax also works great) all the brass. That'll have to be enough shiny. The copper will all be dull brown in ten years.

I'll take a "glamor shot" of the complete Tesla coil primary and secondary -- minus one minor dodge I have to add to the strike rail -- later tonight.

Okay, it's later:

Inquiring minds will want to know what else needs to be finished before it runs. Here's a laundry list:

* primary capacitor - this will be an MMC of course, the most reliable sort there is. I have most of the caps for this, bought as a "finished" (bleh) MMC, but I am not comfortable with the number of caps in each string (determines the maximum safe operating voltage of the complete array) so I will need to add a cap or two to each string. Since that lowers the capacitance of each string I'll need to buy enough additional caps to add another string or two as well. The cap will be built with brass "strapping" terminals mounted on the exterior so that any number of strings may be placed in or out of circuit.

* spark gaps - I have one spark gap completed, but intend to build at least two more. The finished one is the "sucker gap" previously about posted here. I've got a disk, hub, and shaft done for an asynchronous rotary. I've got all the electrodes made, but not yet drilled, for a Richard Quick style gap. I've made nothing else nor thought about the housing for that. I'd kinda like to make a gap like one that Old Nik' was known to have used in his original NYC lab before it burned to the ground. It was just a bunch of brass balls in series. Not the best performing gap, but if memory serves, that was long before he'd invented several new kinds of fast-quenching gap switches.

* control panel - again, I've got most of the expensive parts for this, but there's a lot of finicky work to be done on said parts. I'll need to make new cards for the meters. I'll need to obtain a few specialty parts like a current transformer. The HV transformer is HEAVY so I'm considering building a cart for the control panel rack cabinet (which will be heavy enough without the transformer in it) and possibly putting the transformer in the cart/base. I might leave it in the panel cabinet and use the cart/base for storage of cables and parts. I haven't thought very much about that. And there's surely a ton of small parts I don't have, like controls. Much work to be done there.

* the damned primary tap connectors. These are gonna be finicky to make, but a joy to have and use versus any other primary tap method I've ever seen used by anyone, anywhere. So there. Okay look, that crazy rotating thing with the sliding brushes doesn't count - that's not "tapping". Yeah. So anyway, I think I've got the design worked out. I haven't started collecting various red metals (copper, brass, bronze) yet. And I have to make two of them. Bleh.

* cabling & miscellanea - the ground connections, a ground rod kit with straps, the umbilical between control panel and coil, to name a few examples.

So there is still a long, long way to go before we see first light on this beast.

PS: I noticed just now that pictures in some of my older posts are missing. I'll look into this in a bit. Probably nobody but me noticed anyway.

more progress on "My Last Tesla Coil"

2011-10-06T18:05:00.001-06:00

Here is a sight a few of my friends thought they would never see. The primary deck and all of the primary supports are finally complete. It may not look like much, but it's been a great deal of hand-sawing and sanding to get to this point.

This was also the day I I cut the two pieces of the strike ring and installed them. Those bits - including the strike ring insulators - will come off before I varnish the primary supports.

I have a bit of sanding to do on one or two primary supports before I start varnishing, but they are essentially done and they are installed semi-permanently. I refrained from using glue on them in case one gets broken off at some date in the distant future. With only screws holding them on, it will at least be possible to repair them.

I should have the varnishing of the primary supports done in a week, at which point I will start installing the primary copper. I will probably give them either a clear coat (depending how it looks on a test) or a very light stain and urethane, because these supports are hard rock maple and have a beautiful grain.

Oh by the way, the ground connection to the strike ring is already made, at least as far as the primary deck ground connection that exists so far... can you guess where and how?

PS: oh, and yes, that strike ring constitutes a shorted turn, an undesirable thing. I remember thinking a long time ago that I should do something about that when I got around to cutting it to fit and installing it. And now that I've done that, I still can't decide what to do about the shorted turn issue, because everything I've thought of so far seems ugly. I will probably conceal a break within an insulated sleeve concealed inside a slightly oversized mounting tab.

Tesla coil primary deck progress

2011-09-08T16:50:00.001-06:00

Due to being unemployed for three months and my arthritis acting up badly, I've been largely ignoring hobby projects lately. I've always got an excuse. But today I finished cutting and routing the slots in the primary deck.

The slots are for access to the primary windings from below. One of the things that has always bothered me about most Tesla coils (including mine) was that the adjustable primary tap connection was nearly always brought out from below the primary deck around the edge and over the top. This had two effects: it wasted inductance in the primary circuit (ie; stray inductance which does not contribute to the magnetic field of the primary) and it created a convenient high spot from which corona leaders could start - a place for the secondary streamers to strike.

I have a little cleaning up to do, then I'll touch up the varnish in the slots, do some more finishing work on the bottom, and then mount the primary supports, and finally re-coat the whole top deck.

More as it happens. Stay tuned...

quickie

2011-08-14T00:09:00.002-06:00

Couldn't sleep, so I updated my proposed design for the EML (pssst: "rale guhn" don't tell anyone) projectile package. I added a keyway to hopefully keep armature and cap wedded at least until they exit. Yes, discarding sabots are nice, but let's get the thing working at all first, shall we?

The lighter, white-ish cap on the right is to be made of Delrin (acetyl). The armature proper at rear is made of high-conductivity aluminum, 1100 series or the like. I may be able to find this profile off-the-shelf, saving me a lot of machining. The gaps creating individual fingers are just saw kerfs. The key and keyway will be done with a dovetail mill.

I might see if I can talk my friend with the CNC mill to set up and knock out about 100 pieces of each.

Note that the size is not pinned down yet. The design currently calls for a .75" bore. If I am to use the rails I have in hand, that will be the case. If I can get my hands on some bigger rails (at the price of copper today? Hah!) I would like to have more inductance in the launcher, just to increase the ratio of stray inductance (wasted) and launcher inductance (creates force on the armature).

The fingers are a dodge that's been tried a few times by several labs and which seem promising for square bores having very little pre-load on the armature. At the small scales amateur launchers are working at, and tens of kilojoules input energy, we can't afford to waste much of that energy overcoming stiction from a pressed-in armature. Conversely, since we're in the kilo-amp regime, we don't have as much difficulty with sliding contact resistance as the bigger devices do. Some preloading will absolutely be necessary, but I want to keep it to a minimum.

Fortunately, this launcher has a very high pressure pre-accelerator, compared to most amateur-built devices.

PS: the shape of the void behind the armature -- or, if you prefer, the inside shape of the contact fingers -- would be better if the fingers were thicker toward the armature than shown here, but the armature proper was designed with off-the-shelf c-channel stock in mind to save time and money.

Tesla coil progress

2011-08-13T10:58:00.000-06:00

Due to lack of employment and therefore money (and other reasons I won't get into) the various projects I had been actively working on have ground to a halt, so I turned back to what I'd been pushing before I got distracted by the Mad Scientist Light Switch, Fast Compact Marx Generator, and the Field-Distortion-Triggered Spark Gap Switch: my "Last Tesla Coil".

It was tabled because I lacked the copper strip (about 10 pounds worth) for the primary and the money to buy same. I got lucky and scored (more or less) what I wanted while working on the aforementioned stuff so with that in hand, and certain other difficulties being somewhat ameliorated, I can resume work on it.

So, this post is about the primary coil, its support structure or "deck" as I call it, a grounded "strike rail" mounted above and to the outside of the primary winding, and the various bits and bobs to make it all work.

I'm building this thing with several goals in mind which I don't often see implemented in other Tesla coils, even those built professionally for museums. I want it to have a classic, slightly ornamental, but thoroughly old-world hand-made. I want the build quality to be top notch. I'd like to see it last at least as long as the Griffith Park Observatory coil lasted before some future coiler has to rebuild it. I'd like to see it find its way into a museum or educational facility, so it needs to be well made and sturdy. Unfortunately, I also want it to be portable, so it can't be as sturdy as it would be if it was never to be moved. I want it to have the highest possible performance while retaining spark gap operation. Honest to goodness, for long term reliability and simple lifetime I believe an old-fashioned spark gap coil will last longer than a more efficient and higher-performing solid state coil.

So, for ease of use, the primary will be accessible (for tap connections, using a special connector I've designed) from below the primary deck through any one of eight radial slots. These slots would significantly weaken the deck structurally if it were not for the primary supports which stiffen the deck radially and a glue-lam reinforcing rib around the circumference of the deck.

Here's the glue-lam rib being glued and screwed to the underside of the deck:

The sanded areas on the bottom will be finished with urethane stain/varnish just like the top, after everything that needs gluing is done. Holes for the ends of the radial slots have been drilled, but the slots have not yet been cut out. The black objects are ABS sockets for the uprights which support the primary deck above the base. This "post and two decks" implementation is a very common construction style for amateur-built Tesla coils. It's not as nice looking as a full box or base (of any shape) but it's a lot lighter, it knocks down, and is therefore MUCH more portable. That ring was finished two years ago, and the deck was ready before that. Because of how this thing goes together, it is necessary that construction of the primary deck go in a specific order of steps. I couldn't glue this on until various other things happened, culminating in the need for the copper. Once I obtained the necessary copper, I was able to answer various questions letting me finally finish this bit.

The next bits necessary for this thing are the primary supports, which were last seen looking like THIS.

This week, I finished all of the brass "P-clips" which will fasten the strike rail to the strike rail insulators which are those white cylinders with the button-head cap screws in the ends. Now they look like this:

The clips are all done. Each was individually hand made from brass strip because I couldn't find what I wanted off the shelf.

The supports (which by the way are maple and are going to be f_cking gorgeous when they are stained and finished) have been marked with pencil for the saw cuts which will hold the primary windings. Each support has its cuts offset from those of its neighbors by 1/8 of the distance between two cuts, so that the spiral of the primary will be as smooth and step-free as possible. That is solely an issue of aesthetics, it will not affect performance. The long piece is a special feed-through for the ground connection to the strike rail on support #1. This connection passes entirely through the primary support and deck, is insulated to prevent primary voltage from reaching it (and setting its primary support on fire) and connected to the secondary's bottom ground connection through a brass strap attached to the bottom of the primary deck.

Now comes the "fun" part, and by "fun" I of course mean painful, torturous labor:

Each of the eight primary supports must receive fifteen carefully spaced saw cuts, for a total of 120. Maple is HARD even among hardwoods. As you can see, I have already damaged the tops in a few places. That's what happens when you're exhausted, with aching, twitchy muscles, and you decide to press onward anyway. *sigh*
I have arthritis and other musculoskeletal maladies, so this part is going to go very slowly, but just like those thrice-damned brass clips, if I do a little each day, in a week I'll be done. This is all being done by hand, as I do not own much in the way of wood working power tools, and in any case, this would require a very special (thin) power saw blade indeed, of a sort I've never seen. I am using a "back saw" (normally used in a miter box) to make cuts roughly the same width of the copper's thickness, so I shouldn't need anything fancy to keep the windings from shifting in their supports.

More as it happens, stay tuned.

more delays and worse

2011-06-09T22:33:00.000-06:00

Due to a long spate of misfortune and a short run of stupidity, I will be unable to work on any of my projects, nor have time to write about them, for at least another month, perhaps longer.

I'll resume when I resume.

And no, I am not incarcerated. At least, not yet.

the ever-popular "unavoidable delays"

2011-03-09T19:11:00.021-07:00

Most projects are either on hold or simmering. There is family drama occupying some of my time, and support for my wife some, and little left for hobbies. I have good excuses, but they're none of your business.

I'll report again when I have something significant to talk about and some free time.

Meanwhile, I am puttering about doing little useful things around the shop such as painting Glyptal 1201 onto once-adjustable power resistors, turning them into higher-voltage non-adjustable resistors.

Don't get excited, they're wirewound so inductive as hell (not yet measured, I'll get around to it) and I haven't the faintest idea of their tolerance, but frankly my friends, I couldn't care less. For this application - safety discharge current limiters - some inductance is welcome, and if they're +/- 20% on the resistance I'm happy.

This is for a safety grounding stick project. The idea being that you really (and I can't stress this enough) DO NOT want to ground out the "hot" part of a high voltage system after or during an electrocution incident with your grounded safety grounding stick... if the source you're about to touch it to has an ESR near zero and the capacity to source perhaps 300,000 amps into your ground lead. Whoa. Could be exciting.

So this safety "chicken stick", which will also be used during shut-down procedures, will be equipped with integral power resistors capable of dissipating 1kW continuously and many kilowatts in a single short pulse. They'll be able to take a full (10kV) charge to zero (no, not allowing for dielectric absorption, astute students) in a few seconds whilst avoiding a great deal of noise, energetic photons, and coronary arrests.

The resistors were left over from the old ignitron triggering circuits in the pulser, and the Glyptal (which you should look up, it is extraordinary stuff) was left over from coating the secondary of my "Last Tesla Coil". (which isn't receiving any time either, boo hoo)

The chicken stick has now gone from convenient and easy to use but unsafe to ridiculous to use, but safe. Oh well. I'll be sure to post a pic when it's done.

playing with arc-flash for fun and profit

2011-02-21T14:51:00.002-07:00

I worked on the big pulser (for the railgun) over the weekend.

Lately, job #1 has been to figure out a way to connect the capacitors to the "hot" collector plate. The means had to be inexpensive, reasonably low effort, and capable of handling the full 60,000 amps from each capacitor.

The arrangement I am trying first is to clamp four soft copper straps to the capacitor's bolt terminal with a nut, and clamp the opposite ends under the bolt rings of the old ignitron housings. The rest of the housing won't carry current. Some nice low-rise clamping rings would be better, but I don't feel a pressing need to make six of them (with eight holes each, ugh) right this minute.

I made the straps and installed them yesterday. This involved cutting out forty-eight 1" x 4" x .030" straps from copper sheet (with sheet metal sheers), removing all the corners and sharp edges on the belt sander, and punching forty-eight half-inch holes near one end of each strap.

After that, I felt an urge to know the leakage rate of the capacitors. I decided to perform a "megger" (a high voltage insulation tester) test on all six caps. Now meggers, by their nature, typically put a fair mount of voltage across the unknown. My instrument, an old Genrad unit, puts out anywhere from 100V to 500V DC.

Then, before I could start with the measurements, I had to figure out how I would safely discharge the cap bank after it had been charged by the megger.

Step the first was to wind half a dozen turns in my test leads onto some big ferrites. That should, in theory, keep the hash from the arc out of the detector circuitry of the megger. I would disconnect it by hand before shorting the bank if I trusted my gloves, but I don't. Anybody out there have a set of lineman's (Class 2 or 3) gloves in test-passing condition that they don't want?

Step two was to don rubber gloves, leather gloves, face shield, and ear muffs.

I got my measurement on the low range (100 V) and decided to take measurements at all the other ranges too. I discharged the caps after each measurement. I used a "chicken stick" (insulating shorting stick with copper bar at one end) and an extra little bit of brass lying on the top of the collector plate to strike the arcs to, so the collector plate wouldn't get too badly torn up.

At 100 volts, the 'pop' was impressive, but one does not need hearing protection. A modern firecracker is louder. At 200 volts, I decided the hearing protection was a good idea after all. At 400 volts, I worried what the neighbors might be thinking.

Note that these are 10,000 volt caps. So I was playing with 4% of the total energy this cap bank can store. I won't lie: when this machine lets go with a full charge, I don't want to be in the same room with it. Fortunately I won't have to. I have a nice concrete block wall separating my workshop from the rest of the garage. With that and the remote control panel I'm building, we ought to be golden.

In reality, we don't want this (a short circuit discharge) to happen at any voltage. Without any significant inductance or resistance for a load, the very high currents and high voltage reversal caused by high frequency ringing (since the "load" inductance is so small) stress the dielectric in the caps rather badly. It is to be avoided. These caps don't grow on trees. Ordinarily, when the caps need to be discharged (either to remove residual charge at the end of a shot, or due to an aborted shot) the caps are drained more slowly (and safely) by a large bank of power resistors connected to the cap bank through a pair of high voltage relays.

This is also why I'm testing insulation and clearances to the cabinet and the like. I do not want an uncontrolled arc happening out in the open. Aside from the obvious undesirability and damage, it will also be EXTREMELY LOUD. We're talking broken-windows-loud. Under normal operating conditions, I expect most of the energy to be absorbed or muffled in various ways. In fact, I am going to rather a lot of effort to ensure that happens.

Figuratively speaking, I have in the past had the local villagers light torches, take up pitch-forks, and storm my castle because I frightened them. News media, emergency services, and attorneys were involved. So, let's just say I am not comfortable allowing extremely loud noises to escape from my property these days.

Oh, and I hope to have the shop's old rolling door finally ripped out and replaced with a wall before the first gun firings take place. That should cut down on the blast noise a LOT.

After doing the above, I decided I really need to do a proof test on each individual capacitor, and that means disassembling everything I assembled over the weekend. Well, to be honest, I had other reasons for doing that too. I need to do a proof test on the collector stack, and THAT has to be done with the capacitors disconnected (but with the stack still mounted on the caps) for what I hope is an obvious reason.

Photos to come later.

PS: I often forget what I've described and what I have not. It occurs to me that I have not described the capacitor bank high current connections, which I've been referring to above as "the collector stack" so I'll do that now. The capacitors I have are very early pulse caps using 1960s technology. They are set up for relatively low inductance connections, although improvements have been made in pulse cap design since these were made. The capacitor connections are made between the case and a single 1/2-13 brass bolt located in the center of a round, convoluted, hard rubber insulator. The return connections to the case are made to four brass blocks soldered to the corners of the case. Threaded holes are provided in said blocks to accept 3/8 brass studs.

To minimize stray inductance, the six capacitors are bussed together with aluminum plates separated by a thin layer of strong insulating material. The ground/return plate sits directly on the capacitors' ground return blocks. It has six holes about 4" across to provide clearance for the high voltage connections. A 1/2" thick plate (could be different material, and thinner) of phenolic-linen laminate goes on top of that. It also has big holes to clear the high voltage connections. The "hot" or output plate has big holes too, but they are surrounded by rings of eight 1/4-20 threaded holes so that "something" can be connected there. Originally that "something" was an ignitron on top of each capacitor, which sat inside of a coaxial housing which connected to those rings of holes.

I am still using those housings and the original connections for my new connections, only because they are handy and they eliminate - for now - the need to make a whole bunch of new parts.

Don't Put Your Tongue On That Capacitor, You Don't Know Where It's Been!

2011-02-18T11:08:00.001-07:00

Lately, because the lack of a mill has back-burnered the final bits and pieces for my Mad Scientist Light Switch, I've been working on the railgun project instead. From the very beginning, one of my philosophies for this project was that I was going to everything right. Because time after time, I had seen other amateur railguns which were almost right in most respects, but wherein the builder got one or two crucial details wrong, and it hurt them. The details really do matter, especially when you're trying to take something which works better at large scales and scale it down.

I've been thinking a lot about safety. It's one of those details that a lot of a amateur mad scientists don't think about much. Honestly, it astonishes me that only a small handful of amateurs have killed themselves working with Tesla coils or big pulse caps. And I think I've had just about all the second chances with high voltage and lasers and explosives that the fates are likely to grant me. I'm going to try to avoid unplanned excitement for a while.

Government labs are different. They care about safety, and they spend a lot of money on it. PhDs don't grow on trees. But sometimes even smart people miss a possibility and something bad happens. I shall describe one such incident in hopes that we'll all learn from it.

This is entirely from memory, because I can't find the paper in which it is described. I do have a dead-trees copy somewhere in a box in the basement. Therefore, if I find that or (an online) copy, I'll correct any errors I make and point to a copy you can read.

Back in the 1950s, the biggest homopolar generator in history (so far) had recently been completed at Australia National University. And it wasn't long before this 500 MJ monster was connected to a new railgun. Now railguns want a lot of current, and they have a fair amount of voltage drop across the muzzle- several kilovolts in the plasma armature devices that were in vogue back then. That's a problem if you're using an HPG for your power supply, since they are great at putting out a lot of current into a dead short, but they aren't good at developing much voltage. So, back then, the standard method was to do pulse compression, much as we do today for high powered pulsed-power projects. They would discharge the HPG into an inductive transmission line which had the railgun connected to the other end. Then they'd use a series of short circuiting or opening switches to compress the magnetic field of the line, causing the voltage to soar during the pulse.

The devices used had to be very fast, and had to handle millions of amperes. They included special fuses, exotic circuit breakers, and mechanical switches driven by high explosives.

During preparations for one shot, a heavy equipment cart with metal wheels was rolled over a hydraulic hose which provided high pressure oil to the bearings of the HPG. No apparent damage was caused, and nobody thought anything of it until later. That was because there was no pressure on the hose. But the hose HAD been damaged, and a tiny leak had been created.

All was made ready in the run room, personnel left, and the heavy steel door between the gun room and the control room was closed and locked.

Various power supplies were charged, trigger generators armed, and then the HPG was spun up to speed, which included pressurizing the bearings. The high pressure oil leaked out of the supply hose in a very fine mist. Unknown to the operators, and not visible on the remote cameras, oil vapor was now filling the gun room.

When the shot was fired, the various plasma clouds and sparks promptly ignited the vapor in the room, causing a significant explosion which blew the heavy steel blast door from its hinges, injuring the technician seated nearest it.

The point being: ask yourself a LOT of questions that begin, "what would happen if..." before you ever apply power to anything.

Now I've just completed a 1,000 PSI gas injector (to pre-accelerate the armature before it reaches the rails) which will

One of the "gotchas" I discovered recently was in the breachblock assembly, which gives me a way of connecting things to the back end of the gun, such as a high pressure injector system. The breach closure needs to be easily removed so I can put the armatures in.

It also needs to contain 1,000 PSI of gas pressure. Further, the plug which closes it is 2" in diameter, so now we're seeing 3,140 pounds of force on the breach-plug. Don't worry, I designed it to take that and then some. It's massive. But that was only the forces within the two-part system of breach-plate and breach plug-plate.

Then I looked at the interface -- and more significantly, the gas seal -- between the breach plate and the rest of the gun. What if my original seal design -- with two seals, one being a "backup" -- would fail at the inner seal? Oh, now I have about 20,000 pounds-force trying to tear that plate off of the back of the gun. I'm not sure I want to load the bolts that much. So I eliminated the outer seal! Now if it fails, nothing goes anywhere, it just leaks out undramatically. And I don't expect that inner seal to fail, but no one ever expects the Spanish Inquisition!

The big pulser-without-a-name had been designed with interlocks originally, but they were all defeated, broken, or missing when I got it. I will of course be repairing / replacing them. I need to sit down at some point and work through the entire sequence / time-line of a shot preparation, firing, and safing. Eventually, I'll probably even have a check list and some rotating beacons, because those things not only improve safety, they're genuinely fun for any red-blooded geek.

"Uplink amplifier output switched to dummy load?"

"Uplink amplified output switched to dummy load!"

(that's an in-in-in-joke from my USAF days)

Ya know, all this work would be a lot easier if I had a team and a budget. Sure wish I'd gone to school.

The Beast With No Name

2011-01-31T18:07:00.002-07:00

Around 1995 or so, I received a call from a good friend, asking me if I might be interested in a very large and very heavy high voltage capacitor bank -- capacitors specifically intended for pulse discharge duty -- with its own charging and switching circuitry; in short a complete, turn-key pulsed discharge machine. I immediately said, "yes" and then wondered where I was going to put it since I was living in an apartment at the time.

The thing was immense, weighing just a bit shy of 2,000 pounds. It consisted of two 28" wide rack panel cabinets mounted on a steel skid with four groaning casters. The shorter of the two contained the high voltage power supply, ancillary power supplies for things like control voltages, and a bit of relay control "logic" to interact with operator input and interlocks while enforcing safety rules.

The larger of the two cabinets contained the pulse capacitors (six), a parallel plate current collector stack (nice low inductance design), six ignitrons (one on each cap) in coaxial housings, as well as a bunch of other hulking heavy duty high voltage equipment to trigger the ignitrons, including a nice glass hydrogen thyratron.

This thing was designed in the 1960s and built in the 1970s. It was moderately cutting edge stuff for its time. The company wanted to investigate industrial applications for pulsed power including well frac'ing, and magnetic metal forming. Fortunately, all of the designer's personal notes and sketches, as well as notes from later techs who worked on the machine to rehabilitate it in the early 1990s, were saved and given to me when I got the thing. That has been incredibly valuable.

What happened is that the company's focus had moved on and narrowed - they had made interesting strides in metrology. They were being purchased by another company already famous in the business. They needed to clean house, and SOMEHOW, that two thousand pound white elephant had to go. If I could move it, I could have it. It sat for a bit while I figured out where in the hell I was going to put it, and how. The owner of the company was awfully patient with me.

Well, I had a sort of professor emeritus status with the entity which at that time was right in the middle of morphing from The Dream Park Corporation (bankrupt) into The Diabolic Company. Said organization had just moved into an IMMENSE building which had previously held a K-Mart. I asked, and I was basically told, "any place that won't interfere with the haunted house operation, knock yourself out".

The short version is: I stashed it there.

I got some friends to help me move it.

And we dropped it. But that's a story not terribly relevant here and now.

The astonishing thing was, we didn't do it any harm - er, probably. There was a little sheet metal damage, but it had slid down and landed with a jolt, we didn't really drop it off the dock. The only thing affected was the ignitrons. You see, ignitrons are finicky beasts. They have their advantages (they can switch HUGE charge transfers) but one of their idiosyncrasies is that once they have been used in crowbar or capacitor discharge duty, it is crucial that they not be moved or jarred. When ignitrons ares used in those high-wear modes, the structure which supports the ignitor electrode over the pool of mercury becomes very fragile. And shifting or sloshing of that pool of mercury is likely to damage the ignitor, rending the tube inoperative. True story.

The way we were rolling that thing around the lab and bumping it over doorways to get to the dock, all six of those tubes were toast before we ever got around to dropping it off the dock. Of course, I didn't find that out until later. About the same time in fact that I found out replacement ignitrons cost $1,500. Each.

Not having a spare nine grand burning a hole in my pocket, I decided to use another means of switching. I got rid of the ignitrons to a friend who was collecting mercury anyway. Don't ask. One less hazardous materials disposal pickup (we have an awesome trash department) for me to deal with.

Meanwhile, I needed a way to move this beat around without a Class IV forklift. While I still had the shop and lab space at DiaboliCo's haunted house building, I took to rebuilding it. I removed the power supply and its cabinet, disassembled the ignitron housings, and plate stack, removed the cabinet, and eventually unstacked the six pulse caps. They weigh about 150 pounds each. I'm guessing. It I ever weighed them, I've lost it, though I ought to have the shipping weight in my folder. Anyway. Then I split the big steel skid the whole thing had been mounted on, with a torch. I added more steel, and two new casters, and I had two skids. Then everything was reassembled onto those. Mind you, with a few rare exceptions when I was in a hurry and people were around to help, I was doing most of this alone, all Leedskalnin-style, with levers, blocks, ropes, pulleys and the convenient architectures of the building.

After that, the machine followed me around, all 2,000 pounds of it, through two moves, and has been sitting in my workshop alongside my collection of other Large Heavy Objects, waiting for me to figure out which way to go with it. In the mean time, I've been learning about high speed and high current switching devices. Since I acquire the thing, I've learned a GREAT DEAL.

I've been given one type of switch - which I've yet to fire, tho I am inching closer - and I've designed and built another type - which I've also yet to fire. I'm inching closer to that event too.

Stupid mill.

While the mill has been down I've been working on the parts of the long-stalled railgun project. The parts that don't require the mill, that is. That would including the pulser, or as the previous owner called it, "The Banger" (as far as I am concerned, it still doesn't have the great name it should, yet). From what he told me of their insane escapades, I bet every one of them needed hearing aids. Exploding wires are louder than gun shots, kids.

The first thing I did was excavate the machine itself from a mound of boxes. The new effort provoked a quick round of shop-cleaning this weekend. The second thing I did was remove all of the ancillary parts that were necessary for triggering the ignitrons, about 50 pounds of transformer, capacitors, rectifier, and blah-blah oh yes and a nice thyratron.

I ought to describe how the thing worked before I got my grubby mitts on it and tore half its guts out. It'll be illustrative of how things often go in the pulsed power world. Say you've got a whacking huge amount of energy you want to apply to load with a whacking huge switch. Trouble is, whacking huge switches by their very nature, tend to require a whacking huge something to activate them. Might be a huge pulse of electricity. Or perhaps some compressed air. Explosives are often used. Not kidding. So you come up with this pretty substantial trigger pulse to trigger your whacking huge switch. That pretty substantial trigger energy has to be switched by a substantial switch. The substantial switch needs at least an adequate switch, and so on, finally ending at the big red, jolly, candy-like button on your control panel.

So, this is how this particular machine was originally designed to work:

1. The operator sets the voltage at which they want the machine to fire on a dial (a Simpson relay meter).

2. The operator places the work in the fixture, which is already connected to the machine's output bus.

3. The operator presses the 'Start' button, and I would presume, gets the hell away from there.

4. The machine charges, slowly I assume, and when the meter needle reaches the aforementioned setpoint, the machine discharges in the following sequence:

5. because the Simpson meter contacts aren't good for any current at all, when they make contact, they are used to control another, ordinary industrial control relay.

6. that relay's contacts apply a voltage to a control grid in a pentode

7. the pentode turns on and dumps the charge of a small high voltage capacitor through the primary of a pulse transformer

8. the secondary of the pulse transformer delivers a small high voltage pulse to the

9. hydrogen thyratron located in the capacitor cabinet, which turns on and connects one side of six beefy high voltage capacitors to ground, dumping their charge through the ignitor electrodes of six ignitrons. Inside each ignitron, 100 joules of energy vaporizes a small amount of the mercury in the pool into the vacuum of the tube. The mercury vapor connects the anode and the cathode, turning on the ignitron tube until there is no more current to keep the vapor hot. Functionally, it behaves a lot like an SCR.

Seems a bit complicated, don't it? But that's the way it is. I'm going to try to do this with more sophisticated (and more importantly, obtainable) components, but there is still (and will always be) a long series of steps between you and the big drama. That's another good point worth raising - isolation. You don't want any kind of freak fault to allow the main store energy to get out into places it doesn't belong, such as remote control pendants. That would be bad.

Oh yeah, it came with a crude remote control pendant. I'm already working on a more sexy remote control, uh, "pendant". Well, panel anyway. Okay really, it's more of a box. With a panel that mounts inside the box. Like a miniature rack, sort of. Only with storage inside the lid. I'll try not to let it get any bigger. Also, it's long term project, and won't be done before the pulser-with-no-name is in operation.

Speaking of names, I could use one. It's an idiosyncrasy of mine - I like to name things. If you've got any clever ideas, leave me a comment or something.

Action This Day

2011-01-28T16:45:00.001-07:00

I ordered the remaining parts for the railgun pre-accelerator today. There are a few supporting parts for it which I still don't have. These include:

• the CO2 tank itself - which I'll "buy" from my local industrial gas supplier - about $100 I'm guessing

• a regulator capable of taking CO2 tank pressure and delivering anything from 100 PSI to 1,000 PSI. Of these three items, this will be the most challenging to find at a price I can afford.

• a bit of metal tubing and a few inexpensive (small size) tubing fittings

I think that's it. And what the hey, if I can't get the railgun to work, it'll be a short step to making the most dangerous compressed gas gun in amateur hands. (let's not and claim we did)

No movement on the mill. I'm trying to be patient, but already I'm wondering whether I ought to replace the motor - a heinous thought. But the mill going down is why I started working on the railgun program again. I need something to keep me off the streets and out of the doldrums.

The Mad Scientist Light Switch is on hold because of the mill. I could have it had it done by the end of January if not for the mill problem, I was that close to being done.

The test rig for the fast distortion-triggered, pseudospark mode, spark gap switch (say that three times fast) is on hold because of the mill and some materials not yet purchased. I still haven't figured out a good way to connect the switch to one of the caps in the pulser without pulling the cap out.

The avalanche transistor trigger for the VIG is also on hold due to spending money on other things. Can't do it all at once. Anyone got a cheap source for the Zetex ZTX415 avalanche transistor? $25 seems a bit steep for a single tiny rock. Anyone? Anyone? Bueller? Oh right, I keep forgetting that nobody reads this bloody thing.

Hey! I think I hear an echo.

Lessee, what other programs are "open"? The Tesla coil of course. Haven't touched it because it is at a stage where it needs far too much attention and money. I can't even look at it again until some of these other shorter-term projects are completed. THEN I can (hopefully) knock that thing out and then (hopefully) sell it. It will, at least, be unique in at least one way.

The fast mini-Marx is on hold for lack of money to spend on it. Have to finish some of these other projects first. Cash flow is tight and I'm already spending more money on hobby project than my wife would prefer.

Not much else to talk about today. Until next time...

onward, mischief soldiers...

2011-01-26T18:14:00.001-07:00

The mill is still down. The Variable Frequency Drive (VFD) -- which synthesized clean(ish) 3-phase power from the 240V 1-phase power available in my shop -- has died. It seems likely I brought about its death. I bricked the motherboard with an incautious press of a button in Eurotherm's software, enabling me to write user data over the firmware memory space. I'd like to ask my readers a question:

WHO THE FUCK DESIGNS SOMETHING THAT ALLOWS THE USER, THROUGH SOFTWARE, TO ACCIDENTALLY AND IRREVOCABLY DESTROY THE DEVICE?

Seriously, if you helped design the Eurotherm 605 series of inverter drives, I'd like to hear from you. Just send me your street address, where you work now, your daily habits, and a photograph, and one of my... associates will be in touch.

Where was I?

Goddamned mill is down. See? Now I'm swearing. I'm swearing mad. That's why we're called "mad scientists", at least part of the time. I've got a search going on eBay and am checking a few drives a day. Something will turn up. I'll try to be patient with this, because I prefer to use an inverter drive and not a DIY (nor commercial) "phase converter". But I'll go that rout if I lose my patience. We'll see.

Since I scored the firing valve for the railgun pre-accelerator, I have been busily figuring out the plumbing necessary to connect it to the breach block of the gun, the plenum chamber, and the gas supply. All of this must safely work at 1,000 PSI. I did a lot of homework, read a lot of ANSI and ASME standards, and the corresponding pipe schedule tables, and determined that Sch. 80 seamless steel (not iron) pipe will give me a safety factor of two in working pressures(burst limits are far higher). In reality, I have a safety factor to failure of at least 4X for some components, and 6X - 8X on others.

Yasee, I frequently work with high pressure gas flowing through tiny metal tubes. That doesn't worry me much, because the amount of energy stored by compressed gas depends on both the actual mass and the pressure. The bigger the enclosed volume, the more mass. It goes up fast, and so does the potential danger. The stored energy in compressed gas is incredible.

It may be difficult to get the pre-accelerator valve to fire reliably at low pressures but to be honest, that may not be necessary. I have a hard time imagining any live (electrical) shot that won't have huge friction between the rails. A lot of contact force is required between the armature and the rails to have any hope of getting that armature the length of the gun without being vaporized. The maximum contact force will be determined by what can be pushed down the barrel by 1,000 PSI or less. That was an arbitrary number, but about the biggest I'm willing to try to work with using hardware of my own construction. The breach block assembly is going to be, um, "interesting" to design. If it doesn't leak on the first version, I'll have a heart attack.

For a while there, it looks like I'd lost my CAD files for the railgun project, but thankfully I was mistaken. Found the backup too.
So I'll be looking at that breach block assembly soon.

I also need to revisit how the bus connections are made to the gun rails, and I need to think about how to keep this beast quiet - a muzzle blast suppressor at worst or a complete containment tank for the entire range at worst. Ugh. I'd really rather not. But I've no desire to have the local constabulary knocking on my door (okay, okay, this is the new millennium so busting down my door). Even though I'm fairly certain I'm not breaking any laws (though it's so hard to be sure these days) it is better not to frighten the horses.

incrementa

2011-01-10T09:18:00.000-07:00

I finally got my hands on the solenoid-actuated valve I need for the pre-injector of my second railgun. Now it appears as though the manufacturer has discontinued that model starting this year. I hope my valve never breaks down or wears out. But at least that's one more hard-to-find/afford part that I can cross off my list.

The Mad Scientist Light Switch is finally nearing completion, although YET ANOTHER delay was created when my mill suddenly refused to operate. I suspect the motor drive, which provides 3-ph. power to the mill, has lost its mind after being unplugged for a week. One would think the manufacturer would use EEPROMs to store all the parameters, but apparently not. HOPEFULLY, reprogramming the drive will get me back up and running. I hope the drive isn't toast. That would be depressing, and would take a long time to fix. If all goes as planned (and it never does) I should be done with that thing by the end of January.

Meanwhile, it looks like my first test shot on my switch will be about 30mS pulse time, 100 amps (or a bit less), 3kJ, 0.6C total charge transfer. I'm eager to see the rise time.

I have realized that most of the ideas I had for connecting to the switch were bad, and I've decided to drill an additional six holes in each end cap to allow for bolted connections.

Once those are done (and the switch reassembled) my plan is to use copper pipe and plastic pipe to house the resistor, six half-inch wide copper straps to connect one end of the resistor to the switch, and six more copper straps to connect the copper pipe to the ground / return. Crap: I just now realized that the way the pulse caps are connected to the parallel plate current collectors in the pulser makes it very difficult - maybe even impossible - to connect to the ground return plate. I can either short the two plates at all four edges somehow (which would introduce extra inductance); remove the "hot" collector and insulating board from the stack (a huge pain in the ass and undesirable)... or find another capacitor with which to test my switch. Grumble.

I'm really not interested in testing my switch with all six capacitors connected from the beginning.

Dr. Carl E. Baum dies

2011-01-02T11:14:00.000-07:00

Dr. Carl E. Baum, founder of the Summa Foundation and one of the shining lights of pulsed power research, died December 2, 2010 at the age of 71. I've mentioned his name before. Among many other achievements, he is notable for being the designer of the famed Trestle (ATLAS-1) EMP simulator at Kirtland AFB, New Mexico.

Carl E. Baum was born in Binghamton, New York, on February 6, 1940. He received his B.S. (with honors), M.S., and Ph.D. degrees in electrical engineering from the California Institute of Technology, Pasadena, in 1962, 1963, and 1969.

He received the degree of Doktoringenieurs Ehren halber (Dr.-Ing. E.h.) (Doctor of Enginering honoris causa) from the Otto-van-Guericke-University Magdeburg, Germany, in 2004.

He was stationed at the Air Force Research Laboratory, Directed Energy Directorate (formerly Phillips Laboratory, formerly Air Force Weapons Laboratory), Kirtland AFB, Albuquerque, NM, from 1963 to 1967 and from 1968 to 1971.

From 1971-2005 he served as a civil servant with the of Senior Scientist at the Air Force Research Laboratory. Since 2005 he has been a distinguished research professor at the University of New Mexico, Department of Electrical and Computer Engineering.

You would be hard-pressed to find a paper on electromagnetic pulse or ultrawideband high power signal generation without his name on it somewhere.

You may read his full obituary here.

current, power, energy

2010-12-29T17:47:00.001-07:00

Since my last post, I have repaired the part I thought I'd ruined, and have made further progress on the Mad Scientist Light Switch. Nothing worth showing you just yet however.

I've also inched forward a bit in work toward commissioning tests of my triggered spark gap switch. One of the things I'll need to test and characterize it is a series of dummy loads of various values, all very low inductance, and capable of absorbing around 3kJ in a short pulse.

The resistors I can either make, or in some cases like that shown here, I may get lucky and find something commercial that will serve.

To use any such resistors, I've got to have a way to connect them to the switch, and the switch to the capacitor and so forth. The connections will have to pass a series of increasingly high current test shots, stepping up by orders of magnitude. Here's my first pass at a basic high current, low-inductance connector for the resistor:

The scale is 12 inches / ≈ 30.5 cm long.

I ought to find myself some thicker copper pipe. This stuff is 'schedule L' pipe, with a nominal bore of 1.025". Note how the pipe is bent inward by the clamps to meet the 1.00" OD of the resistor terminals.

There is a thicker series of copper pipe - 'schedule K' - which has a nominal bore of .995". It should mate more smoothly than the L, and if necessary, I can turn out the ID a few thousandths and still have a thicker wall.

Four more slits need to be made in the outer ends of the copper pipe, so as to receive brass inserts which will connect to the switch at one end, and the outer return casing at the other end, subject to my whims and availability of parts.

The switch housing requires 5" ID copper DWV pipe which is available but annoyingly expensive. I need a short piece for this rig, and a long piece (5 or 6 feet) for the big Marx generator.

absentia et noncommunicatimatus or something like that

2010-12-09T06:45:00.000-07:00

In November, I took another long hiatus from any work on mad science due to a death in the family, struggles with medical issues, and commitments to my immediate social circle.

I've resumed work within the last few weeks, but I also just wrecked a part which has me rather depressed and limp-dicked over the whole project.

Work will resume when I feel like it, and news will be posted here when it happens.

dense plasma focus

2010-11-22T11:18:00.002-07:00

A lot of people are researching various technologies which might enable practical fusion-based power generation. Aside from the biggest (in terms of dollars being spent) technology, Inertial Confinement, and the second biggest (Magnetic Confinement), there are a number of other, less popular, but still promising technologies being investigated.

One of these is Dense Plasma Focus. DPF has been around since the 60s and has traditionally been a short-pulse source for neutrons and x-rays. While DPF has long been known to produce fusion for very, very brief periods, few researchers felt it offered much promise for power generation... until recently. However, "not many" is more than "zero", and a few very smart people are working hard on the problems presented.

Over the weekend, I stumbled across a group who have assembled quite a respectable device. They mentioned in one of their posts that they are having trouble getting all of their switches to fire within the acceptable window of 10nS.

I looked at their setup. They are using conventional trigatrons. Conventional trigatrons have, at best, a jitter number around 10nS unless additional finicky methods (such as radiation sources or pre-ionization) are used to reduce the turn-on delay.

Typically, if you care about jitter a great deal, you use either distortion triggered switches, or laser-triggered switches. I'm more than a little puzzled that this team did not do so from the beginning. If they really do need to achieve <10nS jitter between all of their switches, it seems to me they're screwed unless they change the switch type and triggering scheme. Of course, that would significantly increase trigger generator complexity, but if you want faster switching and less jitter, your choices are rather limited.

I look forward to getting on their forums to discuss this, perhaps there is something I'm missing.

further increments

2010-11-17T08:58:00.001-07:00

Progress on my projects is beginning to look a bit like Zeno's Paradox Of Achilles & The Tortoise, in part because my hands and arms and shoulders hurt too much to do any work.

One thing I can do apparently, is continue to build parts for other projects virtually, in SolidWorks. Some of my projects could not be built at all -- at least not by me -- without good drawings, accurate dimensions, close tolerances, and a sophisticated 3D model which warns me where I'm about to screw up before I take any cutters to any materials.

I'm having a bit of trouble with the rail connections at the breach of railgun #2. Various amateurs have tried various things, but it seems an awful lot of amateur railgunners forget that it is inductance in the gun that does all of the work, and all the other stray inductance in your connections serves only to slow and limit the current. In other words, you want low-inductance, low-force connections, cables, and buswork between your power supply and the gun, and you want high-inductance and high force inside the gun ONLY.

I see other railgunners with gorgeously built guns and connections made to a great big expensive capacitor bank twenty feet away through great big SO cables lying on the ground... and I laugh. So much money wasted.

The big guys use RG-220 or YK-198 high current, high voltage coaxial cables - many-many of them, to connect power supply to gun breach. I'm searching for a windfall of same on eBay, but I don't expect to find it. That stuff is expensive and not much of it is made. Its tiny market probably holds on to every piece that is still serviceable until it isn't. So I'll be using bars of copper bolted together, in all likelihood, and that means locating the gun close to the power supply but come on - this is only an 18kJ device for now.

It doesn't help that, like most small-scale, amateur designs, I am using a gas gun pre-accelerator to overcome the standing start problem. Sophisticated tailoring of power supply to armature shape and armature-rail pre-load can overcome the issue, but that's for the professionals. I don't think that's within the reach of most amateurs, and certainly not I. But that injector connects to the breach at the back, meaning the rail connections have to come out somewhere else, and away from any nearby bits of metal.

Hopefully, my virtual Tinker Toys will let me find something that works within my means.

~~~

In the mean time, what I hope are the final parts for the Mad Scientists Light Switch are en route.

more incremental progress

2010-11-11T11:35:00.001-07:00

Nothing exciting to announce, but:

MSLS:
I've worked out most of the details of what the top-of-the-cabinet lamp features will look like. I have most of the parts, and will have to fabricate a few more. Am trying to find a cheap source of lamp socket adapters. The adapters are cheap - a few bucks, but most places want $5 to $8 shipping, EACH.

Anyway, the lamp/fake-vacuum-tube/mystery devices should look relatively bad-ass, some day. I'm unlikely to finish those pretty things until after I get the unit hooked up. The reason it isn't connected to the lights yet is that I have to run a new circuit to the area from the garage's sub-panel, and THAT involves ripping out the ceiling in the other part of the garage. It's something I'd planned to do anyway, I just wasn't ready now. Maybe this weekend, we'll see.

VACUUM SYSTEM:

Work is moving ahead, but slowly. I keep finding old damage (mostly in the wiring) that needs repair, or having minor but annoying things break on me, requiring repair. I'm fixing those now, instead of working on the actual vacuum plumbing.

I identified another probable leak site, and am working to seal it - the adapter plate I made to blank off the top of the stack with two gauges. The gauge mounting assembly is sealed to the adapter plate with a flat gasket, not an o-ring. The gasket stuck out from under the gauge assembly over the adapter plate, all the way around. I trimmed that back and then added leak sealant over the edge of the gasket, joining adapter to gauge assembly all the way around. I also put sealant on two solder joints in the gauge assembly. I'm reasonably confident that it's leak-tight now.

I'm gonna need a new motor. The one that's on there now is not the right motor type despite having the correct horsepower, and as a result, it has inadequate starting torque for the monster mechanical pump it's trying to crank. I can't afford to buy a new motor, so I'll keep looking for one used, at garage sales, flea markets, eBay, and Craig's List.

in which our hero accomplishes very small but significant progress in a project designed to create very small but significant pressures

2010-10-29T06:30:00.011-06:00

In theory, I've been working on the vacuum system. I've been interrupted in various ways, in fact, I had to reassemble the system partially just to make it safe from spilled beers and the like while the Denver Mad Scientists Club invaded my sanctum, well, communio to tell the truth. But I digress again.

Various stuff be happening, and not much of it vacuum work.

BUT! I has a plan. And I am making glacial, but non-zero progress. I have the "stack" partially disassembled, and I need to finish that. I need to take apart one of the valves to get its o-ring sizes (seats and joints) so I can order all new O-rings.

By the by, I don't know how common the term "stack" is in the general vacuum community...

I lie. I just checked the VacuumX Yahoo Group and it's a commonly-used term.

Okay good, cuz I was just about to define it for those reading who might not be familiar with it.

To wit: the "stack" on a vacuum system is most often a vertically arranged sequence -- I'll go from top to bottom -- consisting of:

1. chamber (or a baseplate for a bell jar)

2. big-ass gate or butterfly isolation valve, labeled "Hi-Vac Valve" in the diagram

3. a cold trap and accompanying "optically opaque" baffle, usually only needed if the high vacuum pump is a diffusion pump

4. and at the bottom, the high vacuum pump such as a diffusion pump, which in turn is "backed" by the mechanical pump.

Somewhat amusingly, the mechanical pump is the part everybody notices, because it's big and makes lots of noise. But it is a simple thing, unfinicky, and the least sophisticated element of the system. Normally, it will give the operator no trouble if it treated properly. The diffusion pump and cold trap are where all the action is.

Now if you're wealthy or you're a university or a big company or a government lab, you don't use a diffusion pump, you use a turbo or a cryo or some other high vacuum pump.

(mumble Vonnegut mumble)
HERE IS A PICTURE OF MY TURBOPUMP^*:

And in those cases you probably don't need the cold trap either. Well maybe. But that's outside the scope of this entire blog, I think. I'm thinking I don't want to start trying to teach too much in the way of basic lab techniques like basic high vacuum work, because that's a basic laboratory skill with which every budding (or accomplished) mad scientist should already be familiar. If not, well, there's plenty of books and web resources for high vacuum work, such as the aforementioned Yahoo group. Also, check out The Bell Jar.

So, in my system, I have a 3" oil (DC-704 silicone) diffusion pump (Veeco) backed by an (oversized) two-stage rotary vane pump (Cenco-HyVac). The top of the diffusion pump is bolted to an LN2 cold trap. The cold trap keeps DP vapor out of the high vacuum side of the system, and (an occasionally overlooked but important aspect) it keeps water vapor from your process chamber out of the DP.

Aside: this is why I also have a handy and monstrous dewar I can go fetch large quantities of LN2 in, saving (coronary-inducing) delivery charges, and making it cheaper because of quantity as well. You pay for a base boil-off fee, and the more you buy, the less you pay per liter.

Veeco's cold trap is a rather clever coaxial design with optically dense baffles at one end, a cold finger, and a coaxial (toroidal) gas path. I haven't seen anything like it implemented anywhere else. I will say this: the system appears to be almost overdesigned. It is certainly conservatively designed and if I can ever get it rebuilt, it ought to be a reliable and high performer.
Moving on...

The top of the cold trap, at a point offset horizontally from where the DP is connected below, is connected to a butterfly valve, about 5" bore and only 1" thick body - rather nice. It is a leak-tight high vacuum valve which will hold off an atmosphere against a 5" throat. Now isn't THAT nice? Think: you can pump down the entire vacuum system to some ridiculously low pressure. Open your chamber, insert your process stuff or do whatever it is you gotta do in there, close your chamber, now with the rest of the vacuum system at high vacuum and isolated, you open a roughing valve connecting your process chamber to the roughing pump, but not the the high vacuum parts of the system.

When the chamber pressure is nice and low, you open the butterfly valve and your chamber pressure falls through the floor. Otherwise, you'd be sucking the entire process chamber out through your cold trap and your high vacuum pump. And that, friends and neighbors, is bad design.

So that's what a fully valved system with a chamber isolation valve does for you, if you're lucky enough to stumble into one for a price you can pay.

That is, if you've been good.

If you've been bad, you're still trying to eliminate leaks and repair damage caused by years of abuse and neglect by ~~ignorant savages~~ college students working under the poor-to-nonexistent supervision of college professors or TA's or grads or whoever is - or should have been - in charge of such things.

Ahem. But I'm not bitter. I paid $20 US for the thing although it didn't have a roughing pump and the DP was a mess.

Now where was I? Oh yeah: O-rings. I'm gonna find all the sizes of o-rings used anywhere in the system, and I'm gonna replace every damned one of 'em. And when I replace them, not only am I going to use good Viton O-rings as any fool would, I am going to process them through certain laboratory resources I have at my disposal which will render them extra-good for high vacuum work. I have read in books that it is possible to build a full-featured system with many O-ring sealed joints that will consistently run at 1x10^-8 T. Normally, we think in terms of routine pressures of 10^-6 Torr, maybe 10^-7 on a good day with a tailwind. Well, I aim to prove those books right, and see routine pressure in my chamber below 9x10^-8.

Oh guess what I found in a couple of white papers? One of them was a NASA article by people working with big chambers. Just go ahead and put a thin coating of good high vacuum grease ALL OVER your baked-out O-rings before installation. The grease blocks gas molecules better than the Viton, and it keeps the Viton from loading up on water vapor and other stuff after you've baked it out under high vacuum. Heh. Yeah, you need high vacuum in order to build a high vacuum system right. Don't use vacuum grease if you expect to go much below 5x10^-9. But if you're trying for those pressures, you are already familiar with UHV and you don't need my advice. Or you're very foolish.

So once I have all the o-rings process and bagged and all the parts I think I need to wrap this thing up, I'm hoping I can finish disassembling, re-cleaning, and rebuilding the whole thing in a single weekend, which I probably can if I enlist the help of a good friend who has voiced enthusiasm for said assistance.

I'm just taking my own sweet time leading up to that point.

There, I posted dammit.

^*that is not my turbopump station, it belongs to my employer.
Without the stuff I've stuck on it, it costs about $7,000 -- a bit out of my reach.

vacuum system progress

2010-09-15T16:57:00.001-06:00

I've been focused on the vacuum system lately. The MSLS is being worked, slowly, in the background, but I've hit a mental snag with it, and I need to walk away from it for a while. It's just one of my mental idiosyncrasies, not a problem with the design that I need help to solve.

So, the vacuum system. It leaks. I have been methodically working my way through every single connection in the system, working to eliminate leaks.

Simply by cleaning up o-ring joints, I got the ultimate pressure from 80 mT down to 20.
Still not low enough to even be able to turn on the diffusion pump, but it is progress.

But! Speaking of the diffusion pump, that's where I am in the process. Over the weekend, I removed the DP from the cold trap and inspected it. Here's what it looks like, removed but still assembled:

I was relieved to find that the fluid looks very clean, and the pump as well, for the most part. It has a few bits of black crud left over from the previous owners- stuff I was not able to remove with a glass bead blaster, but which was apparently removed from the valleys of the boiler by boiling action of the fluid. Scrubbing bubbles, as it were.

So that was all well and good, but for the o-ring which seals the DP to the cold trap.

It was flat.

It is not supposed to be flat, it is supposed to be round.

My theory is that some insufficiently-educated or -supervised student (the previous owner of this pumping station was the University of Colorado) allowed the DP to get too hot, possibly as a result of a burned out cooling fan. I already know they were running Octoil or some other low-grade DP fluid in it, because they burned it up, leaving rock-hard carbon-loaded sludge behind - incredibly difficult to remove. That does not happen with the silicone fluid that this pump is SUPPOSED to have in it. (and which it DOES have in it, ever since I rebuilt it) It's also possible that some bozo used a non-Viton O-ring. That would be stupid, since DPs are designed to get hot. In theory, the top end shouldn't get hot enough to hurt a Viton O-ring, but in theory, the pump shouldn't have been mistreated either.

The good news is that high quality o-rings of nearly any reasonable size are dirt cheap and available from anybody, including my favorite, McMaster-Carr. The size I need came 10 to a bag and cost $10 total. They should arrive by the end of this week.

I'm also concerned that the boiler heater is too hot. It is supposed to be a 375 watt heater, but the one that is on it now is dissipating 428 watts. I'm not sure yet how important this is. I'm reading and I've inquired with a vacuum engineer at Varian, but I haven't heard back from him yet.

Once the DP has been tuned up and dialed-in, the next step will be to inspect the butterfly valve which separates the chamber from the cold trap.

It occurs to me that I could be including more pictures to illustrate what the hell I am talking about, and that it might even make these posts more interesting. Okay, I'll see what I can do. In fact, if you're interested in this post's topic, check back in a few days, I will try to snap some photos and insert them into this post at appropriate points.

micro-update

2010-09-03T11:37:00.000-06:00

I've been ill for a bit over a week, but am getting better. I have only small, incremental progress to report for any of the current projects.

I'll cover those bits of progress in a moment, but first I want to talk about a fellow named Jim Sanborn. Jim Sanborn is an American artist known for his complex, often technical sculptures, installation pieces, and outdoor exhibits.

Perhaps his best-known work is Kryptos, a large copper sculpture located on the grounds of the Central Intelligence Agency (CIA) in Langley, Virginia.

However, what Jim Sanborn does that is of interest to mad scientists is his work revolving around nuclear science, including the birth of the nuclear weapon. Recently I was privileged to view his installation - titled "Terrestrial Physics" - at the Museum of Contemporary Art Denver for their show "Energy Effects" which includes work from over a dozen artists. This is where I would link to all the photographs I took of the exhibit. They exist, but as I said earlier, I've been ill, so I haven't vetted, edited, and uploaded them anywhere yet. I'll post something when I do. And if you're not too far from Denver, you owe it to yourself to see Energy Effects before it leaves town -- especially before Jim Sanborn's portion is removed, some time later this month. Call the museum for more information.

That exhibit lit a fire under me, and a long-suppressed desire to build a small particle accelerator has blossomed again. That exhibit also lit a fire under a good friend of mine who also has a penchant for mad science. As a result, he's preparing to build a small Van de Graaff-driven electron accelerator that will fit on a table-top. I'm not ready to go that far right now, but I am interested in building a particle accelerator some day, and that requires high vacuum equipment, and THAT brings me -- finally! -- to the progress report, since my high vacuum system is one of the things I've been working on for the last few weeks, albeit slowly. So without further ado...

1. The MSLS:
I now have all of the parts, which is to say, all of the correct parts, to finish the MSLS. The last one arrived yesterday. I'm going to push forward until the thing is in working order, and I may decide to completely ignore the special effects components for the time being. I need to get this thing out of my hair and more importantly, off of my bench, at least for a few months. Unfortunately, I put some big ugly blemishes on the once-beautiful plastic and brass safety shield for the Jacob's Ladder while I was gluing the top vent baffle in place. I've been working on polishing them out, but I am not very good at this sort of thing, and mostly I seem to be making it worse. I now think I have a handle on what works and what doesn't, so I'll revisit that... eventually. If I feel sufficiently better, it's possible I could have the thing finished except for those repairs, by the end of the Labor Day Weekend.

2. the high vacuum system:
It's not clear whether it leaks any longer. It might not, but I STILL do not trust the gauge electronics I have on hand. I have two complete sets installed now, but the controller I am using to check the electronics built into the system is older than I am and runs on vacuum tubes. I like vacuum tubes, they're hard to kill, although systems built with them are less efficient, take up more space, and generate more heat, than their solid state counterparts. I don't have a great deal of trust for fifty year old capacitors, but that controller does seem to be operating, and best of all, it agrees with the electronics in the unit.

The good friend mentioned above loaned me yet another vacuum gauge - a T/C gauge only, no ion tube - to give me yet another cross-check on the other T/C gauges, and we KNOW that the third gauge is in good working order, because... um... it's been stored in a dirty shed for ten years. Okay, we don't know whether that one can be trusted either.

But if all three gauges agree, we can be reasonably certain they are all correct. Why? Because the odds are very much against all three of them being off from a true reading by the exact same amount.

Unfortunately, the sensor cable for that unit died of old age while it was lying around (or maybe even before that) so I'm going to replace it - also this weekend.

As soon as the above is done, I'll go grab myself some dry ice and a handle-jug of Everclear (for my cold trap) and do another test run to see if I have small leaks. Since I brought the thing home, I have never convinced it to let me turn on the ion gauge tube. Hopefully this time I will be able to turn on at least one of the two ion gauge systems, so that if there is a leak, I can try sniffing for it with helium.

3. I've listed some of my other projects in previous posts. I'm not feeling up to continuing the list with detailed descriptions and discussions for each item, but I'd like to show just how naive and ambitious I am, so I have listed below all of the science-y projects I've done in the past, projects I'm working on now, and projects I'd like to finish before I'm ready for the crematorium:

COMPLETED
* Van de Graaff generator #1 - didn't work, but I was only 12 years old
* electronic thermometer
* triode-driven Tesla coil
* high sensitivity, high stability electronic thermometer for biofeedback study
* 12kW Tesla coil #1
* 12kW Tesla coil #2
* 12kW Tesla coil #3 - used as the basis for DiaboliCo's first Tesla coil product
* 1.2kW Tesla coil #1 - DiaboliCo's second Tesla coil product
* several mediocre pneumatic guns, and one or two quite good ones
* (I bet I am forgetting a few things here)

IN PROGRESS
* MSLS - A Mad Scientist's Light Switch
* high vacuum system
* TLTC - The Last Tesla Coil (my last, at any rate)
* fast pulser - 18,000 watt-seconds of danger
* railgun #2 - powered by the pulser
* fast miniature Marx generator (100kV)
* fast large Marx generator (1MV) for DEW studies

PLANNED
* electrothermal gun - powered by the pulser - fairly easy & quick
* tabletop EMP simulator, powered by the 100kV micro-Marx
* 1MV (?) Van de Graaff generator - for the accelerator below
* 1 MeV proton accelerator - yes, really.
* electron microscope
* DEW studies (TEMP, not HERF)

Last but not least, I once seriously considered building a pulsed radiation source. It would have to be either electrons or x-rays, for what I hope are semi-obvious reasons. A gamma ray machine would be entertaining but let's be realistic here: for one, there is no way in hell I am going to generate an e-beam with enough energy to to make Bremsstrahlung gamma rays in my garage and on a mortal's budget. Second, the shielding is non-trivial. Say you have just enough lead around your operation to stop all of the gamma rays. Guess what comes out the other side? High energy X-rays! So now you need more lead to stop the high energy X-rays. Guess what comes out the other side? Low energy x-rays! So now you need even more lead.

So yeah, I soon thought better of that idea. After all, if I do that, I can't very well look down upon the clowns on YouTube who play with exposed magnetrons or shove fireworks up their butts, now can I? If I had more money than I do, and plenty of space for things like drums full of water, I could do it safely. But I don't, so I can't, so I won't.

Obviously, many of these projects depend on each other. The prerequisites will get built first.

Okay, that list is complete enough that I shouldn't have to do that again.

Any questions?

short update on vacuum

2010-08-08T08:30:00.001-06:00

I am still chasing leaks, but I have bigger worries which must be addressed first.

1. I need to replace the motor. A 3/4 HP motor rated for fan/blower duty is VERY different from a 3/4 HP motor rated for compressor duty. If I carry on with the motor that's mounted on it now, it'll overheat. I had to set up a fan on the motor to keep it cool. I actually have a compressor duty 3/4 HP motor on hand... that doesn't turn. BUT: I believe I know what's wrong with it and that it's easily fixed. I will be looking into that today.

2. There are several switches and pilot lights -- on both the Veeco system and the outboard vacuum tube Veeco gauge controller -- which need to be replaced. One of them is causing erratic readings on a T/C gauge.

One needs the basics working properly before one can presume to troubleshoot something.

Meanwhile, I also have plenty to do on main components for the MSLS, and hope to work on that as well, unless I get diverted onto the honeydew list...

More as it happens. Stay tuned.

further ado about nothing

2010-08-03T06:36:00.001-06:00

Hmph. This evening I reassembled my high vacuum system and not terribly surprisingly, I still have a gross leak... according to the gauges built into the system... the system I bought at an auction for $20... the system which I and a good friend have systematically rebuilt from the ground up... with the exception of the gauge electronics.

Well, I've got the roughing pump running at least. VERY rough. If I could believe the panel gauges (and I cannot) the pressure is bottoming out at just over 100 millibar.

The reason I've revisited the vacuum system so soon is that I have constructed a device (ye gods how I love that phrase) which I wish to encapsulate in epoxy. (it's another special effect for the Mad Science Light Switch)

To do encapsulation properly (ie; without voids) one needs a reasonable vacuum pump. I now have a reasonable vacuum pump again, although I do wish its vacuum were a bit MORE than reasonable. I'll need a few more bits of plumbing to connect the plasma cleaning chamber I'm going to use for this to the vacuum cart, but I ought to be able to get it done during the next week or two.

The hilarious part - to me - is that the electronics in the "trusted" gauge controller are so old, I will have to apply power to them gently -- slowly increasing the voltage from zero, using a variable transformer -- over the period of, say, an hour. This old-timer's trick may preserve the old high voltage filter capacitors inside which need to be "re-formed" after having sat idle for many years. It would be -- unpleasant -- should one of them fail within the chassis. Ick.

I do love old tube equipment though. Using it in the winter saves on heat. And fire bottles have such a cheerful glow.

dV/dt is not your friend

2010-08-02T06:56:00.001-06:00

Honestly, I should not presume to dispense advice on the fine art of building a high performance Tesla coil, when I myself am still making bozo-nono (thanks for the bôn mot Jon!) mistakes such as blowing up capacitors.

Observe:

This is not what a happy, healthy pulse cap looks like, dear readers. This is what a pulse cap looks like after its RMS current rating and frequency rating are exceeded by a few orders of magnitude. Internal solder connections between the individual capacitors in the string melted, then arcs formed, vaporizing the mineral oil impregnant. Technically, the case ought to have a vent to prevent a catastrophic casing rupture, but then technically, I should not have operated the cap in the manner I did.

This was one of two capacitors left over from my days with Dream Park Corporation and DiaboliCo. If I recall correctly, we were paying around $200 each for these. I used one in each ≈1500 watt Tesla coil. They were made to my specs specifically for Tesla coil duty by Plastic Capacitors Inc. They are perfectly fine capacitors, although if you call up PCI and tell them that you want to order a capacitor for a Tesla coil, they might not give you the time of day.

That's because they have lost count of the number of amateur Tesla coil builders who have used their caps, had them fail, and blamed the manufacturer. When DiaboliCo got into the Tesla coil business, PCI were a bit reluctant to talk to me until I started talking quantity, and answered all of their technical questions concerning operating specifications and conditions. As far as I know, not one of the capacitors we sold in any of our Tesla coils (quite a few units, by the way) failed after hundreds of hours of operation. They are good caps if they are specified properly.

As I said above, I ran this cap in conditions far beyond what it was designed for, and it refused to do the impossible. However, the one fun part I got out of all this was what it looks like on the inside, which is the whole reason for this long-winded post.

Incidentally, the ratings on this capacitor were something like (from my notoriously unreliable memory): 0.01µF, 15kVACW (_RMS_ @ 300 kHz) and around 10A circulating current. That last number might be a bit high, it's been over a decade.

I needed two of these for my "Last Tesla Coil" because I have already completed the secondary assembly, which means the designed operating frequency is now fixed. And since I want a large amount of inductance in the primary, that fixes the primary tank capacitance to a very narrow range.

Part of my goal with this coil is to approach the spark-making efficiency (basically, output spark length over input power) of a double-resonant solid state coil -- using a spark gap switched system. It is easy to dissipate half your input power in a badly designed or incorrectly operated spark gap. One way to minimize the power lost in the gap is to minimize the current through the gap. And a great way to do that is to increase the inductance in the primary. Remember, the field intensity generated by the primary goes as ampere * turns, so swapping turns for current doesn't hurt us in terms of transferring energy to the secondary. Also remember that this bloody horrible thing is NOT a transformer, and cannot be modeled as a simple AC transformer. If you knew me 15 years ago, you know that I believed that Tesla coils COULD be modeled as a transformer, albeit a resonant one. Well, I was ignorant. I've learned a few things since then. The point being that we do not care about turns ratio between primary and secondary nearly as much as we would if this were a simple transformer.

So what to do? Well, in the time since I specified and used these capacitors back in the 90s, the state of the art in Tesla coil capacitors improved dramatically, with the development of the "Multiple Miniature Capacitor" -- Capacitor; or "MMC Cap" for short.

The idea came from a simple logical progression. The Tesla coil building community knew what was inside most high voltage capacitors. Pretty much all of them look like this inside, one way or another, although geometries can change for different purposes. We knew that capacitors that _should_ have withstood some RF voltage and current combination failed anyway, and the cause was widely held to be dielectric losses - the dV/dT was too high for the dielectric.

There were relatively inexpensive "commercial / off the shelf" (COTS) capacitors available with very good dV/dt ratings, but their highest voltage was much too low and so was their capacitance, for any reasonably interesting Tesla coil. They had reasonable tolerance - 10%. So why not string a bunch of them in series to obtain the necessary withstand voltage, and then connect several such strings in parallel to obtain the needed capacitance? How much would that cost?

It turns out that it costs a few hundred dollars at the usual prices to build a large MMC cap. By "large", I mean one capable of running a medium to large sized coil, or in the 1,800 watt to 10 kW range.

The best capacitors, determined by trial and error and a lot of deliberate testing were the 942C series from Cornell Dubilier Electronics (CDE). They were (and remain) damned near unobtainable through regular commercial channels. No one stocks them, not even CDE. Lead times are long. Group buys were put together to bring the price down. An amateur science education and "geek shelter" group made a huge bulk purchase and resold the caps at a reasonable markup (to help support their project) which was still cheaper than most individuals could buy them on their own.

Since then, they've become the defacto standard for anyone who can afford them. They turn up on eBay now and then, which is where I purchased this set:

Conveniently mounted and all connected up for me... except that this unit isn't designed quite conservatively enough to satisfy my fetish for reliability, so I'm going to need a few more. Irrelevant - the price on this lot was very good. I hope I can do as well the next time.

pulsed power basics

2010-07-09T08:46:00.001-06:00

Yeah, yeah, I know I haven't written a post in a long time. I've been busy. This blog is not my life, it is reportage on one small facet of my life. Family drama, home life, minor illnesses, fatigue, and actually trying to work on the things I write about here have used up the time I might have used to write blog posts. So there.

~~~

In my travels around the internet, I often stumble across people attempting some of the same things I do. Some times I stumble across people who seem to have grasped some basic physics, but somehow left big gaps in their education and as a result, are promoting completely ridiculous ideas which they feel are sure to revolutionize some industry or other if only people with lots of money (and incidentally, also lacking a decent high school science education -- a real one, I mean, not like what they teach in high school today -- hell, is science even permitted in the classroom today?) would fund their research. Okay, sure, some of those are scams, but some of them are more along the lines of, "hEY guys, this seams leik a grate idea, Y WONT this woRk?"

I could provide examples, but that would be rude.

And then there are all the purveyors of refined bovine waste products on eBay who want you to believe that you can get 50,000 watts of power from a battery-powered device that fits in your pocket. Without it exploding, I mean. I stumbled across that gem earlier this week, and it just cracked me right up. FOUR MILLION VOLTS? Someone call up the guys at all the particle accelerator labs and let them know they've been doing it wrong.

And finally, I know there are a few non-techie types who read my babbling and actually want to understand what I'm writing.

Thus it occurred to me that a very short, simple explanation (more of a review, really) of a few basic physics concepts would be helpful.

First and foremost, this blog primarily discusses "pulsed power". So what is "pulsed power"? Wikipedia has a concise little definition which I like:

      "Pulsed power is the term used to describe the science and technology
      of accumulating energy over a relatively long period of time and releasing
      it very quickly thus increasing the instantaneous power"

An excellent example would be the electronic photo-flash for a camera. The flash's circuitry converts battery voltage to a much higher voltage, slowly charging up a capacitor. Then the capacitor is discharged through the flashlamp in a pulse which lasts less than 0.001 second (1 mS) but which may have a peak power (most of which is turned into light be heating up the xenon gas inside the lamp) of over 1,000 watts.

That definition uses two important terms which we must thoroughly understand before we proceed: energy and power.

Energy is nothing more than the capacity to do work. It is a quantity, a value, which we attribute to many things - subatomic particles, fuels, masses in motion, car batteries.

Two basic categories of energy are potential and kinetic.

Potential energy is energy stored in a system that could do work if it were to be released.

Kinetic energy is energy in motion - energy being transferred from one system to another - in general, kinetic energy is energy which is performing work.

Now then, power is not the same as energy. Power is the rate at which energy is transferred, moved, used, etc.

Both of these parameters can be measured in different units for different purposes, but they each have a single basic unit of measure which is preferred for science and engineering purposes, and which is derived from the other basic scientific units of measure.

The basic SI (International System of Units) unit of power is -- you guessed it -- the watt. Watts are not exclusively electrical in nature, although that is the context in which most of us use the term.

The unit (derived) for energy is the joule. One joule equals one watt of power being delivered / expended per second -- one watt-second.

You can probably see why our power bills don't use joules as a unit of measure. The joule is not a very large unit. If a joule = one watt-second, then one kilowatt-hour equals 3.6 million joules for which you pay, on average, around ten cents if you live in the USA.

You may have already noticed that the thing which binds energy and power together into a relationship is time.

A lot of energy expended over a long period of time results in low peak power. Sometimes that is desirable, and sometimes it isn't. Likewise when a lot of energy is expended over a very short time the resultis very high peak powers. Sometimes desirable, sometimes not.

Very large peak power levels are difficult to arrange for long periods of time. It is very expensive, and requires very large infrastructure to manage. Just take a look at all the huge cables, towers, generators, substations and so forth required to bring a few paltry megawatts to your neighborhood.

Yet it is not too difficult to generate a short pulse of megawatt peak power in a laboratory setting. Sometimes, a short pulse is all we need, such as in the strobe light example provided above.

With very, very high peak powers, short pulses are all that is possible for practical, physical devices built by mere human beings.

That should do it for this installment.

My next one will probably discuss how not to be stupid, using my own most recent stupidity as a singularly painful lesson.

Tesla coil specs: added secondary inductance measurement

2010-06-06T07:23:00.001-06:00

L_s, calculated: 39 mH

L_s, measured: 36 mH

Not bad, all things considered. And my LCR bridge is not in calibration, further it doesn't have the measurement resolution of a modern digital bridge, so for all I know the measurement and calculated inductance are exactly the same.

I suppose I ought to pick up a few inductance and capacitance standards some day...

I have a lot of property maintenance tasks to do, so I doubt I will get the secondary + torus self-resonant frequency measured today. I need to set that up in open space, which is a pain in the ass. Although realistically, an indoor measurement will be close enough for practical purposes.

Tesla coil specs

2010-05-31T08:39:00.001-06:00

I just filled in many of the electrical parameters of The Last Tesla Coil in my earlier post on the subject, for anyone who is genuinely curious / serious about Tesla coils.

(link for convenience)

I seem to have lost some of the measured specs off the secondary, so I'll try to re-measure those in the near future.

Some numbers are not easily measurable, and are only practical to arrive at by calculation. Two such numbers are the capacitance of the top electrode and the distributed capacitance of the secondary coil.

lock bumping

2010-05-30T15:35:00.001-06:00

Security isn't really my main topic of interest here, but it is a hobby and powerful interest of mine. Most anyone else with even a passing interest in security has probably heard the expression "lock bumping" or just "bumping" with regard to a quick way to open a vast number of pin tumbler locks.

Well, Kwikset has just announced a lock with "BumpGuard™ Protection"! They do not anywhere seem to claim that the lock cannot be bumped.

Further, their video makes it look an awful lot like a lever tumbler side bar lock. Only a bit different from a Medeco, and people are always telling me how "Medecos can be bumped" as a general sounding rule, but the only evidence I ever find is a canned video of _a_ Medeco being bumped. I would like to see someone do it in the wild.

Meanwhile, I presume the locksport community is already on this like white on rice and will have something to say sooner (if it gets defeated) or later (if it doesn't, after extended efforts).

I wonder if they make a rim cylinder version... hmm...

See, in my neighborhood, no burgler will pick a lock. This isn't the domain of rich folks with lots of high density portable wealth. Around here, the most determined burglar is likely to try battering a door down or breaking a window or prying open a flimsy garage door. They'll be looking for tools, bicycles, VCRs, computers, sometimes guns.

The majority of burglars just look for unlocked doors! Come on, you other home owners, you're not helping!

Sika™ AnchorFix-2 anchoring adhesive

2010-05-29T22:45:00.001-06:00

Whilst exploring the masonry area of my local big-box homeowner's warehouse store, I discovered the most wonderful adhesive / fill material. It is intended for fixing anchor bolts, rebar, threaded rod, and the like in holes in masonry. It sets very fast and may be fully loaded typically within an hour (depending on temperature).

For the maker / mad scientist / hobbyist / engineer, I would say this material fills in a notional gap between say, JB Weld and a bag of Quickrete™.

It's called Sika AnchorFix-2. It is a quartz(silica) filled, two-part acrylated epoxy packaged in a standard (10oz) caulking tube. The two components of the epoxy are separated by long skinny bags inside the cartridge. These bags are closed with a crimped wire at the nozzle end. To open and use, one simply removes the replaceable cap from the nozzle, pulls the crimped plastic out a bit, cuts it off with a knife, inserts the cartridge into a caulking gun, connects one of the included mixing nozzles and get busy.

The stuff is stronger than concrete. Pot life is strongly dependent on temperature, but this stuff cures fast! At 20ºC it has a working time of only 5 minutes, and in 40 minutes it has reached its full strength for rated loads. Work quickly or in small batches.

It's a bit tough to find detailed technical properties. There are bulletins with its ratings for specific anchoring applications for which it is intended (such as threaded rod or rebar into undercut holes in concrete) but not much else that I've found so far:

http://www.sikaconstruction.com/con-prod-app-aac.htm

And it's expensive. I paid $19 for a 10.1 oz tube at Home Depot. It comes with two mixing nozzles and a cap. I used up both nozzles (clogged both with cured epoxy) in quick succession. But I soon discovered that where the stuff comes out of the nozzle on the cartridge, the two ingredients don't really come into contact - much. The tube come with a cap so you can recap it. Turns out you can simply squirt it out onto a handy piece of cardboard and mix it up with a putty knife. Clean it off your putty knife before it cures or you'll be sorry.

I bought it to anchor the floor end of a piece of reinforcing steel which beefs up the entry door to my garage and workshop. But it appears I'll be able to save the rest for a future job, which is good, given the price tag.

The cured product is quite tough and abuse resistant, and sticks to anything normally epoxy-able. It is FAR less brittle than portland cement-concrete. Within an hour, you need a masonry bit to get through the stuff. Hammer blows don't faze it once cured. It sure doesn't like being heated to the melting point of steel though, heh-heh. (I had to weld on some steel that had this stuff under it). Makes an awful stink, which is no doubt toxic.

And DON'T try to drill it with HSS tools - it's got sand in it, remember?

I've never seen anything quite like this before. It goes way beyond the garden-variety, sand-loaded, butyl rubber masonry repair compound I've been using for decades. Now I'm going to be watching out for applications for this stuff all over my property...

another lesson learned the hard way

2010-05-13T21:28:00.001-06:00

I just snapped off a cheap Chinese-made piece of shit drill bit in a part that had some eight hours of my labor in it so far. The bit was too hard. Pondering how I am going to get that damned drill bit out of my previously-beautiful machined brass part made me a little crazy, but I'm better now. The doctors give me these pills to take, to calm my nerves, you see...

Now where was I?

Gee, those American made numbered drill sets sure are expensive compared to the imports! I wonder why...

project catalog, part 2 of N: The Last Tesla Coil

2010-05-13T10:46:00.001-06:00

Continuing the promised enumeration of all the various projects I have "open", the next one is my "final" Tesla coil.

Over the years, I have built quite a few Tesla coils, ranging from the small vacuum tube design I found in an old copy of Radio Electronics magazine which I built for high school Science Fair, through my first large coil, and culminating in several pretty but rather inefficient coils I designed for Dream Park Corporation and DiaboliCo, which sold quite a few of them to haunted house attractions.

I have never built solid state coils for a variety of reasons, mostly having to do with money.

Furthermore, with the exception of the commercial coils, most of my coils have been ugly lash-ups, with only the barest nod made to aesthetics.

A few years ago, I decided to build one last Tesla coil, this time using all the little tricks I've learned over the years to improve efficiency and performance, and also with the intent to produce a machine of nearly museum quality. Since I have very little experience working with wood, the latter criterion has been difficult to meet, but I've come close, so far.

The overall physical pattern follows one that every Tesla coil builder will be familiar with: a lower deck on wheels which carries the high voltage transformer, spark gap, capacitor, and four supports for the primary deck, the primary deck to which is mounted the eight primary supports and the secondary coil.

Each primary support also has an insulating standoff at the outer edge to hold a grounded brass strike rail. I have already bent the strike rail, and simply need to install it once the primary supports are installed.

The electrical design looks to be fairly efficient on paper. Similarly constructed coils do fairly well, generating streamers 60" - 74" in length from 1,800 watts of input power. The main stumbling block at the moment is the lack of copper for the primary, which will cost some $200 at today's prices. I've managed to miss two dips in the price of copper in the last five years by dint of being broke when the price was low. But now the price is headed back up.

Another goal of this coil is to try out several spark gaps with it, using the best designs others have come up with as well as my own twists, and try out all of them, optimizing performance and tuning each time, to see which one performs best. To that end, I have finished one spark gap already - the "sucker gap":

I've also collected most of the parts for two more gaps - a salient-pole synchronous rotary, and a Richard Quick-style gap.

One thing that I'm doing on this coil which is at least somewhat unusual from most others is the primary connection arrangements. First of all, the primary will be accessed through slots in the primary deck. That allows the primary leads to be shorter since they don't have to go around the edge of the primary deck, reducing wasted inductance. Secondly, the primary connection will be made with a special connector I am designing which will fit neatly between turns, gripping the top and bottom edges of the copper strip with a spring tension clip. This will largely eliminate the projection of the primary connection above the plane of the primary coil, which should eliminate one of the common strike points between secondary and primary.

It occurs to me that the design specifications might interest someone. I've included the basic stats (from memory, there could be mistakes) and I'll post the missing details later, after I have a chance to pull them out of a spreadsheet.

Secondary System

form: 6.25" OD x 38" thin-wall PVC tubing
wire: 1200 turns, 22 AWG double-build motor wire, close-wound
toroid: 30" x 6" fan guard
coating: Glyptal 1201, six coats
L_sec, calculated: 39.5 mH
L_sec, measured: 36 mH
C_sec, calculated: 14.4 pF
C_toroid, calculated: 43.7pF
F_sec, calculated: 105 kHz (loaded)
F_sec, measured:

Primary System

form: eight maple supports w/ strike ring supports
material: 1" x .060" copper strip
turns: 10
diameter innermost turn: 10"
diameter outermost turn: 32"
turn spacing: 0.5"
Lp, calculated: 160 uH
Lp, measured: (I won't be able to provide this number until I build the primary)

HV transformer: 15 kV / 120mA NST w/ anti-resonance+anti-surge "Terry Filter"

C_pri, measured: .02uF

F_pri, calculated: 89kHz (slightly lower than F_sec, for tuning)
F_pri, measured:

slowly I milled, inch by inch, step by step

2010-05-13T08:22:00.000-06:00

Nah, doesn't have quite the same ring to it as "slowly I turned..."

I have managed to put in at least one hour of work per night on the Mad Scientists Light Switch - among other things - since my last post.

The work I'm doing now is a bit fiddly, but the outcome will be the first noticeable (to an outside observer) progress on the thing in months - something very visible.

By the end of the weekend, I ought to be done with the mounting arrangements for the Jacob's ladder, and should have moved on to the final machining, drilling, and tapping, in the base of the Jacob's ladder tube.

I obtained a second "beer sign" NST (7.5kV @ 15mA) from my friend John, and very soon I need to check whether there is room for two of them plus the control electronics. I'll be terribly disappointed if I can't fit two of them in there, because most likely, it won't work without two. With a single NST of this size, there just isn't enough current to heat the arc sufficiently to make the damned thing work.

~~~

Meanwhile, I have the phenolic slab back on the surface plate, with the ends supported by a pair of parallels (flat), a heavy weight (120mA 15kV NST, weighs about 100 pounds) in the center, and two heat lamps warming the whole shebang.

Measurements taken yesterday suggest that the current method is working, albeit slowly. I'm going to give it a full week and measure again. Once I think I've got that piece in the ballpark of "flat", I'll let it sit for another week to see whether it "relaxes" into its old shape.

After that, I'll know whether I want to continue on that path, or just say "fuck it" and mill the thing flat. At least I know now that the latter option is a perfectly reasonable one.

I can't recall whether I've mentioned this before, and I'm too lazy to look, so...

I have a pair of copper bus bars which are perfect for the size of launcher I'm working on. Well, their cross section is perfect, if their length is a little shorter than I'd like. They're roughly 0.75 x 0.5 x 24" or so. But the price was right. The only problem with these rails is that they were previously bus bars, and whoever made them drilled several .25" holes through the thickness. My plan is to ream these holes out to a precise diameter, turn some dowel pins out of 110 copper to a diameter about .002 larger than the holes, and heat-shrink the pins in the holes. I'll order the reamer I need for this operation next pay day. Need to remember to order a few replacement drill bits too...

"the best is the enemy of the good" - Voltaire

2010-05-11T09:01:00.001-06:00

The results of a four day long experiment - to see whether I could remove any of the warp from my 1" thick slab of phenolic-linen laminate - were inconclusive at best, because I failed to measure and write down the amount of warp when I started. I think I removed some of the warp, but I don't think it was very much.

So this morning, using the big surface plate at work, I measured (and wrote down) the amount of warp which is in the piece now . I'm going to take one more stab at bending the stuff over the coming week, and see whether a week long attempt has enough effect on it to be worth continuing.

If not, I'll machine it flat.

It occurred to me to ask, "what if I were to simply mill it flat, what would the consequences be?"

The first consequence is: I'd be removing around 0.1" of material (total) from the material thickness, or about .05" from each side.

That begs the question, how much strength do I lose by going from 1.0" material to 0.9"?

The answer to that turns out to be, "surprisingly little", because of how closely spaced the clamping bolts are. Or rather, one loses exactly as much strength from the material as one would expect, but the "beam" length between any two supported points (pairs of bolts) is so short, it just doesn't matter. I have tried to calculate all of the magnetic reaction forces in the launcher (with the calcs on the rest of the bus-work to follow soon) and it looks to me like the worst case is at the breach connections (which is what I expected).

When I apply the calculated force to the bream (and assuming preloaded Gr.5 bolts), the resulting deflection is so vanishingly small I had to do it three times and re-read the calculator instructions (yeah, there's a calculator for just about everything online, who needs to know any math?) before I believed the number.

My railgun containment structure is gonna be stiffer than John Holmes.

~~~

I haven't worked on the Mad Scientists Light Switch lately because the best fix I can think of to get that project out of the jam it's in requires four to six contiguous hours of focused work - contiguous hours I haven't had.

Actually, I did have the time available last weekend, but I used that time to work on improving the security of my workshop instead - yet another ongoing project with no end in sight.

I'll probably revisit The Light Switch this coming weekend.

taking a warp out of a piece of phenolic laminate

2010-05-06T11:20:00.000-06:00

I have a piece of phenolic linen laminate that I'm hoping to use as either the top or bottom containment plate for EML2 (railgun #2). But it's got a longitudinal warp in it. That is, it's curved from end to end, but not across the width. The material is 36.25" long, 4.75" wide, and 1" thick.

It's the thickness I'm worried about. I could mill the thing flat, but then it wouldn't be 1" thick. In fact, I need to mill the bore portion flat anyway, and I'd like to remove as little material as possible.

While it's possible that I could pull the warp out just by bolting it to the other (presumably straight) parts; I wouldn't count on that, and in any case, I need to be able to verify that the bore is straight and flat in both axes with the launcher disassembled. I could clamp it to the more-or-less-flat table of the mill, mill the required features, and then let it spring back, relying as I said earlier on the clamping bolts of the launcher assembly to pull that plate straight. Of course, it might just pull the other plate and the side walls and everything else slightly out of true too.

For now, it is clamped to my surface plate, which is too small to hold all of it even diagonally. I might have to borrow the surface plate at work over a weekend and see if I can apply heat to it (with a bunch of heat lamps, I suppose) while it is clamped flat. I'm assuming heat caused the warpage in the first place. Of course, I don't know whether phenolic warps TOWARD or AWAY from the heat source. In the case of wood, it's easy, because it's a drying effect. In the case of a thermoset resin, I really don't know.

I've been trying to find out, but no luck so far.

project catalog, part 1 of N: Railgun #2

2010-04-30T12:40:00.005-06:00

It is long past time for me to post an enumeration and explanation of all the various projects I have "open" at the moment - that is, projects I've started but not finished. This post could get long quick. In fact, I think I'll limit this post to one project and likewise write individual posts for the other projects.

Some of these have been in progress a long time, and have been repeatedly tabled and restarted again for various reasons, usually money and spare time, sometimes because I wasn't sure yet how to proceed and had to stop and do some planning or research.

So first of all...

Railgun #2 AKA "EML-2" (Electro-Magnetic Launcher #2), AKA MSRG-2 (Mad Scientist RailGun #2) etc.

Incidentally, I'd like a clever name or acronym for this device if anyone feels creative.

The pretty picture at left is only that so far - just a conceptual design, although it is slowly turning into a real model that I can make real shop drawings from which I can fabricate the real deal.

Originally planned to be a roughly 1 meter class device, the various potentially-expensive parts I already have on hand limit the actual accelerator rail length of the first version of this launcher to about 24" with another 12" or so of pre-acceleration gas gun on the breach end of the railgun section... for now.

Later, when I can afford to purchase high grade rails the full length of the casing, I will convert the gun proper to 36" of rails, bolted to a completely separate injector gun. The materials get expensive if you have to buy them new, so I'm starting out with the various stuff I have already obtained cheaply, and designing so that I can add significant improvements later, re-using some of the more expensive materials.

The design of the launcher itself is still very much in flux, but many of the design decisions are now either made or are rapidly narrowing. Ferinstance, the bore of the first iteration will be approximately .75" square. Round bores are too much hassle and I don't need to build a round-bore device since I'm not interested in a practical weapon - this is strictly an unportable, short range, laboratory device. Oh by the way, the last time I contacted the BATF, the agent I spoke with said he didn't think EMLs / railguns met the criteria for a "firearm", although I never actually pursued the formal "determination of firearm status" (or whatever it's called) paperwork.

Like any other amateur railgun builder, I have little choice but to use a high voltage capacitor bank for the power supply. For over a decade, I have possessed a very large capacitor-based pulse-discharge machine which stores 18,000 Joules in six 60uF capacitors at 10kV max charge. The capacitors are big low-inductance pulse discharge devices, about 100 - 150 pounds weight each. They are significantly derated in terms of both expected current and voltage reversal, and I have all of the designer's paperwork on this machine. It was an experimental magnetic metal forming machine, built in the early 1970s.

It originally used six ignitron tubes to switch the current, one mounted on each capacitor. Ignitrons are rather slow as pulse switching devices go, but they are fine for railgun pulse times, they handle quite large charge transfers compared to triggered spark gaps, and they are rectifiers, which largely prevents voltage reversal across the caps. Furthermore, multiple switches - if they were still here - might have allowed me to play around with distributed energy storage and a segmented launcher some day. Unfortunately, the ignitor electrode assembly of an ignitron becomes very fragile once it has been used in crowbar or pulsed power switching. All six tubes were wrecked long before I got my mitts on it. Replacements are $1,500 each, which is $9,000 I certainly don't have just lying around.

So my plan is to replace all six capacitor banks with one of two triggered spark gap switches I have on hand, hopefully the smaller one I constructed, but if necessary, a larger railgap style switch is available, although it needs work.

In the absence of controllable switching, it's possible I'll need a pulse forming inductor to slow down the current pulse, or alternatively, an impedance matching transformer capable of handling a few hundred thousand amps. This is surprisingly do-able, and I've got a white paper that shows how. However, that thinking concerns standing-start launchers.

I do not plan to operate from a standing start unless I find it just isn't practicable any other way. I'm considering a way to use the armature to help with the switching, but not actually use the armature to do all of the switching.

The rails are copper, 0.5" thick x 1.0" wide. I have these, but have not machined them straight and flat yet. They were used as bus bars, so they have a few 1/4 inch holes that need filling. I plan to ream the holes and press pins (machined from 110 copper) into them. Some big bolt holes are already conveniently located at one end of each rail.

Bore liners will be fabricated from garden-variety Delrin for the first iteration. I do not have the liner material yet.

I'm still working out the breach connections to the rails but as the rails already have some .400 holes in them, you can bet that a 3/8" bolt will be involved somehow. No, I won't be passing current through the bolt.

Top and bottom clamping plates (resist the repulsion force of the rails, and indeed the rails mount directly against these plates) will be phenolic-linen material, 1" thick x 4.5 inches wide x 36" long.

15 - 20 clamping bolts, 1/2-24 x 5" gr.8, will hold the launcher together.

Armatures will be cut from aluminum "C" extrusions readily available from industrial suppliers. There are profiles offered which are startlingly similar to many armatures used during the middle of the railgun development days. I say that last bit because I think we're nearing the end of the R&D period of railguns (some 40 years!). A weaponized practical railgun should be fielded on a Navy ship by the close of 2012).

The injector will run off of high pressure CO2, using a small bottle similar to the one I use for MIG welder. The storage plenum for each shot will be a paintball gun tank. The valve between plenum and injector breach seems to be the Dema 458P.

One aspect of successful railgun design which seems significantly lacking from most other amateur's efforts has been instrumentation. At minimum, we will want to know the pulse supply / rail current, the breach voltage, the muzzle voltage, and the armature muzzle velocity, recorded over the full duration of the shot. Depending on how things work out with the armature-triggered switch idea, it may also prove necessary to have armature position sensing in at least one position - and preferably several - inside the launcher system.

This last desire is a rather challenging trick. You and I can't (well, MOST of us probably can't, and certainly shouldn't) take high speed flash x-ray photographs of our launcher during the shot.

Various ideas have been tried by the pros, and only a few of them seem to be practical for amateurs. For now, I'm thinking about various methods of optical probing with fiber links, so-called "B-dot" probes and even electrical contacts embedded in the liner walls.

Current pulse measurement will probably be done with a Rogowski coil and integrator. Voltage measurements simply require resistive, non-inductive dividers. Isolation and grounding will need to be watched carefully. I have no intention of blowing up my Tek 7934 with EMP, EMI, RFI, etc.

I've left out a great many of the details I've been sweating over. Any questions?

Lately I've begun to wonder whether there would be any point in setting up a PayPal fund for this project. It's going to be a long and expensive one. It already has been, and will continue to be. These things aren't easy or cheap, usually.

and now, a moment of extreme geek humor

2010-04-29T12:34:00.002-06:00

"Yo momma so abhorrent, nature prefers the creation of the Higgs Boson in particle accelerators!"

I have altered this blog. Pray I do not alter it any further.

2010-04-28T11:22:00.002-06:00

I have three things for you right now:

Thing One:
I've modified my personal/profile info for this blog somewhat, due to ~~paranoia~~ my renewed interest in privacy concerns. This shouldn't change anything for most of you since, as far as I can tell, there are only half a dozen people reading it, and you all know who I am.

Thing Two:
I have found another important reference book for makers. In my opinion, this title ranks with Machinery's Handbook (which, if you haven't already got a copy, means that either you do not work with metal at all, or you DO work with metal, but your entire tool collection consists of a bag of hammers).

This "new" classic title is the Tool & Manufacturing Engineers Handbook, published by the Society of Manufacturing Engineers. Mine is the 3rd edition, ©1979, and as far as I can tell, none of the basics for which it is a reference have changed since it was written. I picked it up dirt cheap, used, all three thousand trade format pages of it, through Alibris, a clearinghouse for small booksellers which I highly recommend.

Thing Three:
Here is a reference page about nuts and bolts, containing handy demystification of basic concepts needed to use them skilfully and properly:

http://www.gizmology.net/nutsbolts.htm

So useful I archived it, as I generally do with useful web pages or sites, as they have a nasty habit of disappearing abruptly, leaving you without the info unless you saved it yourself. Disk space is cheap, your time is not.

temporary, brief, inexpensive, but highly irritating setback

2010-04-28T08:51:00.001-06:00

Remember, while I may be designing Railgun #2 in my spare time, the actual physical shop work going on remains the MSLS until all aspects of it are finished and working. So that's what I was working on last night when I was hit with another "learning experience". This is about the fourth time I've had this particular learning experience. The first two or three times this happened, the "lesson" I learned was the wrong one: I thought I wasn't doing it right. This time, I learned the correct lesson, which I will share with you now:

Joining metal with solder is to be used for two purposes and two purposes only:

* electrical connections which do not see high stress or current

* copper water pipes

Period, the end, full stop. (though I won't be surprised if someone comments with an application I hadn't thought of.)

More directly, solder shall not be used for mechanical fastening of any sort, for any load, in any application, any time, anywhere.

This time, I will put the two parts back together with fucking machine screws, and plenty of 'em. I am done fucking around with this silly part, given that I have already spent the equivalent of the more expensive method I was trying to avoid when I decided to try to put three bits of scrap brass together rather than machine the whole thing from billet. Which is what I will do "next time" if something like this ever comes up again.

~~~

PS: Have I mentioned before that I swear a lot? Get used to it. I'm a man's man.

** spits on floor **

railgun R& D work

2010-04-27T11:57:00.003-06:00

Because I am an ADHD/OCD type, I generally work on multiple projects at a time. This has held me back in the past, as I tended to start new projects when I had existing projects already going, and often I would turn aside when some obstacle slowed me down.

Now that I have a few handles on these tendencies - not to mention medicinal crutches, I'm finding I can finally achieve focus on a single project.

However, I've still got multiple projects open right now, so I still do Research & Development work on longer term projects (such as Railgun #2) while I'm doing fabrication work on things with completed designs (such as the Mad Scientist Light Switch project).

So, with regard to the RG #2 design... there are many design variables, and I've been trying to find answers which enable me to balance them against each other and determine the optimum values based on the materials I already have. Most of this I've done by reading white papers, doing a few back-of-the-envelope calculations, and a bit of model work in SolidWorks.

So here are some conclusions I've worked out recently on the design of RG #2:

* at 10kV, I've got all the voltage I need. The Dahlgren and Kirckudbright EML facilities seldom used voltages higher than 10kV. A fairly promising student-built "bench scale" EML at the Naval Postgraduate School performed well on 20kJ @ 10kV and that was without any of the performance-enhancing tweaks which have become common practice since that gun was built.

* the first bore is going to be right around .75" square

* My pre-accelerator is going to be scary dangerous all by itself - it will be able to operate reliably from just over 100 to well over 1,000 PSI. Dump valve is rated to 1,200 PSI maximum working pressure. I have yet to do any friction calculations for the armatures, but for sliding fits, the acceleration is genuinely frightening. At least the pre-accelerator section won't need to be very long.

* pre-accel gas will most likely be CO2

* the best looking pre-accel valve I can find, in terms of high flow and pressure numbers, is the Dema (manufacturer) model 458P. Funny thing, that looks to be exactly the same valve that Sam Barros used on his second gun. I couldn't find a better one - at least, not one I could ever afford. I've started an eBay search for this item.

* the dimensions of the rails I have are just fine for what I want to do. I'll actually be cutting one of them shorter! (they came as two different lengths)

* augmentation methods still look promising, especially on smaller ("bench scale") guns of the size I am able to build.

* I have found where to source extruded aluminum C-channel, at a low price, which has a nearly ideal cross section for easily fabricating C armatures

* bore liner material is still likely to be Delrin for the first try. Something with higher arc resistance may be necessary near the pre-accelerator / driving rail interface - G7 perhaps.

* the 1" x 4.75" x 36" LE phenolic material I have on hand will work beautifully for one of the clamping plates. I will need a very similar piece for the other clamping plate. This stuff is expensive, I was lucky to get the first piece for nearly nothing. I will have to shop carefully ( cut-drops from eBay or local plastics suppliers) to get the second piece for an affordable price. I've started an eBay search for the second piece of this stuff.

* the afore-mentioned LE phenolic clamp plate material is not flat. It appears to have been slightly bowed by heat. To address this, I can think of three options:

1. bolt it to a piece of milled-flat I-beam and adjust the bolts and nuts until it's pulled straight. Note that I have been thinking of picking up a chunk of I-beam to mount the gun on anyway.

2. attempt to straighten it by heating it while it lies on my surface plate.

3. mill it flat, removing a fair amount of thickness in the process (probably around .125 total)

* the clamping plate material I have on hand is stout enough to allow me to have various gun designs within a single set of plates, and should handle perhaps 4X the energy & power I can deliver now. If I get more caps later, I ought to be able to build a longer, more sophisticated launcher, still using the same expensive parts. I'll have to be careful with my machining decisions so I preserve the ability to change the gun design without changing those plates.

* I am going to compare performance using both:
1. standing start firings with really carefully machined armatures with clever geometries and precise pre-loading (the way the pros do it)
-as well as-
2. pre-accelerated firings with really basic C armatures and a more relaxed bore fit in the same gun.

So, a lot of design decisions are getting finalized, and I've even begun to construct a preliminary model in Solidworks utilizing the parts and materials I have on hand, assuming additional parts and materials are obtained, and utilizing an integrated pre-accelerator section within the same gun housing.

I'll post a pretty CGI model pic eventually, when it looks like it's closer to the final design. Don't hold your breath.

What? Oh that? That's nothing...

2010-04-21T22:47:00.001-06:00

Yep, things are proceeding apace, as this here photo will attest:

More writing when I have more to say. Ain't I a tease?

micro-Marx addendum

2010-04-14T11:42:00.000-06:00

One of the issues I've been struggling with is the relationship between switching medium (hydrogen gas), gas pressure, electrode radius, and charge voltage.

The variables get hairy quickly, despite the work by a fellow named Paschen which makes some of it easier.

Well this morning, I found a paper ("Sparking Formulae for Very High-Voltage
Paschen Characteristics of Gases", A.E.D. Haylen, IEEE Electrical Insulation Magazine May/June 2006, vol.22, No.3) which provides the P*D curve for Hydrogen, which at 10,000 volts (stage charge voltage) is 550 Pascal-Meters.

Note that this is by definition between infinite parallel plates. (Generally, since plates with infinite surface area and no edges are inconvenient to fabricate, round plates having a diameter >> gap distance, and with the edges curved into Rogowski or Bruce profiles, are used.)

Setting aside electrode radius temporarily (because it is a variable with much less influence than the P*D product for a given voltage) I should now be able to get into the ballpark for my gap distances (which are, almost of necessity, fixed) so I can finish up the insert (mounting panel) model & drawings.

We care a lot about the stage gap distance for several reasons:

1. the gap distance has to be large enough to hold off the stage charge voltage at the operating pressure, BUT:

2. spark channel inductance is actually important, because at the high currents and high dI/dT we're talking about here, the spark channel is very small, which leads to surprisingly increased inductance per unit length of spark channel, so we want all gaps to be as small as issue #1 will allow, AND:

3. we don't want to waste any more energy (as Joule heating) than necessary in the gaps. Shorter spark channel = lower resistance.

I'll be pretty busy today and tomorrow, and unavailable for the next week or so after that, so I probably won't visit any of this again until late next week.

brief status update

2010-04-14T06:49:00.001-06:00

Work continues slowly on the mad scientist light switch. Some of the work is rework to fix mistakes, but I'm still pulling ahead slowly. That remains a priority just so I can get all of the parts off of my bench.

I'm also working simultaneously on the micro-Marx generator. I am rapidly gaining respect for the guys who designed the new breed of fast Marx generators in the 60s, 70s, and 80s, because I now realize that every time you want to change one little thing, EVERYTHING else in the system changes. Like Tesla coils, Marx generators look simple as a schematic on paper, but turn out to be surprisingly complex in actual physical design and the calculations required to determine said physical design. The best performing examples are finely tuned devices. I am aware that I'm probably smarter than the average man on the street - tho I'm no genius - but the guys who do this stuff for a living are considerably smarter than I.

I'm in the process of evaluating two physical layouts for the insert panel which will hold the spark gaps on one side and the caps on the other. The winner will offer the best tradeoff between number of stages which will fit into the available space and self-inductance. I was surprised to notice for the first time a few weeks ago that the design I am working from as a stepping-off point creates a helical current path when the bank erects. I can't help wondering whether that is deliberate or accidental. As near as I can tell, any stray or self inductance in the system is undesirable. Stray capacitance quite the opposite, as it can be used to decrease erection time significantly.

I've also changed the sphere gap size, which required an extensive redesign in SolidWorks and as I still do not have a SolidWorks platform at home I'm limited to working on that task on my lunch hour and after work hours. It's taking a while. Any readers out there feel like donating a semi-modern Windows box for me to use as a CAD (SolidWorks) platform?

I am attempting to learn how to calculate all of the stray inductances and capacitances so as to determine the device's characteristic impedance. I'd like to include such design parameters in my spreadsheet-based "Marx Calculator". Good thing I'm studying calculus this year. At least in theory. Free time seems to be at a minimum lately.

There are some physical machining tasks for the micro-Marx housing and liner which will not change no matter what happens to the innards. I can start on those whenever I have the MSLS done.

So things are happening, they just aren't happening quickly, and I haven't any photogenic progress to show off for the moment. I might be able to post something along those lines near the end of next week. I want to put up some images, notes, and calculations from the micro-marx design I am working on.

Fast Micro-Marx Generator revisited

2010-04-02T00:06:00.001-06:00

////// EDIT: pretty pictures added 4-2-2010 @ 9:08 AM //////

Eventually, this blog is going to earn me a visit from some three letter agency, I'm sure. That is, in part, why I am blogging about these fun amateur science projects so publicly. So if I disappear suddenly, please do drop an inquiry to the warden of the Thompson Correctional Center, Thompson Ilinois. I understand that's where political prisoners are going to be going soon.

But I digress, as usual.

I've been a bit ... not myself lately, and struggling with various challenges. I haven't accomplished much actual work on any projects (until recently) although I have been slowly setting up certain machining operations and preparing the way for various things to be done when I'm better rested, not stressed out, wide awake, and so forth. There will soon be parts at stake which have many hours of time in them. Fucking them up is not acceptable.

Oh yeah: I swear a lot in this blog, as I am attempting to write in a more chatty tone, and since I swear a lot in real life... there you have it.

I have tattoos also.

I should buy a motorcycle.

Anyway. Today I was obsessed with thoughts of the Fast Micro-Marx Generator, and not ready to proceed on the Mad Scientist Light Switch parts... so I did a lot of thinking when I wasn't having to use my brain to do my job, which was most of the day to be honest. Over my lunch hour, I dug up some papers by various really smart people about spark gaps and clever Marx generator techniques and peaking gaps and so forth and so on. There's a lot to think about for a guy who isn't very strong in math.

When I got home this evening I was ambitious for the first time in weeks, and I set to work. One of the assemblies that makes up the FMMG (cough, get used to it) is the housing which is, incidentally, also the current return and coaxial shield of the Marx Generator. The housing shall be constructed mostly from a length of 4" diameter copper pipe. The ends were saw-cut as I received it, and after cutting a piece close to the final length needed, the ends will need to be made square and circular preparatory to having flanges fitted to each end.

A problem arises: how to turn this pipe. Although I can chuck it in the lathe, it is soft, thin-wall copper and I don't have a way to support the other end.

The answer is a mandrel which can be supported at the far end by the tailstock center, fitted with two disks to support the tubing, one at the headstock end which supports the tube where it is clamped by the chuck. The other disk is fixed to the mandrel an inch or so inside the tail end of the pipe using, oh, three set screws. The OD of the disks are chosen to be only a few thousandths less than the ID of the pipe. Et Voila! The tube is now supported firmly, gripped by the chuck firmly and may be carefully turned. Which reminds me, copper is notorious for being a S.O.B. to machine, I must look up the tips and tricks and figure out what kind of tool I'm going to need. Yay insert tooling!

Tonight, I made the mandrel. It's a piece of thick walled steel tubing (pipe) fitted with inch or so long bushings turned to fit neatly into each end. The pipe had four quarter-inch holes drilled into it. The bushings were welded into place through these holes as well as around the flange left on the end. The the welded ends were annealed with a torch prior to turning. The ends were turned down, center holes were drilled into each end, and it may now be turned between centers. Except since I haven't got the headstock accessories required to do that, I suppose I'll fix one end to the headstock disk with a bunch more set screws. Hooray. I want the disks removable so I can use this for another tool some day.

Anyway, the mandrel proper is finished, but lo and behold, the fucker has a taper - that is, one end is larger in diameter than the other. A fairly significant taper. This means that the center of rotation of my lathe's head stock is not aligned axially with the center of my tail stock. This is annoying, but correctable, by adjusting the tail stock. I'll want to mount up some other hunk of metal rather than carving the walls of my mandrel ever thinner. Harrumph. This is annoying, but not surprising. I've never turned anything long before, so I wouldn't have noticed (although I've suspected for a long time).

After I decided I was done working in the shop, I cleaned up, came inside, had dinner, and then set to working on my Marx generator spreadsheet. I've started a section to help me do the calculations for the distributed capacitance and inductance of all the current paths, some of which are tortuous. If I do this thoroughly and correctly, I should be able to estimate the characteristic impedance of the thing. That will be important to know later.

After I built models in Solidworks, I was dismayed to find that my doorknob capacitors take up quite a lot of room and that I might not be able to fit as many stages (ten) as I'd originally hoped:

That's the current state of my Solidworks model for the Marx bank itself - what I've come to call the "insert" since this is the bit that goes inside the housing. It was nice and tidy and complete, with all of those spheres (the stage gaps) mounted to the panel with screws and a peaking gap design mounted at one end... until I discovered a few things about peaking gaps. Then I had to tear it apart and I haven't had time at work to finish changing the configuration, design, and dimensions. I don't have a machine at home on which I can run Solidworks.

At any rate, I now realize that the output peaking gap has to be a separate assembly with a different gas pressure (and different gas, most likely). Making it a separate unit frees up room inside the Marx housing for another stage. The end with the two holes and an empty space is where the peaking gap was. We are now looking at eight stages with a charge voltage of 10kV, so an erected voltage of 80kV max. Not the 100kV I'd originally been hoping for, but c'est la vie.

The peaking gap, IF it is connected directly to the output of the Marx bank, will steepen the Marx output rise time, as it provides a slight delay ensuring that all stages in the Marx have erected before connecting the output to the load. It is possible to construct special peaking gaps which go from non-conducing to full conducting in extraordinarily short times (pS regime) by using special geometries, gas mix, and pressures versus what's used in the Marx bank.

But there are other clever tricks in the pulsed power business that I am tempted to try. One of them is the intermediate storage transmission line and peaking gap. Funny things happen when you do this right, although the dimensions - especially length - can become inconvenient quickly. Here's a block diagram of how the real pulsed power boffins typically obtain ridiculous peak powers by pulse compression:

Now _if_ (and that is a big 'if') I decide to try such advanced shenanigans, I will have to maintain a constant impedance throughout the entire structure, Marx bank, intermediate store, peaking switch, etc. So it behooveth me to know what the characteristic impedance of my Marx bank is to begin with. Knowing that, I can figure out the necessary scale required to do the more advanced stuff (and see whether it's practical or something I really want to get into). Also, it will enable me to make that peaking gap switch work better, and ultimately, match the output of the pulser to some (still hypothetical) load.

The original purpose of this Marx generator was to generate a fast enough (and high enough voltage) trigger pulse for my distortion-triggered spark gap switch:

A very fast, very high voltage trigger is needed to achieve proper field distortion triggering and hopefully operate the switch in pseudospark commutation mode.

However since I acquired the spiral generator (which I've previously blogged about):

...I'm not sure I need the Marx for that purpose any more. But I do still need to build a small prototype and understand its behavior thoroughly before attempting the larger device that I plan to use for UWB impulse radiating experiments.

Thus, the new tables in the spreadsheet, and many measurements to be taken from 3D models and parts to plug into the spreadsheet formulae.

Over the weekend, I'll probably go back to working on the Mad Scientist Light Switch project.

That's how it goes,
everybody knows.

PS: Watch this space, I'll try to backfill some pictures, after I get them taken, or at least throw in some screengrabs of SolidWorks models.

learn from my mistakes

2010-03-20T10:09:00.001-06:00

I am absent-minded. Always have been, since I was a child. Some of it seems to be connected to the ADHD, and some of it is sleep- and hydration-dependent. I get a lot worse when I'm sleep-deprived or dehydrated. But I digress. Again.

A few weeks back, I needed to machine a part. The stock was centerless-ground hardened shafting, made of some high grade steel and probably as hard as a Nickolsen file. I cut off the bit I needed with an abrasive cut-off wheel mounted in my angle grinder.

I intended to detemper it with Ze Torch before machining it, since I don't have carbide cutters, nor cooling, nor all day.

Then I set it and the part it would go into aside for the night, as I was getting tired.

In the morning, I secured the part in the mill and attempted to machine it.

Carbide could have done it, although I'm sure I was running the wrong speed-n-feed for a carbide tool even if I owned one.

But my cobalt steel end mill wasn't having any of that nonsense, and now I need a new cutter:

Perhaps I'll buy a carbide end mill to replace it, unless I get sticker shock.

incremental progress on MSLS

2010-03-20T10:03:00.003-06:00

A few nights back, I felt better than I had in weeks, so I worked in the shop for several hours. Job One of late has been to finish the Mad Scientist Light Switch. The part I've been working on is a brass mounting flange to adapt a 1" copper pipe elbow to the flat steel wall of The Device.

Here is a picture of the parts nearly completed:

A is the elbow and reducer fitting. Both have been cut down to reduce overall height and also to move the Jacob's Ladder as close to the wall of The Device as is practical. They have been polished and then sprayed with acrylic to preserve the shine.

B is the nearly finished adapter flange, made from random bits of brass scrounged from my parts & materials bins. As seen here, it still requires a bushing to fit the pipe correctly.

C is a hunk of brass from which the aforementioned bushing will be machined.

~~~

Since I wrote the above several nights ago, the bushing has been finished and installed. Just for the hell of it, I decided to try heat-shrinking it on instead of soldering it, just to see if I could hold the required tolerances on my piece-of-crap lathe. As it turned out, I could. The OD of the flange was .0015 larger than the ID of the bushing. OD piece went into the freezer, then I brought it out to the garage and hit the ID piece with the heat gun to hotter than I could stand to touch it. Dropped ID of bushing onto OD of flange, ET VOILA! they are now one piece of metal.

Next I just need to continue boring out the center until it is large enough to accept fat high voltage leads, drill five or six holes for mounting the flange, and it will be done. Exceptin' that I need to make another simple steel ring flange to go inside and help stiffen the flange mounting so it doesn't flex like crazy or bend the side wall of The Device.

After that, I can either put together the little 6V power supply I need, or I can resume working on the Jacob's Ladder safety housing.

Pulsed Power Possibilities Pregnant With Promise

2010-03-18T22:37:00.001-06:00

I hope you'll forgive me a little alliteration if I promise not to do it too often.

Recently I was playing around with the Rijke Tube and, while looking around my shop for larger pipes to try, I was suddenly seized with the possibility that a couple of constructions I have sitting around - my own distortion-triggered spark gap switch and a stack of doorknob capacitors -- might well fit inside of a few combinations of insulating tubes and conducting shells, to make coaxial geometries. When I tried some, I was surprised to find that a bunch of unrelated parts happened to fit together in very useful ways, all un-planned.

What follows are just idle ideas, with no actual projects or applications in mind - yet.

First, consider the following image:

At lower left we have a stack of barium titanate (er, I think) capacitors, 6 x 500 pf @ 20kV = 3nF per section, 3 sections in series, for 1nF @ an absolute maximum of 60kV. However, operating a stack like at its rated voltage in pulsed power duty would start killing caps right away due to voltage reversal. What we want to do is derate the caps about 50% and run them at 10kV / stack for 30kV total. For those following along at home, that's 900 joules. The acrylic insulating tube came from one of six ignitron housings from The Big Pulser.

I still need suggestions for a name for that thing, by the way, if any of you three readers are feeling creative. Large scary devices ought to have names. Or perhaps a ridiculous acronym. But I digress.

At upper right, we have the distortion-triggered spark gap switch which I built. It's a close fit, but i believe there is room for a wrap of mylar or similar film with a good dielectric strength and a low dielectric constant (and er, absorption too, please).

Now, I want you to imagine a very low self inductance, fairly fast pulser with capacitive store and switch in a close-fitting coaxial housing, possibly with a non-linear pulse compression transmission line between the output of that supply and the load. I'm not sure whether I'll ever be competent to try pulse compression techniques, but the first half of that seems a slam dunk, aside from getting the trigger signal to the switch, which is something I'm going to have to think about if I ever want to try this. Clearly, the self-inductance would be ridiculously low if you did the terminations right.

In any case, that pipe is destined to become the micro-Marx generator, so if I ever want to fool with that idea, I'll need another bit of this somewhat expensive stuff.

Next, have a look at this:

That's the same switch as above, fitting into one of the ignitron housings from The Pulser. Again, there is enough room for an insulating film between switch and housing. This would be a handy geometry to apply to a parallel plate transmission line with minimal inductance used by the switch connection. Again, pure happenstance. I don't have an application in mind, I'm just thinking out loud.

Last but not least, this is just a lash-up enabling me to evaluate an idea. I've promised to build someone a pulsed magnetic field device. I want as much rise time as I can get from first a cap string and later the micro-Marx. I want all the inductance in the one turn work coil of course. This wouldn't be made on that cardboard coil form, it would be enclosed in some resilient fiberglass-reinforced epoxy, possibly a two-part urethane or silicone. A hard epoxy will shatter when the coil expands a little very very suddenly.

None of these are projects I'll be starting on any time soon, I have plenty on my plate already.

More news soon in another post.

three steps forward, two steps back

2010-03-18T13:10:00.001-06:00

Very little progress has been made on any projects due to er, "health problems".

However, I just received my first shipment of 10 used doorknob capacitors at a very reasonable price. These have slightly lower voltage and slightly higher capacitance than the ones I intended to use before, however, that might not matter much to me since A) I didn't have enough of the other caps, and the new caps will serve as well:

I've been planning to run a stage voltage of 10kV, and a peak voltage rating on the caps of 18kV shouldn't be much more of a risk than the 20kV units, but I'll have more stage capacitance.

I'm going to look at the possibility of doubling the stage capacitance, although that is probably unnecessary for my applications.

I have not tested these caps for value and dielectric withstand yet, tho I plan to do so tonight or tomorrow night. Hopefully they are okay.

If they test out okay, I plan to pick up another set of ten within a few weeks, and more after that if I am able. The seller appears to have quite a few available. These things are useful for ALL of my pulsed power projects, both current and future, so having enough for both the micro-Marx I'm working on and future pulsed power projects would be A Very Good Thing.

~~~

The Mad Scientist Light Switch for my workshop lights is on hold temporarily. I made a foolish mistake on a brass part I was turning, and I now have to spend quite a bit of time getting it corrected before I can proceed. But I shall get it done some time soon, and then I can mount the bits that hold the mini Jacob's Ladder, and after that, work toward finishing up the art deco plastic bits for same. Eventually, I might even work on the functional bits that go inside. I think need to locate another nice old style indicator lamp, tho. Time to hit J.B. Saunders again.

More in this space as it happens.

more mad scientists light switch project

2010-03-03T07:19:00.002-07:00

In an effort to get even more parts off of my workbench, I pushed a bit further on the MSLS ~~today~~ last ~~week~~ month. (I, um, sat on this post a while, through sheer absent-mindedness)

If I could have kept up that pace - just a few hours a day, maybe only two, I might have had the thing finished in time for the monthly Mad Scientists party, but my energy levels and mood have been unpredictable at best, so I couldn't make it happen. At least it was fairly photogenic by the time the Feb party came around.

One part which has been needed for some time, and which I've been thinking hard about, was a purty way to attach the pipe elbow which supports the miniature Jacob's Ladder to the side of The Device. For some time I've been assuming this would turn out to be some sort of custom-turned brass flange, and today I brought it closer to fruition:

Now what prompted me to get to where this photo is is a bit of a story:

I had in my Brass Parts drawer, a blivit that was perfect (especially after some quick mods on the lathe) for soldering into a copper pipe, but which didn't have sufficient diameter to provide good solid rigid support for the thin sheet metal I need to mount it on.

But I had in that same drawer a bit more of the .125" alpha (I think - must remember to test a scrap cutting) brass sheet sheet from which I cut the trigger plane electrode for the triggered spark gap switch (more about that in an upcoming post).

What you're looking at above is two bits of metal, joined together by a bolt, several springy washers, and a nut.

Prior to assembling that mess, I first faced the two surfaces on the lathe to ensure they are very flat, and then tinned the faced surfaces with as thin a layer of silver solder as I could manage. Once I tightened the bolt and nut, the compression of the spring washers would force the two together as soon as the solder melted, and force any excess out of the joint, leaving a very thin layer of solder indeed. I did this as an experiment to learn about making very low resistance connections between brass (later, will also try copper). I did a four-wire ohms measurement on this joint with my six-decimal-place lab grade VOM, and measured the joint at around .00015 ohms or thereabouts. That's really not all that impressive. If I were to pass 100 kA through that joint -- and I assure you I could do that fairly easily, for a very, very short period of time -- the solder joint would dissipate 1.5 million watts peak. Erp. I might need better solder.

So that bit being done now, there's another bit that will be either soldered or heat-shrunk onto the part you see here which has threads on it. The threads will be turned off first. I am sorta half-heartedly working on that now. My lathe is a piece of shit made in China in the 1970s, worse news it was abused by its previous owners, so it can be challenging to hold any sort of tolerances on it. But we'll see. If I blow past the dimension I need, I can always fall back to soldering it on.

strange days

2010-03-03T07:04:00.002-07:00

The Resident Mad Scientist has seen some mad days indeed as those who know me can attest. Recent months have had their ups and downs, and the laboratory and workshop have suffered not only neglect but sometimes - gasp - even usage for other purposes. So this entry is going to be a stream of consciousness collection of random thoughts and information tidbits and handy shop tips.

(note from later: handy shop tips not included - will add to future post. Or, you could just buy all of Guy Lautard's books)

None of my hand tendons have bothered me much for weeks. I attribute this - scientifically or not- and aside from skipping a day or two due to sheer senility - to soaking them at least 5 minutes a day in a hot pot of Yan Jing Pharmacy's "Die da Jiao" hand-merinade -- sometimes twice a day on weekends. Judy also threw in some other stuff in there - Chinese "muscle oil" and something else.

Well, either whatever was wrong with those tendons healed itself abruptly or the hand-soak did it. I don't much care which. Sure feels good. You have to use good lotion afterward tho, this soup has a lot of ethanol in it to keep various essential oils in solution, so it has a tendency to dry out the hands something fierce because of the high alcohol and light weight essential oils content. Even if you have callused hands as I do, you'll only get something even worse than dried-out, cracked skin: dried-out, cracked calluses. They're the dermal equivalent of glacial crevasses. Cracked calluses can be painful enough to be crippling. Ask me how I know this - yeah. So learn from my mistakes. If you are reading this, you probably use your hands to make your living. Take good care of them. Enough of that digression.

~~~

As a result of the above, and as I've mentioned in a previous post, I've been cautiously allowing myself to do more and more work in my shop again.

Both the shop and lab got a once-over-lightly right before the monthly Mad Scientists Party, which didn't hurt. Well okay, in the case of the lab, it was more like a major operation, but it all got done, and now I once again have room on my bench to work -- sort of.

More clearing off and putting-away and identifying and storing and cleaning-up is needed (dammit) before I'll be able to make this one damned measurement which I've become rather obsessed with, then I can move on to other things, although technically I already have, since I am already simultaneously working on more bits for the brass fitting that joins the miniature Jacob's Ladder to the side of The Device.

This whole thing is going to be rather nicely put together once it is complete... and utterly pointless. I should make it so that it is easily removable. Someone else, somewhere, may want an utterly pointless thing after I am done with it. Which utterly pointless thing will be easily adaptable as a light switch for someone else I might add...

I've been toying with the idea of adding an override switch to the box. I might have to do that for full NEC compliance anyway, and doing so will provide an ultimate override should some one-in-a-million thing go wrong (what? with one of MY designs?!?) during our duration on this planet or at least at this address. Actually, now I think on it a bit more, there's no way in hell this thing will ever comply with code, but I think I have a hack for that in case I ever get inspected and someone doesn't like it.

high vacuum

2010-02-23T11:53:00.001-07:00

I'm still working on parts for the Mad Science Light Switch.

BUT: another project of mine that I've been ~~working on~~ (talking about) for years is my high vacuum system. This is necessary for a short list of other projects and experiments I want to do including mirror coating, a 500kV linear accelerator (maybe 1MV if my buddy builds an identical Van de Graaff generator), and maybe a small electron microscope. I also want to try my hand at making my own vacuum tube at least once - a triode. IF that French guy on YouTube can make slick small ones by the dozen, I ought to be able to make a single clunky large one that works. What a shame I cannibalized that induction furnace I used to have. Would have been great for baking out the tube elements before sealing the tube.

I've left the high vacuum system sitting for a while, because it either has a leak or it has bad gauge electronics, and I couldn't tell which without an additional set of gauges connected to the system.

Well, I had the extra gauges (one thermocouple, one Bayard-Alpert ion gauge) and I had an old (vacuum tubes!) Veeco gauge controller. I needed a way to mount the gauges on the input (top of the "stack") port where the chamber base bolts on, and I needed some funky cables to connect the gauges to the controller.

Some years ago, I built an adapter blank-off plate which would accept a gauge holding flange, and which would in turn bolt onto the system. After that, I got side-tracked with other things.

Recently, I ordered nearly all of the parts and connectors I need to finish hooking up the gauge controller to the second set of gauges. I still need to order a few more minor items next pay-day.

Once everything is hooked up and (hopefully) working, then if the new gauges agree with the old, then I have a leak.

If the new gauges disagree, then I have to fix or adjust the original gauge electronics in the vacuum system.

The circuitry isn't TOO complicated, but I almost wish it were a leak. Except that chasing high vacuum leaks is no fun (I've been doing it for weeks and weeks at work), and troubleshooting electronics to the component level is something I should get good at again - I am somewhat out of practice.

I guess I'll just have to wait and see what happens.

The Mad Scientists Light Switch

2010-02-03T23:20:00.003-07:00

Good evening Mr. and Mrs. America, and all the ships at sea.

While attempting to clear my workbench in preparation to make a somewhat precise high voltage low current measurement, I realized I had more stuff than real estate. My bench was rapidly becoming eleven pounds of shit in a ten pound bag.

One of the projects that's been occupying space either on said workbench, or on the riser shelf above my workbench, or parts of both at the same time, is: "The Mad Scientist's Light Switch". Kindly imagine while re-reading that phrase that it is being shouted through a large public address system with a fair bit of reverb thrown in for good measure.

I thought for sure I'd written a piece in this very journal about it, but a quick review last night turned up nothing but a short reference to it. Perhaps I was thinking of my partial coverage of it over at my Flickr page.

So anyone reading that one little mention here must have thought I'm a bit 'round the bend. They'd not be far wrong of course, but that's entirely beside the point.

In order for me to make room on my bench, I needed to get this damned project -- I shall call it the MSLS for short -- off my bench. As it happens, the MSLS was designed from the start to mount on the wall. Of course, back at the start, it hadn't been quite so complicated. But I'm getting ahead of myself.

For starters, if I could get this thing set up on the wall temporarily, I'd have lots of bench space. So I decided to push on this a while since making a small amount of progress on it would enable any progress on measuring that damned capacitor.

Not that the capacitor has anything to do with any projects whatsoever - for the moment - but it could, and I'll have more idea of how likely that is... AFTER I AM ABLE TO TAKE THIS DAMNED MEASUREMENT.

*ahem*

So anyway, here's what this MSLS thing is, how it all got started, and er, caught a case of Creeping Featuritis.

A really long time ago (decades), an old U.S. Air Force buddy offered to give me a "great big old fashioned knife switch" - hereinafter referred to as the Giant Ass Knife Switch, or GAKS, which he had found somewhere, probably eBay. After hearing his hyperbole about this thing, I eagerly accepted his offer, although he didn't actually have it with him at the time... *cough*

Some time passed and I asked him about it. He thought about it a while, and eventually realized it was buried in a closet which was blocked with quite literally several tons of amateur SETI ground station receiving equipment in three full-height racks... I just sighed and told him to let me know when (if) he ever saw the inside of that closet again.

Several years passed.

We got together for the first time in several years, and I was shocked to hear that he had decided to dump the entire amateur SETI project, had sold all the racks of equipment, and was in the process of cleaning out the previously-blocked closet. He promised to find that switch and get it to me the next time we got together.

Several more years passed.

Did I mention at the outset that this was going to be a long story? No? Well, it is.

To make a painfully long story slightly less painful but not shorter at all, I got my hands on the thing quite a few years ago, found that it was quite as beautiful and impressive and archaic as promised, and it has been languishing around my workshop ever since, looking for a purpose. It's got three poles, a solid slate base about 5/8" thick, lots of copper and brass, nicely constructed.

It could probably handle several hundred amps closed, but there are limits to how much voltage across its contacts (or current through them) that I'd be willing to use with that switch with my bare hands on it. Now with a bit of rope and an arc flash suit, that would be a different ~~experiment~~ story.

ANYWAY, between the moment I first heard the rumor of this GAKS's existence and the moment I got it in my perspiring hands; I had bought a quite nice, but slightly smaller, Big Ass Knife Switch. We'll call it the BAKS for short.

MSLS, GAKS, BAKS. Got all that? Good.

I had the idea that it would make for a great (tho fantastically dangerous) actual light switch for the lights in my workshop. It's not for nothing that Frankenstein is always shown throwing the big switch that powers up the laboratory AFTER he has located the handle with a lantern, torch, or candle. He's also usually wearing thick rubber gloves up to at least his elbows. I used to have a pair of lineman's gloves like those until I outgrew them. They're not for show, and they're very expensive.

You can do things with GAKS that aren't exactly safe to be close to unless you are wearing good protective gear. And don't forget, a lot of the old special effects for those 1950's mad scientists movies were real working, mad science equipment like Tesla coils and huge jacob's ladders built by Ken Strickfaden. His equipment actually WAS dangerous, there wer no optical effects, the camera recorded what was actually happening on the set. At one point, Boris Karloff refused to go onto a set because he thought Strickfaden's equipment was too dangerous.

But I digress. Sorry, favorite topic.

I am actually fairly safety conscious. I've had a few close calls, and I've listened carefully to every one of them. Usually the communication medium has been pain.

So it occurred to me, I could set up the knife switch with six volts across it, hide a short circuit under one side of it so that it shorts the two poles when it is "closed". Then use the six volts to control a relay capable of operating all of the lights that I intend to put into the re-modeled workshop.

Ah, that'll b about four times as much, and more evenly distributed, than it has now. It will be bright. It needs to be. I've learned this lesson over and over again. I have a twin 40 watt fluorescent fixture under the riser shelf of my workbench. It's priceless.

So anyway, anywhere from 5 to 10 amps of modern, T5-HE daylight and warm mixed lamps with high frequency electronic ballasts (no visible flicker, probably consumer about as much energy as my old starter type fixtures do now). I'll need a good sized relay, but I probably have one lying around, no problem.

Well okay, I gotta come up with a 6 volt power supply. Needs to be well-isolated from the power line - so add an isolation transformer...

Gee, it sure would be cool if that big impressive knife switch could actually DO something, or pretend to, when you threw it.

An old friend from fandom once described to me a device which made some impressive noises when activated, and I think it was related to a light switch. I thought about this for a while.

It occurred to me that a small Jacob's ladder would be a cool addition. But you know, Jacob's ladders are cool and all, but it would get seriously annoying once you were done admiring it. I'd better add a timer relay to turn it off after five seconds or so. Amusing when you turn on the lights, and then you can forget about it.

You can see where this is going.

The design got complicated. From the very beginning, the aesthetic I had in mind was half steam-punk and half atomic era science fiction. Big ribbed insulator bushings! Round dials mounted with brass screws! Baked-on crinkled-enamel cabinetry! Heat sinks! Fins! Cooling louvers! High voltage anti-corona spheres! A hand-crafted jacob's ladder with custom fittings to allow mounting in a big copper pipe elbow, with turned brass electrode holders, brass electrodes, and art deco finned finger guards to keep curious fingers from offing their owners! Uh, an extra neon sign transformer because a single piddling little 5mA NST couldn't generate enough heat in the arc to create the hot air that causes the arc to rise.

And so on and so forth. Plus, the wires all still need to be connected to interesting looking things, so that they appear to actually do something, preferably something arcane and vaguely dangerous looking. I want a physicist specializing in high power electronics to look at this thing and say, "okay, I can see it is beautifully constructed, and probably quite dangerous... but what in the hell does it DO??"

You have the general idea. If you've had the patience to read down this far, you deserve a small reward:

This is the main control box, hereinafter known as The Device. It will contain the neon sign transformers, 6VDC supply, isolation transformer, timing relay, and some associated gewgaws. Eventually, it will also contain a solid state sound record/playback chip, a small Class B amplifier, and a small speaker, so it can make interesting sounds (think Krell Mind Amplifier turning on, or turbines spooling up).

The upper left and upper right quadrants above the meter will contain a pilot lamp and an override switch respectively, in case something goes wonky with the power supply or 6V wiring, I can still turn on the lights in the shop. The finned bits on top will be extended upward a little with some contrasting tubing (they are actually power resistors).

The mini-Jacob's ladder looks like this, so far:

It includes customized copper piping and the obvious elbow. Not yet installed is a finned art deco vent at the top that will allow warm air to escape and keep all but small conductors (held by very stupid people deserving of electrocution) away from the 3,750 VAC (to ground) inside. I've cut the parts, but still have some very careful drilling and gluing to complete that bit.

The Jacob's ladder will mount through a custom brass flange (which I am working on now) to the side of The Device.

Please note the custom fabricated shelf brackets with pulp science fiction inspired "lightening" holes. The wooden shelf itself is not yet complete - it will get some kind of edge treatment - routing at minimum, possibly a metal rail of some sort.

The conical insulators on the front panel will carry a whole 6 volts through some big impressive coiled wires to the BAKS. (The GAKS was too big to fit where I wanted to mount a light switch.) Something like this:

Note that you're looking up from roughly my 5'10" nose -- the top of my workbench and drill press pulley housing are visible at lower right and at the extreme left edge of the image is a pass-through about 7 feet in height.

The lower connections of the BAKS will be wired to yet more impressive but ultimately non-functional greebly bits.

And that's the way it is.

slowly I returned, step by step, inch by inch...

2010-01-27T08:37:00.000-07:00

Since I started being more careful with where I allow pressure to be applied to my hands (at work), pushing fluids and NSAIDs, and taking a break from my home workshop, my hands (and more to the point, the tendons therein) have finally begun to feel a bit better. I shall continue the regimen.

Meanwhile, I have begun a cautious return to my workshop, allowing myself to do light tasks such as:

1. cleaning off my workbench

2. improving the organization of my hand tools (which includes removing anything that is not a hand tool from the pegboard at the back of my workbench to make more room for actual tools)

3. setting up for a high voltage breakdown test in the capacitor from my previous post. That must be done with some care, as the voltage may have to be raised quite high to find the breakdown voltage of the capacitor. That means taking up a lot of space for safety. And that is what lead to #1.

4. cleaning and tidying up the shop in general - we're going to be hosting the February Mad Scientists Club meeting, and I want to make some progress on a few minor projects between now and then. It's time we brought back the show-n-tell that used to be common at these things. Also, the shop is a mess.

A bit on the breakdown test of the odd wax-filled capacitor: the test will be done in such a way that any short-circuit current will be limited to some microamps. I am blessed with a fair collection of ridiculous value high voltage resistors with fairly precision values (example: 990 MΩ, 0.3%) which should make that relatively trivial.

I'll measure the (DC) current directly through a 100µA meter movement. I don't believe I am likely to damage the insulation under such conditions.

I haven't a ready-to-hand DC high voltage supply above 5.5 kV, so I will have to jury-rig something out of parts. Nothing I haven't done before.

Gee whiz I wish I hadn't let that non-current-limited laser power supply transformer (16kVAC) slip through my fingers a few years ago...

mystery capacitor

2010-01-25T07:29:00.001-07:00

Some years ago, I was given an apparent high voltage capacitor. Last weekend, I dragged it out to investigate it further.

It looks very old in terms of materials and construction, and it also looks like a one-off. It might be home-made, or it might have been custom made in the 1940s, it's hard to say.

That's a six inch ruler sitting on the paper in front of it.

The case is constructed of folded and soldered sheet brass and painted with black enamel. The case has considerable taper from the vertical on all four sides. The top is a thin sheet of phenolic laminate, fastened to the lips of the case with four small brass machine screws and nuts.

As is obvious from the photo, the connections are two wide copper straps.

Inside, the case is filled with an unknown wax:

I melted and burned a small sample, and it smells like neither paraffin nor beeswax. The wax is a dark amber color.

The capacitor measures about 25 nF at 1kHz and a low test voltage. I may try again at a higher charge voltage and frequency, but I don't expect the value to change much. No doubt the wax becomes lossy at very high frequencies, but this thing doesn't look like an RF capacitor, it looks like a pulse capacitor to me.

I have no idea what its maximum working voltage might be, although the construction and dimensions suggests it can't be good for more than a couple of tens of kilovolts.

I hope to get around to setting up a test rig this week to measure its breakdown voltage safely. The general idea with that is to limit the breakdown current to such a low value that no damage can possibly be done to the dielectric. We especially want to avoid carbon tracking through the wax. No arcs allowed!

Toward that end and others, I really ought to build myself a general purpose high voltage power supply some day.

The case being tapered (ie; it has draft) suggests that the builder might have been thinking of removing the molded capacitor once cooled. Hard to know.

I will post more data on this beast as I generate it.

flux compression generators

2009-12-10T20:41:00.003-07:00

Oh how I would love to build a flux compression generator some day. Unfortunately, there aren't many practical actuation means other than explosives, and those tend to garner unwanted attention unless you have a chunk of land from inhabited places on which to set up a test range.

If only I owned my own private, preferably remote island somewhere. With fresh water of course. And extinct volcano making up most of it. And one nearly shear face which, in the right light, looks an awful lot like a skull. And an underground water passage leading to the inside of the extinct volcano...

idled, sidelined, shelved, or suspended

2009-12-01T07:54:00.002-07:00

I've managed to do several minor injuries to myself, including tendinitis in my hands, which require that I minimize the use of my hands while I heal.

This is somewhat challenging, given what I do for a living.

So I will be taking it easy in my workplace to the extent possible, but I will not work at all on my personal mad science projects for the next few weeks at a minimum.

I'll also be minimizing time spent typing on keyboards, which will reduce my posts everywhere.

Don't expect many posts for the next month.

Any significant developments, for better or worse, will probably be posted here (even if I have to dictate them) not including medical info I choose not to share with the general public.

I am a tool user with opposable thumbs... and tools

2009-11-24T14:43:00.000-07:00

By now, I should know better than to say, "I can't..."

A solution to the question of how to machine the ends of thin-walled tubing without having either a steady rest or a six-jawed chuck has been presented, nearly simultaneously, by two different friends.

Plugs.

A plug shall be machined to support the tubing (internally) in the chuck at the headstock end.

The tailstock end will be supported by a second plug, riding on an arbor which extends through both plugs and is supported on a center in the tailstock. The tail end plug will be recessed in about an inch or two from the end so I can sneak in a boring bar to turn the ID at one end. I don't need a turned ID on the other end, I'll only have to do that once.

I'll probably fashion the plugs from aluminum.

What keeps the arbor from moving toward or into the spindle, I don't know yet. Perhaps a few set screws. Something tells me I might have to try to be clever. I'll figure something out. No, I don't have a dog plate and center for the spindle.

I'm just embarrassed I didn't think of this answer on my own. Of course, it would be nice to have more and better accessories for the lathe so I don't have to keep making custom tooling to hold my parts...

subcontracting and machine tools

2009-11-19T09:26:00.001-07:00

Well, it seems I won't be able to use my lathe or my steady rest to machine the housing and liner for the micro-Marx after all. The work won't fit in my steady rest, and I can't do the work without using a steady rest.

I am in the process of modifying my steady rest so that it will take slightly larger diameter work, but there is only so far I can go before the functionality of the rest is affected. I think I can get about 1/4" larger diameter from it.

The mechanical design for the mico-Marx is approaching the point where I will need the exact final dimensions of certain parts. For example, the length of the liner - which is shorter than the length of 4" copper pipe which will form its housing - constrains the maximum length of the Marx generator, which must have room for N stages of capacitors, spark gaps, and connection straps.

I will talk later today to a friend who owns a small machine shop business, and who allows me to occasionally bribe him with good beer to use his better quality facilities. Perhaps one of his lathes can handle these things.

I've finally admitted to myself that buying the lathe I did was a mistake. Given the information that I and my adviser had at the time, it seemed like a decent deal, but since then, I've found out just how much easier it is to work with a quality machine tool.

Even the Chinese lathes (which mine is) that are being sold now are 10X higher quality than the one I have, which appears to have been made in the 1970s.

There are companies (I've read that Grizzly Industrial is one) who have good working relationships with a factory in Taiwan, and that after receiving the machines in the states, they do a lot of finishing, tweaking, and adjusting in large regional plants. According to some, this results in a pretty good quality machine for much less money than you would pay for something made in the USA or Europe. They also add a premium over the apparently identical machines sold by Harbor Freight, Enco, MSC, and McMaster-Carr - their prices are about 5% - 10% higher for identical machines.

I think when I can afford it (which may be a very long time from now) I will buy a Grizzly lathe (complete with all the accessories mine was missing, dammit) and sell the one I have.

In the mean time, I will muddle along with the lathe I have, making occasional upgrades where possible (as I have already begun doing) so that it will fetch a reasonable price when I sell it.

Besides, I really need a band saw and a 3-in-1 metalworking machine much more than I need a higher quality lathe, and I can get by with the lathe I have for quite a few years.

tooling & fixturing

2009-11-18T10:21:00.000-07:00

It would seem I am going to need some new tooling to clean up and machine the 4" copper pipe ends (forms the coaxial housing/current return) and the 3.5" phenolic liner.

I need a roller steady rest. I have a steady rest for my lathe, but it is of the old-fashioned brass-tipped sort that expects you to use it on a turned and semi-polished surface of the work, with grease. I am not convinced that's practical on either piece I need to work on. But I might be able to make some roller bearing tips to fit my existing steady-rest.

I think I can make it. I am highly unlikely to find one that already fits my lathe. After all, there don't seem to be repair parts for it available from anyone. I suppose I could try contacting the SEIG factory, where I suspect it was made... in the '70s or '80s...

Also, I really, really wish I had a 4-jaw independent chuck, a 6-jaw scroll chuck (which is specifically intended for chucking thin walled tubing) and oh, what the hell; a Lee Valley shop apron, a Parlec swivel base for my mill vice, a pair of vee blocks, a hot tub...

thoughts over the weekend

2009-11-16T08:48:00.000-07:00

For the past several days, nothing of interest happened, so naturally I'm going to write about that. This is just a summary of the thoughts I've been thinking over the last few days on these topics.

Although I am budget-bound, there are a few tasks I can push on my projects with the materials and means I have on hand.

There are plenty of things which need doing on the Mad Scientist Light Switch project.

The high powered spark gap switch project is still waiting for a trigger generator, which may or may not end up being the VIG I've been blogging about recently.

I am still going to forge ahead with the micro-Marx generator, because I realize now that it will be very useful to see which device will do a better job of triggering my switch.

Now I really wish I had been able to make the switch housing out of something transparent such as cast polycarbonate so I could (perhaps) capture on a camera whether or not the thing is achieving multichannel operation. I may yet have to attempt making a new housing, assuming I can ever find a chunk of the appropriate size. There are ways to test for multichannel capability (mostly a consequence of very high dV/dT), such as applying the trigger pulse to a fine wire stretched parallel over a grounded plane electrode, and photographing the streamers or sparks which develop. You can't see it with the naked eye. It might not even be possible to photograph the phenomenon with any camera I currently have access to. Yet another thing I need to look into. How were those multichannel test photographs taken?

Both trigger pulse generators will also be tested (eventually) with the railgap switch I received from Jon Singer, lo these many moons gone by. I will start by attempting to use it in its original configuration, which appears to have been a multi-point Trigatron configuration, although we don't really know. There doesn't seem to be much documentation for that ancient device available from anyone anywhere, and what little there is, I already have. To do that, I would have to construct a 4-way power divider to split the trigger pulse into four lines for the four trigger ports on that switch. Upon brief reflection, that appears to be somewhat challenging, but I know it can be done, because I've seen devices in white papers which did precisely that.

Later, I will almost certainly attempt to modify that switch to utilize field distortion triggering with a single linear trigger electrode (most likely either a knife edge or a fine tungsten wire). But I have lots to do and learn before I begin messing about with that switch.

a digression on tools, part of a series

2009-11-15T11:50:00.009-07:00

Any good mad scientist requires tools. And not just any tools, but good tools, and quite often, some very special purpose tools.

Now since I am in a generous frame of mind -- and not in any case the sort of hermit mad scientist who hides in a remote castle with only a misshapen lab assistant for company, but one who actively shares mad science information with other mad scientists -- ahem, a generous frame of mind I say, I wish to impart a small bit of shop wisdom to my many (three?) readers. When this sort of thing happens, it will be beginner's stuff - I ain't no Guy Lautard.

Today's topic: Zen & The Art of Oilcan Appreciation: An Inquiry Into Value

Almost any self-respecting mad scientist keeps machinery about. Preferably machinery constructed using forbidden knowledge and dark arts, but machinery nevertheless. And all machinery has moving parts, rather by definition. And as a general rule, moving parts require lubrication.

It would be nice if there were one lubricant - whale oil, dolphin cartilage extract, babies tears, something exotic - that would work for all applications, but alas the laws of physics, no matter how far we attempt to bend them in our laboratories, eventually win out.

For moderate bearing loads and high bearing speeds, we use a thin, very slippery oil, for example: spindle oil on a mill. For high bearing loads and low bearing speeds (ie; vehicle wheels) we use heavy grease.

We'll talk about grease some other time. As I said, we're talking about oil cans today.

Now, as with all things in life -- transistors, lawyers, lawn mowers, poker chip sets -- there are very bad ones, bad ones, mediocre ones, good ones, and very good ones. The curve of quantity over quality is a gaussian distribution, nyuck-nyuck. Writer Theodore Sturgeon generalized this as "Sturgeon's Law" - ie; "Ninety percent of everything is crap." This is simply a cynical way of restating one of the things we all learned from reading "Zen & The Art of Motorcycle Maintenance": that quality is rarely found and often not even noticed or appreciated.

And now a digression within a digression: if you have no idea what I am talking about - if you have not read "Zen & The Art Of Motorcycle Maintenance: An Inquiry Into Values" by Robert M. Pirsig, step away from the computer, go find a copy, read it, and then come back to this. This article will wait.

As I was saying, because of the law of supply and demand, quality costs more, because as we all know - most people would rather purchase crap because it's handy or cheap, so that's what most companies make: crap.

Now you might reasonably think that the lowly oil can would hardly bear mentioning, as shop tools go. Why am I not instead expounding upon something more interesting and cutting edge such as the virtues of cobalt alloy end mills vs. TiN coated High Speed Steel end mills?

Get it? "Cutting edge"?

Ahem.

Look, if you have a "laboratory" of some sort, you may not possess much machinery. But if you also own a "workshop" - a place where objects are transformed into, and intermingled with, other objects - the hard way - then you will find yourself oiling all sorts of mechanisms on a semi-chaotic, semi-periodic basis.

I have a workshop.

Now over the years, I have owned several oil cans with several different lubricants in them.

Since I began to obtain machine tools, I started to think about using exactly the right recommended lubricants for various bits so as to prolong the tool for the next generation. Good machine tools, (of which I own one, ahem) can last for generations and can be handed down for a long time if you take care of them .

As I was saying, over the years, I have owned several oil cans. Most of them have sucked, in that they didn't pump oil ONLY to the end of the spout, they leaked in various places and soon became covered in a thin film of whatever they contained mixed with tiny bits of everything else you find around a shop.

Except for one oil can, which I shall tell you about now.

The Goldenrod brand of oil cans, manufactured by the Dutton-Lainson Company in Hastings, Nebraska (68902) is of superior quality to other oil cans I have owned. Here are the reasons:

1) it doesn't leak. It ~~puts~~ pumps the ~~lotion~~ oil on ~~it's skin~~ the job where it's wanted ~~or it gets the hose - stop it goddammit! they'll hear you!~~ without leaking anywhere else. I attribute this to the gobs of what appear to be a white epoxy-like substance joining the nozzle to the flexible hose (which hasn't leaked either, yet) and said hose to the threaded bit that goes onto the pump output. The latter bit is a removable threaded fitting (if the spout gets damaged, you can replace it, other cans have this feature too) and said fitting has a gasket between its edges and the pump. It doesn't leak there.

2) The label is classy and seems impervious to oil. I realize this is a pretty cutting edge idea here: a label for an oil can... that is, now get this: impervious... to oil! Isn't that cool? It's some metallized mylar which hasn't shown any sign of changing its character since I bought the thing a few years ago. Every other oil can I've ever had came with a cheap paper label stuck on with something that is probably best not thought about, and if I did not peal it off after bringing the thing home, it would soon become saturated with oil - mirabile dictu! - and slough off, on my hands, my gloves, etc. This is good marketing. If your label stays on, it not only impresses people, it continues to advertise your product while keeping it looking classy. If your label turns to gooey, oily sludge... especially if the logo or name can still just quite be made out... not so much.

3) It is better constructed than others. Upon examining seams, it becomes apparent that the Goldenrod can is fabricated from a thicker gauge of sheet metal than the other oilers I own. Since this one is a larger capacity (10 or 12 ounces, I believe) it comes with a stout handle. The outside of the flexible spout remains dry. The moving bits are made from MUCH thicker sheet metal than the moving bits on the other cans.

It occurred to me just now to go hit the manufacturer's website, which states that the company has been in business since 1886:

It seems they make some cheaper oilcans whch look suspiciously similar to the ones I own but don't like. Hmm. In any case, I recommend purchasing their "Industrial" line, which appears to be the one I own and love.

Here is a picture of two oil cans I own:

I've had the one on the right for a couple of years. I've been keeping 30W motor oil in it. I'm going to move its contents into the anonymous can on the left, because I expect it (the black one) to leak, and a thicker oil will leak less, or or at least more slowly. Then I will fill the Goldenrod can with spindle oil for the mill. There is a little capped oil cup on the mill for this purpose which must be filled occasionally. This Goldenrod can is perfect for that job. It even has a special tip which is specifically designed for lifting up the dust caps on oil caps. Did you know what those funny shaped oil can spout tips are for?

The one on the left is brand new, and it is label-less because I expended a good thirty minutes, several ounces of mineral spirits, and at least a pound of "elbow grease" getting the damned paper labels off. Except for the style and construction of the flexible spout, it is identical to another leaky oil can I've had for about five years.

Unfortunately, the last time I found myself in a hardware store, I remembered that I needed a new oil can for spindle oil. I was overtaken by the imp of the perverse and I bought the oil can on the left. Had I thought about it for a moment, I would have remembered that I wanted more oil cans like the Goldenrod I already have. (they come in a variety of shapes and sizes) But I was a well-trained consumer, and bought the expedient thing without thinking much about it. Until later. And thinking about it prompted this blog entry, which is gone on altogether too long already, so I'll keep the rest of this short. I fully expect the anonymous black can on the left to leak just as badly as its twin brother - another can made in a strikingly similar way. If it does, I expect I will replace them both with Goldenrod cans. You see, I want you dear readers, to learn from my mistakes.

So if you're in the market for an oil can or three, Gomez recommends "Goldenrod" brand oil cans by the Dutton-Lainson Company!

I have not been compensated for this recommendation in any way by anyone. No one at Dutton-Lainson knows I'm going to write this.

_{But you know, if you guys at Dutton-Lainson do end up reading this, and you think I'm a great guy and should be rewarded for pimping your products, you're totally welcome to send me some free stuff.}

I'd like to talk a bit more about lubrication in general. The two chief complaints one hears from visitors to a metal-working shop is A) how filthy it is (especially true of welding areas or places where grinding or sanding are done; and B) how every surface in the place seems covered in oil.

Hint: it's supposed to be; oily, that is. The dirtiness can be contained, managed, periodically reduced, but never eliminated.

But all those lumps of cast iron and machined steel are made from alloys that will rust if not painted, and wear if not lubricated - often on the same surfaces. Wear surfaces such as bed ways are of necessity unpainted and made of alloys which can rust. So it's good practice to run moving bits of of machinery through their entire ranges of motion right after lubricating, to distribute that nice film of protective oil over the ways, v-grooves, etc.

It takes surprisingly little time to attend to every lubrication point on a machine, but it takes only a few hours of running time after lubrication has run low between two surfaces to ruin those bearings forever.

So if it moves, oil it.
If it doesn't move, paint it.

VIG / spiral generator update

2009-11-12T21:14:00.001-07:00

Since my last post, I have learned that there are two likely failure modes for this spiral generator in particular:

1) the dielectric of the striplines could have failed (ie; puncture).

2) the krytron used to switch the striplines could have worn out.

I have just tested the dielectric by the relatively simple expedient of applying the specified charge voltage and measuring the current.

The unit is specified to operate between 5kV and 7kV. I have on hand a small HVDC supply nominally 5.5 kV @ 0.5 mA, however it develops 6.6kV across an open circuit.

The unit charged up to 6.6kV in a few seconds, and the current at 6.6kV was 0.1 mA, which is exactly what it is supposed to be.
This is good news.

The only way I can think to test the krytron is to actually try firing this thing. According to the manufacturer, that requires a +250V pulse with a 10nS rise time. That's a darned fast pulse, but it should be do-able with a small, fast pulse capacitor, a FET, and some tight construction geometry.

10nS equates to 100 MHz, so presumably I can use ordinary RG58 cable for this, but I wonder whether there is something better. I was once given a rule of thumb by an RF engineer regarding preservation of pulse shapes and signals: if the shape of the signal is important (ie; single- shot rise times, square waves, etc) use cable and connectors rated for at least 2X the maximum frequency of interest and preferably 10X!

The trigger input connector is a BNC. 50Ω BNCs are good to 2GHz or so, and RG-58 (foam dielectric) is good to about half that, depending on how much attenuation one is willing to endure. Looks to me like I just need to figure out how to generate that pulse.

Then there's the minor issue of how to observe the output. The unit comes with a 1000:1 output monitor with 50Ω output impedance, which is tres convenient, but how well my storage oscilloscope will capture this remains to be seen. My scope's bandwidth is only 500 MHz, and for now, I don't even have a sufficiently fast vertical plugin for said scope, but I think I can borrow one from a friend.

Another interesting fact I learned is that the generator has only 1,500 ohms of output shunt resistance (measured across the output). This is the manufacturer's spec. The resistance looking into the charging connector - at 500V, the maximum voltage my megohmeter will deliver, is 135 MΩ. I suspect the "keep-alive" supply for the internal krytron is derived from the charge voltage through a resistive divider.

And finally, I'm going to start referring to these things as Vector Inversion Generators, because it seems that more pulsed power researchers use that term than "spiral generator". In any case, "VIG" is faster to type.

Aw, what a cute widdle pulser you are!

2009-11-05T22:00:00.000-07:00

My new toy from eBay arrived today. Click the subject for a link (previously provided in an earlier post).

I was surprised to see several things:

1) The casing is made from PVC glued together with PVC cement. I suppose it's serviceable, but it's cheaper construction than I had expected, mostly based on other products I've seen produced in the pulsed power industry.

2) It's much lighter than I had expected. It is immediately apparent that the interior is not filled with oil, as I had suspected it might be.

Bear with me while I think out loud. The addition of oil (and the accompanying and mandatory evacuation so as to eliminate all bubbles) into a laminated structure like a high voltage pulse capacitor or a spiral generator could do several things, some good, some maybe not so good. To start with, it displaces all air. It decreases small variations in voltage field intensity which tend to encourage dielectric breakdown. It increases capacitance. I'm not sure whether that's good or bad. Do we care about the capacitance, speed-wise? After all, this thing is a transmission line, and as such its inductance ought to at least partially cancel out its capacitance, shouldn't it? I need to think on that. Another thing the addition of oil is likely to do is raise the loss tangent which, for transmission lines propagating nanosecond-regime pulses, is probably a bad thing.

The most common high voltage insulating oils for instruments (as opposed to the power industry) would be mineral oil and castor oil, maybe a few synthetics.

One might also use sulfur hexafluoride gas, but I would not bet lunch that there is any SF₆ inside this thing.

I note with pleasure, that the unit comes with a 1000:1 output monitor (50 Ω expected impedance).

I cannot recall whether I previously mentioned that the charge input is via MHV connector (as opposed to the preferred SHV). Boo! Well, at least they're less expensive. I shall go shopping for one or two now.

Jon

2009-11-04T21:59:00.000-07:00

I know this very smart guy who is into all kinds of things, but since this here is a mad science blog, and we mostly talk about pulsed power and high voltage and that sort of thing, I shall mention Jon in that context: he makes lasers. From scratch. Very fast lasers. Nitrogen lasers. With very high peak powers.

The nitrogen lasers' output is short wave UV. Powerful pulses of coherent short wave UV can be used to "pump" other lasers using organic dyes for the lasing medium.

If this sounds interesting, go check out his blogish sort of thing:
http://www.jossresearch.org/tjiirrs/005c2.html#FirstLight

He's into REALLY cool "mud pies" too. Seriously, check out his pottery work.

Spiral generators

2009-11-02T07:44:00.004-07:00

I haven't posted anything in a while for a good reason: the micro-Marx generator has been on hold while I waited to see whether I won a spiral generator which recently popped up on eBay. "What the hell is a spiral generator?" I hear some of you say. Please visualize me putting on my pedant's hat while I throw together a description cribbed from a patent.

A spiral generator is an interesting fast pulse generator which works in a way similar to a Marx generator: a number of opposing electric vectors (put simply, directions of electric charge) are suddenly rearranged to be in series addition.

As near as I can tell, spiral generators were first reported by Fitch and Howell (PROC. IEEE, Volume 111, No. 4, pages 849-855, 1964). The earliest patent on a spiral generator that I've found (1967) was assigned to a fellow who was working for Tobe Deutschmann Labs at the time.

I'll try to describe how these things are constructed. First, imagine a rolled-up strip transmission line (stripline) with an additional layer of insulation between turns. Each of the N turns of the spiral consists of both an "active" and a "passive" stripline. The "active" stripline is the one you have before you rolled the whole thing up into a coil (a spiral seen from the end, thus the name). The "passive" stripline is formed by the inner conductor of one turn and the outer conductor of the next turn, separated by that additional insulation layer. This image should help make this clear:

The shaded gap between the inner (A) and outer (B) conductor strips represents the active stripline formed by the two strips, and the unshaded gap represents the passive stripline.

The outer end of the active line is connected to a switch (the rectangle with two semicircles at the upper right). Also shown at upper right are the biasing resistors and a trigger pulse isolating capacitor associated with switch operation. That's a distortion-triggered spark gap switch, for which there is no standard electronic symbol.

The remaining ends of both lines are open. The output terminals are normally connected with the reference or ground terminal being connected to the outer active stripline and the high voltage output terminal (C) being connected to the inner passive stripline. Note that in this rather misleading drawing (which I stole from an old patent) the connection of the outer stripline to the reference terminal is not shown. In fact, this drawing makes it look as if the active stripline is shorted to itself at the center, which is NOT the case in a real device. No, I don't know why this drawing looks that way. It does not agree with the patent description. I guess the patent examiner was asleep at the wheel when this patent was awarded. I couldn't find a better image, and haven't the time to draw one. If I do, I'll replace this one and maybe edit the post for clarity.

Initially, a DC charging voltage Vc is applied to the active stripline causing both striplines to be charged to the voltage Vc with the electric field vectors of the active lines oriented in one direction and the electric field vectors of the passive lines oriented in the opposite direction. In other words, before the switch closes, the charge across the insulating layers A-->B points in one direction (toward the outside, let's say, just for example) and the charge across the insulating layer B-->A points in the other direction (toward the inside).

The generation of the high voltage output pulse is caused by the alignment of the active line vectors with the passive line vectors. When the switch closes, a wave travels down the active line, reflects off the open circuit at the inner end, and travels back to the switch, reversing the polarity of the active line relative to its initial charge. Now all the charges point in the same direction, causing (for a very, very brief period of time) the voltage between the active and passive lines to add, or double.

The voltage across each active and passive line pair is then ideally (assuming perfect materials and construction and thus no losses) -2V_c. If there are N pairs of active/passive lines then the total output voltage ideally is given by the equation:

-2NV_c.

The time for the spiral generator to erect is given by the equation:

t = 2πND/v

where N is the number of turns, D is the mean diameter of the spiral and v is the propagation velocity of electricity through the stripline (aka local speed of light). Note that the speed of electrical propagation through materials is always slower than the speed of light in a vacuum, although said slower speed can be maximized by the use of transmission lines. A stripline is a form of transmission line. That "wave" I mentioned is moving at the propagation velocity

-=o=-

Now then, I just bought a very fast high voltage spiral generator from eBay. The manufacturer doesn't say that's what it is, but a white paper I've got which mentions this specific product refers to it as a "commercial spiral generator". The discussion in the white paper, the connections to the device, and its physical shape assure me that's what it is inside. Not that I care all that much, I'm more interested in what it does.

Now as far as I know, all spiral generators require a very fast switch to initiate the inversion-inducing pulse. Yet this particular device requires a pulse of only 250 V to fire, which doesn't sound like the trigger pulse for a spark gap switch. I suppose it could contain a solid state switch. Spark gap switches have finite lifetimes and usually require periodic maintenance to reach their rated life, but there doesn't appear to be such a device attached to this unit. Since the case is sealed, I'm somewhat skeptical that it contains a spark gap switch inside.

This makes me curious about what the guts of this device look like, but I'm unlikely to ever know since PAE seems to have built it from phenolic laminate epoxied together. By the way, this is one of the few exceptions to my usual hatred of products which cannot be opened or serviced. These things cannot be serviced in any case, and in all other respects, PAE's design is entirely reasonable given the voltages involved and the requirements of functionality.

And finally, we get back to why the micro-Marx generator has been on hold. Its primary purpose in life was to trigger the distortion-triggered spark gap switch I built. This item, assuming it is in working order (and I have convinced myself that it most likely is) will trigger said switch better than anything I am capable of constructing. It is faster than any Marx generator I'm likely to build, and has adequate output voltage. The Marx would have had 2X the output voltage of the spiral generator (probably desirable) but would have a much slower rise time (undesirable).

I find it very amusing that the patent I took the above image from uses a distortion-triggered spark gap switch to fire the spiral generator, when spiral generators are often uses to trigger large spark gap switches.

I will probably go ahead and build the Micro-Marx eventually, because I intend to build a larger one some day and I need to learn the practical construction issues of these things at a smaller scale before I attempt the big one.

But for now, the need for the small Marx is much less urgent than it was, unless this new toy turns out to be an $80 paperweight. Of course, if it's in working condition, it'll be the steal of the decade for me.

(note to self: call up PAE and ask 'em how much these things sell for and also what the stored energy is)

EDIT: they cost about $2,600, and have 40nF of capacitance, which equates to about half a joule of stored energy. As pulsed high voltage power supplies go, that's wimpy, but as trigger generators go, that's plenty.

I expect my next post will concern my testing of the PT-55N after it arrives. How the hell am I going to trigger the thing? It wants a trigger input of not less than 250V with a maximum rise time of 10nS (!) and a minimum pulse length of 150nS. The voltage is no big deal, the pulse length is no big deal, but that rise time might be tricky. Perhaps a fast FET, a low-inductance capacitor and some careful construction geometry will do the trick.

-=0=-

I am also still working (albeit very slowly) on the "mad scientist light switch project".

raw materials

2009-10-21T20:13:00.009-06:00

This is where we begin. This is just a quick shot of some of the materials which will (er, might) be used in the fast micro-Marx.

The design is still in flux as I write this. The dimensions of the raw materials I happened to find (at a price I could afford, sort of) constrain parts count depending on the parts. But I have two types of capacitors I could use for the first implementation of this thing.

The next time I take a picture like this, those yellow disks (ceramic capacitors) might be replaced by green cylinders (different ceramic capacitors). The four inch copper pipe will become the coaxial shield (or current return). It also serves as the pressure housing - up to perhaps 200 PSI if I do all my joints right. (use good silver solder and lots of surface area). Unfortunately, terrifically fast rise times are achieved by higher pressures and higher E/d in the gaps than I can manage, due the pressure limit of the pipe. Four inch drawn copper DWV pipe has a rated working pressure of only 257 PSI at 100ºF. Although working pressures from industry standards include safety factors (typically 2X - 2.5X to burst pressure) of their own, I prefer to design on the conservative side rather than on the hairy edge. No unintentional bombs, thank you very much. I've seen what kind of damage energy stored by compressed air can do.

The long sheet to the right is the same ≈.25 phenolic-linen laminate. This will be used to fabricate the support plate onto which the spark gaps, capacitors, and resistors will be mounted. That long, narrow plate will slide into a liner, made from the brown tubing, which is ≈.25 wall, 3.5" ID phenolic-paper laminate. This stuff will hold off 180 kV at least.

One end of the liner will be closed off with a turned plug, permanently epoxied into place, and hopefully degased/debubbled in a vacuum chamber, if it will fit in my bell jar. Must remember to check dimensions of bell jar vs. estimated housing length.

Said plug to be made from the 0.5" material from the big square at lower left.

Not shown is the material - to be determined - which will make up a long narrow insulator passing through the end cap and allowing the dielectric of a cable to enter the housing and plug into the load side of the peaking gap, mounted on the marx panel. Clear as mud?

It'll be clearer once I put up some drawings, after I start to figure things out, and get the drawings to match the real world objects at hand.

The other odds and ends are bushings/spacers for the switch spheres, the spheres themselves (which are likely to get larger) and other miscellanea. There's a box of screws somewhere on my messy bench that belongs with that lot.

Work bench needs cleaning, that's why you're seeing this on a cart.

More as it happens.

EM weapons

2009-10-17T21:59:00.004-06:00

Interested in EM weapons? Enjoyed reading the GBPPR archives?

Guess what? Almost everything (including built projects) in those pages was done wrong. Quite a lot of the theory is wrong. Some of the devices just aren't reliably effective (ie; ANY device built from a microwave oven magnetron). Some of them present more of a danger to the user than the target.

I just re-read -- for the first time in maybe ten years -- Winn Schwartau's classic piece (ch.10 of "Information Warfare") on "HERF Guns & EMP/T Bombs". There's a lot of good stuff in there, but nowadays, I have the background and knowledge to see that there is more sensationalism than solid science in the writing. He got some important theory wrong, and missed a lot of important details.

EDIT, A DAY LATER: it just occurred to me that that might have been deliberate. It's not my intent to criticize his work, I just want to point out to the budding mad scientists & junior hackers out there who might be researching this subject that what's in that document isn't gold-plated. I hope to meet him some day, he seems like an interesting and smart guy.

Besides, he managed to get lots of corporate wigs excited about information security at a time when most people other than hackers weren't thinking about it much, promptly founded an information security company just as the furor peaked, and made bank. How many information security consulting companies have I started up? Right then, moving on...

David Schriner's TED using a ground plane horn and built into a VW microbus is probably the best "amateur" effort I've seen to date, and I haven't been able to find out a whole lot about the pulser he used. It looks crude. I don't think it's an UWB system, which means it is MUCH less effective against partially shielded electronic commodities such as personal computers. If there is one thing I have learned from my EMC testing experiences (and from talking to the engineers who are experts in RF susceptibility) it is that you can almost never predict what frequencies will get inside a partially shielded enclosure such as a car, a mobile phone, a laptop, or even say, a 5ESS / 5E-XC switch. (Yes, some COs are shielded, but not as many as you'd think. According to one engineer I know who worked in the telco industry, a great many COs, and the racks inside them, are not NEBS-compliant.

The frequencies to which a particular piece of gear is susceptible depend on many factors - the various combinations of slot widths (long gaps in the case) holes around connectors, the lengths of the various "antennas" connected to the equipment (power cord or charger, network cable, USB, etc) and so on. The orientation of gaps - horizontal or vertical- also dramatically affect susceptibility if the radiation is polarized, as it usually will be. (by the way, it is difficult tho not impossible to build a circularly polarized EM-weapon, but it generally involves levels of sophistication - such as EFCGs - that the amateur is unlikely to have access to.

Impulse plane-wave impulses launched from sophisticated antennas driven by a UWB pulsers appear to offer the best chance of "getting into" target equipment. And lo, that's exactly where the AFWL is putting a great deal of their EM-weapons money.
HPM is almost dead because it is too narrow-band.

If I ever dabble in this sort of thing, I will probably try a ground plane TEM horn like Schriner's device, because it's easier than a proper Impulse Radiating Antenna (a relatively new invention, with most of the research being driven by a single company/man working with the people at Sandia & AFWL). Those things look just barely possible, but rather unpleasant, to build by even a well-funded amateur, whereas TEM ground plane horns are almost as directive and only slightly less broadband.

kind regards,
Tyler Durden

PS: your helium-neon laser still doesn't count as a HERF gun, kid.

(thanks for that laugh, Tim, I'll never forget it)

2009-10-17T19:11:00.002-06:00

It occurred to me that export of very fast, very high current, high voltage switching devices (like the one I am currently building) is ITAR controlled, as these are (at least in theory) "dual-use" items.

So all I gotta do is sell it to someone who lives outside the USA, and in fairly short order, Federal Marshals (or FBI? I don't know who acts as the actual law enforcement hand at the end of the DTCC's long arm of the law).

So, ah, yeah... let's not.

(not that I ever had any intention of parting with something I've put so much time and effort into)

"37. Triggered spark-gaps having an anode delay time of 15 µs or less and rated for a peak current of 500 A or more

"38. Modules or assemblies with a fast switching function having anode peak voltage rating greater than 2 kV; anode peak current rating of 500 A or more and turn-on time of 1 µs or less"

don't think of it as spam...

2009-10-17T19:09:00.000-06:00

think of it as supporting an underemployed mad scientist's crazy hobbies.

FOR SALE:

Customized StorCase DataSilo four-bay SCSI (U320) RAID/JBOD desktop mini-tower, dual cooling fans, reasonably quiet (but not silent).

Fitted with U320 CU-68 (AKA "VHDCI") type connectors. It is possible to change these. I might even have some alternative connector plates and internal cables which I would throw in for free if you asked nicely. I would have to dig for those.

Contains four StorCase DE100 (black) metal removable disk drive carriers + frames, with green LED SCSI channel indicators and key locks. I personally think these are some of the sexiest and well-made removable disk carriers any company has ever made. Which is why I bought them.

Each carrier comes with an U160 or U320 73GB 10K RPM Seagate Cheetah (or equivalent , I think one drive might be a Fujitsu) disk. Of course, you're free to put larger disks in. I haven't used this system in a while.

I also have spare carriers and I think some matching frames as well. Possibly some complete frame/sled kits, I'll have to look. Those are still fetching $40+ each on eBay, so they won't be given away for free. But if you just want a single 5.25" helf-height removable disk unit for your existing PC, drop me a line.

The DataSilo case has been painted and sealed with gray/granite Fleckstone paint. Picture on request. Looks very handsome with the black disk trays, IMO.

Will come with one high quality cable, twisted-pair (inside round jacket), solid metal connector housings, U320-rated, CU-68 (both ends) cable; and one U320 active LVD terminator, also solid metal housing.

This storage system is in perfect working order. I will provide StorCase documentation files (PDF) BUT NO AFTER-PURCHASE TECH SUPPORT. Know how to use and configure SCSI disks and disk arrays before you purchase this.

No reasonable offer refused.

The ATLAS-1 EMP simulator, AKA "The Trestle"

2009-10-16T13:07:00.001-06:00

"ATLAS" in this case stands for Air Force Weapons Lab Transmission-Line Aircraft Simulator #1

If you've ever wanted juicy details on this now-defunct Cold War program, here is an amazing video with a complete history of the project. Although I've read half a dozen white papers that discuss various elements of this system, I've never seen this much clearly-presented information about the program in one place before. The video is not particularly technical so don't be put off if you're merely curious and not an electromagnetics techno-geek.

http://www.ece.unm.edu/summa/notes/trestle_movie.html

Here are some titillating techno-facts about The Trestle:

* $60M in 1970-1980 dollars

* the simulator envelope (bounded by aerial wires suspended between tall poles) is 1,300 feet long from ground wedge to terminator tower

* 6.4 million board-feet of glue-laminated larch and fir

* 60,000 fiberglass bolt fastenings

* 600 foot wide bowl in the ground, 120 feet deep, excavated from an existing arroyo

* 185 foot high termination tower

* the big obvious wedge (a tapered "ground plane" which forms part of the simulator transmission line) is 240 feet high

* twin pulsers capable of generating single-shot pulses of 8 MV amplitude (4MV each, opposite polarity) with nanosecond-regime rise times and pulse lengths.

I would give -- gosh, I don't know where my limits might be -- body parts probably; to get a tour of that facility before they tear it down. Inactive since 1990, I'm almost certain they intend to tear it down soon, or they wouldn't have gone to all the trouble of building the high accuracy 3D models for the Historical American Engineering Record, or this video. :-(

riding on the shoulders of giants

2009-10-16T11:31:00.001-06:00

Or to flog the metaphor a bit further: tugging on the coat-tails of giants.

I want face time with the heroes of pulsed power, I'm either gonna have to kidnap a bunch of scientists, or somehow get into an IEEE Pulsed Power Conference. These a few of the luminaries:

David Platts
Carl E. Baum
Everette Farr
J.C. Martin
Ihor Vitkovitsky
John Pappas

Because most of the pioneering work was done during (and because of) The Cold War, some of those names won't be around much longer.

Of course, as soon as I revealed to any of them that I can't do real math I expect they'd lose interest in talking to me rather quickly. Everything these guys write is full of calculus. I really need to get on that task. Maybe I can get the time to do it by not washing dishes or putting away clothes or cleaning up after the cats for a few months. (j/k)

Dr. Baum especially is someone who should receive more recognition for his work. He's scary smart, and spent most of his career working on EMP simulation and other cutting-edge pulsed power work relating to nuclear weapons, nuclear weapon effects simulation, controlled fusion experiments, and so on. He has received several scientific awards, and he used the monetary portion of them to start up The Summa Foundation, a philanthropic organization to promote scientific and educational activities in the field of electromagnetics.

2009-10-14T09:28:00.001-06:00

Brain has returned to being obsessed with the "micro-Marx".

There's a perfect piece of copper pipe on eBay for this project. I'm hoping it won't sell and that I can contact the seller next month. There's not much call for that sort of thing these days. Most sensible and non-wealthy people use PVC for their DWV pipe. So I have high hopes there.

When I am sitting at the corner of our big surface plate at work, measuring pulley after endless pulley, I think of nothing else. I was just doing that, and will return to doing it shortly, but I had to get this crap written down before I forgot all of it again.

When I can't actually do any real work (ie; in my shop) toward a project, such as last night, I make pictures in my head like SolidWorks and try out ideas of how things might fit together, how a joint ought to work for technical reasons versus the limitations of my fabrication capabilities (thankfully few, these days, toot-toot!)

Out of all this head-crunching have come many more unanswered questions. I put them here as much for my own convenience to refer to later (and I'd better pull this onto my home computer, as I don't trust online services to continue to exist) as for any education or entertainment you all might derive from it:

1. Now that I'm no longer using (suboptimal) PVC pipe for the liner, how will the Marx panel be mounted / held rigid within the phenolic liner?

2. Should the peaking gap be included within the Marx device proper, or be an optional external device which can be attached to the output or not? There are advantages and disadvantages to both designs.

3. What will the connection between the Marx panel and the output insulator (or the peaking gap as may be) look like?

4. If the peaking gap is internal, then the output will be directly into RG8 or optionally, a short insulated conductor that goes directly to the trigger plane of the switch. In that event, what will the Doran-style HV receptacle for that cable look like?

5. Is there room inside the liner and on the Marx panel for the trigger generator? If yes (and I doubt it) it would mean I could (probably) design it to charge itself off the stage charge voltage.

6. Would it be better to add more stages for other possible applications? If I do that, I will almost certainly have to move the trigger generator outside the housing. I probably should anyway. It's not very big, although I would have to add some SHV connectors, which are expensive, and there'd be this dangling dongle on the outside, which offends my delicate sensibilities. I really ought to try to obtain another TM-11 trigger generator or similar device.

7. How wide an operating range is practical for these compact Marx pulsers? More trigger voltage for the big switch is better... only up to a point. 10X the operating voltage seems to be a typical maximum. If I add stages to the Marx, I don't want to find myself incapable of operating below some maximum trigger voltage (probably 100kV). The operating (or stage) voltage is determined by Marx operating pressure (which is adjustable) and spark gap distances (which are not).

#8: Part of the design is to use pure(ish) dry nitrogen as the switch gas. Using N2 vs. air should increase the UV output of each switch. _IF_ I have enough stage energy to generate significant UV from each gap at all, using N2 will make it better. High UV emission from gaps (to illuminate adjacent gaps) is extremely desirable for fast erection time.

To that end, where shall I obtain the N2? Should I just buy a tank of some blowing my budget into low earth orbit, or is there an easy way to remove most of the oxygen from a dry air stream? The latter would be convenient, since I recently purchased an automatically regenerating commercial air drier-compressor combo for next to nothing.

I think the answer to this one will be found via the Lasers List.

In short, it is easy to do these things informally, if all you want is big sparks. It is non-trivial to do these things very well. I wish I had a bunch of physics and math and a EE (or more preferably a PhD) under my belt. I might be able to do this sort of thing on a consulting basis. You should know that there's a metric butt-load of DoD money to be made in the pulsed power field.

I need to figure out how to turn ($THING_I_LOVE_TO_DO) into dollars.

You probably don't know who Bob Widlar was...

2009-09-28T11:17:00.000-06:00

I have a friend who I think is a LITTLE bit like him.

But I would give a lot to have known Bob. He was one of those painfully smart, incorrigible, larger-than-life characters... like Richard Feynman. Like Feynman, there are many "Widlar Stories".

He is most known for having invented the op-amp integrated circuit at National Semiconductor in the 1960s. For you non-electronics types, that is a BIG deal.

And yet, it is not the only great thing he accomplished in the field of semiconductors and integrated circuits.
I won't go into all that. If you care, you can go look him up.

I just wanted to mention his name, just as a few other people occasionally do, because his name should not be forgotten.

Aha!

2009-09-16T14:27:00.000-06:00

I ordered some parts I don't need.

I thought that one would wish to have the peaking gap (an untriggered gap switch on the output of my Marx generator intended to sharpen rise time of the output pulse) fire last. While that is true, the way I intended to ensure it was wrong. I made the peaking gap electrodes larger in radius (increases firing voltage) and a bit further apart (increases firing voltage).

It turns out that -- for complicated reasons I won't go into here -- the savvy designer wants SMALL diameter electrodes and surprisingly SHORT gap distances for peaking gaps on the end of Marx generators.

"The generation of 50pS pulses with a 60kV charge in a single channel switch of very small dimensions has been reported in 'Ultrafast Gas Switching Experiments', Proc. 9th IEEE Pulsed Power Conference, pp.491-494, 1993."

Which paper I have subsequently dug up, but not read yet. 60kV in 50 fucking picoseconds? Unpossible!

So I wasted a few bucks, but the thing has just become quite a bit simpler to construct. I do have to rework my models though.

micro-Marx work update

2009-09-15T21:15:00.001-06:00

The Micro-Marx Generator design is coming along slowly. Because I am trying to use as many off-the-shelf parts as possible, machining them or modifying them as little as possible, the process is somewhat tedious and slow. If all goes well (cough) it is possible I will have the entire design finished by the end of this week. We'll see. I've already started on the housing, although the base/insert assembly - the bit with all the electrical parts - will need some tweaking to fit the housing once I've got the housing design finished. This thing is gonna be slicker than snot on a glass doorknob.

Unfortunately, it is also very likely to be a lot more expensive than I had at first surmised. See, it really needs to go into a housing made of a very good conductor. The proven design from which I am taking my inspiration for this project used copper pipe. It needs to be something I can solder to, which lets out steel and aluminum tubes, which are lousy conductors anyway. Skin effect can come into play with very short rise time pulses, and skin effect increases dramatically as conductivity goes down. So I'm all for using copper pipe myself.

There's just one itty bitty problem: the device can't be made to fit into a piece of pipe smaller than 3".

A liner of a non-conductor with pretty good dielectric withstand (able to hold off say 100kV for 1mS) has to go inside that, so the OD of the liner and the ID of the housing need to be close to each other, and the ID of the liner has to be large enough to accept the smallest practical configuration of the Marx generator assembly. I also need a way to hold the Marx assy firmly when it is inserted into the liner. I have some ideas about that.

I'm hoping to use gray PVC conduit for the liner. PVC isn't the best choice due to its high dielectric constant and loss tangent (both of which degrade rise time), but it's the only plastic I can work with easily in terms of machining shapes and gluing pieces together. I am not gonna go in for custom epoxy castings. I thought I had a line on some styrene tubing that would have been much better than PVC due to its lower loss tangent, but it turned out to be too large in diameter. Fiberglass & phenolic tubing of the required size can be had, but the price is too dear for my budget.

So, the current plan is to make the housing from 3" Type K copper pipe and the liner from 2.5" gray PVC conduit, plus some additional slabs and pieces of PVC machined and cemented together. It is unfortunate that the dimensions required force me to use the heaviest-wall copper pipe which is Type K. It's more expensive than the others since it uses more copper.

The spot price of which, by the way, is on the rise again, currently at $2.75/lb. Dammit. I should have bought the coil I needed for the primary winding of my Tesla coil when it was still hovering around $1.25 a pound in December of '08!

I'd like to use copper pipe fittings for the output end of the housing, but it remains to be seen how well that will work. The output connector will be designed to use a standard coaxial cable with a modified UHF connector, the insulation extended by 4" - 5", and a banana plug and jack on the end of the center conductor. I might also consider some sort of direct connection to the switch this thing is being built for, but I want to have the option of using a coaxial cable later, as I may want to use this pulse generator with other equipment.

I'd also like to use some off-the-shelf, solder-on brass bolt flanges for the ground/input end, but those look expensive, and frankly, I don't like their dimensions. I am already imagining that I may have to make my own flange and end-cap for the ground/input end. I sure do like that bolt circle feature of the DRO on my mill. Incredibly handy feature, that.

3" copper pipe isn't carried in the usual homeowner stores locally, but I will call around to the professional plumbing / pipe fitter supply houses. Online suppliers sell it for around $24 a foot. That sure beats McMaster-Carr's price, but I hope to find a cheaper source locally, including some of the surplus yards or building material recycling places such as Resource 2000 in Boulder or Bud's Warehouse in Denver.

Mo parts, mo work, no money, mo problems

2009-09-11T11:43:00.000-06:00

I forgot to order the gap spheres for the peaking gap between the last stage of the Marx and its output.

Oh well, since I also ordered too few spacers for the finger guard on the mini-Jacob's Ladder that's going on the Mad Scientist Light Switch, I figured I'd just order several things at once. It's pay day.

I remember thinking I needed a carbide milling cutter for something, but now I can't remember what for, so I don't know what size to order. Hopefully I will remember soon, I have a filled shopping cart that is getting stale. Oh yes: milling G-10 (epoxy fiberglass slabs). Now why did I want to do that? Oh yes: anti-surface-tracking grooves in the Marx support slab, between the gap spheres. Hmm, I might have to make that thing thicker. I'd like to only do it once, so I might use a different material from what I have on hand, but I'd rather not spend any more money than I have to. Hmm.

L0stboy moved to 303 so I thought I was gonna have another fresh new Mad Scientist type mind to bounce ideas off of and maybe even work together on something. He's just accepted a new job in DC and will be moving soon. Figures.

I thought I'd be able to use Macor machinable glass ceramic for a few projects... until I checked the prices. Holy crap, you would think that in the 20+ or so years since it hit the market the price would have fallen some. I could use it for some tiny things I expect to need, maybe. Example: trigger assembly for the miniature trigatron which fires the first stage of the Marx.

A 3" long piece of .25" diameter Macor rod is "only" $20. >_<

~~~

Discovered another esoteric book I'm gonna try to get my hands on: "J.C. Martin On Pulsed Power". Currently used copies are selling for $200. I'm hoping that if I'm patient, I'll eventually find a used copy I can afford. J. C. Martin was a pulsed power researcher at the Atomic Weapons Research Establishment, Aldermaston UK.

And I should probably re-read Vitkovitsky, as I might actually understand more of what I'm reading this time around. Holy shit, new copies of THAT are $200 now. I think I paid $75 for mine, back in the early 90s.

Micro-Marx

2009-09-11T07:36:00.000-06:00

I am starting to realize just how involved this little micro-Marx generator I'm building really is. That is, if I want to do it _right_.

Kindly indulge me while I polish up this dusty old thing called my ego.

I can do this Marx thing "right now" (well, not really, but fairly soon) or I can do it "right". By doing it right, I believe I will be the first amateur to build a "fast" (that is, sub-100nS) Marx.

Anybody who can solder can string together some resistors, capacitors, and bent bits of wire for spark gaps, and make big sparks. The rise times will be on the order of uS, and the rise and commutation times will be scattered statistically all over the place. (this phenomenon is known in the vernacular of the business as "jitter". Are you taking notes kids? There will be a quiz.) But the people who are achieving rise times anywhere below 1uS, with low jitter, those guys are scientists. They've got PhD's and budgets.

And I am hardly a past master at this stuff! I have constructed precisely two spark gap switches in my life time. The recent one, and another, more ghetto construction I built around 1986, based on a design from good old Bob "Information Unlimited" Iannini. I think it's in that first book of projects he wrote but I'm too lazy to go look it up right now. I might have bought the plans from his mail-order business. Only pictures and a few parts from it remain today. It was nothing to be proud of.

Some twenty three years passed between the two projects, during which I did a great deal of reading on the subject.

I think the new switch, if not yet, will soon be something to be proud of. But in order for it to be "something to be proud of", it has to be demonstrated to do something above-n-beyond the switches that other amateurs have built lately. At least one of these - a rather good looking design by a fellow in Canada - has experienced a casing failure.

Oh, and have I mentioned that when this project is complete, and I've figured out wherever the upper boundary for charge transfer is on this switch, I will possess TWO very high powered, very fast switches? The other is a so-called "rail gap switch" in that its electrodes have a linear rail-like geometry.

There are projects one might consider doing with double-pulse machines.

Does anyone know where I could pick up a neutron detector, cheap? Or a really good backstop for 10.6µ light? Naahh, you don't want to know.

dammit

2009-09-09T20:49:00.000-06:00

Based on some reading I've been doing lately, the hole in the trigger plane electrode of my switch is almost certainly too small. The rule of thumb is that the ID diameter of the trigger plane electrode should roughly equal the distance between main electrodes.

I am going to try it anyway. I can always make the hole bigger, or make another one like it with a bigger hole.

"black science" contractors - small followup

2009-09-09T08:34:00.000-06:00

Well, well, well.

It seems Physics International is still around! PI and their debt (!) have been passed from DoD contractor to DoD contractor for the past fifty years, like a smelly diaper nobody can find a trash can for.

In following the ins and outs of all these contracts, programs, acquisitions, mergers, divestitures, and so on, I kept stumbling across SAIC.

Holy crap that company is dirty. And armpit deep in just about everything black. They've been involved in CIA remote viewing programs for chrissakes! Their fingers are very far into the Trailblazer pie. Trailblazer, for those who haven't been keeping up, is the NSA's runaway train that was supposed to update all of their monitoring infrastructure to be able to snoop modern developments like cell networks, the internet, and so forth.

Telcordia Technologies is a key player in Trailblazer. Originally Bell Communications Research, or BELCOR, Telcordia is practically a front for NSA operations now. It was a division of SAIC from 1997-2004, during which time it was run by Stephen Learnheart who, before he held that position, was the NSA's deputy director of SIGINT!

Remember the "secret CO rooms" associated with Verizon and NSA? Operated by Telcordia and stuffed with their technology.

Now, I'm hardly a conspiracy theorist, but the more I follow these threads, the easier it is to believe that everything really is being run by a dozen fat old balding white men wearing $3K suits and smoking $50 cigars in a Tempest-shielded room ala The X Files.

they come and go

2009-09-09T07:21:00.006-06:00

Does anybody here remember Veradyne?

Vera! Vera!
What has become of you?
Does anybody else in here
Feel the way I do?

~~~

Okay, so I couldn't quite get the Pink Floyd lyrics to work.

Some time around 1976 or so, a guy named David Platts (and presumably at least one or two other researchers) working at Los Alamos National Laboratories developed a miniaturized Marx Generator which exhibited very high performance. Mr. Platts wrote it up in an internal paper (of which I happen to possess a copy) titled "10-Joule 200 kV Mini Marx". As far as I know, it was never published outside of Los Alamos National Labs. I found it in a public-facing web directory of technical papers that disappeared from the web after 9-11.

The work was performed under the auspices of the DOE. (duh)

In 1978, a company called Veradyne was formed in Burbank California to produce this pulse generator commercially. David Platts was not involved as far as I can tell, and no patent was ever applied for by either David Platts or Veradyne. (Veradyne does have one patent assigned to it, for a weird little coaxial spark gap switch of a geometry I've never seen used anywhere, for reasons I can easily understand - it is a silly and impractical design).

Since that time, EVERYONE has copied (and here and there, even improved upon) that original design. Companies like Physics International (no longer extant), Applied Pulsed Power Inc, R.E. Beverly III & Associates, and Applied Physical Electronics have made many millions of dollars of profit from Marx pulsers based upon Mr. Platts' and Veradyne's generator.

I often wonder what becomes of these little (sometimes not so little) companies which are formed to fulfill the contract requirements of some shadowy government program (BTW, most of this stuff has to do with either nuclear weapons research or high energy RF developments for EMP simulation, broadband radar or EM weapons) and then dissolved after the program is over.

Most of the more successful ventures got bought up by larger and older government contractors. Ferinstance, Physics International, who developed a switch design that has been used EVERYWHERE, by the hundreds, in a whole raft of huge pulsed power machines, was bought by Primex Technologies, who later sold that division to Maxwell Laboratories, which eventually divested it to Titan Corporation, a once-huge defense corporation, some of whose employees were implicated in the Abu Ghraib scandal (!).

Eventually, Titan was purchased by the immensely powerful defense contractor L3 Communications, its various divisions redistributed among L3's existing products and services. Titan's pulsed power division remains as a small L3 shop called the Pulse Sciences Division. The sole product which has remained essentially unchanged through all of this is the T-508 Switch.

That switch was developed in the 1950s by Physics International for one of the early nuclear weapons effects (NWE) gamma ray simulators. You can see an ass-load of T-508s here. It is still sold today, with a few physical changes designed to make it even cheaper to produce, but essentially unchanged in terms of how it works, size, shape, capacity, performance, etc. The model number remains the same.

Meanwhile, the scientists who invented and developed these devices, which have made other people so much money... soldier on for $65,000 - $75,000 a year.

how scientists call bullshit

2009-09-08T22:08:00.000-06:00

"Claimed efficiencies of 46% using the Reflex Triode setup by Didenko [33], is somewhat controversial."

translation: "We think Didenko is full of shit."

Say Bill, what ever happened to that "Mad Scientist's Light Switch"?

2009-09-01T20:51:00.000-06:00

Funny you should ask!

I will tell you.

It's on hold because the next step I need to take requires 15 little screw spacers, which I feel certain I ought to be able to buy somewhere, and which I really do not wish to machine by hand one at a time.

I can find them for several dollars each.

I can find them cheap... if I buy 500 at a time.

I was more than a little surprised that not one single electronics or industrial supply house with whom I deal regularly has the exact thing I need.

Marx generator and switching news

2009-09-01T07:37:00.000-06:00

I can't wait for my "micro-Marx" parts to get here. And to find time at work (on my lunch hour) to start the drawings. Yesterday's lunch I was out of the office (for a good reason, but out of the office nevertheless), today I have a dentist appointment (oh yay, my favorite thing) that I MUST NOT forget. I have managed to forget the previous two, and NOBODY is happy about that. But I digress. If I can maintain momentum on the micro-Marx and get it done in roughly the same time frame as the TSGS (a few months) I will be ROCKING hobby project progress far in excess of what I have accomplished in the previous six years (ie; since we moved in here). This is a fine thing.

I have also realized that this micro-Marx may in fact be capable of multi-channeling (multiple current paths per switch firing) my big railgap switch, which is still intended for the pulser. I need to think about how I would determine whether multi-channel operation occurs. And the answer to that is starting to look like an o-ring-sealed, thick polycarb lid covering the gigantic hole I would mill into the top of the casing. Holy shit. I can't believe I am considering the heavy modification of what may be a piece of history.

I wanted to provide a link to a page of information about the pioneering company (and its president) that built that switch, but there doesn't seem to be one.

For now, suffice it to say that custom HV pulse capacitors and exotic high power switching devices made by his company played a role in a few early (and quite secret, at the time) pulsed power projects at Los Alamos National Labs and Sandia National Labs in the 50s and 60s. I'm gonna have to think about whether I "should" make that modification. It may be that the power levels I have at my disposal are so much less than the capabilities of that switch that I just won't have to worry about it. Of course... if I'm wrong... I might end up having to repair one or both electrodes, which would suck. I do want to try something better than the apparent trigatron style triggering method it appears to have originally used.

OTOH, I think one of the switch's former owners modified it slightly by making two holes in the case. I do not know this for a fact. But I'd have done something like them anyway, for gas inlet and outlet. There are no other ports in the thing. It is a very curious object, about which very little is known. I should scan the literature I have for them and put it online, I've never found that info anywhere else, therefore it is precious. Oh, and I need to ask him to remind me how he got his mitts on it.

A micro-Marx generator is now on the slate

2009-08-31T09:58:00.000-06:00

I decided I could at least afford to buy the parts for the innards of the micro-Marx generator I'm building, even if I might have to either wait on the copper housing, or gin up something less desirable out of sheets of copper clad PCB soldered together.

I re-found the 1/2" brass balls I'd been having trouble finding, and ordered them.

I also ordered some little nylon spacers, and some screws I will need.

I will fabricate almost everything else. I have the capacitors.

The insulating base will be cut from some FR-4-like substance cannibalized from the P.O.S. Chinese laser engraving machine I brought home a few weeks ago. Almost no sign of that thing remains, and very few cubic inches within my garage were lost after I was done. I'm glad I saved the bits I did, I'm already using them.

Anyway, that sub-project (this is part of the DTSGS project) is now officially underway, and that's a boon for my morale. All of the parts won't arrive until late in the week, but I can start making the baseplate as soon as I have drawings, but because this thing is as finicky as the switch was, they actually do have to be GOOD drawings, which means I have to do them at work on my lunch hours. I haven't even started.

2009-08-26T11:41:00.000-06:00

"The authors report an estimated pulse energy of greater than 15 J in a 100 ns pulse with a peak power of about 100 MW. They state: 'Accurate measurement of pulse energy was difficult since the laser energy produced a plasma in the thermopile used.' "

What that translates to is that the "thermopile" (a device routinely used to measure laser power) got "blowed up good" -- sustained physical damage, destroying the measuring head. Ironically, blowing up sensor heads during the development of a new laser design is a common sign of great success.

~~~

I have a knack for finding/saving useful parts for which documentation is at best hard to find and at worst, flat-out nonexistent. For instance: I have a little trigger / igniter circuit board out of a very old American Laser Corporation model 60 argon laser head I used to have. That was an ex-laser even before I got my mitts on it, and long ago I cannibalized bits out of it and gave the rest away as a collection of laser head parts.

I saved the little igniter board because it was a self-contained trigger pulse generator that "might come in handy some day", with all parts including trigger transformer. As it happens, I now find myself in need of a small trigger pulse generator. I pulled it out and began wondering what voltage to feed it to see whether it still works or whether I need to replace one of the parts on it. (all except for the unknown trigger transformer are dirt cheap, readily available parts. The trigger transformer is very unlikely to have failed.)

Fortunately, the thing is simple enough (12 parts total) that I can trace and redraw the schematic without much difficulty. Armed with that and the component values (thankfully, all but the trigger transformer have readable numbers or color codes, and I can guesstimate most of what I need to know about that transformer by a few simple measurements). Naturally, one of the parts was discontinued 20 or 30 years ago, but it's a general purpose UJT, so if by chance that one turns out to be bad, I can probably find a replacement. For all I know, the whole board may be perfectly good. I just need to know how much DC to feed it. It will be in the neighborhood of a few hundred volts.

Sometimes saving parts for ten years pays off. I have been wondering how I was going to lay hands on a decent trigger transformer.

Oh yeah: I had been half-hoping it would turn out to be practical to trigger my new spark gap switch directly from a trigger transformer, if I could just find one with enough voltage and a fast enough rise time. After tracking down almost every manufacturer of such devices in existence and reading all the spec sheets, it turned out that there is zero chance of multi-channeling such a switch from a trigger transformer. They are all too slow (due to the inductance of the ferrite core) and the highest voltage unit I could find was about 40kV. That might be enough, but I'd really rather have about 60. (sounds like a lot, but the current and energy for these trigger pulses is really small.)

Thus, last night's mention of making a small, fast Marx generator to trigger the switch. But the Marx itself needs a triggering device for its very first stage, and that will be a cute (tiny) little trigatron I'm gonna build, which will in turn need a high voltage trigger pulse of its own (which can be fairly slow rise time, ie; a trigger transformer will be just fine).

This is the nature of high energy pulsed power systems. You start with whatever whacking huge exotic switching device is required to handle whatever it is you're doing with the main energy store. But such things nearly always require very fast, very high voltage trigger pulses, so then you have to have another, somewhat less exotic device to create THAT, and so on, backwards, with each preceding triggering or switching device being less exotic, slower, and of a lower voltage than the latter, until you arrive at a simple push button, relay, or logic input that starts the whole chain reaction. Of course, the entire chain reaction is typically over in a small fraction of a second.

For instance, the original design for my big pulser, as I originally received it, worked like this:

1. Operator presses the "go" button, which starts the main capacitor bank charging, along with several other high voltage capacitors used in the various triggering stages.

2. When the desired voltage is reached, a Simpson Relay Meter closes a pair of light duty contacts.

3. The meter relay contacts close an intermediate relay having contacts rated for higher voltage and current.

4. That relay turns on a small pentode (a type of vacuum tube often used for light duty switching).

5. The pentode switches a small capacitor (charged to a few hundred volts) to a pulse transformer, the output of which is a pulse of a few thousand volts which is delivered via HV cables to a hydrogen thyratron inside the capacitor cabinet.

6. The thyratron fires, delivering energy stored in each of six capacitors, to each of six mercury ignitrons. There was one ignitron on each main capacitor. The ignitrons were the final, big switches in the original design. They are obsolete, lower-performing, fragile, and very expensive devices, and all of them were inoperative before I got my hands on the thing. Since those particular tubes cost $1500 each, I'm replacing them with a single railgap switch, but that's another story.

ARGH! I think I'm losink my mind!

2009-08-25T20:56:00.001-06:00

On the other hand, teenagers have built these things, why haven't I done it yet?

'What's that?", you say?

Hah! Wouldn't YOU like to know!

Shhhh, don't tell anyone: I'm doing a first-pass analysis of a medium-voltage Marx generator, with a mind to estimating costs of the parts I don't have lying around, on a count-a because I think I can do it quite cheaply.

It may be a pretty fast one, I think.

I need it to trigger the spark gap switch. Field distortion triggering requires very high voltages (6X - 10X the voltage across the switch is not uncommon) at as fast as a rise time as is possible, but generally below 1µS. Sounds unlikely, but it is possible.

metals

2009-07-19T18:49:00.000-06:00

fact: there are no high conductivity (ie; 90+ % IACS) alloys that are also hard(ish) or strong(ish).

Beryllium copper is often used where a hard conductive alloy is called for, or phosphor bronze.
Both have lousy conductivity compared to plain old copper, and the former is not something you want to machine, because the dust is carcinogenic.

"Reasonable" conductivity - say 50% - 70% IACS - that isn't nearly as good as copper but is at least better than say, iron, is fine for most applications, where the currents are relatively low.

What a shame that I don't deal in low current applications, at least not hobby-wise.

When the amps are in the tens of thousands or higher, you start paying attention to the conductivity of your alloys.

If you don't, Bad Things can happen, such as rapid disassembly of non-moving components.

When the conducting volume is high, you don't care either. For instance, the collecting plates of my pulser, which connect the six pulse caps together, are made of 1/4" aluminum sheet. Dunno the alloy, but I'd bet lunch that it's garden-variety 6061-T6, and that's 40% IACS. But there is so much of it, it just cannot possibly matter.

Unfortunately, what I need a high strength & high conductivity alloy for is a small volume part that willy have a great deal of current (10,000 - 100,000 amps) passing through it, albeit for very short times.

There is no good solution, really.

Well.

If what I do were easy, everyone would do it.

And we can't have that.

the task, the process, the reward

2009-07-16T22:27:00.000-06:00

The Triggered Spark Gap Switch is nearing completion.

Not to be confused with Mad Scientist Light Switch, which not only is not nearing completion, it isn't even breathing hard.

*ba-doomsh*

Let's not talk about that.

For weeks now, the status of the spark gap has been: "complete except for: 1) locating grooves in the inside surfaces of the end caps to center the housing pieces and end caps on each other; and 2) drilled and tapped holes in the end caps for the gas fittings."

As of tonight, both grooves are done. One gas fitting hole is drilled and mostly-tapped. I say, "mostly" because the tap I have is a normal one, but this hole is a shallow one, and is for an NPTF fitting besides. I ground mine down a little (because a new _standard_ tap to replace it is cheap) but it wasn't enough. So I have a "short projection" NPTF tap en route. $14 grumble.

And the gas fitting holes turn out to be finicky. You see, I did not plan their locations vis-a-vis the available real estate inside the housing but outside the spark gap main electrodes. AND I didn't allow for screwing up the spark gap main electrode holes, so they currently require a nut and washer. That may change, but for now, the gas fitting holes have to be outside the washer.

That's about the right diameter if I do something terribly clever to make the nut and washer go away. We shall see what we shall see. Meanwhile, I had to put a little hole behind the back-side curve of each main electrode into the end cap, drill and tap the hole for the fitting quite a bit outboard of that, and then drill a small connecting holes between them, at an angle. I have done the first one successfully. I am, as they say, well chuffed.

I shall post a picture or two, another day.

Now I am going to consume about an ounce of Hennessy (Privilege VSOP, not the top of the line I know, but on the other hand, I can afford it. about once a year.) begin reading Yet Another Pratchett Distraction, and fall comatose within minutes.

triggered spark gap switch work

2009-05-12T20:45:00.002-06:00

1. I have a pretty good idea of what to do about the projecting studs from the main electrodes.

The solution will add some complexity, but should ultimately be a win.

2. I think I am going to add a circle of several blind threaded holes to the exterior surfaces of the end caps for mounting and electrical connections. One needs to make rather good connections when one is expecting to conduct tens of thousands of amperes, even for a fraction of a second.

3. I have the all-plastic gas fittings in hand - free samples from one of our vendors at work. If you ever need fluid power (pneumatic, hydraulic, etc) products in the Colorado area, give Fiero Fluid Power a call. They have competent and friendly sales staff, a large selection of product types and brands, huge stock, and very competitive prices.

4. Last night I set up one of the end caps in the lathe for turning the relief that will axially locate the housings. It's a dodgy proposition, because there is just enough chuck jaw to pick up the curved maxima on the perimeter of the end cap, made a bit worse by the insertion of small tabs cut from sheet copper to protect the aluminum from the lathe jaws. Tonight I will see whether there is room for me to (instead) mount the end cap onto the turning tool backward (since I am now working on the inside face, not the edge), probably with some washers to protect the inside surface from the cap screws. The washers will be the key, as they will be very close to where the cut needs to go. If that can be made to work, it will be a lot more secure than trying to grip the radiused edge in the chuck jaws. I would be heartbroken if I wrecked one of the end caps at this late stage.

I didn't proceed past setting it up though, because I was tired and foggy-headed, and I wasn't entirely sure how to proceed as far as getting the edges of the cut in exactly the right positions. I am used to doing things rather ad-hoc and this is one of the first projects where I have had to really observe accurate dimensions, parts placement, tolerances, etc. It's also the first time I've ever made a bunch of parts with lots of holes and things that needed to line up - fabricating them separately - and had everything fit perfectly the first time. So apparently, planning ahead (drawings) and careful, methodical work actually pays off. Who knew?!? ^_^

Anyway, this morning I think I figured out how to proceed as far as getting the cut located correctly.

The more I do this kind of stuff, the more respect I gain for skilled machinists.

To sum up, when all of the above work is complete, the switch will be... oh crap, no it won't.

5. I forgot I still have to devise a connection method for the trigger electrode. Grumble. I have to think about that some more. And maybe try to find more pictures of the old Physics International T508 switches which I am partly imitating to see whether they offer any clues. I might be worrying too much about certain things.

Right then, when the above 5 items are done, the switch will be ready for testing. At that point, I will probably stick it on a shelf and forget about it for a while as I try to get the Mad Scientist Light Switch project finished.

~~~

“Do what you love to do and give it your very best. Whether it's business or baseball, or the theater, or any field. If you don't love what you're doing and you can't give it your best, get out of it. Life is too short. You'll be an old man before you know it.” -- Al Lopez

~~~

"You wake up in the morning and lo! Your purse is magically filled with twenty-four hours of the un-manufactured tissue of the universe of your life. It is yours. It is the most precious of your possessions. No one can take it from you. It is un-stealable. And no one receives either more or less than you receive." -- Arnold Bennett

~~~

"Recognize that neither technology nor efficiency can acquire more time for you, because time is not a thing you have lost. It is not a thing you ever had. It is what you live in. You can drift or you can swim, and it will carry you along either way." -- James Gleick

please indulge me in a small digression

2009-05-11T15:06:00.002-06:00

I wish to tell you a little about Nuclear Weapons Effects simulators and the companies that build them.

NWE simulators exist because governments need to know:

A) whether their existing nuclear weapons will continue to work if needed

B) how new weapons and defensive systems will behave when exposed to NWE.

The data required to answer those questions can only be acquired from real-world phenomena usually found only in the proximity of a nuclear detonation. International treaties prevent actual nuclear detonations these days, so other means have to be found to generate frighteningly intense bursts of gamma radiation (among other sorts) comparable to those attained from a nuclear weapon, but without the big explosion, messy & inconvenient radioactive fallout, and so on of the real thing.

If you are a very smart person (or a small group of very smart people) in the right place, at the right time, you can form a company with the express purpose of converting millions of federal dollars into terrifyingly powerful, if vanishingly short, bursts of photons - intense gamma radiation, magnetic fields, etc. This happens quite often in the field of Nuclear Weapons Effects simulators.

Lately, I've been trying to find information about these companies, who built such imposing and little-known devices as Columbus I & II, Aurora, REBA, SLIM, Nereus, HARP, PROTO I & PROTO II, EBFA/PBFA, Decade Quad, GAMBLE-II, Black Jack, SNARK, RADLAC-I & -II, CASINO, HERMES-I, -II, & -III, Hydra, Atlas, Saturn, The Z Machine, and many others.

But most of these companies existed only for five or ten years, and then something happens - a crucial partner dies in a car accident leading the surviving partner to dissolve the company, the company is bought by another company, various things. Actually, a LOT of them seem to get bought or merge with another company. The history of these companies and their projects reads like a mixture of old bible stories (ABC begat KLM, KLM begat XYZ, etc) and modern industrial conglomerate acquisitions.

And that is how L3 Communications (a major defense contractor formed from ten former business units of Lockheed Corporation when Lockheed merged in 1996 with Martin Marietta) has suddenly and only recently become the largest contractor for Big Scary Government Doomsday Devices in the world. Unless you count Sandia National Laboratories itself as if it were a contractor - which in essence it is, since it is operated for the U.S. Government by - guess who? Lockheed Martin.

Not all of the results had to do with simulating the radiation effects of nuclear weapons. Some experiments were designed to verify or develop equations of state for exotic states of matter such as plutonium being compressed into a hyperdense state during the detonation of an implosion fission device.

Experiments were also performed with an eye toward developing directed energy weapons (see also Strategic Defense Initiative or "Star Wars") and toward means of discriminating between real and dummy re-entry vehicles (see also Penetration Aids).

Then too, it was recognized fairly early that very powerful pulsed power machines might be capable of triggering nuclear fusion, and research along those lines is still being performed on some of the largest pulse machines in the world. Unfortunately, it took quite a long time before the data supporting the possibility of inertial confinement fusion became generally known in the physics community because of the classified nature of most of the work performed on these machines.

The energies involved are so staggering, they are difficult to comprehend even in terms of magnitude comparisons to familiar events. The numbers beggar belief. For example, Sandia National Labs "Z Machine" delivers 1,000 times more energy than the most powerful lightning bolt nature has to offer, and does it in 1/20,000 of the time. In electrical terms, it delivers 20 million amperes of current in a 290 trillion watt (290 terawatts) pulse over 100 nanoseconds. For that short period, 'Z' is delivering 80 times the power generating capacity of the entire world. In physics terms, Z converts that electrical pulse into 2 million joules of Bremsstrahlung x-rays in a 4 nanosecond pulse. 8x10¹⁵ watts. Enough to implode a tiny deuterium-filled target and raise its temperature high enough to undergo nuclear fusion.

Is it any wonder I am fascinated with pulsed power experiments?

more switch

2009-05-11T08:21:00.000-06:00

The DTSGS is nearing completion.

However, the viewer is cautioned that appearances are deceiving. It looks far more finished than it actually is. There is much work yet to be done before it can be tested.

Hell, to test it, not only do I need to finish it, I need to build other devices for measurement and instrumentation.

Nevertheless I am, as they say, well chuffed.

Oh yeah, and I need to figure out a way to trigger it. When you throw a really big powerful switch, you need to hit it with a really big, powerful stick.

Grog shape metal! Ungf!

2009-05-06T23:24:00.002-06:00

This is my progress so far in fabricating the end caps of the switch:

further progress on the DTSGS

2009-05-06T07:29:00.000-06:00

I made significant progress on the spark gap switch last night while Judy was finishing up with her last patient and before we headed down to the hospital. I drilled all the bolt holes in one end cap, and also counterbored them.

Why did I counterbore the holes? I have changed the design on the fly, as is my habit. My original design used button-head cap screws with the heads on top of the end cap, leaving a conductive projection, which is Undesirable for high voltage kit, because corona forms on projections, and if an undesired spark to ground (ie; flash-over or arc fault) occurs, it will start from a sharp projection. This is why you see strange rounded and smooth shapes (including donut-shaped "corona rings") applied to high voltage equipment. The smooth surfaces of large radius smooth out the electrical field, preventing the sharp voltage gradients which encourage corona or unwanted sparks. This is why the end caps have rounded edges in the drawing. So I decided to use socket-head cap screws countersunk into the caps so that nothing projects at all.

I also rounded off another corner of that end cap. The end caps were cut as squares from 1/2" plate, and will subsequently need to be turned on the lathe to their net diameter. Turning those big corners off on the lathe is stupid, excruciatingly slow, and the heavy interrupted cut makes damaging an insert more likely. If I had a bandsaw, that would be the best way to cut out circles like this, but I don't. So I decided to mill off the corners. It's slower than using a bandsaw, but at least I can leave a nearly circular edge for turning on the lathe. After that I had to quit because Judy finished with her last patient and it was time to grab some dinner on the way to the hospital to visit a friend.

I realized I don't have an easy way to hold these 1/2" thin plates in or on the lathe, since there can be only a tiny little 1/4" hole in the center for the main electrode bolts - not enough for a tapered arbor (which I don't have anyway, and would have to make even if it were thick enough to support the work).

So I've decided to make a custom fixture which will have the same bolt-circle pattern of threaded holes as the end caps, but a smaller diameter, leaving the edges free to be rounded off. Then I can simply bolt the end caps to the fixture and mount it in the 3-jaw chuck. This bugs me because it will take more time, but I can see no other solution. I dislike making single-purpose "white elephant" tools, but if I must, I must.

DTSGS parts progress

2009-04-27T21:50:00.000-06:00

So, a quick snap of some of the (mostly) completed parts so far:

Soon, some Delrin rod will arrive, after which I shall be able to verify its diameter (I have learned not to trust catalog-provided dimensions), then make the bolt circles in the end caps and proceed with end cap fabrication. The end caps are going to be interesting. I begin to understand why companies like Maxwell Laboratories or Titan PSD charge many kilobucks for these devices.

I am beginning to wonder whether I could make money at this. I think I do not know enough yet about the field to act as any sort of consultant or manufacturer, however. And I'm weak in math, physics, and EM theory... okay, never mind that.

More as it happens.

oops, I may have screwed up

2009-04-23T12:10:00.001-06:00

Last night I sat down with pencil and paper and ruler and compass (none of which have ever been known to crash, or refuse to allow me to change a dimension, or do anything else which a poorly designed piece of software might do) and pipe...

...and figured out that there is something wrong in the dimensions of my own design somewhere. Or rather, I believe that my design, as embodied in the 3D model, is jest fine thenkyewverymuch, because I already did interference and clearance checks on it using SolidWorks' built-in tools. BUT, I seem to have lost the shop drawings I originally brought home, and I ordered the aluminum based on how I THOUGHT the thing was built and dimensioned... and I think I screwed up / misremembered, because 4" aluminum appears to be about 1/2" too small.

I know right? Me? Not remember something accurately? UNPOSSIBLE!!

*ahem*

So I'll try to look into that today, if work and life allow me the time. I also need to get in touch with $PSHRINK today, dammit.

Oh, but to return to the spark gap project for a moment, I think I can solve the issue only two ways:
1. scrap the parts I've already cut from the 1/2" aluminum plate I just bought and get some slightly larger
-or-
2. use smaller diameter clamping rods to hold the thing together, so as to ensure they will clear the outside of the trigger plane electrode AND not sit outside the radiused edges of the end caps. Blah.

#1 may or may not work. I've been making a lot of design decisions on this thing by waving my hands in the air and making educated guesses based on past experience. But I didn't like the idea of making any of the parts weaker. You see, when a sealed spark gap fires, switching a whole bunch of power, some of that power gets converted to heat in the spark itself inside the switch, which expands the gas quickly, creating a powerful shock. (the same mechanism is the reason lightning causes thunder)

So the other thing I did last night was calculate how much force the clamping rods could withstand if I made them 3/8" diameter instead of 1/2". I am using acetal (Delrin) rods. The short version is that six rods of 3/8" acetal could withstand an instantaneous force of 5,964 pounds total, ignoring whether and how much the threaded holes in the ends and the fasteners threaded into them can withstand. I need to look up some information on machine screws pulling out of various materials and how one handles that.

Then I decided to look at how much PSI overpressure would have to exist on the exposed (inside the o-ring seals) end caps to cause that total force on the rods. The area of that circle is 1.767 in2, so the pressure required to create the above force is 3,375 PSI, which is pretty damned high. Not bad! And no, we don't have any idea what the peak pressure inside the switch will be for a given shot, because it will depend on a whole bunch of hard-to-calculate factors, including whether the switch is operating in multi-spark, single-spark, or pseudospark mode. (you probably don't want to know)

Now, there is another component which sees that pressure spike, and that is the cylindrical housing. We need to do the hoop stress calculation to see how much strain the walls of the housing can withstand, based on an internal pressure of 3,375 PSI. The housing material I have lying around is some variety of polyurethane, and I don't know which variety, or how hard it is. The harder it is, the stronger (not always true for many materials, but true for PU, I looked it up). I thought PU would be a good material for this, because it is more flexible and less brittle than other materials, and ought to withstand shock very well (literally, by stretching a little bit and recovering). So I did the hoop stress calculation for a 1/2" wall thickness tube of the lowest grade PU (4,500 PSI) given my dimensions. And it seems it can withstand 1,791 PSI.

Er, oops. So the limiting factor isn't the damned rods anyhow, it's the housing. Mind you, I wouldn't have used the 3,375 PSI internal pressure number anyway. You don't design something without a safety factor. For general engineering where humans aren't TOO likely to get hurt of the design fails, you use at least a 2X safety factor, which would reduce our maximum operating pressure to 1,687. For human-rated devices (airplanes, cars, bicycles, crutches, or things which will be used near people and which might explode, sending a few hundred pieces of shrapnel into the flesh of nearby persons, cough-cough) you use a 4X safety factor, which suggests 843 PSI, based on the rod's strength. That also gives us a 2X safety factor on the housing. Hmm. Maybe. Sure would be nice to know what kind of actual pressure pulses I can expect. Testing this thing is going to be interesting. I wish I had some kind of very fast (ie; capacitive or piezoelectric) pressure sensor I could connect to the switch through one of the gas fill ports. Then I could connect such a sensor to my storage scope during test firings and see what kind of pressure pulse is generated by what kinds of energies in which operating regimes. I know that the heat released must go as I2*R, and R will go up with the amount of gas in the switch that the arc has to pass through, so it seems likely that pseudospark operation - which is done at very low pressures - ought to generate the least amount of pressure. And it's where I want to be anyhow for other reasons like not eroding the fuck out of my electrodes.

I would explain more of what is going on with all this in terms of design features and why it has to be the way it has to be, but I very much doubt any of my readers care. Mostly I am just "thinking out loud" here, for my own benefit.

Mad Scientist Light Switch parts pictures

2009-04-22T07:43:00.002-06:00

Pictures of the miniature horn gap (AKA "Jacob's Ladder"), or rather, the parts I've completed so far, are up on my Flickr pages here.