On metal scraping

A deeper, more thorough treatise on the subject
of hand scraping may be found here:
https://qdmetrology.com/scraping-for-precision-machine-tools/

 Hand scraping is a technique for removing very small amounts of material from metal surfaces, almost exclusively various irons - either bearing surfaces on machines or reference surfaces on things like straightedges and iron surface plates.  Scraping makes a nicely ground surface uglier... but better.  It is used to accomplish two things: to make the surface even flatter than can be achieved with a surface grinder, and to improve oil retention on the surface, which is super important on machine ways.

fig. Uno
This is oil pattern scraping, performed as the finishing
step. Scraping done in the roughing step appears less regular.

 The amount removed is typically in the range of 0.0002" to 0.001" (about 5 to 25 microns) depending on the operator's technique and whether they are at the roughing or the finishing stage. 

 The amount of material removed is not uniform across the surface; rather, the process targets and removes only the microscopic high spots, which are identified by lightly pressing the work against a reference surface which has been coated with a very thin layer of a marking compound, typically Prussian Blue.  This is called "printing" the work. 

 

 The high spots pick up the marking compound, the low spots don't, and you scrape away the blue parts and repeat until it shows lots of small light marks of roughly equal size and shape, all over.

 Fig. 1 shows a nice job of Moore pattern scraping, applied from several directions.  This is on the saddle ways on the knee of my Bridgeport.

fig. OMG NO

 
 And, looking out of the left side of the bus, we have the disastrous "scraping" that was done on the cross-slide of my '70s Chinesium lathe.  It is to cringe.  Would love to have met the pendejo responsible, (so I could punch him in the nose) but he's long dead, what can ya do?

 The marks are random and VERY deep, and it's not clear whether they were made with a power scraper  with a dull blade... or a cold chisel!  Those gouges along the edge... I dunno man, I just don't know.  Es mui mal.

 Either way, they are unreasonably bad and the ways seem to be as flat as the Devil's Golf Course, but I haven't got a reference straightedge nor the scratch to buy one in order to verify that belief.

 I think I'm at the point of actively wanting a roughly 18in "master" straightedge.  All scraped and ready to go, they are fantastically expensive - a few hundred for a suspect Chinesium one, north of six hundred simoleons for a freedom casting.  Sadly, Charlie's Secondhand Tools is closing, so I'll have no chance of finding one there.  I don't patronize the flea market, it's all stolen goods.  I'll have to keep an eye out for property disposal auctions and such like, and of course, eBay, cuz ya never know.

 There are folks who sell straightedge castings, fully stress-relieved, for you to grind, lap, or scrape to your heart's delight.  And even those are kinda spendy because they're cast in small batches, then there's labor and electricity for the heat soak.  The iron is thrown in for free.  ;)

 

 This might be my best option.

There is an old colloquial expression for what I'm attempting to do with this lathe; "making a silk purse out of a sow's ear".  The inference is that it can't be done, because of what you started with. :/

You have failed me for the last time, Bosun Gibbs

  So I want to bore out some ridges inside my lathe spindle...  This requires my boring bar to reach in from four and half to six and a half inches (and I hope no further) inside.  I recently welded an extension onto my 3/4in HSS boring bar for just this purpose.

 Now an old bogeyman raises his head again: the cross slide of my lathe, as well as the compound rest, have more slop in them than a soup kitchen. This means that when cutting forces try to push the bar down at the business end, the end with the cutter moves instead of staying still.  At best it will chatter and give lousy surface finish, at worst it'll break something.  The longer a bar one tries to use, the worse this situation gets.  The only real answer I know of is to take lighter cuts and have as stiff a setup as the machine is capable of.

 Various repair attempts have been made to both slides, by the former owners and by me with help from a smarter friend, and while these attempts did improve one adjustment screw at one end of one gib, it hasn't done much for the fact that both gibs are bent and badly finished on their bearing surfaces, and the dovetails were left in a horrid state after the most-recent cut was made (at the factory?) and they are now, not to put too fine a point on it, lumpy.  It it likely that stoning the lumpy dovetails will not make them worse...  There is certainly no worthwhile scraping on them now to worry about unintentionally removing.  The bearing surfaces are a dog's breakfast!  Flat or not flat, without scraping it will not hold an oil film worth a damn...*

The bottom of the cross slide showing horrid bearing
surface, worn end of gib, and worn gib screw.

 There are high spots on the body of the cross slide that I will try stoning as well.  I might even attack those dovetails too (with a triangular stone) because the factory's idea of scraping is just dismal; I could literally do better... if I had a straightedge to print them with.  I could print the bottom bearing surface of the cross-slide since it is coplanar (ish) with the bottom surface, just hangin' out in the breeze, no way covers or any of that fancy stuff.

  aside: In this blog, I routinely use the word 'print' to refer
to two different things; 3D printing, and also the practice of
"printing" and hand scraping precision surfaces of machines.
In this post, I am referring to the latter definition. 
There are excellent tutorials on printing and scraping on
YouTube, I especially recommend RobRenz & Keith Rucker.
It is worth noting that no automated process or machine is
capable of producing a surface as flat as hand-scraping, and
scraped surfaces hold oil far better than ground surfaces.

 

shows the gib screw hack engaged
with a notch filed into the gib

  In fact, it occurs to me that the cross slide and its mating surfaces on the carriage would be a good place to learn / practice hand scraping.  I can't make these surfaces any worse than they are now, and the cross slide is not too heavy to repeatedly put on and take off the plate between scraping.  But the carriage is much too big to come off the lathe, nor can it be printed on a plate anyway; I would need a proper (scraped or lapped flat) straight edge to print it in place, and I haven't scored one of those yet.

 Okay so let's talk about the gibs.  The dovetail ways on machine tools are adjustable in order to remove nearly all slack and off-axis motions.  The adjustment is made with a tapered insert called a "gib" which acts as one side of the dovetail pair.  When it is moved back and forth lengthwise, the taper causes the dovetails to get tighter or looser.

 The gibs are adjusted with, and held in the correct position by, gib screws, which have very big heads and a sharp corner at the base of the head, intended to pick up a notch ground into the gib.

adjustment screw hole and worn gib-
it's obvious there is very little metal
under the screw head

 But the screw and the notch can wear, especially if not checked regularly for tightness.  If the gib is allowed to move back and forth with the slide, it is disastrous not only to the work you're trying to do, but also to the gib and adjustment screws themselves.

 And when the gib and screw get worn, attempting to tighten it only pushes the end of the gib outward against the slide, jamming motion or at least making it hard to move near the end of travel...

  Now, in theory I could fill in the notch with some weld or braze, regrind the notch, and replace the screw.  And I'd have something just as bad as what originally shipped.  Well who wants that?  That level of quality is what brought me to this place...

original gib screw and replacement hack

 Some years (decades?) ago, a friend showed me how to do a better job, replacing the gib screw with a hollow tube having an angled finger brazed onto it, and projecting well out of the screw's counterbore and much further into the gib's slot than the original screw head.

 The gib gets an angled notch to receive the angled end of the finger.  This angle will pull the end of the gib closer to its slot when tightened.  Tightening is accomplished with a SHCS that sits inside the hollow tube.  This replaces the wide head of the original screw with something that more positively engages the end of the gib.

  So one of my tasks is to make three more of those gimmicks, one for the cross slide, and both screws in the compound rest. 

 I have a little more to say about the bearing surfaces of the cross slide and the nature of scraping, but this post is already long enough, I think I'll put that in the next post... maybe even later today.

__
* I'm aware that modern, high-accuracy CNC machines use ground ways with no hand scraping on them.  They can do that because they have high-accuracy custom-made surface grinders AND because the resulting machines use pressurized lubrication systems which maintain the oil film (and also use a fair bit of oil in a day).

statuses

 I still have not finished the outriggers for the hoist cart, mostly due to lack of money.  I need to go to the scrap yard and buy steel after December or January.

 The cross-slide and compound rest of the lathe are still disassembled and spread across my workbench, waiting for me to be sufficiently clear-headed and non-achy to work on the gibs and a new end-screw doodad, with confidence.  Grateful to the fellow Mad Scientist who showed me the way with the gib screw improvement / repair gimmick.

 I have mice in the shop / garage.  They are wary little bastards and have ignored some bait, but I got one yesterday and I will get them all eventually.  'Tis the season, annoyingly.  The very old design of Victor mouse traps - the ones with the metal bait pan, NOT the plastic bait pan, work best; they are humane, and the body won't poison wildlife / cats / raccoons if they somehow get at it, or when I put traps outdoors... poison is super-effective at getting them all, but aside from the above problem, I dislike the notion of dead rats hiding an moldering in walls or in inaccessible crannies.  So traps it is.  Also, F live traps, the world is not short on mice, and I'm not going to the effort to release them in a park every time I catch one.

 The bikes are out of the shop now, for better or for worse.  Mostly better.  Can't ride in the snow and cold anyway.  Nice to be able to use the mill again. >_>

 I'm pondering building a larger discone antenna, but a used Diamond or Tramm might be a better use of my time/money.  The first one I built isn't terrible, but is just as wildly uneven in VSWR as most other discones.  I think I have blogged about DIY antennas here before, not sure.

 It turns out, if you make the active element top AND the ground radials from a bunch of resistors in series, you can have nearly flat response from a discone in both reeive and transmit.  You can also build it such that it will handle 200w xmit, too. 

 Granted, the discone is a "unity gain" antenna with no directionality, but I like them for the wide bandwidth.  Still pondering this project, and the two options.

 Also radio, I really need to get more / larger air variable caps for my MF/LF loop.  Right now, it only goes down to a little below the AM broadcast band.  I need to go to more ham swap meets I suppose.

 And some time this winter, I intend to attack the 3D printer which is currently inop from upgrades, oops.  Fixable, just needs some time and patience and hopefully no expensive parts... winter is the best time for 3D printing for me, because we do not have air conditioning, we have an evap cooler in the summer, which means very high humidity.  Despite having a full tent for the printer, humidity is problematic and bad for any exposed filaments.

 I think that's it for all the open projects, not counting neglected art projects. :(
I still owe this blog more entries concerning the lathe work I'm doing, I'll get to those soon, I think.

work halted due to weather

item: Our weather has been cold lately.  It is very cold today. 
item: The shop is not insulated to much of any degree, so it is expensive to heat.
item: I can't work in cold temps due to stiffness (lack of dexterity == danger in a metal shop) and arthritis pain.

When we get a day or two of warmish weather, work will resume.
We really need to keep that thermostat low.

deck the table with vows of folly

...or something like that.  Happy belated Woolfnoot, by the way.

 Today's avoidance of writing the various lathe articles consists of me writing up this nice and easy and safe article about tweaking my drill press.

 Once upon a time, I bought a Craptsman floor-mount drill press, cheap.  And while it is more evidence that 'You get what you pay for', it has been much better than no drill press, especially for the money... and at the time, I did not know a knee mill was about to fall into my lap.

 One of the things that bugged me was that - somehow - the table rim and the table were at different heights on opposing corners, despite the surface evidence that the whole shmear had been decked with a five inch fly cutter while they rotated it with the mounting spud.  As should be, in fact, but why then was it not flat?  On one corner, the rim was higher than the table, on the opposing corner it was lower than the table, and on the other two corners, the table and rim were about the same.  I mean, how do you even do that by sweeping it with a five inch fly cutter??  And the table itself had a curve in it. (aha, your first clue, padawan)

 I decided long ago that this was some kinda post-post-cooling, precipitation-having, slow crystal forming, meta-magi-lurgical phenomenon, and tried to work around it.  But man, it has bugged me ever since.

 The other day, as I was feeling The Urge (to putter / work in the shop) but not feeling the confidence to work on the lathe gibs (which are non-replaceable) my eyes alighted on that drill press table, and I decided it was time to Do Something About It.  It occurred to me that if there was some kind of thermal stress issue in the casting at the plant, then by now it should be as relieved as its gonna get without a long time in an oven, so I theorize that if I only skim off a few thousandths (which is all that seems to be required) any remaining internal strain shouldn't warp it too much.

Honestly I did not like this setup, but it worked

 Long story short, I fixtured the table in the mill vise using the mounting spud between two vee blocks, because we very much want the table surface square to the mounting spud, otherwise if you rotate the table, things go up and down, no bueno.

 I milled it with a cheap Enco (RIP) fly cutter holding a cheap Enco lathe holder for triangular inserts.

 The inserts were, amusingly, TiN coated, which really only helps you when machining aluminum, and a little on gummy stainless alloys.  But the profile on them is best for ferrous alloys not aluminum.  D'oh!

 In any case, HSS was a non-starter because cast iron is so abrasive.  All the speeds and feeds charts have blank spots or "not recommended".  So it was a cheap triangle insert tool or nuthin'.  The ones I had available have a quite rounded nose, and this seems to have contributed to a nice surface finish.

Enco's finest!

 I used 300 SFM based loosely on the chart and not having the foggiest what sort of iron I am dealing with.  As usual for me, cutting fluid was Tap Magic™ applied to work with brush.  I really oughta get a misting system.  With all my extra money.

 The first pass fixed the high spot but failed to pick up the low spot.  I rotated the table in the jig on its mounting spud by 90º and made a second pass which covered all but a tiny low spot on one rim edge that did not seem worth re-running the whole table again to address.

 The widest swath I could get on the fly cutter was about four inches plus a bit, the twelve inch table thus requiring three cuts.  Or is that three passes per cut?  Passes?  Cuts?  Don't ask me, I am not a real machinist.  In this case, 300 SFM worked out to nearly 300 RPM.  Ugh; that is not fast.  And it meant that, if I wanted a nice surface finish, the feed rate had to be glacial.  Ahem.  I don't have a working table feed.   So between arthritis in hands and shoulder joints and stuff, I had to take breaks after each cut.  So three breaks per pass, six breaks total, a meal break, and a couple of hours total to slowly turn a crank handle do the machining, not counting maybe ten minutes to make and check the setup.  The surface turned out okay-ish, but because I cannot turn the crank as smoothly and consistently as a power table feed would, the tool marks are very uneven.  "That'll do, Pig, that'll do."

the job was painfully slow, but at least simple

 

Surface finish?  Single point cutters for the win!

 Between both passes, about .009in was removed.  The one high corner might have lost eleven or twelve.  Afterward, the overlap between cuts passed the fingernail test.
 

 Mucking about with the indicator on the mill showed the surface flatness to be no worse than the mill's table ways at least. ;)

 I also put it down on the plate for yucks, and I could not get a 0.0015 feeler gauge under the edge anywhere, so that's a good sign.  I think it's good enough for a drill press table.

 Oh, I could go around and break all the newly-machined corners in the slots and groove - somehow - but none of them seemed eager to cut my finger so I'll just let time take care of that.  I did do a quick stoning on the outer edge since it is exposed.  It's at the end of my bench and I walk past it all the time.

 Anyway, I think we're done here.

Oh, PS: What do you suppose is the purpose of that peripheral groove?  It's not to catch coolant, since the T-slots are wide open.  And what purchaser of a Craptsman drill press is gonna run coolant on it?  Mr. Nobody, that's who.  Something for woodworking that I don't know about?  If anyone has any ideas, let me know.  It might just be a fake feature, making the table more of a "mug's eyeful".  I can't help wondering if I'm ignorant of something though.

The Shadetree Cyberpunk

  There once was a time when folks wrenched on their own cars either to get it to do something the factory hadn't intended (now we call this 'hacking') or just to do the work that a shop would normally do, but which they could not afford.  Collectively, these were often referred to as "shadetree mechanics", since they didn't have a nice cool (hah! presumably) shop to work in, but used a tree for shade whilst working.

 This has changed somewhat.  Before you can repair, troubleshoot, hack, or hot-rod a modern vehicle, you must first arrange an audience with, and subsequently establish diplomatic ties with, the vehicle's crown prince computer and all umpteen of his lesser computers.

 In reality, we only need communicate with the seneschal, who speaks the diplomatic language 'OBD-II', specifically the BMW/Mini dialect.  The seneschal then communicates with the other systems over, get this, four different networks gimmeabreak.

 All of this just so I can get all of those computers to tell me what they think is wrong with their house.

 So a new tool is needed for the shop; a software app and an OBVD-to-BT hardware link.  And they might not even be right - Prince ECU listens to each of his advisors and reports to his seneschal what he thinks is going on.  But they could all be seeing 'the elephant' from a different point of view, leading the seneschal to lie, technically, if unintentionally.  And I have to pay to see this play out.  And then backcheck whatever its story is.

 Ya know, I liked wrenching on cars when they had a carburetor and a very simple electromechanical ignition system.  In the 1970s, you checked to see if it had fuel, and if it had spark, and if it had compression.  If you couldn't troubleshoot an engine in an hour, and probably fix it, either you weren't all that good, or it was very badly broken indeed.

 These days, you need the equivalent of an Electrical Engineering degree just to troubleshoot the vehicle, well before you get into turning wrenches.  Fortunately, I have that.  I just resent and complain about using it.  And also about the cost of admission (oh how many scanners have I owned so far??) because I am a broke-ass retiree and I have no disposable income.

I'm telling you, many parts of the future are really cool, while other parts are really stupid.

I hope I don't need a round part or feature any time soon

 Er, hang on, don't I need two brand-new round parts to fix both gibs?

So I decided to dig into the problem of slop in the cross slide and compound rest of the lathe...

There's... a lot going on here.  I wish there wasn't.
The cross-slide isn't even in this shot, it was over on the plate.

  I'm putting off writing it all up because for one thing, the story isn't finished yet and for another, there's just so much to talk about, by which I mean, there's just so much wrong with the damned thing.  And most of these problems should have been addressed, by me, soon after I brought the thing home.  Had I done so, later work with it would have been much easier because it would be a much stiffer tool.  It's important to keep your tools stiff.

 My excuse is that I got busy and so did the lathe, as I had things to do, and none of those things, preferentially, was tearing my lathe apart and doing fiddly things with it.  This excuse worked for nearly twenty years.  But now, it's really buggin' me, man!

 Note that it's entirely possible to do good, close-tolerance work on a clapped-out 1970s SIEG PoS... _if_ you are patient.

 The timing is perfect since I'm no longer building a railgun laboratory, and my alleged job shop has no customers.

 More as it happens, I guess.  My back is improving but still hurts, so work continues to be slow.

to everything, turn, turn, turn, there is a season, turn, turn, turn

I'm not religious, just well-read. 😛

 I am slowly working on a big post about the lathe, concerning the gibs of the cross-slide and of the compound rest.  It is a sad affair.

 Writing anything right now is slow, because I'm dealing with ferocious back pain.  I hate painkillers which kinda sucks for me, and I avoid NSAIDs and similar just because I like my liver healthy, and side effects like severe tinnitus.  <_<

 I've lots I'd like to say about lots of things, but I'd rather take another hot bath.

 Have some foreshadowing.  I'm genuinely sorry that it isn't level:

for a hint, that's granite in the fuzzy foreground/bottom

Dear Metalworking Toolmakers: you are wrong about taps. (and so was I - mostly because of you)

 First, let me explain.  No, there is too much.  Let me sum up.

 I am making two handles which require a very deep (≈1.5in) threaded (10-32) hole.  The material is some unknown high strength aluminum.  One of the ways I have broken taps in the past is tapping holes not even that deep, although that was also a smaller hole/thread, so I am slightly charry about this task.

 So I go looking for the deep-thread-making kinda tap, which I thought, due to a variety of sources, was called a "plug tap".  Bzzt, wrong!  Searching for same constantly turned up taps close to what I was looking for, but the results were spotty.

 So it turns out, "plug" refers only to the taper on the end of the tap.  While it is quite common for extra-long taps to have a plug taper on the end, it does not define or name the overall tap; just that taper.

 Extra-long taps, which I am seeking, come in two varieties; Extension taps, and Pulley taps.

 Extension taps have a shank which is smaller than the thread-cutting part, so the shank itself can pass into the threaded hole the tap is making. The downside is that the shank is much smaller and therefore easier to break.

 Pulley taps have a larger shank, making the shank less likely to break.  They are designed for tasks like pulley hubs, where the threaded depth is unlikely to be greater than the thread length of the tap.

 That's it, that's the only difference between those two; shank size relative to the thread's major diameter.

 And while the plug taper is the most common taper found on long taps, they can have any taper, any number of flutes, any of a variety of coatings, and so on.

 So, manufacturers and resellers: fix yer damn web stores. 😛

The Unbelievable True Story Of The Amazing Vise Jaw Caps*

*(title inspired by YouTube - I should add "Shocking!")

 I use vise jaw caps to keep from indenting my workpieces with the diamond pattern in the hardened steel factory jaws.  Clearly, I should probably make a "regular" set of vise jaws which are soft low carbon steel.

 But see, sometimes I need something softer than that because the work is really soft or highly polished, is made of plutonium, or what-have-you. 😇

 The time-tested solution for this is to have a variety of soft jaws for not only bench vises but even for lathe chucks, if your chuck supports removable jaws. (mine does)

I have on hand, courtesy of Hoarder Fright Casinos, a pair of soft jaws in very soft urethane, and a pair in some variety of nearly-dead-soft-but-not-quite-pure aluminum.  Both were super-duper-cheap.  Both are equipped with ceramic magnets to hold them against the vise jaws. All the magnets are broken.  I half-assed a repair on the aluminum caps with JB Weld, but ceramic magnets are brittle...

 I kept telling myself I'd make some from brass, when I had the right stock...

 Now, years ago, I salvaged an L-shaped strip of what I assumed was brass, from the side of an unsalvageable rusted and worn chinese magnetic vise, where it was used as a non-magnetic "stop" for the work.
And then I forgot about it for years, even when I was thinking about jaw covers for the bench vise, because I have been senile since I was fifteen years old...

 I unforgot it this morning and dug it out.  Not only is the cross-section shape exactly the desired shape for vise jaw caps, saving me a boat-load of machining time, but the cross-sectional dimensions are perfect for my vise too! 

 Very well, let's make some vise jaw caps.  Uh, gee whiz, this stuff sure cuts slowly and feels "slippery" with a brand new, sharp hacksaw blade.  Whatever it is, it is not "brass"; it's either phosphor or Al-Si bronze, which is very nice (but hard) stuff indeed. (I mean 'hard' relative to cupric alloys, not 'hard' like tempered steel - completely different scales!)

 Well now I have a quandary; the important property of the jaws is that they be soft - or at least, softer than steel.  And while this stuff is much harder than any variety of brass, it still meets that qualification.  But it will put dents in copper(s) and brasses.

 Fine then, I am resolved to make (later, not soon) a pair of thin copper soft jaws which simple wrap around the vise jaws.  And so I will also make these...

 I want these to be held in place with magnets like most bench vise soft jaws.  Ideally, I'd mill out a couple of pockets in each cap, stick some thin rare earth (NIB) magnet disks in (top surface below surface of cap) with a dab of JB Weld (magnetic!) under and over the magnet disk.  Then, when cured, mill off the top coat of the JB weld so it's flush with the surrounding bronze, and the magnet is not exposed.  That should keep the magnet from breaking, in theory.  I'm not even sure rare earth (NIB) magnet disks are available thin enough, but I'm about to find out. They absolutely do.  0.5mm!

 I considered flexible magnetic tape, but it's not very strong, and it wears and breaks and gouges easily.

 So I nearly finished one jaw this morning before I'd finished my first cup of coffee.  All I had to do was cut it and square up the ends on the mill, and break the edges and corners.  This material creates chips which are half powder, and half tiny chip, indicative of a hardish material.  It machines fine with HSS and brushed-on Tap Magic, and a surprisingly slow SFM; like 100-200.  So I can't tell which kind of high-strength bronze it is, but it makes sense in the original application that someone wanted something that wasn't magnetic, but also wouldn't wear much, at all, ever, because it was for a surface grinder, where tenths (1/10,000th) matter.  It's not as soft as brass, but it sure won't get dinged or bent easily!

 The two circles show the approximate location and size of the bored / milled pockets for the magnets.  The lines are where the pockets for the jaw mounting screws end, so I make sure to miss them.

 Today's efforts were brought to an abrupt close by severe unexpected back pain, so not much more work likely to be done out there today unless that improves.  Ow.