the bizarre camera problem

<this is not a Moving Picture>

 So I've traced the camera problem to my temperature control system.  If I turn it off, the problem goes away.  This is not unreasonable for a suspect, as SSR switching is notoriously fast and therefore nasty.

 Neither the R-Pi nor the printer is connected to the temp control system in any way, so this could be either radiated crap or coming in to the devices via the common power strip.  Okay, but why the hell does the connection between R-Pi and printer matter??   o_0

 I've honestly considered that the aluminum coating on the inside of the printer tent could be acting like an antenna or at least a big capacitanc, coupling between the wires for the printer and Pi, and the wires for the temp control fan and heater.  Hmph.

 What I want to do for a solution is to find an unused Tripp-Lite™ Isobar™ unit and plug the temp control system into one channel and the printer and Pi into the other...

 But since I'm fairly certain all of mine are in use (uh, I could give up the stereo?  or um... yeah, not) and even used ones are not really in my budget, (what budget?) the probably-optimum solution might have to wait.

 The next thing to try is to get a bunch of 77-ish split ferrites or a couple of big toroids I can loop power cords through.  If that doesn't work, I'll look for a used 3-channel Iso-bar unit and plug the three offending items into that rather than all together on the same outlet strip as everything else.

 I'm also going to consider installing shielded & grounded cable between controller and heating element, since that is the thing which is constantly getting switched on and off.  The wire gauge can be tiny because 120V and small wattage.  Existing wire is twisted a lot, but apparently that's not enough to stop.  The might be a common mode thing or radiate RFI.

 PID + PWM is really good for controlling an SSR and heating element, but unless it is zero-crossing switched - and this one apparently is not - they are kinda notorious for generating noise.  I had problems with this professionally maybe a hundred years ago with process controllers that drove SSRs, but I plumb forgot about that until now.

 I wonder whether I could get away with putting say, a 0.01µF capacitor directly across the load terminals of the (fake, chinesium) SSR I used.  Might it destabilize the thing somehow? It would help take the edge off (geddit) the switching transitions.  I gotta find the manual and see if there is a way to make the controller switch on zero crossing or not.  Problem is, I don't really keep any parts of any sort on hand any more.  There are some odds and ends lying about, but they are rarely useful.  And a former housemate stole a few things from me which I didn't catch until much later.

 Today I also swapped out the disastrous tangle of a very weird old EIA outlet strip for a normal one, much longer, without the outlet cables/splitters involved with the old one. O_O

 This is not terrible progress given I am also working on other things today, like the plumbing disaster in the shop toilet and a spot of house cleaning.  Since the Ritalin apparently stopped working nearly a decade ago (how is that even possible?) I am completely scatter-brained most days, trying to do too many things at once.