SE anode cap: Inconsistent connection with inside of CRT?

[Boctor]

I have one annoying quirk left in my reborn SE/30: It seems the anode cap prongs don't always hold a perfect connection inside the hole in the CRT. Subtle bumps or the usual thermal expansion/contraction are sometimes enough to upset the connection, introducing a sort of visual noise where the gun is dimmer on some rows than others. Not a collapse, not noise like snow, and not the usual picture distortion. No noises, no arcing, nothing violent whatsoever, just the tube seemingly not getting enough current. It goes away when I very gently nudge the rubber of the cap. It seems that this one connection is just marginal, and I'm worried whether this is something that can be fixed, or if there is some kind of residue or grime in the inlet of the CRT. Of course, I'm not going to take this as a sign to go haphazardly spraying isopropyl and deoxit, and risk making everything worse.

My AB and neck board are cleaned, reflowed, recapped, and even re-potted with the appropriate neutral cure silicone to match the stock AB. The power supply is the modular Meanwell one and outputs steady, correct voltages with more than sufficient current. The logic board is a brand new Reloaded build. The CRT itself is in very good shape, I can't even see a menu bar burnt in yet and I think it was barely used or NOS when I replaced it around 2017. Nothing being changed about any of those affects the wonky per-row brightness of the gun, so I am almost dead certain it's the anode cap being flaky.

Could the red wire from the flyback be getting interference from somewhere else, like if it's too close to the yoke? That does not seem probable to me as I see them in proximity in photos all the time, but I'd appreciate expert advice from someone who's more familiar with CRTs.

 

[Boctor]

Discharged and thoroughly cleaned out the terminals on both the tube and the screw-in clip itself. The wire in there should be going through the pronged piece's hole, making it clamp tightly between the screw and the clip, right? After working on it, it still seems defective sometimes, until I give the side of the bucket a good slap. I can get it to come back more easily when I turn the brightness all the way up, since then the gun needs more energy. I noticed some dark green stuff on the thin wire inside of the suction cup while cleaning, but it looked more like paint or coating than copper corrosion to me.

I'm almost sure the issue is either the wire inside the rubber cup being frayed/corroded or somehow otherwise marginal, or the screw not squeezing it into the pronged part tightly enough. Like I said before, using an insulated tool to very gently nudge that cup is the only thing that immediately makes the wonky per-row brightness either come back in full force or disappear completely.

I have a spare analog board for parts somewhere. I think one of the ferrite rods on that one got damaged, but everything else is good. I may take a closer look at it for reference, or swap parts if there's no way to salvage this one, though I'd find that hard to believe. The wire's right there, I can see it through the rubber.

 

[luRaichu]

Yeah intermittent HV is not something I've ever dealt with on a picture jug but at this point I would just try to swap in a different flyback if I couldn't figure out how to redo the connection on the anode cap.
No, your red HV wire can't pick up interference. Kilovolts Don't Care

 

[Boctor]

Yeah, I fooled with the position of the wire between tests to make sure since it seemed implausible, thanks for confirming. Strangely, I went back inside to correct the height on the analog board, since I'd set the geometry slightly wrong years ago and kept forgetting to set it back. After the cleaning I had some slight weirdness that hasn't come back since I jiggled the cap a bit, but after the vertical adjustment I can't get it to go wonky anymore unless I use something insulated to really push down on it.

Obviously it's still only as "fixed" as someone claiming they corrected something via compulsive Windows restarts, but I'll do burn-in tests and use the Mac as normal in the meantime to see what's up with it. If I can still get it to misbehave just by using the computer normally, I'll take a closer look at my spare analog board. On the bright side, no arcing/corona discharges ever. There's always LCD conversions as a last resort, but I would never give up the fight to preserve a perfectly good CRT. This SE chassis has been in my possession for so long (2007?) that I remember getting it for a reasonable price.

Here's what the artifacts looked like, though this was taken when I forced it to behave as badly as possible by nudging the anode cap. When it happened organically, it'd usually be so slight that you'd only see two zones on the picture where the gun's brightness went wonky for a few lines, with the middle staying mostly normal. The neck board, analog board, and anything else I was okay with gently prodding during operation did not cause any sort of changes to the picture, it was always that darn screw-in and terminal in the suction cup.

 

[Boctor]

This issue has not returned, and I no longer see the squiggle/sinewave of the gun when the CRT is powered down. However, I am still noticing what a few others have in the past: Bloom/bulging, where the horizontal raster widens when something white is on screen, which is more intense when the brightness pot is turned up, but goes away when it's slightly dimmed (I usually use my Mac with the brightness slightly lower than maximum anyway so I rarely saw it. I don't know anyone local with a reference Mac to check mine against, either!)

I'm considering whether my flyback transformer could be bad, my CRT could simply be a bit worn out from age, or the cutoff should just be turned down as a simple fix, as this doesn't really harm the machine. I have complete and working backup analog/neck boards if needed. Turning down the cutoff seems the most sensible, but there are no visible (to my eyes) retrace lines, so I'm worried it may be something worse or more complicated.

Figures I'd start noticing imperfections in other areas after I finally reloaded my LB, but that's what makes a project a project. I'd just reach for my accelerated IIci if I wanted zero maintenance!