PVM partial vertical collapse (edit: problem is back...)

[jd213]

Starting having problems with my 20L5 last night. First major problems I've had with it after about 7 years of owning it.

- Problem first appeared when I turned it on last night
- Went away after a power cycle, then suddenly reappeared after about 30 minutes
- Vertical collapse at top of screen with a single white-ish line (can see some colors from game graphics so it's not solid white)
- Black, severely pincushioned area below white line, so maybe partial vertical collapse and partial horizontal collapse?
- Front panel buttons become nonresponsive, flicker, then disappear after the vertical collapse
- When flickering, the buttons get brighter than before, they were extremely dim when I first got the PVM
- Problem goes away after a power cycle, then returns after 5-10 minutes
- Occurs on both RGB and composite inputs

Since the problem isn't a total vertical or horizontal collapse, could this just be a capacitor problem? Maybe the power board? Figured I'd have to do a recap someday...

[tongshadow]

Definitely should check the main/power board. I worked on 3 L5 units and they all had bad capacitors in these areas (check the picture):
https://i.imgur.com/rN5HmK3.jpeg
That area gets so stupidly hot that the small capacitors are slowly destroyed overtime. The capacitors on the D-Board can also go bad, but it seems your case is power related. You'll need a good ESR meter to properly check them.

[jd213]

Awesome, thanks. Not looking forward to the disassembly and reassembly, but I can handle recapping. Hopefully I can get started sometime this week.

Figures that this happened just a few months after I stopped putting cooling fans on top of the case (I was like "eh, some people say they're not necessary so it'll be fine").

Guess I'll find a place to mount them internally since I have to work on it anyway...

[tongshadow]

These L5's get so hot that the plastic bracket holding the D board bends or, worse, breaks right away. Nothing that some JB Weld cant fix for sure, but for something that cost 1500$ back then I figured they would use more definite and reliable solutions.

[jd213]

Yeah, I thankfully have some experience repairing plastic toys and model kits, but hopefully the plastic doesn't become a problem in the first place.

[jd213]

Got Savon Pat's cap kit (had some of the caps I needed already but didn't feel like going through Digikey's caps, always seems to take me hours). I noticed on some old Reddit threads and arcade forums that he used to provide Chinese caps, but that seems to no longer be the case as the ones I got were mostly Nichicon with a few Rubycon. The kit seems to have most of the caps that tongshadow circled, but there were a couple that weren't in there so I'll replace those as well, pretty sure they're all ones I extras of.

Getting the boards out was slightly frustrating (especially the ground wires on the C board, even after spraying some silcone grease on them, was afraid my hand would slip and I'd break the yoke). C board spacer plastic broke but I see from Retrotech USA vids that it's not necessary.

D board bracket didn't break, but the D board itself wasn't screwed in. Seems that someone had serviced this PVM before, there's a some replaced caps, some bodges, and a couple added resistors (one on the D board at the top in the 4th photo one on the C board):














Hopefully I'll be able to do all the recapping tomorrow.

Would anyone recommend doing an internal fan mod? Looked around for info but pretty much only found people saying it's not necessary.
If internal modding would be too much trouble, I might just keep using the USB powered computer fans I previous had on the top to draw air out, and maybe put some on the back and/or side for drawing air in.

[tongshadow]

That board was a 100% sure serviced by an authorized Sony technician. They all use the same type of silicon glue to isolate or keep components in place, and also Elna/Nichicon caps. I dont think the issue is there as the monitor was working fine and the caps *look* fine (would still check though).

[jd213]

Thanks, I kinda guessed that was the case.

I've already replaced all the caps from Savon Pat's kit as well as all the ones in the areas you circled for me. I also noticed the other day that the 1000uf/180v cap on the G board was bulging a bit, so decided to replace that one as well, so have to wait for a replacement to come in. The old one was really hard to get out, but thankfully the vias seem fine and still have continuity.

If the problem isn't fixed after that, I'll start looking around for an ESR meter.

[jd213]

Finally put aside some time to finish recapping (was busy with work and also paranoid that I'd mess something up, so kept putting it off).

The PVM seemed to still have some kind of partial collapse when I first turned it on, but it displayed an image fine after I power cycled it once, and it seemed fine after a 2-hour gaming session (consisting of Rockman 2 and Castlevania, temporarily had it on push-up bars to keep it off the carpet until I move back onto my desk):



Unfortunately the front panel buttons are still as dim as before the recap, but I don't mind too much, so long as they don't burn out.

Now I just need to find a good way to keep the PVM cool. Maybe I'll try just using velcro strips to attach some of the USB-powered PC case fans I have lying around.

[Bassa-Bassa]

Sorry for the noob question again, but are recap kits for consumer Trinitron sets still available and how should I look for? I have a similar vertical collapse issue as pictured here and still kind of lost in my ignorance. The set has a really nice picture despite its size and age other than this, so it'd be a real shame if I had to dump it.

[jd213]

I think cap kits are generally only available for popular monitors. Shotgun replacement probably wouldn't be too bad for a consumer set.

You could look up the service manual, then look up all the electrolyte caps and buy them on Digikey or Mouser (former is generally cheaper in my experience). For the really big caps (more than 1000uf or 100v), it might be a good idea to make sure you get the right physical size. I had to re-order one for my PVM since it was too wide to fit. Retrotech USA has a video on buying caps, so maybe take a look at that.

[tongshadow]

Caps are always the weakest link on these PVM monitors, good job.
Anyone serious about vintage equipment repair should always have a good ESR Meter in hand.

[Bassa-Bassa]

Thanks for the reply, jd213, now I have a much better idea of what I'm facing at.

[jd213]

Caps are always the weakest link on these PVM monitors, good job.
Anyone serious about vintage equipment repair should always have a good ESR Meter in hand.

Yeah, was also thinking about getting a component tester one of these days, will have to look into it.

Thanks for the reply, jd213, now I have a much better idea of what I'm facing at.

No prob, let us know how it goes. Recapping isn't too hard with a decent desoldering gun like a Hakko, although the solder can give me trouble on some vias on Sony boards, I guess the ground planes can absorb the heat away from the Hakko or something.

There's also places like RetrotechUSA's Discord that have more experienced people willing to help, you might need to be in his Patreon to access it though, not sure.

[jd213]

Looks like I celebrated too soon, the problem showed up again after about 3 hours of gameplay last night:

https://drive.google.com/file/d/1VrvjTZ ... sp=sharing

Mostly the same as before, the picture partially collapses, and the front panel button lights flash between bright and dim. As shown in the video at the end, power cycling the PVM restores the picture, but it collapses again after 10-20 seconds.

So I'm guessing that the problem was capacitor-related, but it wasn't the underlying problem. So maybe some component is causing stress on one or more capacitors, and replacing them was only a temporary fix.

Any ideas what I should be looking at?

[tongshadow]

Does it look like this?
https://www.youtube.com/live/hXTItA2g1Ac

[jd213]

Ah, it does indeed look similar to what happens at https://youtu.be/hXTItA2g1Ac?t=4810 , mine doesn't shut off quite so fast though. Mine doesn't come back on by itself, either.

When I tested it again this morning, the collapse was there on the first power on, but cleared up after 1 power cycle. Left it running with the scroll test from the 240p Test Suite, and it seemed fine after about 10 minutes. So maybe it takes a little while for the failure to occur.

Will try conatcting Retrotech USA and RetroRGB about this, thanks.

[jd213]

Retrotech USA got back to me already, he thinks that it's likely one or more of the 3 ICs on the neckboard. Pat's cap kit only included 1 replacement, according to Retrotech USA the red one is likely to burn out first. Will give it a go, probably won't have the chance to work on it for at least a few days though.

[jd213]

Unfortunately replacing all 3 of the ICs on the 20L5 neckboard didn't resolve the issue, I'm still getting partial collapse after a few minutes.

Not sure what to do at this point, I can maybe try checking voltages and readings with a multimeter/oscilloscope. Maybe finally get a component tester. But any suggestions would be very welcome.

[jd213]

Here's another video of the collapse, maybe a problem with the vertical or horizontal IC?
https://drive.google.com/file/d/1o5eXMc ... T6nGv/view

[jd213]

Got a component tester + ESR meter on the way from ebay.

Retrotech USA suggested that it might be something like a voltage regulator or resistor around the horizontal or vertical ICs.

I'm guessing that one or more of the replaced capacitors has also gone bad due to something else failing and causing increased load.
(since the PVM appeared fixed and lasted for a couple hours after the recap, but then the problem re-appeared and is now coming back after only a minute or less after powering on)

[tongshadow]

I'm guessing that one or more of the replaced capacitors has also gone bad due to something else failing and causing increased load.
(since the PVM appeared fixed and lasted for a couple hours after the recap, but then the problem re-appeared and is now coming back after only a minute or less after powering on)

You have to be sure, so measure the new capacitors you put it and see if they're fine. Maybe you got bad new capacitor, so rule out this possibility before moving on.

[jd213]

Right, will definitely test any capacitors I take out and any I plan on putting in.

Will probably do the same for any recap job I do from now on, doesn't look like it will take much more time.

I also still have the old caps from the 20L5. Didn't label them, but I'll first try testing them to see if any of them were bad.

[jd213]

Tested my old caps with the TC1 tester. They all seem mostly OK. The biggest one I removed, a 1000uf/180v one (which wasn't in Pat's cap kit but was bulging slightly so I decided to replace it), read 842uf with 0.6% Vloss and 0.14ohm ESR, so still within 20% tolerance but getting pretty low.

Some of caps had slightly high ESR, but maybe not bad enough to cause a problem? I've saw an instructables.com article that said that ESR needs to be higher for ~10uf caps used for decoupling around voltage regulators, so perhaps these are still within normal range.

Will try desoldering resistors and voltage regulators, then test them next.

Here's all the readings I got (original rating = read uf, Vloss, and ESR) from the old caps if anyone is curious:

1000uf/180v = 842uf 0.6% 0.14ohm
33uf/160v = 40uf 1.0% 0.13ohm
2.2uf/250v = 2.339uf 0.6% 0.31kohm
1000uf/25v = 908uf 2.1% 0.03ohm
1000uf/16v = 939.9uf 3.1% 0.26ohm
47uf/25v = 51.1uf no Vloss shown 3.5ohm
100uf/50v = 95.78uf 0.8% 0.42ohm
100uf/50v = 95.43uf 1.1% 0.59ohm
100uf/50v = 95.58uf 1.1% 0.66ohm
10uf/50v = 11.06uf 2.7% 1.2ohm
47uf/50v = 42.7uf 2.0% 0.48ohm
470uf/16v = 421.1uf 1.8% 0.24ohm
470uf/16v = 380.1uf 3.4% 0.53ohm
100uf/15v = 93.93uf 4.9% 2.2ohm
100uf/25v = 89.64uf 2.5% 1.3ohm
100uf/25v = 94.98uf 2.2% 0.73ohm
100uf/25v = 89.64uf 2.3% 1.3ohm
10uf/50v = 10.16uf 4.1% 4.4ohm
10uf/50v = 10.74uf 2.0% 1.2ohm
10uf/50v = 10.83uf 2.1% 1.2ohm
10uf/50v = 10.96uf 1.7% 0.63ohm
3.3uf/50v = 3.299uf 4.4% 5.2ohm
22uf/35v = 21.76uf 0.6% 1.7ohm
100uf/16v = 88.98uf 3.5% 1.2ohm
100uf/16v = 92.95uf 2.6% 0.79ohm
100uf/16v = 87.82uf 5.9% 3.1ohm
1uf/50v = 1.009uf 1.6% 6.7ohm
22uf/200v = 25.13uf 1.1% 0.59ohm
1000uf/25v = 920.7uf 1.9% 0.16ohm
1000uf/16v = 939.6uf 3.4% 0.24ohm
neckboard resistor: 39.5ohm

[Josh128]

If you can get it running with the case open, use an insulated / plastic stick to tap around the vertical deflection circuitry and/or all over the mainboard and the neckboard to see if you can affect the picture-- if so, its a bad solder joint or joints. Bad joints can get better or worse as things heat up, but probably the MOST common failure in CRTs of this age, especially ones that have been moved around.

[jd213]

Hmm, could be a possibility. I didn't see any joints that looked bad when I inspected the boards with my magnifying glasses before, but there might've been something I missed or something that's hard to see visually.

Doesn't look like getting the boards into service position is too hard going by the service manual, but I'm not sure if service position shows a picture of any kind, as it looks like the main board is only connected by one or two cables, so I'll maybe try tapping around on the parts side of the main board first. Thanks!

[tongshadow]

The main board is double layered, and in my experience they rarely, if ever, have cold solder joints.

[jd213]

I found that a voltage regulator, IC2505 (BA05T) on the D board isn't properly regulating voltage, its output is 10V (when input 12V) when it should be 5V. Hopefully that's the only major problem I need to fix. Looking at the schematics, it seems that most components after this regulator wouldn't be affected by extra voltage, except maybe the TC74HC4538AF-TP2 chips (IC1503, IC1507) on the D1 board. These would be a pain to desolder and replace, so hopefully they're fine. Will order replacements just in case.

The 2 surface mount electrolytic caps (one on the D1 board, one on the D board) seem to be completely dead, they only measure about 50pf out of circuit. Some of the bigger caps on the G board measure a bit low in capacitance, still within 20% but may as well replace them. Going to test some more capacitors before I order replacements.

[jd213]

Finally had the chance to finish desoldering the remaining caps. Was a bit scared to desolder the big chungus 330uf/450v on the G board, but it fell right out when I used some off-brand Chipquik from Digikey. As noted in this thread: https://forum.digikey.com/t/can-t-find- ... or/19332/3
These terminal lead caps don't seem to be made anymore, so guess I'll have to use some quick disconnects or maybe try straightening out snap-in leads with pliers.

Seems like nearly all of the larger capacitance ones are down by 15% or more, with one or two down by slightly more than 20% (like C1550 on the D1 board reading 783uf when its nominal rating is 1000uf). Didn't find any that had shorted, though.

Not sure how many hours this PVM had on it when I got it in 2016, but might've been higher than I thought. The picture always looked great, but the front panel buttons did go dim right after I got it.

[tongshadow]

The picture always looked great, but the front panel buttons did go dim right after I got it.

That's gonna be the capacitor C1601 and others around the area near the speaker:
https://i.imgur.com/rN5HmK3.jpeg

Those small capacitors on the G Board often go bad due to the heat and being small (absorb more heat).