K4900 - picture shrank and faded out after recap

ccie38296

Well-known member

Donor 3 years: 2019, 2024-2025
Joined
Jul 29, 2019
Messages
785
Reaction score
923
Hello!
I have a K4900 4-pot that came in a Williams cabinet I bought a while ago for restoration. It did work when I bought it although the picture was washed out. I decided to recap the chassis while doing the other restoration work. I also tested the tube on my B&K and it tested very good so I did not use any of the other functions on the B&K.

I followed @mecha's website for reminders of things to reflow including the jumpers, neck socket, signal headers and ceramic resistors. I replaced all caps (including the filter cap) with a kit from AP&R. I double-checked all my work to ensure cap values and orientation were correct, and they were.

I connected my TPG and powered it for the smoke test and for a few seconds the image looked really good. Bright and crisp. Then, it started to shrink vertically, almost like the image was zooming out. Then it started to fade out. I cut power, let it sit a few seconds and powered up again. This time the image was again shrunken vertically and very dim.

When I shut it off this time I caught a whiff of hot component but couldn't isolate it. A few of the resistors in the vertical area like R312, R310, and R315 look like they might have gotten hot - there's some discoloration on the casings and/or the leads, but they all tested correct for their values on my DMM. And they're all in that corner of the board that clear does get hot, so maybe they were already like that.

I consulted the Randy Fromm flowchart for the section for shrunken image:

IMG_7597.jpeg

R313 is the big 470 Ohm vertically-mounted ceramic. I did reflow that one, but it is measuring right on the money on both the component and the solder side.

I confirmed that the caps were all oriented correctly when I installed them and just rechecked them all with the ESR meter and they all seem OK. For Q302 and Q303 I'm seeing 0.5v drop in diode mode from B->C and B->E.

I didn't observe any actual smoke and no other components seem to be visibly damaged. I definitely smelled something getting hot when it was on though.

Any pointers on what to check next?
 
I also just remeasured the yoke since the flowchart mentioned a vertical deflection problem. Yoke measures 40.5 Ohm on Yellow-Green and 3.4 Ohm on Red-Blue
 
Well, I think I probably found the problem. Dammit. Seems double-checking polarity wasn't enough.

IMG_7612.jpeg

I powered it back on and the image actually looked fairly stable but after 45 seconds I saw a whiff of smoke and spotted the cap bulging.

Guess I'm raiding the other 4900 cap kit I had on hand for the moment. Will report back after fixing my boneheaded error. Hopefully nothing else got cooked in the process. 🤞
 
Well, I think I probably found the problem. Dammit. Seems double-checking polarity wasn't enough.

View attachment 860780

I powered it back on and the image actually looked fairly stable but after 45 seconds I saw a whiff of smoke and spotted the cap bulging.

Guess I'm raiding the other 4900 cap kit I had on hand for the moment. Will report back after fixing my boneheaded error. Hopefully nothing else got cooked in the process. 🤞
yes I strongly encourage notating which way the polarity marking faces on each cap you pull out and the microfarad and voltage values. and if you're easily distracted at times like I can be then it's probably best to do surgery by yourself. unless you become truly enlightened like I have in the last 15 years and can manage both with absolute precision. lol

also word to the wise for anyone reading this, never under any circumstance think of trying a shortcut of yoinking all the caps at once. always do them one at a time.
 
yes I strongly encourage notating which way the polarity marking faces on each cap you pull out and the microfarad and voltage values. and if you're easily distracted at times like I can be then it's probably best to do surgery by yourself. unless you become truly enlightened like I have in the last 15 years and can manage both with absolute precision. lol

also word to the wise for anyone reading this, never under any circumstance think of trying a shortcut of yoinking all the caps at once. always do them one at a time.
Good advice for newbies. There's a thousand ways to slice the pie when it comes to installing a cap kit on a monitor. You can first start by checking if every capacitor on the chassis was installed in the correct orientation from the factory, desolder them all at once and put them in all at once. I always recommend APAR's cap kits because he uses the best quality stuff but I've done a lot of free work for local people and I used to always make them buy cap kits from Ian Kellogg because it made my life super easy doing a cap kit. He put his kits on a labeled strip so if you desoldered all the caps you could easily pop them all in, personally I was able to cap some sanyo monitors and have them back in the machines in under an hour with ian's kits. I think Ian burned his house down though. I'd always recommend going one by one like mecha said if you don't have a well labeled kit, aren't experienced, or have a bad memory and are a goober. There should always be a final check too, doesn't hurt to double check your work. I also wish security001 would do the labeled strip, best invention since the cap kit itself!
 
So far I've been lucky and only installed 1 cap backwards. C42 on a 19K7000. It ran like that for quite some time before I yanked it to install another flyback for testing and realized my mistake. Not sure how but I guess I got lucky there. But yea, happens to the best of us.

also word to the wise for anyone reading this, never under any circumstance think of trying a shortcut of yoinking all the caps at once. always do them one at a time.

This... but it didn't save me in the above.
 
Update: After swapping the incorrectly installed cap with a fresh one, the picture looks great, is the correct size and is stable.

The cap that was installed wrong still tested a good ESR and at the rated value, but it was physically bulging on both ends so I did not attempt to reuse it. I'll backfill the correct part into my spare cap kit.

I (thought I) did all the suggested SOPs here. I always print a fresh cap map and confirm the polarities match the markings on the chassis (or note discrepancies) and confirm that all the existing caps are installed as indicated. I test all the new caps with my ESR to try to ensure I'm not putting in any new bad caps. Then work one at a time confirming the rating of the one I removed against the kit's sheet and the new on going in, highlighting each cap on the sheet as I work on it and marking it completed on the map and the sheet when it's done before moving to the next.

When finished, I go over them all again to confirm the installed rating and polarity. Clearly I zoned out or was tired by the time I got to that step this time.

I do all the right steps, but hey we all make a mistake now and then. Hopefully this one will help me be even more diligent in the future. I'm just glad I didn't cause any other damage with my goof (so far as I can tell).

Thanks for all the tips!
 
Back
Top Bottom