Sony PVM-2044QM vertical foldover

[decoder_82]

Hi,

I am trying to repair a Sony PVM-2044QM. The full service manual is here: https://archive.org/details/sony_PVM-2042QM_2044QM_Service_Manual
When the monitor is turned on, the top part is a little bit distorted and white lines are displayed. See picture below.
After a few minutes the lines slowly goes up and disappear completely.

I think the issue is called vertical foldover from what I read
https://www.repairfaq.org/samnew/tvfaq/tvvfold.htm

From the website it says "Probable issue is a charge pump that generates a high voltage spike to return the beam to the top of the screen".

So far I have checked/done the following:

- Checked for cold solder joints
-  Replaced the following capacitors: C651, C501, C559. C502. C535. C519. C520. C535, C520. C547. They either had high ESR value or obvious sign of leakage.
- Reflowed solder joint of IC501 (UPC1377C) which is managing horizontal and vertical deflection.

What should I check next ? What kind of failure would solve itself with the monitor heating up ?

Any advice would be appreciated.

Thanks

[Jeff eelcr]

Not a model I worked on but Sony had some 3KV capacitors that had problems.
Jeff

[decoder_82]

Not a model I worked on but Sony had some 3KV capacitors that had problems.
Jeff

Thanks for the reply ! What were the symptoms on your Sony ? Can you give details about the problem with the 3KV capacitors ? Were they open / short /out of range ?

[inse]

To verify whether it is a fold over or blanking problem, can you supply a crosshatch test signal?
Usually you can identify bad electrolytics with a hair dryer and cold spray, did you try that?

[decoder_82]

To verify whether it is a fold over or blanking problem, can you supply a crosshatch test signal?
Usually you can identify bad electrolytics with a hair dryer and cold spray, did you try that?

Yes here is what I get with crosshatch signal. Looks more fold over issue to me.

I am not sure about the hair dryer and cold spray technique, can you explain more ? Thanks 

[inse]

I am not sure what it is, the geometry looks good, obviously nothing of the picture is missing, the fold-over faults I remember looked different.
Play with the vertical size and position adjustments to get a picture without overscan.
The hair dryer method is simple: heat up the chassis with the hair dryer while watching the screen (a mirror is helpful) if you notice a change, identify the culprit with selective shots of cold spray.
Usually aged electrolytics will recover a little when heated, other components may react as well.

[Jeff eelcr]

I do not remember the failure mechanisum but it was many units TV, monitor, consumer and industrial and broadcast.
Jeff

[decoder_82]

Thanks indeed the hairdryer method makes sense in this case as the issue seems related to temperature. I will try and provide an update here if I managed to identify the issue.

[commongrounder]

It might be instructive if you were to engage the underscan mode and/or the h-v delay (aka cross-pulse) mode to see if the fold over still appears. The line I see looks like part of the vertical interval, so wouldn’t affect the video image content areas. It does not look like a retrace blanking problem to me.

[inse]

The folded area consists of a part of the image in general, do we agree on that?
The line that appears on the screen here is not part of the image or an entire line of boxes would have disappeared.
This is how I see it.
Reduce the vertical size as proposed already and we might get a clearer picture
;-) pun intended

[commongrounder]

I would agree that the fold over is part of the overall image, if you include the vertical interval in that. The interval contains 20 lines, around half of which are used for the vertical sync pulse. The pulse itself should be visible using underscan or cross-pulse display modes on monitors. I agree that part of the image would be distorted and inverted if there was fold over caused by vertical sweep distortion.  I’m wondering if the cross-pulse delay circuit itself might be the cause. Underscan circuits usually involve relay switching of the actual horizontal and vertical sweep circuits to decrease CRT yoke drive, and shrink the picture.

[decoder_82]

It might be instructive if you were to engage the underscan mode and/or the h-v delay (aka cross-pulse) mode to see if the fold over still appears. The line I see looks like part of the vertical interval, so wouldn’t affect the video image content areas. It does not look like a retrace blanking problem to me.

When I engage the underscan the fold still appear.

[decoder_82]

I tried hairdryer method. I actually used a hot air station on low temp with 2.5mm nozzle to be more precise.
It seems the issue is near the heatsink of Q511. I will replace all the electrolytics capacitors in the area.

[inse]

I am glad the trick worked for you.
As I am not so familiar with Sony design, can you show the affected part(s) in the schematic when done?

[decoder_82]

Yes no problem, I will post the details once I will have confirmed that I correctly identified the fault.

[decoder_82]

Quick update, I replaced caps and reflowed component in the area in red in PCB layout below. I could not really use the diagram as the part of interest was split in two pages in my PDF.
This resolved the issue. Thanks for the help !

[inse]

  Congratulations!
Was the affected circuit part of the vertical deflection?
I couldn't figure out from the print.

[decoder_82]

Yes here is a diagram more useful. From what I see, Q511 is labelled as vertical output. Connected to it, the caps C547, C549 and C591.

[commongrounder]

Congrats on the successful fix. Makes perfect sense.

[decoder_82]

The only thing that I found was strange is that just measuring the value of the capacitors I replaced they were not that off. For instance C547 measured 3uF instead of 3.3uF on my multimeter. This is within the tolerance I think. Maybe it doesn't behave well with higher voltage ? is that possible ?

[inse]

ESR is a factor that cannot be determined with a capacity measurement.

[decoder_82]

Yes I know. I did try to measure ESR but on small caps like this (3.3uF) my meter did not work. It indicated OL. I had the same with the new cap.