I found a TV-Set in a dumpster, and checked if it's still working.

Indeed it is somewhat functional, but the picture is distorted:

(screenshot of the TV)

Depending how bright the picture is, it get's more or less vertically "compressed", where the upper half is more compressed than the lower part. The brighter the picture, the more compressed it becomes.

There is also a combing effect, and the picture seems horizontally slightly shrunk.

When the picture is very bright, as in almost white, the screen flashes, and sound disappears.

Also, the brighter the picture, the more 50Hz noise I can hear from the speakers.

I suspect either the V-control IC to be broken (TDA3653B), or one of the capacitors connected to it.

Or, some of the AC/DC capacitors might be dried out or otherwise broken.

Any suggestions?!

  • \$\begingroup\$ I have the wiring diagram for the TV set, so identifying parts isn't that complicated. Should I replace all capacitors on the AC/DC converter part? \$\endgroup\$ – polemon Mar 24 '11 at 8:53
  • \$\begingroup\$ Could it be that it's TDA3653B, and not TBA3653B? \$\endgroup\$ – zebonaut Mar 24 '11 at 20:33
  • \$\begingroup\$ @zebonaut: Yes of course! (corrected) \$\endgroup\$ – polemon Mar 25 '11 at 5:17

What makes me curious is the 50 Hz noise whenever the picture is bright. Tow possible reasons come to mind:

1) Interference from the (broken?) vertical deflection stage or backwards from there to the power supply and then on to the audio amp

2) Ripple from the power supply (should be 100 Hz, though.) The total power increases for bright pictures, hence the ripple voltag would increase as well.

Does the TV have a traditional power supply that uses a 50 Hz transformer or does it have a switch-mode power supply?

Do you have a scope? Are you able to probe the various supply rails, especially the one that powers the vertical deflection circuit?

Here's an extremely good app'note with tons of typical oscillograms and ICs that are likely very similar to the ones your TV uses: http://www.nxp.com/documents/application_note/APPCHP4.pdf


Some more hints after I have checked my old funny book that has pages after pages with pictures of ill TVs and list their possible causes (ISBN 978-3772350580); parts refer to Fig 4 in the TDA3653 data sheet:

  • Check electrolytic cap in the vertical amp's supply (220 µF). Also check the decouplig resistor (4,7 Ohm)
  • Check electrolytic cap in the vertical oscillator or flyback circuit (100 µF).
  • Check big cap at deflection coil (1500 µF) and 1,2 Ohm resistor.
  • Check feedback circuit (4,7 µF)
  • \$\begingroup\$ We have 240V/50Hz in Germany. It has a switch-mode PSU. I don't have a scope. \$\endgroup\$ – polemon Mar 24 '11 at 10:03
  • \$\begingroup\$ If it's a SMPS, I guess the 50 Hz noise is interference from the vertical deflection circuit. Will be kind of hard without a scope, but I would test(/replace/try out) (a) the capacitors that stabilize the deflection circuit's supply rails and (b) any capacitors in the signal or power paths between the deflection IC and the deflection yoke (coil). Also, check for bad solder joints around any parts of the deflection circuit. I've successfully repaired quite many TVs just by re-soldering some pins. \$\endgroup\$ – zebonaut Mar 24 '11 at 10:22
  • \$\begingroup\$ I checked all solder pins, all are OK. I replaced two capacitors adjacent to the TBA3653B, the rest I need to order. But isn't it like there's unstable power supply since the TV starts to flicker, loses color and sound at bright pictures? I believe all CRT guns at full power (white = all CRT guns at full power) seem to drain too much power, which would suggest an unstable power supply (broken capacitors). Or am I on the wrong path here? \$\endgroup\$ – polemon Mar 24 '11 at 10:31
  • \$\begingroup\$ Yes, also likely. Even without an oscilloscope, measuring the voltages from the power supply with a DMM might give you hints. It is also likely that one voltage is regulated, and the others are just obtained by rectification of other secondary windings in the SMPS main transformer. Also, stuff might be synchronized with the picture (i.e. with the vertical skipping pulses). I'd have to dig into the schematic/service manual to get a better idea. \$\endgroup\$ – zebonaut Mar 24 '11 at 11:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.