Old LCD-TV (Sony Wega KLV-20SR3) Problem

[Teaser]

Hello,
First of all I'm new here and this is my first thread. I came here hoping to find help repairing an older LCD-TV.
It is the Sony Wega KLV-20SR3. It has the good old florescent lamp for providing light.
Now here is the problem:
It does not work at all Kinda says it all ha? Well First I thought that it could be the usual Capacitor problem in the power circuit but after opening up the TV-Set it turned out that all capacitors on all boards actually "look" perfect. Now I don't have a capacity meter so I can't determine if a capacitor is bad.
So then I suspected that the florescent lamp could be dead. These actually have a life cycle and the device is kinda 10 Years old or something like that.
But then I noticed that two LEDs (red and green) which are for determining if the TV is on or off start to flicker as soon as one connects the power cable. The ON/OFF button on the device does not change any thing and has no effect.
Here is a photo of the whole circuitry. The LEDs are on the right edge of the screen (Up on the right in the attached photo).
Please, any idea what could be the problem?

 

[Nigel Goodwin]

Well First I thought that it could be the usual Capacitor problem in the power circuit but after opening up the TV-Set it turned out that all capacitors on all boards actually "look" perfect. Now I don't have a capacity meter so I can't determine if a capacitor is bad.

You don't need a 'capacity meter' (or even a capacitance one ), it would be of no use - you need an ESR meter, which is the only way to test them.

However, in this case it's not needed, as while you mention the 'usual capacitor problem', there's no such thing on Sony sets, they use higher quality capacitors and they pretty well NEVER fail.

So then I suspected that the florescent lamp could be dead. These actually have a life cycle and the device is kinda 10 Years old or something like that.
But then I noticed that two LEDs (red and green) which are for determining if the TV is on or off start to flicker as soon as one connects the power cable. The ON/OFF button on the device does not change any thing and has no effect.
Here is a photo of the whole circuitry. The LEDs are on the right edge of the screen (Up on the right in the attached photo).
Please, any idea what could be the problem?

What do you mean by 'flicker'? - in common with most sets these ones have a self-diagnosis function, and flash the LED a specific number of times, then stop, then repeat - on these models it goes from 2 flashes to 14.

If it's flickering rather than flashing, then it's probably a PSU problem, and it's not firing up correctly - have you checked the supply rails from the PSU?.

CN6101 pin 2 (3 pin connector) should be 5V, even on standby.

CN6101 pins 1, 2, 5, and 6 should be 14.3V (with the set running only)

CN6102 pins 1, 2, 3, and 4 should be 16.5V (with the set running only) and feed the inverters for the CCFL backlights.

However, you should bear in mind this set is 11 years old, and not even widescreen

 

[Teaser]

Hello,
thx for ur answer.

What do you mean by 'flicker'?

They are flickering, non stop.

CN6101 pin 2 (3 pin connector) should be 5V, even on standby.

No it doesn't deliver . The voltage measures 1,66v between pin 2 and 1 (or 2 and 3 (1 and 3 are both ground)).

CN6101 pins 1, 2, 5, and 6 should be 14.3V (with the set running only)

CN6102 pins 1, 2, 3, and 4 should be 16.5V (with the set running only) and feed the inverters for the CCFL backlights.

These are not measurable as the TV won't run

However, you should bear in mind this set is 11 years old, and not even widescreen

I know, I got it for free and I would really be more than happy if I could get it repaired

 

[Nigel Goodwin]

You might try replacing D6104, the 5V standby rectifier - it MUST be a fast recovery rectifier - I've seen a number of sets where this diode has gone slightly leaky causing similar problems.

Failing that, check around IC6100 and the rest of the standby PSU - but it's normally the rectifier.

 

[Teaser]

You might try replacing D6104, the 5V standby rectifier

I didn't find such a number. Could u please point it out?
There are only the D6102 and D6103! However the D6102 is kinda acting like a wire when I do the connectivity test using the meter (in both directions! ). Any suggestions on which diode I should buy? There is no writing on the D6102, only an "R" and an arrow!

 

[Tony Stewart]

count how many LED flickers . **broken link removed**

2 = Brd A, 3= Brd B, 4= Panel, 5= EPROM etc

 

[Teaser]

You might try replacing D6104, the 5V standby rectifier - it MUST be a fast recovery rectifier - I've seen a number of sets where this diode has gone slightly leaky causing similar problems.

Failing that, check around IC6100 and the rest of the standby PSU - but it's normally the rectifier.

I didn't find such a number. Could u please point it out?
There are only the D6102 and D6103! However the D6102 is kinda acting like a wire when I do the connectivity test using the meter (in both directions! ). Any suggestions on which diode I should buy? There is no writing on the D6102, only an "R" and an arrow!

I fixed it!! It was the rectifier like u said (D6102). I took it out and replaced it with a "BAW75" which should be a high speed diode. I noticed the noise in spite of that though. What actually was a bit funny is that the 1N4007 which suppose to be a normal (not fast switching) diode did not make that much noise, if at all, as I tried it using crocodile clips before I soldered the BAW75. I just did not trust the 1N4007

 

[Nigel Goodwin]

count how many LED flickers . **broken link removed**

2 = Brd A, 3= Brd B, 4= Panel, 5= EPROM etc

Perhaps you should try reading the rest of the posts

You don't count 'flickers' - you count 'flashes' - in this case the set wasn't getting far enough forward for the processor to even start running.

 

[Nigel Goodwin]

I fixed it!! It was the rectifier like u said (D6102). I took it out and replaced it with a "BAW75" which should be a high speed diode. I noticed the noise in spite of that though. What actually was a bit funny is that the 1N4007 which suppose to be a normal (not fast switching) diode did not make that much noise, if at all, as I tried it using crocodile clips before I soldered the BAW75. I just did not trust the 1N4007

D6104 is on the secondary side, D6102 is on the primary, and powers the IC.

Either could cause the problem, well done

While a 1N4007 will work enough to test it, it's likely to have a short live as it will waste much more heat than a fast diode.

 

[Teaser]

D6104 is on the secondary side, D6102 is on the primary, and powers the IC.

Either could cause the problem, well done

While a 1N4007 will work enough to test it, it's likely to have a short live as it will waste much more heat than a fast diode.

Thanks a lot for your help! I must admit that I noticed that that Diode had a short but I thought my meter was crazy cuz I didn't know that diodes could completely short when they break so I didn't pay attention to that. Following ur answers I started some searching in the internet and I now understand more about the issue.
BTW, getting this TV fixed encouraged me to open up an old LCD-Screen that I have which was also broken. Now I have a few questions about this, may I ask them in this thread as well?

 

[Nigel Goodwin]

Thanks a lot for your help! I must admit that I noticed that that Diode had a short but I thought my meter was crazy cuz I didn't know that diodes could completely short when they break so I didn't pay attention to that. Following ur answers I started some searching in the internet and I now understand more about the issue.
BTW, getting this TV fixed encouraged me to open up an old LCD-Screen that I have which was also broken. Now I have a few questions about this, may I ask them in this thread as well?

I would suggest opening a new thread, titled with the make and model.

However, a LOT depends on what make it is - for example, I'm a Sony approved service agent

 

[JoeJester]

It wouldn't surprise me if that diode was connected to the a.c. return, as nearby ground strikes can and do short out diodes connected to the reurn side.

 

[Nigel Goodwin]

It wouldn't surprise me if that diode was connected to the a.c. return, as nearby ground strikes can and do short out diodes connected to the reurn side.

Sorry, but your completely random suggestion has no bearing on the case in question

The diode in question, which is a VERY common failure, is the permanent supply for the standby PSU chip, which takes over from the start-up supply once it's running.

I suspect it's just a crappy diode, as the exact same one fails on the secondary side as well.