Philips LCD 42PFL5432D/37 - Bad PSU

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I connected the Aux coil from the transformer back into the power supply, but no luck the light still flickers.

Right.
That's because on T904 (the standby transformer), the auxiliary winding gets its power from the main winding. Basically IC907 (standby PWM controller) "tells" Q908 (the standby MOSFET) to pulse on and off very fast. The pulses (ac voltage) of Q908 are sent through the main winding of T904, which results in an output voltage at the secondary side of T904 and the auxiliary winding.
But we already have disabled the primary side of the standby circuit, remember? (the removal of J5 and R974 disables Q908 and IC907). Therefore, you don't need T904 connected at all for this mod, just two external power supplies - one is the 12v computer PSU feeding the standby output, and the other is the ungrounded power supply feeding the positive side of electrolytic capacitor C958 (this makes it seem as if the voltage is supplied from the auxiliary winding of T904).

I have a 12v 1500mA linear power supply that should do that trick if not. It is unregulated so the power that comes out of it is generally 14-16v, but as you stated that shouldn't be an issue since the two ICs can handle far beyond that. I guess if push comes to shove, I can always stick a regulator IC on the end of it. Unfortunately, I don't have any 5v power supplies that fit the bill. Most of the ones I do have are cell phone chargers, but they only put out 300-500mAs, I'm not sure if that is enough.

Well, I looked through the TEA1507 (IC904) datasheet again, and it says that the minimum start-up voltage is 10.3v. Therefore, use the 12v, 1500mA linear power supply. I don't think a regulator IC will be needed.

You're correct, I was wrong what I stated in the post above about where Pin 8 connected. I jump lines.

No worries. I do that too sometimes .
I think this is why I didn't see the potential problem that was caused by the missing auxiliary winding voltage.

By the way, those capacitors you used for the replacement are huge! What brand and series are they?

[Xeniczone]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Ah, what a difference. I can see things much better now .
Thank you.

No problem. You're helping me remember.

I looked at the schematic you posted, and everything I posted above seems to be correct.
Pin 8 of IC902 (the PFC controller) is VCC, and it is connected to SMT transistor Q914, which is controlled by optocoupler IC909, and Q914 is supplied with power by the auxiliary winding of the standby transformer, T904. Without it, you get no PFC, which too could mess up the operation of the PWM controller.

You're correct, I was wrong what I stated in the post above about where Pin 8 connected. I jump lines.

I think the idea of feeding a positive voltage to the positive terminal of electrolytic capacitor C958 should solve the problem. But again, the voltage you feed to that spot needs to come from an ungrounded power supply only. Otherwise you can do a lot of damage to both the external power supply and the TV power supply.
I *think* a 12v power supply should work for this. Perhaps try and start with 5v first, and if that doesn't work, then go up in voltage. Technically, IC904 can take up to 22v, and IC902 can take up to 25v. However, it's the other components in the circuit that may be of limited capability, so that's why it's better to start with something low.

By the way, for the power supply to work, you will also need that 12v computer PSU you had connected to the standby output before.

Before I tried that, I gave this a try. In theory, it should have worked right? I connected the Aux coil from the transformer back into the power supply, but no luck the light still flickers.

I have a 12v 1500mA linear power supply that should do that trick if not. It is unregulated so the power that comes out of it is generally 14-16v, but as you stated that shouldn't be an issue since the two ICs can handle far beyond that. I guess if push comes to shove, I can always stick a regulator IC on the end of it. Unfortunately, I don't have any 5v power supplies that fit the bill. Most of the ones I do have are cell phone chargers, but they only put out 300-500mAs, I'm not sure if that is enough.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Here is the datasheet I'm using, it looks a lot better than those pictures you have.

Ah, what a difference. I can see things much better now .
Thank you.

I looked at the schematic you posted, and everything I posted above seems to be correct.
Pin 8 of IC902 (the PFC controller) is VCC, and it is connected to SMT transistor Q914, which is controlled by optocoupler IC909, and Q914 is supplied with power by the auxiliary winding of the standby transformer, T904. Without it, you get no PFC, which too could mess up the operation of the PWM controller.

I think the idea of feeding a positive voltage to the positive terminal of electrolytic capacitor C958 should solve the problem. But again, the voltage you feed to that spot needs to come from an ungrounded power supply only. Otherwise you can do a lot of damage to both the external power supply and the TV power supply.
I *think* a 12v power supply should work for this. Perhaps try and start with 5v first, and if that doesn't work, then go up in voltage. Technically, IC904 can take up to 22v, and IC902 can take up to 25v. However, it's the other components in the circuit that may be of limited capability, so that's why it's better to start with something low.

By the way, for the power supply to work, you will also need that 12v computer PSU you had connected to the standby output before.

[Xeniczone]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I will test some of your suggestions out tomarrow and report back, I got a day off tomorrow.

Here is the datasheet I'm using, it looks a lot better than those pictures you have.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Okay, apparently power to the IC904 (main PWM controller) also comes from pin 5, the auxiliary winding of T905 (the main transformer). Not quite sure why there would be two supply circuits for IC904. Either way, it seems possible for IC904 to draw power from both standby and main transformer auxiliary windings. Perhaps one of them is for when the TV is in standby mode.
Diode D907 is for main transformer aux. power, whereas D910 is for standby aux. power. Interesting...

Anyways, I still think this missing voltage is the cause for the main power supply not starting up. The standby auxiliary winding is connected to a lot of things, so it really must be important, otherwise the designer(s) wouldn't have bothered.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Good catch, but I'm not sure I understand the purpose of the aux coil there. Based on the wiring diagram it doesn't look like the coil does anything couldn't pin 5 and 6 on T904 be shorted out and receive the same result. Forgive me on this one, I don't know as much as I should about transformers.

Pin 6 on T904 supplies IC907 (the standby circuit PWM controller) with power. Without it, IC907 would not work and there would be no standby power.
Initially, when the TV is plugged in, IC907 receives power through pin 6 (Drain) and charges capacitor C957 (which is connected to pin 1, VCC, of IC907). Once C957 reaches a certain voltage, IC907 uses that to "pulse" Q908 (the standby MOSFET). This pulse generates a voltage in the auxiliary winding of T904 (pin 6), which then gets rectified through D908 and charges C957 and thus provides IC907 with constant power. IC907 then again pulses Q908, and the whole cycle keeps repeating. Meanwhile, the pulses from Q908 also generate a voltage on the secondary of T904 - this is what produces the 12v standby voltage.
All of this is relevant to your next question...

Also, I have trouble understanding why this would cause the backlight only to flicker once the firmware has booted.

Well, at least from what I traced, the PFC controller seems to be controlled by a bunch of SMT transistors (Q914 particularly) which are located near opto-isolator IC909. It also seems that IC909 is "supplied" with power through R975, which if you trace back comes from the standby auxiliary coil. Moreover, what I just found out, based on this schematic,
https://www.badcaps.net/forum/attach...6&d=1298647899
it seems that IC904 (the PWM controller of the main power supply) is also supplied by power from the standby auxiliary coil through Q912. This would explain the pulsing of the backlights, since the backlights are powered from the main 24v power supply, whereas the firmware and other logic-related power is likely supplied from the 12v standby power supply. The main PWM controller pulses because it too initially starts up like IC907 (i.e. it charges C921 through pin 6, Drain).
Also, when the PFC is not working, the voltage across the big 220uF, 400v primary capacitor is likely only about 170 volts (if you live in a country where the line voltage is 120V rms, of course). With the PFC circuit enabled, this voltage should be ~385v. Feel free to check me on this by measuring the voltage across the primary cap (be very careful, though!). If it's close to 170v when the TV is powered on, your PFC is not working.

Pin 8 on IC902 got to IC910 (also an opto-isolator) not IC909 that you're referring to.

Okay, let's make sure we're on the same page here. I'm using these pictures as a reference when I determined that IC902 (the PFC controller, supposedly) is connected to opto-isolator IC909:
https://www.badcaps.net/forum/attach...3&d=1298297504
https://www.badcaps.net/forum/attach...6&d=1298206864
If I'm not mistaken, one of the pins on IC902 should be connected to SMT transistor Q914.

Double check me on all of these items and let me know what you find.

[Xeniczone]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Good catch, but I'm not sure I understand the purpose of the aux coil there. Based on the wiring diagram it doesn't look like the coil does anything couldn't pin 5 and 6 on T904 be shorted out and receive the same result. Forgive me on this one, I don't know as much as I should about transformers.

Also, I have trouble understanding why this would cause the backlight only to flicker once the firmware has booted.

Pin 8 on IC902 got to IC910 (also an opto-isolator) not IC909 that you're referring to.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Very interesting. I too have had problems with our Philips 42 inch TV. I traced the chirping back to the same transformer, T904. I started by replacing the 4 larger capacitors because they looked swollen and all that jazz. After nothing I tested came back bad or nothing I replace made the PSU work, I decided to try this idea out.

It does seem to work but there is a problem with the backlight. It flickers. I believe it has something to do with the PWM not functioning correctly, but I'm not sure 100% how it works because I can't find it in the wiring diagrams. The reason I think it is the PWM is because when first turning the TV on the backlight doesn't flicker until the software initializes (shows HDMI in the corner and such) then it starts flickering. Plus trying to turn the brightness up/down in the menus has no effect on the TV.

Interesting results.
When I wrote that suggestion for the external power supply, I did spend a good chunk of time observing the circuit in order to make sure an external power supply can be inserted without having to have anything on the primary of the standby circuit function. Now that I look at it again, though, it seems that there is indeed a voltage line derived from the auxiliary winding on the standby transformer. It goes through SMD resistor R977, then gets rectified through SMD didode D929, filtered through cap C958 and then goes out to optocoupler IC909 and surrounding circuitry.

See if the voltages on the main power supply pulse with the same frequency as the backlights. If they do, then I think this may be the problem - i.e. missing auxiliary voltage on the primary side of standby transformer.
Possible solutions...
Pin 1 of IC907 is Vcc. According to the TEA1507P (IC907) datasheet, Vcc can be up to 20V. So if there is a way to get a voltage inserted at the positive leg of C958, I think the power supply may then function properly. Unfortunately, you cannot insert any external voltage source for this because everything on the primary side has a hot ground. Perhaps one of those unregulated, ungrounded 9-12v chargers may do the trick.
I'm also wondering if a voltage divider from the primary would work. Perhaps if the current required for IC909 and the surrounding circuitry turns out to be fairly low.

I will keep at it and try to find a solution. Plus try to make the solution look better so you don't have a computer psu sitting next to your TV.

Yes, I agree that this solution is not very pretty at all. More of a last-resort type of thing. And if the above turns out to be true, you will have not one, but two external power supplies! Now that's a bit too much, even for me .

Just to note, I'm doing this for the thrill of it. My dad has already bought a new TV to replace this.

Yeah, that's fine. It's normal to see other folks here (including myself) fix things that are usually not worth fixing.

*EDIT*
I think I may have figured it out...
It looks like the purpose of optocoupler IC909 is to send a signal to the PFC controller in order to run the PFC circuit. There's a wire from pin 8 on IC902 (the PFC controller) that goes to IC909 and the transistors around it.
Get a voltage inserted on C958 and you may have a working power supply. (Again, just make sure it's an ungrounded power supply like one of those cheap, linear wall-warts.)

[Xeniczone]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I concur.
The other clue that gives this away is that there's only 2 transformers in this power supply (one of them being the broken one you mentioned) and each has a single output only.

So what's next?...
Remove J5 and R974. This will disable the primary side of the standby circuit. Then finally wire in 12v from an external power supply (see attached image below). Note that ground (GND) from the external PSU also needs to be connected.
I suggest a PSU capable of at least 2A. Also if there's any loose solder like in the pictures you posted, make sure to clean it up.

Very interesting. I too have had problems with our Philips 42 inch TV. I traced the chirping back to the same transformer, T904. I started by replacing the 4 larger capacitors because they looked swollen and all that jazz. After nothing I tested came back bad or nothing I replace made the PSU work, I decided to try this idea out.

It does seem to work but there is a problem with the backlight. It flickers. I believe it has something to do with the PWM not functioning correctly, but I'm not sure 100% how it works because I can't find it in the wiring diagrams. The reason I think it is the PWM is because when first turning the TV on the backlight doesn't flicker until the software initializes (shows HDMI in the corner and such) then it starts flickering. Plus trying to turn the brightness up/down in the menus has no effect on the TV.

I will keep at it and try to find a solution. Plus try to make the solution look better so you don't have a computer psu sitting next to your TV.

Just to note, I'm doing this for the thrill of it. My dad has already bought a new TV to replace this.

[gdinon]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I soldered the two new diodes in. Turned the tv on and no more chirp. Then all the sudden the white lights came on. I looked at the screen and it was back to normal. Yay!!! I screwed everything back together and three of us picked up the tv to mount it on the wall and plugged in all the video cables and power. Turned it on and nothing. It started chirping again. Darn. All that work. There must be something still wrong that caused it to go bad again. I took it all apart and tested the diodes. They are still good. The power must have gone beyond that now and now something else could be bad. Got to do some more testing.

What did you find? I am in a similar situation. Replaced the 3 caps, powered it on, and 1 blew. The diodes are bad but I would like to know what you found that caused it before I replace them.

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I soldered the two new diodes in. Turned the tv on and no more chirp. Then all the sudden the white lights came on. I looked at the screen and it was back to normal. Yay!!! I screwed everything back together and three of us picked up the tv to mount it on the wall and plugged in all the video cables and power. Turned it on and nothing. It started chirping again. Darn. All that work. There must be something still wrong that caused it to go bad again. I took it all apart and tested the diodes. They are still good. The power must have gone beyond that now and now something else could be bad. Got to do some more testing.

[Dgtech]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

I'm glad you see a difference. Replace them, test it and report back.

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

New diodes came in. They measure 300 and only going one way.

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Don't know what the numbers mean with my DMM, but with one diode I get 11 both ways and the other 29 both ways.

[Dgtech]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

What type of reading do you get both ways? Is it 0 ohms both ways? Using the diode test, what voltage reading do you get both ways, anode to cathode, cathode to anode?

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Can I test the voltage without the TV? I don't have the TV now. Just the powerboard. I can plug it into the power but I'm sure it won't turn on without something jumping certain pins.

Attached is where I noticed some discoloration under two diodes. I took the diodes out and tested with DMM on diode mode. I get a reading both ways on both diodes. I believe that means they are bad. I'll get them replaced. They are part# P6KE120A. Data sheet says they are uni directional.

Has anyone ever fixed their board? Seems no one has actually fixed their but just has advice.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

That's likely left over flux from when the transformer was soldered at the factory. Did you replace the small caps yet? Also did you try measuring the standby voltage at connector CN902 (the one next to capacitor C955)? There should be 12v there when the PSU is working properly.

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Yes the transformer does look good. I meant the solder. Looks like they're kinda burnt. That's all.

[momaka]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Does my T905 Transformer look bad.

Hard to say exactly. Looks fine to me. Check the solder joints on the other side - i.e. see if any of them look dry or cracked. Also, change the small electrolytic caps like you mentioned and let us know if the problem remains.

[trixter]

Re: Philips LCD 42PFL5432D/37 - Bad PSU

Just replaced ABCD and still chirps. Next I'll try all the small caps and see.

Does my T905 Transformer look bad. It sounds like the chirping is coming from this. It looks this way on both sides.