AUO t-con - Help me identify and replace transistor please!

[tom66]

Re: AUO t-con - Help me identify and replace transistor please!

Transistor gain could cause it to get hot.
If the gain is too low, the transistor is no longer saturated and is acting as a fairly high impedance.
This means any current will make it heat up fast.
For example 30mA load, which would be a good guess for quiescent current on 15V, this would lead to a power dissipation of 0.36W ((15V - 3V)*0.03A) which in a SOT-23 package, with a thermal resistance of around 250°C/W, would heat the device to 90 deg C above ambient.
That's just an example. numbers are guesses.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

That was my thinking T30M31 (re. your edit).

Resistance collector pad to ground V0: 5.5k, V2: 5.9k. So actually slightly higher on the V2 (bad board).

lookimback, I did the resistance check you mentioned before your edit, collector to ribbon connectors. Nothing was shorted/low ohms except where the AVDD connections are supposed to connect - same on good board.

I'll have to take a break until tomorrow on this one, when I will do more resistance checks and maybe try that jumper unless anyone thinks it's a really bad idea.

Thanks for all the help so far!

[T30M31]

Re: AUO t-con - Help me identify and replace transistor please!

Transistor gain would not be an issue with it getting hot. It must be something down stream of the transistor that is overloading it.

Probably not a good idea to jump around the transistor if it is overloaded. I would stick with resistance measurements of all that is fed by the collector of this transistor.

After thinking on it a bit more, the gain could be an issue. If the transistor is used as an on/off switch it may not be conducting enough to supply the load and this would cause excess power dissipation. If that is the case the jumper might be an OK test. Before trying the jumper I would compare the resistance from collector to ground on both TCON boards. Since the V0 board does not have the transistor measure from the pad where the collector would be connected.

If the V2 board has a lower resistance it is probably an overload condition. And as lookimback mentioned the TCON needs to be connected and power off when making this resistance check.

[lookimback]

Re: AUO t-con - Help me identify and replace transistor please!

Since your voltages are normal until you connect the TV, I would suspect that the part of the circuit which is shorted is not completed until you connect the TV. Try connecting the board to the TV (with power off) and check for continuity between the pnp transistor and the connectors for the ribbon cables. If you find something with low resistance then check without the TV connected.

Edit: maybe not test between the pnp transistor and the ribbon cable connectors but actually trace the output of the transistor as far as you can and test from there.

[lookimback]

Re: AUO t-con - Help me identify and replace transistor please!

Probably not. Might be other differences in the VO board that aren't physically aparent. That transistor doesn't necessarily have to be the faulty component. I had a main board recently which had a mosfet getting hot but it wasn't the bad component, the CPU was shorted to ground. It is going to be something on that path though. Something which comes after it I think.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

Thanks for the replies guys.

The PNP transistor is getting quite hot yes. Nothing else is particularly hot. The new ic is a little warm, but it had thermal pads top and bottom before which I haven't replaced yet so I think that is normal.

T30M31, as the good board is the V0 board without that transistor, it's not possible to compare readings on those components as they are not present on the V0.

Measuring resistance to ground from AVDD (collector of the PNP) it is actually just slightly lower on the good board.

If the transistor doesn't have enough gain, would that be consistent with it getting hot?

I'm tempted to try jumpering across R807 to cut out the transistor entirely and replicate what the V0 board is doing, just to rule out the transistor as the problem. Good idea or not?

[lookimback]

Re: AUO t-con - Help me identify and replace transistor please!

Voltage comparisons too, I forgot that one.

[T30M31]

Re: AUO t-con - Help me identify and replace transistor please!

Does the PNP transistor get hot with the panel connected?

Have you checked R805, R806, C817, and C821? These all look like they are part of the Q812 circuit. It would probably be helpful to compare voltage readings on these components with the good TCON.

The PNP transistor that you substituted may not have enough gain but that is only a guess since I don't know the specs on the original.

The V0 boards looks like it used a zero ohm resistor/jumper at R807 in place of Q812. The Q812 circuit looks like it configured as a time delay circuit to sequence power.

[lookimback]

Re: AUO t-con - Help me identify and replace transistor please!

You have a good board so you could compare readings of individual components between boards. Meter in ohms or diode test mode dependant on type of component. Find something that's significantly different, pull it and test it out of the board. I've had quite a bit of success doing that. I don't really have that much understanding of how a lot of this stuff works. I just learn it as I go.

Also, if voltages are dropping when you connect it to the tv, it sounds like something is drawing the current to ground, or to somewhere other than it should be anyway. Still a shorted component somewhere. Did you notice anything hot after it had been connected to the TV for a minute?

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

A progress report.

I put in the mosfet as suggested by T30M31, and replaced the BD8156EFV.

Connected up the board without the panel connected and all voltages check out ok. Great!!

But when I connect up the panel, I get no picture and the voltages change.

Voltages without/with panel connected:

AVDD 15.4V / 3.4V
VGHC 24V / 28.3V
VGMA_REF / 15.3V / 3.4V
VGL -6V / -6V

Without panel the pnp transistor I put in at the start of the thread has 15.5V, 14.8V, 15.4V (E/B/C). With panel connected collector drops to 3.3V while other pins stay the same.

The panel is fine. It works with the other t-con I have, so there's still something wrong on this board.

I will be testing more later, but any ideas would be greatly appreciated.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

So tom66 after your hint about the boost converter not switching I figured maybe MOSFET Q811 wasn't getting the switching signal to it's base. I traced that back to U801 (see pic) which had been hiding under a thermal pad and I'm starting to suspect it may well be faulty. It is a BD8156EFV (datasheet attached).

I'm not sure how exactly to test if it's faulty but it is getting 12V to VCC power supply input pins. Can anyone guide me on fault-finding for this IC? It's cheap enough to buy though will probably have to wait for delivery from China.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

D812 is not short tom66.

[tom66]

Re: AUO t-con - Help me identify and replace transistor please!

I'd check D812 isn't short, yes, I think that's the converter.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

Awesome T30M31! Thanks so much for taking the time to look those up. I found someone selling the DMP3056LDM on ebay for £1.57 free postage, so not much to lose there if it doesn't work. Have ordered one.

Still very interested in others' opinions about the other missing voltages and if it's related to Q302 or something else.

[T30M31]

Re: AUO t-con - Help me identify and replace transistor please!

TN245, it looks like a P-channel MOSFET would work as a high side switch the way Q302 is connected on this board. Q302 appears to be in an SOT26 case and I have found 2 possible replacements. Links to both are include below and you can check physical dimensions using the data sheets provided on the Farnell website.

I don't know if Q302 is related to the missing voltages you referred to in your last post but the difference in resistance between Q302 pins 3 and 4 for the good and bad boards does indicate a problem around Q302. The lower resistance could be due to R330 or C307 but I think Q302 is most likely the problem.

http://uk.farnell.com/diodes-inc/dmp...t26/dp/2061420

http://uk.farnell.com/taiwan-semicon...t26/dp/1864596

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

Well, I'm a bit lost now. Can you give me some pointers tom66 or others?

Tom, I figure you're talking about the boost converter around the inductor L812 (6R8). I'm not sure how to figure out what's going wrong with it though. There are all kinds of voltages missing around the board. The only suspect component I can find is that Q302 MOSFET.

Some more voltages, compared with readings from the good board:

            Good Board       Bad Board
GAVCC1 	        3.3 	        0
LVDS_OUT        3 	        0
V3D3 	        3.3 	        0
V2D5 	        2.4 	        0
Q811 (RSS065N03)
S, G, D         0, 0.7, 11.7 	    0, 0, 12
U802 Pins 1, 2, 3 	3.2, 0, 2.4 	    0, 0, 0
U804 Pins 1, 2, 3 	0, 1.7, 3.2 	    0, 0, 0

Any thoughts? Any more readings I could take to narrow things down?

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

Ah, ok then, thanks. Which part of the circuit should I be looking at?

Just checking though.. you followed that we're talking about a different component than the transistor at the start of the thread? The six-pin mosfet connected pretty much directly to Vin. From my very limited understanding of the circuit I figured it would be pretty critical.

[tom66]

Re: AUO t-con - Help me identify and replace transistor please!

No, I think you have other faults on the board, with that transistor out you should still have 15V AVDD and a distorted image on screen.

[tn245]

Re: AUO t-con - Help me identify and replace transistor please!

Thanks T30M1. Q302 seems a little larger than the SSM6J212FE. It's just under 3mm long while the SSM6J212FE is 1.6mm. See attached.

lookimback, thanks for the tips. It is helpful. This was the first AS15 I did. I've since done others that have gone much better and worked first time. Replacing my worn out soldering iron tip helped a lot, but so did some of the techniques you mentioned. Getting better all the time.

tom66. Thanks! And Q302 being short would cause that right? Would you go along with it being a P-channel mosfet in this application?

[tom66]

Re: AUO t-con - Help me identify and replace transistor please!

If AVDD = VIN (more or less) the circuit is not switching. AVDD should be around 15V and this boost converter is then used to generate VGH and VGL.