ACER AL2423W Monitor - Dead Inverter

[Trinite]

Hi all, I have an Acer AL2423W LCD monitor that has been plagued with capacitor problems. First, I got the notorious no signal/persistent "ACER" logo problems, which progressed into a flickering screen that would only go away when the monitor had warmed up after a few minutes. So I tore it open to find numerous sorry excuses for capacitors bulging and about to explode (Teapo and Capxon). I replaced all these on the PSU and main logic board with Nichicon and United Chemicon capacitors. After doing so, everything worked fine for a few months.

Unfortunately, I overlooked the lamp inverter, as it was underneath a shield and so I assumed it would be fine and not worth looking at. I regret that now. I got back from a trip to find that the backlight on my monitor would not come on. Tearing it apart again, I found 3 bulging NKCon capacitors on the board (seems like another garbage brand). I replaced these with Chemicoms and reinstalled it. However, the backlight would still not power on. So I checked the fuses, none of them were open. Checked the power, board is getting 19V, per spec.

I assume that the caps took out the MOSFETs with them, as there is no voltage coming from the output side of any of them. It does seem odd though that all six MOSFETs would go out, so I'm thinking it could also be the PWM chip. Any idea which would be the more likely culprit? I could replace the MOSFET chips if I can find an equivalent replacement for them. The PWM chip looks like it would be too much of a pain to bother replacing (has 20 closely-spaced legs). There is also a window comparator chip on there, but I don't think it's a likely candidate.

The inverter board is a TDK model XAD306LR.

The MOSFETs are NIKO-SEM P2804NVG N&P-channel enhancement mode FETs.


Any idea where I could find a replacement/equivalent inverter board, or MOSFET chips, if they are the likely culprit? The monitor has a superb MVA panel, and I'd hate to get rid of it and downgrade to an inferior TN panel. Any help would be greatly appreciated.

[jetadm123]

Re: ACER AL2423W Monitor - Dead Inverter

Photos would help. What cap values/brand did you use for the inverter board?


Check mosfets on inverter board for shorts:

With your DMM set on resistance, place probes across:

D1 and G1
G1 and S1
D1 and S1

Reverse probes and perform the above tests again.

Perform same tests for D2,G2 and S2.

Perform above tests for all 6 mosfets.

[retiredcaps]

Re: ACER AL2423W Monitor - Dead Inverter

Can we also get photos? Please post clear focused pictures after reading

https://www.badcaps.net/forum/showthread.php?t=1868

Please do not post pictures inline as they slow down the loading of pages.

For best picture quality and clarity take your boards to a window on a sunny day, turn flash off, and use macro mode. Take a top down view of all your boards (front and back). Make sure the photo is legible so that we can read the PCB printing clearly. A shutter speed of 1/125 or faster will produce nice clear focused pictures. Try to get a photo that is 2000x2000 resolution or as close as possible.

Here is an example of the pictures we want.

https://www.badcaps.net/forum/showpo...94&postcount=1

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

MOSFET 1

D1- and G1+ = Infinite resistance
G1- and S1+ = ~14.5 MΩ
D1- and S1+ = ~11 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~36 MΩ
D1+ and S1- = ~11.5MΩ

D2- and G2+ = ~9 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~9 MΩ

D2+ and G2- = ~11 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~11 MΩ

MOSFET 2

D1- and G1+ = Infinite resistance
G1- and S1+ = ~40 MΩ
D1- and S1+ = ~7 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~16 MΩ
D1+ and S1- = ~13 MΩ

D2- and G2+ = ~8 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~8 MΩ

D2+ and G2- = ~12 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~12 MΩ

MOSFET 3

D1- and G1+ = Infinite resistance
G1- and S1+ = ~15 MΩ
D1- and S1+ = ~7 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~37 MΩ
D1+ and S1- = ~12MΩ

D2- and G2+ = ~8 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~8 MΩ

D2+ and G2- = ~12 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~12 MΩ

MOSFET 4

D1- and G1+ = Infinite resistance
G1- and S1+ = ~15 MΩ
D1- and S1+ = ~8 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~36.5 MΩ
D1+ and S1- = ~12 MΩ

D2- and G2+ = ~9 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~9 MΩ

D2+ and G2- = ~9.5 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~9.5 MΩ

MOSFET 5

D1- and G1+ = Infinite resistance
G1- and S1+ = ~14.5 MΩ
D1- and S1+ = ~7 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~34.5 MΩ
D1+ and S1- = ~12.5MΩ

D2- and G2+ = ~7.5 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~7.5 MΩ

D2+ and G2- = ~11.5 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~11.5 MΩ

MOSFET 6

D1- and G1+ = Infinite resistance
G1- and S1+ = ~14.5 MΩ
D1- and S1+ = ~9 MΩ

D1+ and G1- = Infinite resistance
G1+ and S1- = ~33 MΩ
D1+ and S1- = ~11.5MΩ

D2- and G2+ = ~9 MΩ
G2- and S2+ = ~10 KΩ
D2- and S2+ = ~9 MΩ

D2+ and G2- = ~11.5 MΩ
G2+ and S2- = ~10 KΩ
D2+ and S2- = ~11.5 MΩ


Those are the values I obtained for each. I'm not sure if the infinitely resistive state between S1 and G1 on each chip indicates an open circuit or if it is simply the nature of the semiconductor design. I'll have pics up tomorrow. The caps I used are 3X Chemicon 35V 220uF.

[jetadm123]

Re: ACER AL2423W Monitor - Dead Inverter

Those are the values I obtained for each. I'm not sure if the infinitely resistive state between S1 and G1 on each chip indicates an open circuit or if it is simply the nature of the semiconductor design. I'll have pics up tomorrow. The caps I used are 3X Chemicon 35V 220uF.

The mosfets look to be ok. The circuit is generally setup where if one mosfet shorts out, then it shuts down the entire circuit. It will usually also blow one of the fuses, but in your case, the good fuses were a good sign.

You replaced the caps with 220's. Reading some of the other threads about this monitor, it seems that the original cap was 180uf. If yes, I don't know how senstive this board is to the change in value.

In your first post, you thought the mosfets were bad because there was no voltage from them. Where were you measuring this voltage on the mosfet?

Also, post some top and bottom photos of the power supply.

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

The original caps were 180uF. I don't know if it is an issue or not, because they appear to only be power filter caps, but I will have to have a closer look.

I measured the output on the MOSFETs at D1/D2 and am only getting a few millivolts output. S1 and G1 are getting 0V, but I think S1 is tied to ground anyway. S2 and G2 are both getting 19V.

Got the pics up. The back has nothing on it, only traces.

[jetadm123]

Re: ACER AL2423W Monitor - Dead Inverter

A few more comments/questions:

1) there looks to be a 20-pin chip in your inverter photo. I think this is the CCFL controller. What is the part number?

2) when the monitor is powered up, do you see an image on the screen if you shine a flashlight on it?

3) does the power led come on and stay on?

4) is there a flash of light (like the backlights trying to come on) on the screen when you try powering it up?

5) some members have reported swollen caps on the logic board. Have you observed any?

6) any other observations are always helpful.

[kc8adu]

Re: ACER AL2423W Monitor - Dead Inverter

check the fuses near each cap.blue/green parts marked fxx .

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

check the fuses near each cap.blue/green parts marked fxx .

Checked and double checked. Tested for continuity and voltage downstream. Not the fuses. Trying to figure out if the MOSFETs are blown or if the PWM is not providing a switching signal to the MOSFETs.

Edit: Look over the PWM data sheet (OZ964SN) and it indicates that it has "open lamp protection". Since I am testing voltages without the lamps attached, could the PWM be detecting this and not operating? Looks like I might have to retest with lamps attached, though when I last tested the monitor fully assembled, the lamps did not strike.

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

Okay, this is where things get weird. I hooked the lamps back up, attached my DMM probes to one of the lamp outputs, turned it on and I get a quick overvoltage beep from the DMM (rated for 600V). So that means the lamps are getting 600+V of power, at least briefly. Not wanting to damage my multimeter, I moved the probes to the MOSFET output and turned it on again. This time, the lamps struck briefly and went out (about 1/4 sec). I get a quick flash of around ~37V on the DMM, which is within spec for the MOSFET output. All 6 MOSFETs show about the same. So I think that rules out the MOSFETs and the PWM. It would seem that the PWM is powering the lamps briefly, detecting some kind of fault condition and shutting off the circuit.

So this could mean either one or more of the lamps are burned out, or there is some other component failure on the board, such as the comparator (LM393). I suspect, though, that it is the open lamp protection activating. Any way to test the lamps electrically, or am I going to have to open up the panel itself and visually inspect?

[jetadm123]

Re: ACER AL2423W Monitor - Dead Inverter

The suggested method on this forum for checking CCFL lamps is to obtain one good working lamp from another source and start substituting it in place of your current lamps one at a time. If you get to the point where all lamps light, then you've isolated the bad lamp. Of course, this method assumes only one lamp is bad.

Another suggested item to measure is the resistance of the inverter transformer secondaries. They should be approx. within 3% of each other.

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

Another suggested item to measure is the resistance of the inverter transformer secondaries. They should be approx. within 3% of each other.

All appear to be around 3K Ohms.

And to answer some of your other questions I hadn't gotten to:

Yes, there is an image on the screen if I shine a flashlight on it.
The power LED stays on.
I did have swollen caps on the logic board, but I've since replaced those.

[retiredcaps]

Re: ACER AL2423W Monitor - Dead Inverter

Okay, this is where things get weird. I hooked the lamps back up, attached my DMM probes to one of the lamp outputs, turned it on and I get a quick overvoltage beep from the DMM (rated for 600V). So that means the lamps are getting 600+V of power, at least briefly.

Most of us here never ask people to measure the AC voltage at the ccfls because the extremely high voltage (600 to 1000V).

[jetadm123]

Re: ACER AL2423W Monitor - Dead Inverter

Most of us here never ask people to measure the AC voltage at the ccfls because the extremely high voltage (600 to 1000V).

Good point. I should have mentioned this in my earlier posts. There have been several members who have burned out their meters because they did not voltage overload protection.

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

Yeah, I was misinformed about the CCFL voltage. I thought it was around 500V, but that is only the voltage between each pole of the transformer and ground. Between both poles, the voltage is over 1000V.

Anyway, did some more research, and looks likes the problem is a burned out lamp. Unfortunately, it looks like it is going to be rather difficult to get to.

[retiredcaps]

Re: ACER AL2423W Monitor - Dead Inverter

Between both poles, the voltage is over 1000V.

Anyway, did some more research, and looks likes the problem is a burned out lamp. Unfortunately, it looks like it is going to be rather difficult to get to.

Even the quickest flash of the backlight tells us something without having to measure the AC voltage. Thus the questions from jetadm123.

The lamp may not be burned out. It could be the wiring. See photos below to help determine which.

http://s807.photobucket.com/albums/y...klight%20lamp/

[Toasty]

Re: ACER AL2423W Monitor - Dead Inverter

Sounds like bad lamp shutdown.

Just under the CCFL connectors is a small gold plated dot. It is a test point and is connected to the SMD chip cap nearby (brown/red) that connects to one side of the CCFL. Probe that point while cycling power. You have to wait for the shutdown circuit to cycle, so wait 10-20 seconds between power on/off cycles. An peak-hold meter is perfect, or an analog meter for checking this, as a regular DMM does not usually respond fast enough to get a reading.

Run through measuring them all a couple times. The "oddball" reading is the likely suspect.

Strike voltage for U shaped tubes in 24" is 1500-1800 volts.

Toast

[alexanna]

Re: ACER AL2423W Monitor - Dead Inverter

Sounds like bad lamp shutdown.

Just under the CCFL connectors is a small gold plated dot. It is a test point and is connected to the SMD chip cap nearby (brown/red) that connects to one side of the CCFL. Probe that point while cycling power. You have to wait for the shutdown circuit to cycle, so wait 10-20 seconds between power on/off cycles. An peak-hold meter is perfect, or an analog meter for checking this, as a regular DMM does not usually respond fast enough to get a reading.

Run through measuring them all a couple times. The "oddball" reading is the likely suspect.

Strike voltage for U shaped tubes in 24" is 1500-1800 volts.

Toast

Toasty.
I am not sure if this is what you're talking about, but I have had some luck monitoring the voltage on the return wires of the ccfls, just recently I have a rash of ccfls with wiring problems that have been due to burning or poor soldering on the return wires.
What I have noticed on a ccfl that has wiring that is just barely hanging on by a couple of strands of wire, will have an obviously lower voltage during the 2second to black problem. This is really noticeable on the bar graft of the DMM .If you find one that is different than the rest, simply plug it in to a different connector on the inverter see if the voltage follows the ccfl.
I haven't run into a problem damaging my DMM yet.
Is the gold dot on the return side of the ccfl?

Al.

[Toasty]

Re: ACER AL2423W Monitor - Dead Inverter

CORRECTION: The brown/red chips are resistors, not caps. Look at the first picture. You can see the dots and where connected. They are at each tube connection and potential. On the other side of the resistor is a safer place to measure.

Bar graph on your meter, perhaps. But, not on mine or a lot of folks. That's why a peak-hold or analog meter was suggested.

The CR items in between and around the transformers are diodes that form the over current and over voltage protection and provide feedback to the OZ964 PS-PWM. On the back side shot the circuit traces running in between the gray caulk globs are those circuits. They eventually end up at either Pin 2 (voltage fb) or Pin 9 (current fb).

Toast

[Trinite]

Re: ACER AL2423W Monitor - Dead Inverter

Might try an analog meter, but I think my best bet is a visual inspection. Ah, the joy of breaking into a large metal box filled with extremely delicate components.