Dell 3008wfp PSU problem

[leapius]

Hi guys, I'm hoping you'll be able to help me fix the PSU in my Dell 3008wfp monitor.

Symptoms: Will occasionally show a picture for about a minute then go blank. After this the screen will come on again each time and then quickly go blank (1 second). There is no picture and no backlight when this happens but the power light stays blue.

I have confirmed it is definitely the PSU at fault as I've temporarily replaced it with a known working PSU and the monitor works perfectly.

I have read the 3008 Schottky mega thread but I'm not sure it's the D22 diode that is at fault in my case as the power light stays on and it does come on (more like 2 seconds to black). The monitor is not dead to the world in my case.

All components look healthy with the naked eye and Nichicon capacitors are used on this PSU. I thought maybe it could be a dry joint but my untrained eye couldn't find anything suspect.

What can I do to fix this? I have soldering skills but only a lowly standard multimeter.

Many thanks for any help

[Lumberjack777]

Re: Dell 3008wfp PSU problem

This is a new one on me. It's good you have/had access to a known good PSU. That helps to narrow the problem down quickly.
Excellent pictures. That really helps a lot. It definitely sounds like it could be a solder joint that is seperating as the board expands a little, due to the heat. I found what appeared to be tiny gaps on the soldering joints of 2 smd components. I circled them in red. D17 and especially D36.

If that's not the problem then you'll need to diagnose which rail is going out. For that, what would help is if you get that good PSU; put it back in and take voltage readings on the outputs. If you need any help on that, just come back and ask.

Hopefully its that tiny gap on one of those smd diodes. Unless I'm seeing things

[leapius]

Re: Dell 3008wfp PSU problem

Thanks for the quick repsonse lumberjack. You must have eagle eyes to spot that! I will reflow those joints tomorrow and get back to you with results. They seemed ok to me but I am no expert. A lot of the joints seem a bit dull and I am aware this PSU gets red hot. I can take closer photos of particular part of the board if you'd like? I have access to the good PSU and can run tests with it as required.

Many thanks.

[Lumberjack777]

Re: Dell 3008wfp PSU problem

Your welcome leapius. I think it's also possible on those smd diodes that there is a stripe on the cathode side that is giving the appearance, or an optical illusion, of a gap.
BUT..... if you could get a jewelers loop or some type of high powered magnification I would go over all of those solder joints with a fine tooth comb. Because your symptoms sound like a solder joint that is seperating due to heat expansion of the pcb. *cheers*

[retiredcaps]

Re: Dell 3008wfp PSU problem

Excellent photos.

Long shot. On the back of the pcb there seems to be a lot of white residue. This could be a reaction of some sort of flux with humidity? It is best to remove it by using isopropyl alcohol and clean toothbrush. If left on long enough it will eat away at the traces.

See

http://mrmodemhead.com/blog/fluke-77...per-corrosion/

[retiredcaps]

Re: Dell 3008wfp PSU problem

For that, what would help is if you get that good PSU; put it back in and take voltage readings on the outputs.

+1 on the above idea. It will be easy to compare the two readings (good and bad) and hopefully lead to a clue.

[Lumberjack777]

Re: Dell 3008wfp PSU problem

On the back of the pcb there seems to be a lot of white residue. This could be a reaction of some sort of flux with humidity? It is best to remove it by using isopropyl alcohol and clean toothbrush. If left on long enough it will eat away at the traces.

That's interesting RC. I just checked an extra PSU I have for the Dell 3008. It absolutely does NOT have any traces of white residue. Even with magnification.
The top side uses the white thermal paste on the heat sink. Unless maybe somebody got some on their fingers and transferred it to the bottom side? If its that same salty type of corrosion you linked to, that would be a tough one.

[leapius]

Re: Dell 3008wfp PSU problem

Hi guys, sorry for the late reply, I'm actually late for work right now so don't have time to post closer photos of those diodes but I looked with my jewelers loop and they seemed ok (no gap) possibly the light angle on that photograph. I do like the dry joint theory as this PSU does heat up a lot.

If you can tell me what outputs to check I can give you readings from both PSUs for comparison when I get back from work. There does seems to be a lot of that residue and I do actually have some isopropanol in stock so I'll clean it up as suggested retiredcaps.

Again thanks loads for the help.

[retiredcaps]

Re: Dell 3008wfp PSU problem

If you can tell me what outputs to check I can give you readings from both PSUs for comparison when I get back from work.

Put your black probe on chassis ground. Put your red probe on all the connector color pins. Report them like this.

CN800
pin 1 says 12V - I get 11.99
pin 2 says 5V - I get 4.99

etc

[leapius]

Re: Dell 3008wfp PSU problem

Put your black probe on chassis ground. Put your red probe on all the connector color pins. Report them like this.

CN800
pin 1 says 12V - I get 11.99
pin 2 says 5V - I get 4.99

etc

Thanks, I will do this tomorrow and get back to you. When you say 'pin 1 says..' how do I know what it should be? By the wire colour? Or do you mean what the good PSU says and then what the bad PSU says on the same pin? There are three output blocks to check.

[retiredcaps]

Re: Dell 3008wfp PSU problem

It doesn't matter how you number them. Just keep both sets of measurements in the same order so you can compare them.

[leapius]

Re: Dell 3008wfp PSU problem

Ok got it, do the boards have to be screwed to the metal casing to earth them? I ask this because the only way to get to the psu is by flipping it (and the a/v board) over but then I can't connect the tcon and inverter to it and the psu wouldn't turn on as a result. It would be easier to remove the metal cover so I can get to the points and leave the psu and av board in the same orientation. I guess I would need some sort of insulator to protect the underside of the board from contacting the metal of the panel, would anti-static bags work?

Lumberjack - I took a closer pic of D17 and D36 (see attached) I think the solder is ok on second look.

[Lumberjack777]

Re: Dell 3008wfp PSU problem

So it was the stripe on the cathode side. Not a gap. I kinda figured on second look. That would have been too easy!

Ok, I don't have that board in front of me because I'm at home. So I'll need to explain this from memory. About what retiredcaps is saying.

Check the top side of the board near the connectors for voltage markings. If it doesn't have any that's ok. What retiredcaps was saying is that you use the same numbering system for both power supplies.
Here is the way you take these voltage measurements on a Dell 3008. The psu and the mainboard are housed in a metal container. You can seperate that from the lcd panel. The connectors from psu to mainboard are still plugged in. You'll need to plug in the on/off power ribbon into the mainboard. Plug in a power cord of course. And also input some form of video signal from a computer. (vga, dvi, hdmi, etc)
The connector to the lcd panel does not need to be plugged in. Nor the inverter. (somebody correct me if I am wrong) But you can plug in the inverter to check for lights if you want to.
Without the on switch being on you are going to find 5v dc from one of the outputs. That's the standby power and it goes into the mainboard. All measurements are taken with the red probe on the connector lead. Black probe on a screw on the mainboard.OK, now when you turn the on switch on, the mainboard will send the PS_ON (power supply on) signal to the power supply board. That's when everything comes alive and all the voltages will be present. Some outputs will be 24v so make sure your multimeter is set above a common setting of 20v dc.

From your symptoms it sounds like the bad psu will give those voltages for about a minute then something is going to quit.

I now turn it over to others that can help you track it down, that are much more qualified than me. retiredcaps, budm, and a few others.
I hope that helps you understand how these readings are taken. **cheers**

[leapius]

Re: Dell 3008wfp PSU problem

Lumberjack, thanks loads for the detailed explanation - very helpful. I thought you'd need the inverter and tcon plugged in for the psu to stay on you see.

I will get the readings done in a few hours when I get home from work and post my findings.

[retiredcaps]

Re: Dell 3008wfp PSU problem

I guess I would need some sort of insulator to protect the underside of the board from contacting the metal of the panel, would anti-static bags work?

I use cardboard.

[leapius]

Re: Dell 3008wfp PSU problem

Well I seem to be falling down the rabbit hole here. I think I should point out that the good PSU comes from a 3008 that works normally but has a cracked screen (which I have temporarily).

I tested both PSUs last night and they both gave the same voltages on each pin of the three outputs. So I thought maybe I should heat the bad PSU up and put it under more load so I connected it to the cracked panel and left it powered on for about 40 minutes (partial picture obviously).

Came back and at this point it was too hot to touch but still displaying normally..

So I put the 'bad' PSU back into the original panel and powered it on and it works fine (for the 2 hours I tested it). This morning it went blank after about 20 minutes. After one power cycle it has currently been on for over 30 minutes so far (as I type this).

I'll give a quick run down of events from the start:

• Monitor stays on for a few minutes then goes blank. Each subsequent power cycle monitor goes black after 1 sec consistently.
• Try a working inverter. Nothing else touched. Monitor still goes blank in the same way.
• Replace old inverter and try a working PSU. At the same time I stick case fans over the TCON to pull some heat off through the top vents of the PSU casing. Works completely normally for the period I tested (about a day).
• Do the tests mentioned above.
• Put original PSU back, leave fans off. Goes blank after 20 mins. Immediately power cycle and currently on for over 30 minutes and counting (fans still off).

Can anyone understand what on earth is going on? I thought maybe a dry joint on the TCON (the Gateway 30inch seemed to have a lot of problems with this iirc) but this thing is red hot right now and the picture is still on. When I first opened the monitor all the cables with properly in place (factory glued down with no movement).

Many thanks for any help.

[leapius]

Re: Dell 3008wfp PSU problem

Quick update: I turned it off and put the fans on. Turned it on again about 90 minutes later after it was completely cold. Went blank after ~5 mins. Turned it off/on, went blank again after ~3 minutes. Off/on again - currently still on (for about 2 hours so far). Kept the fans on this time and as before, the blue LED always stays on but the blank events occur at seemingly random intervals.

I'd love to test the PSU when it is in this 'blank' state (without power cycling it) so I can see if the voltages are different but it's inaccesible when connected to the panel and without the panel I don't know if it's gone blank again.

[jetadm123]

Re: Dell 3008wfp PSU problem

Have you tried swapping the main board (board that you plug your video cable into) from the cracked screen monitor into the defective monitor?

[leapius]

Re: Dell 3008wfp PSU problem

Ok, I again moved the bad PSU back to the monitor with the broken panel. This time it went blank with renewed regularity (virtually every minute or so at the moment).

From this I think we can assume it is indeed the PSU that is at fault as the symptoms are confirmed to have moved to the other panel when the PSU is swapped in isolation (knowing that all the other boards on the broken panel monitor are working). The question now is how to test the voltages once I know the screen has gone 'blank'.. There is no way to know if I unplug the panel. Any ideas? Thanks!

[leapius]

Re: Dell 3008wfp PSU problem

I have now re-tested the bad PSU as before and have found something I think might be interesting. Pin 11 on connector P801, which goes to the a/v board is acting weird. (see attachment for reference).

On the good PSU this stays a stable 18.18v and never fluctuates.

On the bad PSU I oberved it fluctuating (over about a minute) when I turned it on in the following sequence:

19.11v
19.65v
19.18v
19.71v
19.21v
18.8v
19.18v
18.86v

At that point it seemed to stabilise. The other difference between the two was pins 1-5 on P803 (going to the inverter) are all 24.3v on the good PSU and 25.1v on the bad one. However, this does not fluctuate and I'm guessing that this might be ok.

The only other thing I've noticed is when the screen goes blank and I turn off/on again the blue light momentarily flickers once when I turn it on. In normal operation this does not happen. I noticed this effect when I was doing the above testing so maybe the PSU had actually malfunctioned at this point.

What do the experts think? Thanks.