HP w20 LCD Display Inverter, running out of options.

[CapBlown]

Having issue with HP w20, 20" LCD, Manufactured 2006.

Display works perfecly about 15 minutes then it suddenly brightens a bit and immediately after this backlights are turned off. When monitor has been operated to this point, and I restart the monitor it opens up, immediately brightens a bit, and backlights are turned off again. Letting it cool down for half an hour and I get another 15 minutes before baclights go off.

Problem must be on inventer board as image is still there. So I visually inspected all capacitors, they seem to be allright, could it still be a capacitor issue? Or could it be a dry joint?

I resoldered almost everything on inverter board, especially transformers as I found out looking other threads that my problem is close to 'two seconds to black' -symptom. Now tested with backpanel off and monitor operated nicely about 4 hours, then backlights turned off again.

Is this a heat issue? Those four 1000 uF and two 1500 uF caps are very hot. As ventilation was now so good, I tried to cool them down with cold spray. It did not help at all. So I froze all other components with it too... problem remains.

I took another approach... after monitor cooled down, i took a blowdryer and heated up the components on inverter board for minutes. Then I turned the monitor on expecting it to shut down immediately. But it worked for half an hour now before backlights went off.

Obviously the problem is with inverter but I don't know what parts should I take for closer inspection. Maybe I'm going to resolder transformers again?

Any other suggestions will be appreciated. I'm confused.

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Having issue with HP w20, 20" LCD, Manufactured 2006.

Display works perfecly about 15 minutes then it suddenly brightens a bit and immediately after this backlights are turned off. When monitor has been operated to this point, and I restart the monitor it opens up, immediately brightens a bit, and backlights are turned off again. Letting it cool down for half an hour and I get another 15 minutes before baclights go off.

Problem must be on inventer board as image is still there. So I visually inspected all capacitors, they seem to be allright, could it still be a capacitor issue? Or could it be a dry joint?

I resoldered almost everything on inverter board, especially transformers as I found out looking other threads that my problem is close to 'two seconds to black' -symptom. Now tested with backpanel off and monitor operated nicely about 4 hours, then backlights turned off again.

Is this a heat issue? Those four 1000 uF and two 1500 uF caps are very hot. As ventilation was now so good, I tried to cool them down with cold spray. It did not help at all. So I froze all other components with it too... problem remains.

I took another approach... after monitor cooled down, i took a blowdryer and heated up the components on inverter board for minutes. Then I turned the monitor on expecting it to shut down immediately. But it worked for half an hour now before backlights went off.

Obviously the problem is with inverter but I don't know what parts should I take for closer inspection. Maybe I'm going to resolder transformers again?

Any other suggestions will be appreciated. I'm confused.

Well, you did a good deal of troubleshooting. I wish I had some solid answers for you, but I don't. I do have a few suggestions, however.

One relates to the design. The inverter has been described as a constant current source (I think of it as a current limited source) driving the CCFLs. Actually, in this case it is TWO sources, each driving three CCFLs. Obviously, suddenly the current through one or more of the CCFLs rises. This 'trips' the current protection and the inverter shuts down. NOW, is this an inverter or a CCFL problem? Your experiment would seem to indicate that the inverter heating up is NOT the problem. And the age of the monitor makes a dying CCFL a possibility.

One approach would be to replace one or more of the CCFLs with either another CCFL, or with a resistor. I understand about a 120K ohm 5 or 10 watt resistor should do it. If substituting a resistor for a specific CCFL solves the problem we can be pretty sure it's a dying CCFL. But there are a few things to check first.

1. Does the voltage to the inverter change, causing the shutdown? That's a relatively easy thing to monitor.

2. Does the brightness control voltage suddenly change, triggering the shutdown? Again, easy to monitor.

3. The hot capacitors are worrisome. Why are they hot? Is it from the heat sinks, or are they handling more surge current than designed? What brand are they?

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thank you PlainBill.

I'll try to figure out how to arrange voltage change measurement on inventer. I have only a simple multimeter, limited tools and no schematics whatsoever so I need to find out safe measurement points. I've no idea on where brightness control voltage can be measured.

Hot capasitors are Rubycon, they are between 2 heat sinks (Attached image), and I don't think heat sinks make them hot.

I resoldered lots of connections earlier today and put the monitor back together. Now it took about 50 minutes before backlights turned off so something is maybe better now as it used to serve only 15 minutes.

Something else I noticed, high frequency twirling sound comes out just before backlights turn off.

P.S I've fixed only some motherboards with blown caps and DC stuff mostly so I may be out of my league with this TFT and AC powers in it. I rather wear rubber boots and keep my other hand on pocket when touching active device .. or measure stuff when they're off board... ;-)

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Thank you PlainBill.

I'll try to figure out how to arrange voltage change measurement on inventer. I have only a simple multimeter, limited tools and no schematics whatsoever so I need to find out safe measurement points. I've no idea on where brightness control voltage can be measured.

Hot capasitors are Rubycon, they are between 2 heat sinks (Attached image), and I don't think heat sinks make them hot.

I resoldered lots of connections earlier today and put the monitor back together. Now it took about 50 minutes before backlights turned off so something is maybe better now as it used to serve only 15 minutes.

Something else I noticed, high frequency twirling sound comes out just before backlights turn off.

P.S I've fixed only some motherboards with blown caps and DC stuff mostly so I may be out of my league with this TFT and AC powers in it. I rather wear rubber boots and keep my other hand on pocket when touching active device .. or measure stuff when they're off board... ;-)

Hot Dawg!!! An audible clue. A high tech piece of equipment such as a 'Sonic Sniffer' might help locate the source of the 'high frequency twirling sound'. If you don't happen to have one in your shop, perhaps you can find a substitute in another room of your home. A cardboard tube, such as teh core from a roll of paper towels or toilet paper will do the trick. Hold one end to your ear and aim the other end at possible candidates for the source of the noise. (The most likely suspects would be the inverter transformers and the main smps transformer.)

As far as monitoring the inverter voltage, I was referring to the inverter input voltage. It would be another clue as to what is happening. If it drops, the conclusion is something is drawing too much current from the 12 V supply. If it increases, that would indicate something may be loading the 5 V supply. (this is AS the display is overly bright.)

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks again PlainBill.

I measured the inverter input voltage and it stayed at 12.62 V for almost 2 hours. Then all of a sudden backlights went off while voltage gets up to 14.11 V.

On standby mode, voltage is 13.08 volts. With hot or cold monitor.

So I get too much voltage, but why? What could cause this? I have no idea.

I can't locate the source of twirling sound. Tried with the core from a roll of paper towel but the frequency is so high it's all over my head.

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks again PlainBill.

I measured the inverter input voltage and it stayed at 12.62 V for almost 2 hours. Then all of a sudden backlights went off while voltage gets up to 14.11 V.

On standby mode, voltage is 13.08 volts. With hot or cold monitor.

So I get too much voltage, but why? What could cause this? I have no idea.

I can't locate the source of twirling sound. Tried with the core from a roll of paper towel but the frequency is so high it's all over my head.

The usual SMPS design used in monitors has a single transformer with two secondaries, one for 5 volts, one for 12 volts. The design regulates the 5 volt supply very closely; the 12 volt supply will vary slightly depending on the load on the 5 volt supply (the higher the load on the 5 volt supply, the higher the output voltage on the 12 volt supply.

It would be interesting to know if the 12 volt supply increase was because the inverter shut down, or caused the inverter to shut down.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

I measured 5 Voltage line (5.13 V) and it drops down to 4.95 V when 12 V increases to 14.11 V. So I decided to focus on inventer as it's the scary thing for me.

I got back to freezing components and finally find out that freezing two high voltage caps gives me 10 more minutes operating time. Tested this several times and when I moved these caps a bit, I suddenly got 30 minutes more operating time. At least one of them must be broken. So I resoldered them again and now the damn thing stays on for 1 second only. I think I better buy and change all eight caps before I try anything else.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Are these 33J 3KV caps 33pF 3000V? Hard to find if so.

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Are these 33J 3KV caps 33pF 3000V? Hard to find if so.

Available from Digikey.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill but they have 18€ shipping bill.. a bit much for this project. I think I'm going to put this on hold until I find local seller with cheaper delivery options or scrapped monitor with similar parts.

[Bobdee]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill but they have 18€ shipping bill.. a bit much for this project. I think I'm going to put this on hold until I find local seller with cheaper delivery options or scrapped monitor with similar parts.

Hello
I have read a few times by members on this forum that these high volt caps rarely fail, but when they do, they burn or bulge out, meaning you can see the failure (no 100% guarantees mind)
So it might save you some money by taking Bills earlier advice and check those CCFLs first

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill but they have 18€ shipping bill.. a bit much for this project. I think I'm going to put this on hold until I find local seller with cheaper delivery options or scrapped monitor with similar parts.

That's the reason users are encouraged to place their location in their profile.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill and Bobdee. Updated my profile.

I may try to replace CCFL's with 120 K resistors just to be sure, have to go shopping first anyway. Though I know that this monitor has been used only for couple of years in light use and it's been sitting in warehouse half of it's life. So CCFL's should not be even close to end of their lives.

[pedro]

Re: HP w20 LCD Display Inverter, running out of options.

Hello
I have read a few times by members on this forum that these high volt caps rarely fail, but when they do, they burn or bulge out, meaning you can see the failure (no 100% guarantees mind

Bear in mind that the HV cap is operating at only 2x AC line frequency. ESR is rarely if ever an issue there. If they fail non-catastrophically the symptom would be a high ripple on the HV DC input to the switcher. If your DMM is capable of reading AC on top of DC this will be fairly obvious.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Hi again.

As I got my fingertip burned on one of those HV caps, and found a way to get them cheap, I ordered replacements just in case. I changed "hot" ones but nothing much was achieved.

- They don't get hot anymore, I can touch them after 1 hour of usage. Better quality perhaps?
- Shutdown sequence changed a bit. Screen does not get lighter anymore just before it shuts down. It just turns into black.

Also I think that I finally located the twirling sound: It probably comes from B20100G. That is probably a schottky power rectifier. According to wikipedia these things may have relatively high reverse leakage current which increases with temperature and may lead to a thermal instability issue. Could this be a reason why those bigger caps get so hot? Maybe I should try to change this rectifier?

I know I should inspect CCFL's too but they are so well hidden. I even got those 3 film layers out of the panel, but could not get into CCFL's. Darn thing is packed too well I just can't get into it.

It's also strange that when I soldered those new HV caps (monitor has not been used for days) in I got just a sec or two of operating time. But after letting them cool down for about 10 minutes after soldering, I got it workin for 20 minutes with backpanel on.

Now screen stays on more than one hour when backpanel is not on, and only 20 minutes when it's on, so this problem must be caused by heat. Hottest part inside is B20100G.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

It seems I still need some expert help with this one.

1) In the beginning monitor stayed on for 15 minutes, then backlights went off.

2) After resoldering and applying freezer to components I thought that I finally located possible problem to HV caps on bottom side inverter. Cooling them got me extra time so I ordered new HV caps and replaced them but not too much improvement was gained so I thought that there must be cold joints nearby and resoldered all I could.

3) As 1000uF and 1500uF caps were really hot after 10 minutes and schottky was making noises, I changed 1000uF, 1500 uF and 470 uF caps to LOW ESR versions and replaced B20100G schottky with new one.

4) Now I have 1,5 hours of operating time with perfect image even when backpanel is attached... and then again backlights are turned off. It's better now than the original 15 minutes... but I still don't know the problem. No audible clues anymore.

5) Letting the monitor cool down for about 15-20 minutes I get another 1 to 1,5 hours of operating time, no problems whatsoever and then again 2 secs to black.

6) When I try to restart the monitor when it has failed the 2 seconds to black problem occures until 15 to 20 minutes have passed, then it starts and stays on again for 1,5 hours.

I'm going to mental asylum with this one. Help please... Should I start inspecting CCFL's? Is it safe to measure voltages there with normal multimeter?

P.S. In attached image red ones resoldered, green ones replaced.

[retiredcaps]

Re: HP w20 LCD Display Inverter, running out of options.

3) Is there a cap underneath that heat shield? Just right of the big cap on its side. If yes, did you replace that one?

4) Just curious if what settings you have for contrast and brightness. Is it at "torch" mode where brightness is 90+? If yes, try 35 for brightness and 35 for contrast to see if you get longer running time?

3) As 1000uF and 1500uF caps were really hot after 10 minutes and schottky was making noises, I changed 1000uF, 1500 uF and 470 uF caps to LOW ESR versions and replaced B20100G schottky with new one.

4) Now I have 1,5 hours of operating time with perfect image even when backpanel is attached... and then again backlights are turned off. It's better now than the original 15 minutes... but I still don't know the problem. No audible clues anymore.

P.S. In attached image red ones resoldered, green ones replaced.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

3) Is there a cap underneath that heat shield? Just right of the big cap on its side. If yes, did you replace that one?

Thanks retiredcaps, there is indeed something underneath it but I did not know what it was. There is no any markings on it. I remember I measured it earlier today thinking by the looks of it that it must be a diode but it eventually acted like a capasitor. I did not pay too much attention to it. Do you have any idea what capacitance it should have? I have no schematics.

4) Just curious if what settings you have for contrast and brightness. Is it at "torch" mode where brightness is 90+? If yes, try 35 for brightness and 35 for contrast to see if you get longer running time?

Yes, I tried a torch mode and I tried to reduce brightness to 20. There was no noticeable difference in operating time. I will try this again now as I've changed lots of parts since I tested this.

[retiredcaps]

Re: HP w20 LCD Display Inverter, running out of options.

What does the circuit board say? If it starts with a C it is a capacitor. Something like C204.

Thanks retiredcaps, there is indeed something underneath it but I did not know what it was. There is no any markings on it. I remember I measured it earlier today thinking by the looks of it that it must be a diode but it eventually acted like a capasitor. I did not pay too much attention to it. Do you have any idea what capacitance it should have? I have no schematics.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

It's L106. Some sort of choke perhaps?