Benq G2255 doesn't power on - amber power light when plugged in

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

ok, with diode tab lifted, at the setting of 1V X 1 multiplier, I get 1.57V vpp for R837 and 1.73V vpp for R841.

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Doing tests with pin 5 jumped, both lamps on bright, right?

We're looking for about 12Vpp.

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

This is where we were:
With pin 2 lifted, meaning that we are not using the LI1 monitoring circuit, the lamps would light without jumping pin 5. You could invert the lamp plugs and both lamps lit without hesitation as long as they were only monitored by LI2 (p3). This is very important, because it ruled out a bad lamp.

This is where we are now:
With either pin lifted, the lamps come on sometimes, and sometimes they don't. It doesn't matter if LI1 or LI2 is doing the monitoring. This indicates that it could be a bad lamp. It's like we're teetering on the edge of pass/fail fault monitoring, but now it's on both LI1 and LI2.

Question:
How do we get back to where we were? What all changed between when you ran the tests? Was it putting the original mosfet back in? Is a subtle change on the primary side, kicking us over the fault threshold? Which goes back to why did the lamps light up for 10 seconds (w/o anything jumped) when you put the under rated mosfet (AF) in? Did it in fact blow?

If we can't get consistent test results, we're never going to be able to rule anything in or out as the problem. You can put the diode tab back down, but please don't change anything else until you confirm or modify what was said above.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Hi Will,

I understand that we are getting wild results, but it may be because of putting the original mosfet back in (as for the underrated one, my mistake was to not keep it, I assumed it was blown and threw it out, but because the original was the original I kept it).

Could it be that if there is a bad lamp (when they light up, they look normal to me, but it is only a white screen, so it's hard to tell if they are both perfect), is it getting worse with the tests?

Maybe we should go to the lamp route, and use Budm's method with one of the F13TT lamps I have. I could plug it into the CN801 and see if the lamps go out?

What do you think?

Edit: Also, I am not sure that my oscilloscope is very precise, sometimes it gives me results that I cannot replicate, so please bear with me and thank you for your patience.

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

To test the power to the lamp, Budm's method is the best option I've seen unless you had extra ccfl bulbs laying around. Not sure if the bulb you have will work. You would have to get Budm's input on that.

Another thing I'd like his experience on, is what are the odds of the inverter section of the power board, and the main board failing at the exact same time. The only common
links are the enable and dimming connections. Is it possible the driver chip doesn't like the 3.9V that you have jumped to it for dimming. In other words, is it possible the problem we're looking for doesn't exist when proper signals are received from the main board.

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I'm adding a couple of diagrams from the MP1009 datasheet and some comments. This is pertaining to my above comment about the signal from the main board. Do you follow what I'm talking about on it?

The mosfets may test differently, the lamps may be suspect, but when you disable current monitoring, both lamps come on immediately without any flicker. We haven't found any bad parts yet, and we've got to consider that there may not be any.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

To test the power to the lamp, Budm's method is the best option I've seen unless you had extra ccfl bulbs laying around. Not sure if the bulb you have will work. You would have to get Budm's input on that.

Another thing I'd like his experience on, is what are the odds of the inverter section of the power board, and the main board failing at the exact same time. The only common
links are the enable and dimming connections. Is it possible the driver chip doesn't like the 3.9V that you have jumped to it for dimming. In other words, is it possible the problem we're looking for doesn't exist when proper signals are received from the main board.

There is a user in another thread that says that it would work, but Budm has not responded yet. https://www.badcaps.net/forum/showthread.php?t=19987

Your other point is quite good, in that yes, what are the odds? Unless when the power board went out, it damaged something on the main board? So would a proper test of this is to remove the 3.9V for dimming? I have it connected with a 1K resistor to the 5V, should I try and connecting it to the 3.3V?

Or are you saying we would have to find the problem on the main board first to get it to work?

[Dumah Brazorf]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I'll would rather use a 10k resistor for dimming instead.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I'm adding a couple of diagrams from the MP1009 datasheet and some comments. This is pertaining to my above comment about the signal from the main board. Do you follow what I'm talking about on it?

The mosfets may test differently, the lamps may be suspect, but when you disable current monitoring, both lamps come on immediately without any flicker. We haven't found any bad parts yet, and we've got to consider that there may not be any.

Oh yes I understand very well what you are talking about. Just not sure how to rule this out without getting the main board working....I was able to get the firmware in a zip file for this monitor from a Russian website. Maybe we could try to flash the firmware and see if that fixes it?

I have a Uno board and could unsolder the Eprom chip and try to program it? I realize that it is a big task for which I'm not sure I would succeed, but what the heck! Let me know what you think.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I'll would rather use a 10k resistor for dimming instead.

Well Dumah, I'll give this a try as soon as I get that diode resoldered (and find a 10K resistor). Thank you for the tip. I'll report back as soon as I have tried it.

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Yeah, try what Dumah said and see what the voltage goes down to. If that doesn't get it down, we'll use a resistor divider. Solder 2 equal resistors (value doesn't matter, just over 1k) soldered end to end. Another resistor of value several hundred to 1k soldered at the junction of the other 2, making a "T".

One end of the equals to ground, other end to 5V. The resistor in the middle to dimming. That will cut your voltage in half. Try it at 5V and 3.3V.

It's possible that the MP10091 is getting that oscillation for burst dimming from somewhere else. The problem is that we only have the datasheet for the MP1009. We know the pinout has changed, and we're left guessing what else has changed.

The only common connection between the main board and the inverter section, are the enable and dimming lines. I guess a short on the main could have zapped the driver chip via those lines, but the chip seems to be working. Otherwise, you had 2 random failures on 2 different boards at the same time.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Ok, I noticed that we have 3 resistors one on the blue, one on the green and on the white all with 5 volts. Not too sure what wire is the dimming is it the blue? The green would be enable and what is the white?

I think Will you said pin 7 is the dimming, so I am trying to trace pin 7 back to the connector but I get lost along the way?

I resoldered the diode and made a 10K resistor jumper, so I am ready to go, but want to do it the right way. Do I replace the 1K resistor on the blue wire with the 10K?

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I think the connector is like the one in the pic. BRT_ADJ would be pin 1. Jump the 1k from blue to 3.3V

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Will, 2 things. Do you mean the 1K resistor or the 10K resistor on 3.3V?

Also, in the diagram you just attached (which is right on with my connector) we have a 1K resistor on the power saving mode (pin 5)...should I remove that, maybe that is causing us the headaches, no?

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Oh, oh, things are getting weirder. I decided to switch the blue wire on to 3.3V with the 1K resistor and try it out.

Before making the test, I measured voltage on the blue pin with board upside down...I got 3.228V. While I was doing that I was hearing like a very faint crackling noise so started to investigate what it could be. Then after a few seconds I saw a bit of smoke coming from under the board (the top side) so I unplugged everything.

Turned the board around to inspect it and this is what I found. C817 had started to roast (see picture).

I haven't touched anything, just measured resistance at C817 and I get approx. 450ohms like before.

What's going on?

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

I was talking about 1k from 3.3V to blue power saving. I can't find anything that says what the normal signal to power saving should be. Looks like your original jumpers had 5V going to it. Just move it over to 3.3V (orange). You can try powering on with and w/o the jumper to power saving.

I believe Dumah was talking about using a 10k resistor to lower the voltage going to BRT_ADJ. We need to get that voltage down to around 1.5-2.5V to see if that changes anything. If necessary, use the resistor divider I mentioned above

On post 215:
That's not good. C817 goes between the pins on the transformer secondary. I would remove it and check resistance between 7 and 8 on the secondary. Also check for shorts to ground. What could have caused that?

[Dumah Brazorf]

Re: Benq G2255 doesn't power on - amber power light when plugged in

You're not powering the board without always connecting the ccfls, don't you?

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

Will: Ok, that is what I had, the 1K resistor connected from 3.3V to blue.

Dumah: Yes I did power it on without the CFL's just wanting to check what voltage I was getting on the blue wire. I shouldn't...I thought that the power to the CFL's was turned off when they are not connected?

Not sure what to do now. I removed the C817 and checked the resistance between 7 and 8 of the secondary and I get 453 ohms, just like before. That capacitor (charred) still measures 12pf. I inspected all my soldering, the last thing I had done was to put down the top pin of the diode and solder it and don't notice anything special. I have some 5 pf HV caps, and one of them measures 14pf (if my transistor tester is accurate). Should I put that one in place of C817, or the old 10PF that I still have, or the Frankenstein cap that I had made but dismantled?

[will62]

Re: Benq G2255 doesn't power on - amber power light when plugged in

453 ohms is ok for the secondary. Good catch on Dumah for recognizing the cause. I wouldn't put ole crispy back in there. I believe that cap and the inductance of the secondary form a tank circuit. It's value determines the resonant frequency of the circuit. That's the point that the max power can be absorbed for hitting striking voltage. If you increase the cap value, you lower the resonant frequency. The opposite for lowering cap value. It's over my head to tell you how a 2pf difference would impact lighting the lamps. Better rebuild Frankie, or we'll be wondering about it down the line.

[rddube]

Re: Benq G2255 doesn't power on - amber power light when plugged in

ok, I will rebuild Frankie and test with the CFL's plugged in. Will report later.

Thank you for your help and patience!