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

[will62]

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

The conical shape may be something related to MPS's Nu-Pulse technology, not sure.
Now we're going to bypass the chip's fault protection and take a longer look at the TG and BG signals (same as earlier). Need to jump pin 5 to ground. An alligator clip across C803 should work, if it'll hold. After the clip is on, measure resistance between pin 5 and pin 15. Should be near 0 ohms. Run the test for about 5 seconds to get a better look at the waves. Don't want to run it too long because shutdown may be due to over-current. See if you can display the frequency on your scope.

Questions
1. Do you have a hole on the back of that panel so you could see the backlight if it came on? If not maybe you could see some glow at the corners where the cables go.
2. Does your DMM have frequency function? If so, you could try touching the top of the transformer (not the pins, just the top plastic) with your red probe (black to ground) while you are running the mosfets, to see if you pick up the oscillation frequency.
3. Did you check your probe on the scopes 1kHz signal? There's a loop at the top of the scope to hook on to. Pulse nice and square, or conical? (set horizontal to 1 ms).

[rddube]

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

Ok, so did the tests with the oscilloscope as you explained Will, and this is getting to be fun!

First off, pin 11, when I put power on, I get a wave not perfectly square, but sort of conical with about 12-13v amplitude for approx 1 second and then it goes flat.

I waited a little until pin 7 & 9 were almost at 0, redid the test on pin 11 and got the exact same result.

Waited another while until pin 7 & 9 to 0 volts, then tried pin 14. Got about the same wave (conical) of about the same amplitude but for maybe 1/2 a second, somewhat shorter time than pin 11, and then flat. Did the test a few times and got the same results.

So, what is next, I'm starting to like this. I left the setup as is.

[rddube]

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

Thought I had found a resistor with wrong value, but my mistake. So this post is not important, please ignore as I can't delete it.

[rddube]

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

Ok, tks Budm and welcome back.

Tks for the tip, will do my tests and report.

[budm]

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

The lamp connection can be used in any directions, they are just 2 two individual lamps.

[rddube]

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

Ok, will try my best without blowing up the house!

Resistance across C808 is 460K ohms.

[will62]

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

Not sure if connector swap will create a problem, but let's not add another variable to the mix. Lay panel face down. Set power board top side up on panel. Connect your ccfl wires as normal. Now flip the other end of power board over, exposing bottom of board. Insulate with cardboard between panel & PB.

BG and TG can never have power at the same time without blowing mosfets. There is some dead time between switching.

Also, please check resistance across C808.

[rddube]

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

If panel face down, then the CFL's will be inverted in the power board, is that ok (the lenght of the cfl wires are not long enough to plug them in their respective sockets)? Also Will62, the voltage measurements I did in previous post was without the CFL's connected...and you can see that pint 14 is 0V. Should I do voltage readings with the CFL's connected before doing the oscilloscope measures and report?

[will62]

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

Yes. Panel face down. Use a sheet of cardboard to insulate between panel and power board. I assume that you are using a power strip near your setup, so that you can turn panel on/off quickly w/o having to plug/unplug PB. After you run tests, leave the setup as is (more tests to follow).

[rddube]

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

Do I need to do these tests with the CFL's connected? That would be a little tricky, because the power board needs to be side up when I connect them?

[will62]

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

Time to dust off your DSO138. We need to see if the Mp10091 chip is sending a PWM signal to the mosfet gates. You should probe pin 11 (TG), and then pin 14 (BG). Need a nice square wave on both. If the chip is working, but going into shutdown, the signal won't last more than about a second, so you need to have everything connected and eyes on scope when you turn the power on. Between tests, either drain caps or wait until pins 7 & 9 are near 0V. To avoid accidentally shorting pins and frying the chip, I'd recommend that you trace out to the nearest component and probe there.

There are tons of videos about using the DSO138 on youtube. If you haven't used the scope very much, maybe watch a few videos. We are looking for a square wave of ~12V at a frequency of ~50kHz. A vertical of 5v/div and a horizontal of 10-20us should be OK.

[rddube]

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

Does it measure the same swapping the probes of the multimeter?

Yes it measures exactly the same thing when swapping the probes 80.4 ohms both ways.

[rddube]

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

Ok, so checked the Fet and it turns out it is not cooked, it still gives good readings.

So I rechecked all voltages on the controller chip, all 3 resistors in place, and here is what I get:

Pin 1: 5.85V
Pin 2: .136V
Pin 3: .136V
Pin 4: .092V
Pin 5: .001V
Pin 6: 1.206V
Pin 7: 3.58V
Pin 8: 3.92V
Pin 9: 15.15V
Pin 10: 15.13V
Pin 11: 13.4V
pin 12: 13.6V
Pin 13: 5.89V
Pin 14: 0V
Pin 15: 0V
Pin 16: 5.54V

Tried the power board again connected to the CFL's with all 3 resistors in, nothing happens. Rechecked the FET and it is still good? Rechecked my soldering of the FET to make sure it as good, and all is good.

So not sure what gives.

[will62]

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

Yeah, the ones on ebay from China were closer to the fets that you had. The TI one is closest that I could find from Digikey (60V, 7A, 23m, 2.1w). Let it sit for a while. Then get back and check all your voltages, and recheck the fets to make sure they're blown. Check them the way the guy in this video does. (G=p2, D=p8, S=p1)

https://www.youtube.com/watch?v=RkWy1EirEu8

[rddube]

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

Yep, just checked and that is the way I installed it i.e. pin 1 in the right place. They say in the description for this Mosfet:

This dual SO-8, 60 V, 23 mΩ NexFET™ power
MOSFET is designed to serve as a half bridge in low-
current motor control applications.

Probably not the right one?

[will62]

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

Drain 1, pins 7&8 are tied together, as are drain 2, pins 5&6. They should have 0 ohms between them (7 to 8, 5 to 6). That should tell you where pin 1 is.

[rddube]

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

Well, total disappointment. Got the Digikey order in, installed the CSD88539ND Mosfet, checked resistance between S1 and D1 - 350K ohms, same with S2 and D2, and gave it a try. Lights did not even flash, nothing.

Checked the resistance between S1 and D1 and now get 1.4 Mega ohm...so it's shot.

Rechecked the data sheet to make sure I had pin 1 in the right place, but on the datasheet they show a dot but there is none on the mosfet (photo attached). I assumed the line meant the top of the Mosfet which I am pretty sure gives pin 1 where I indicate in the photo.

So, this Mosfet doesn't work....shucks.

[will62]

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

Rddube,
Responding to your questions/points:

1. On the Ohms with and w/o caps, understood now.

2. Post 9&10: Over my pay-grade to try to explain it (or understand it), but the tl431 controls the optocoupler's path to ground. It varies the intensity of the led. That feedback returns to the high voltage driver IC, which in turn adjusts the PWM to the mosfets. Someone posted earlier that the blue wire may be the power saving signal.
I think they were right. It might just be a high or low signal from the main board, which adjusts the tl431. You read ~.77V, which is enough to "turn on" most zeners. You'll never read voltage (0.04V) on the right side, until the left exceeds the "zener/knee" voltage. They may have used it to protect the main board.

3. Russian forum translation: What I got out of that, was that the voltage regulator was for 1.8V, and he was getting 2.5-2.8V (regulator hot). He also ended up having to replace the TSUMU58 chip to fix his problem.

4. You may not have a problem with the 3.3V on the main board. At the regulator, you read 3.3V in and 1.8V out. The 3.3V is generated on the power board. If you had a short to ground (or ohms too low) on the main board, I'd expect to see a voltage drop at the regulator.

Now, when you get a chance, go back to that 1.8V regulator, and check resistance from the 1.8V to the TSUMU58 chip pins 53,74,104 &126.

Thanks

[Dumah Brazorf]

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

Does it measure the same swapping the probes of the multimeter?

[rddube]

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

The 80 ohms does sound low. Above, you mentioned that the caps measured 85 ohms when removed. That sounds way too low for a good cap. You might want to remove those again and recheck resistance (3V to gnd).

Also, please check the area on the back of the board, circled in the pic. Looks like something scraped across the board there.

Hi Will62,

Tks for hanging in there. What I meant, was that the resistance was 85 ohms on the 3.3V to ground even with the caps removed. The caps I checked with my multi component tester and they tested fine...I also checked their resistance and it was in the kilo ohms, so they are not the culprit. So it has to be something else along the line of the 3.3v rail.

As for the picture, here is a closeup of the area with another photo of the underside after cleaning it up with isopropyl alcohol.

Will, could you review posts 9 and 10 and tell me what you think?

I am also attaching a pdf document of a conversation I found on a Russian website that seems to be discussing a problem with the voltage regulator, but it is a translation so not too sure what some of the words might mean, like fieldwoman? The number on that voltage regulator is very close to mine which is 210151DA and under that number I have 1GC0091. Might spring some thoughts.

Still pondering..........