Viewsonic VA902b dim lights, then black

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

When I jumper R505 the lights come on really bright (slight buzzing sound on the board also) and then off after 2 seconds... Per PlainBill's instructions I did not jumper the other items.

That would indicate the over-voltage protect circuitry is limiting the output voltage. But it also indicates another problem exists, probably the output current rises too high. I would look in the area directly below CN502, specifically at R511, R513, R512, and the two components below R512 and D508. What is the discoloration in that area?

PlainBill

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

Lecture time. There won't be a quiz later. However, the last two paragraphs are the next test.

An inverter is designed using a feedback loop, and a number of protection features. The usual feedback loop senses the current through one or more CCFLs and adjusts the output voltage to maintain that current. This design uses the current through the CCFLs connected to CN501 and CN502. Since the CCFLS connected to CN503 and CN504 are effectively in series with these two, they are also part of the loop. The current sense signals are fed to the Isen input of IC501 through R506 and R507.

Over-voltage sensing is accomplished by capacitive voltage dividers C527 / C530, C535 and CC525 / C523, C536; AND resistive voltage dividers R501 / R502 and R503 / R504. D503, D505, D507, and D508 rectify these outputs and feed them to a filter consisting of R505 and C507.

Over-current sensing is a mess. As closely as I can figure it, positive going pulses from the return lines on CN501, CN502, CN503, and CN504 are rectified by D501, D502, D504, and D506 and then pass through D509, D510, D511, and D512. The pulses charge capacitors C531, C532, C533, and C534. When the voltages get high enough, they individually turn on Q502, Q503, Q504, and Q505. When all for of them are turned on, they reduce the gate voltage on Q501 by pulling down the voltage on R508.

Now, this design has me scratching my head. The obvious question is why did they use this method of building a NAND gate (other than a surplus of 2N7002s). What's more, what is the 'normal' operating voltage of the gate of Q501?

Well, as my Grandpa would say, in for a dime, in for a dollar (Grandpa wasn't a big spender). The next test is to short the gate of Q501 to ground.

Again, this is a TEST, it is not a fix!!! I have a great deal of hope that this will result in all CCFLs lighting at normal brightness. Then you have the task of figuring out which signal line is the culprit.

PlainBill

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

Got it!!! Initially Q501 disables the current feedback, so the drive voltage to the lamps is limited by the OVP voltage. As the CCFLs conduct, Q502 - Q505 turn on, turning Q501 off. The inverter then stabilizes at the proper current through the CCFLs. If a lamp is bad, or the return line is shorting or any of the dozen or more components involved is bad, it never stabilizes, and you get 'two seconds to black'.

PlainBill

[sceva]

Re: Viewsonic VA902b dim lights, then black

I would look in the area directly below CN502, specifically at R511, R513, R512, and the two components below R512 and D508. What is the discoloration in that area?PlainBill

Looks ok, I tested the resistors and diodes around there and they appear ok.

Well, as my Grandpa would say, in for a dime, in for a dollar (Grandpa wasn't a big spender). The next test is to short the gate of Q501 to ground. PlainBill

I removed the earlier jumper on r505 and then shorted Q501's gate to ground and got the normal 1/2 backlights and 2 sec to black.

I tested each of the gates of Q501-Q505 when the backlights come on.
Q502 - 2.4V
Q503 - 2.3V
Q504 - 2.5V
Q505 - 2.3V
All of these went back to 0 when the ccfl's turn off.
Q501 was 0 and went to 4.99 when the ccfl's turned off.

Got it!!! Initially Q501 disables the current feedback, so the drive voltage to the lamps is limited by the OVP voltage. As the CCFLs conduct, Q502 - Q505 turn on, turning Q501 off. The inverter then stabilizes at the proper current through the CCFLs. If a lamp is bad, or the return line is shorting or any of the dozen or more components involved is bad, it never stabilizes, and you get 'two seconds to black'.
PlainBill

You enjoy this too much!

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

Looks ok, I tested the resistors and diodes around there and they appear ok.


I removed the earlier jumper on r505 and then shorted Q501's gate to ground and got the normal 1/2 backlights and 2 sec to black.

I tested each of the gates of Q501-Q505 when the backlights come on.
Q502 - 2.4V
Q503 - 2.3V
Q504 - 2.5V
Q505 - 2.3V
All of these went back to 0 when the ccfl's turn off.
Q501 was 0 and went to 4.99 when the ccfl's turned off.

You enjoy this too much!

Not that much. I thought I had the problem nailed. I'm now sure that I have the design concept right, but have failed to identify the failure mode.

The jumper on R505 eliminated the OVP. At that point the CCFLs went to full (excessive?) brightness. Then the inverter shut down. Why? I don't have an explanation for that.

OK, let's assume that the OVP signal from one of the CCFLs is too high. There are four ways to eliminate that. None are particularly attractive, and I'm not the one doing the work. Choose one of the following.

A. Repeat the test I had you skip earlier. One at a time short the following components. R502, R504, C523, C530. Short one of them, test the monitor. Move the short to the next component, test again. Success will be if it works only if one of the above components are shorted.

B. Unsolder pin 9 of IC501, and short the pin (NOT the trace) to ground. Test.

C. Remove, one at a time, C525, C527, R501, R503. Remove one end of the component, test, solder it back in, move on to the next.

D. Use the diode test function to test both junctions of D503, D505, D507, and D508. Also check the resistance across each junction. Readings on D505 should match those on D507; D503 should match D508. If there are no discrepancies, remove all four of them and test the monitor.

B. seems easiest, but I'm not sure your soldering skills are up to it. A and C may not provide valid results - the circuit would be unbalanced. D is a lot of work. I'm leaning to A as being the quickest.

Have fun.

PlainBill

[sceva]

Re: Viewsonic VA902b dim lights, then black

A. Repeat the test I had you skip earlier. One at a time short the following components. R502, R504, C523, C530. Short one of them, test the monitor. Move the short to the next component, test again. Success will be if it works only if one of the above components are shorted.

Tried this and no change - ccfl's about 50% and 2 sec to black.

B. Unsolder pin 9 of IC501, and short the pin (NOT the trace) to ground. Test.

did not attempt, if I destroy the ic we are done!

C. Remove, one at a time, C525, C527, R501, R503. Remove one end of the component, test, solder it back in, move on to the next.

A little success. When C525 or C527 were removed the ccfl's came on at about 80%, but still 2 sec to black. No change with R501 or R503.

D. Use the diode test function to test both junctions of D503, D505, D507, and D508. Also check the resistance across each junction. Readings on D505 should match those on D507; D503 should match D508. If there are no discrepancies, remove all four of them and test the monitor.

Readings matched as you said. I did not try and remove them yet in case one of the other tests tell you something... but will if need be.

You helped me with one before with 2 sec to black and their were a couple of caps blown on the sense circuit. This one seems similar, except for the lamps not lighting up 100%. (I am thinking out loud...) But since jumping R505 did make the lights very bright, can we assume that there is enough power? And that would mean the problem is in the IC or sense circuits?

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

Tried this and no change - ccfl's about 50% and 2 sec to black.

did not attempt, if I destroy the ic we are done!

A little success. When C525 or C527 were removed the ccfl's came on at about 80%, but still 2 sec to black. No change with R501 or R503.

Readings matched as you said. I did not try and remove them yet in case one of the other tests tell you something... but will if need be.

You helped me with one before with 2 sec to black and their were a couple of caps blown on the sense circuit. This one seems similar, except for the lamps not lighting up 100%. (I am thinking out loud...) But since jumping R505 did make the lights very bright, can we assume that there is enough power? And that would mean the problem is in the IC or sense circuits?

Yes, but something very strange is going on. Here's a quick summary of the operation of the inverter controller. At startup the controller is in ignition mode. It drives the transformers at full power until one of two events occur. Normally the CCFL current will rise until it reaches the design value. At that point the controller reduces the drive to the transformers, then switches to run mode. Or if the output voltage rises too high, the controller reduces the drive to the transformers. If the current reaches the design value, the drive to the transformers is reduced and the controller switches to run mode. If the current never reaches the design value, the controller shuts down. If teh OVP circuit triggers in run mode, the controller shuts down.

There are some guidelines I use in troubleshooting. Look for an explanation of the failure. Assume a single component is at fault. Don't assume simultaneous failures unless they have a common cause.

The tests you did indicate the OVP line is forcing the inverter controller to reduce it's output. This is happening before the CCFLs come up to full brightness. That indicates a problem in the voltage sense circuitry. A problem with either a resistor in the resistive voltage divider or a capacitor in the capacitive voltage divider would cause this. But your test indicated none of them were totally responsible.

What's more, when you shorted R505 the lamps came up to full (excessive?) brightness. What's more there was a buzzing, and the controller shut down within 2 seconds. That indicates the system never stabilized.

One question - what is the voltage at pin 7 of IC501?

PlainBill

[sceva]

Re: Viewsonic VA902b dim lights, then black

Yes, but something very strange is going on. Here's a quick summary of the operation of the inverter controller. At startup the controller is in ignition mode. It drives the transformers at full power until one of two events occur. Normally the CCFL current will rise until it reaches the design value. At that point the controller reduces the drive to the transformers, then switches to run mode. Or if the output voltage rises too high, the controller reduces the drive to the transformers. If the current reaches the design value, the drive to the transformers is reduced and the controller switches to run mode. If the current never reaches the design value, the controller shuts down. If teh OVP circuit triggers in run mode, the controller shuts down.

OK, so that is why when there are bad caps on the 12V PS you get 0 sec to black, and when the problem is with the inverter you usually get the 2 sec to black while it is going thru its startup sequence and failing.

There are some guidelines I use in troubleshooting. Look for an explanation of the failure. Assume a single component is at fault. Don't assume simultaneous failures unless they have a common cause.

Another good tip. Thanks.

The tests you did indicate the OVP line is forcing the inverter controller to reduce it's output. This is happening before the CCFLs come up to full brightness. That indicates a problem in the voltage sense circuitry. A problem with either a resistor in the resistive voltage divider or a capacitor in the capacitive voltage divider would cause this. But your test indicated none of them were totally responsible.

What's more, when you shorted R505 the lamps came up to full (excessive?) brightness. What's more there was a buzzing, and the controller shut down within 2 seconds. That indicates the system never stabilized.

One question - what is the voltage at pin 7 of IC501?

PlainBill

Pin 7 is .93V when off, 1.33V when on, and stays at 1.33V when ccfl's turn off.

I tested some other voltages on the IC while I was there:
Pin 1 and 3: 2.46V when the ccfls are on, and goes to 0 when the go off. Are these not getting high enough?
Pin 15 & 16: 14.7V, drops to 12.4V when ccfls come ON, and back to 14V when they turn off.
Pin 8: 4.8V when they come on, and stays at 4.8V
Pin 9: 3V when they come on, and 0V when they go off.
Pin11: .33V when they come on, and 0V when they go off.

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

OK, so that is why when there are bad caps on the 12V PS you get 0 sec to black, and when the problem is with the inverter you usually get the 2 sec to black while it is going thru its startup sequence and failing.

Ahh, sorta. In general, 'two seconds to black' indicates the 'logic board' is functioning properly; this can be confirmed by the flashlight test. ') seconds to black' (nice phrase, by the way) can have a number of causes, including REALLY bad power supply, problem on logic card, blown fuse in inverter, or the REALLY cute one: design choice. When the inverter controller is enabled it starts charging a timing cap. When the voltage on that cap reaches a certain point AND the inverter controller hasn't shifted into run mode (or a fault exists), it shuts down. Typically that time is 2 seconds. But the actual time is at the discretion of the designer. If they chose a cap 1/10th of the nominal value, you get 1/4 second to black - which probably isn't long enough for the phosphors in the CCFLS to reach full brightness.

Pin 7 is .93V when off, 1.33V when on, and stays at 1.33V when ccfl's turn off.

I tested some other voltages on the IC while I was there:
Pin 1 and 3: 2.46V when the ccfls are on, and goes to 0 when the go off. Are these not getting high enough?
Pin 15 & 16: 14.7V, drops to 12.4V when ccfls come ON, and back to 14V when they turn off.
Pin 8: 4.8V when they come on, and stays at 4.8V
Pin 9: 3V when they come on, and 0V when they go off.
Pin11: .33V when they come on, and 0V when they go off.

The voltage on pins 1,3, 15, 16 are probably normal. It only takes a couple of volts to turn on the drivers, and these are pulsing at 30 Khz.

I'm now interested in the voltage at pin 12 when the power LED is off (pin 7 at .93 volts).

I'm loosing track of who has what equipment. Do you have a scope?

PlainBill

[sceva]

Re: Viewsonic VA902b dim lights, then black

I had seen Timer on some schematics but did not understand how it worked. Thanks again.

Pin 12 is .05V when pin 7 is at .93V.

I don't have a scope. Been looking around for an inexpensive used one, but haven't landed one yet.

[PlainBill]

Re: Viewsonic VA902b dim lights, then black

I had seen Timer on some schematics but did not understand how it worked. Thanks again.

Pin 12 is .05V when pin 7 is at .93V.

I don't have a scope. Been looking around for an inexpensive used one, but haven't landed one yet.

Sorry, I should have asked this at the same time. What is the voltage on pin 12 when the CCFLs are on? It should be in the range of 5 volts.

PlainBill

[sceva]

Re: Viewsonic VA902b dim lights, then black

And now for something completely different...

I turned it on, testing pin 12, got 5V when the lamps lit, but the ccfl's went out twice as fast as usual. I turned it off, and back on, and this time got 5V, and 2 sec to black like normal. The next time I turned it back on, NO backlights... I tried several times, no change. then I unplugged it for a couple of minutes and tried again... still no backlights... I am still getting 4.8V at pin 8, so the on signal has not changed. Still getting almost 15V on 15 & 16, but pins 1 & 3 are showing 0V when they were 2.4V before... Arrrgh.

[thomaskevan]

Re: Viewsonic VA902b dim lights, then black

And now for something completely different...

I turned it on, testing pin 12, got 5V when the lamps lit, but the ccfl's went out twice as fast as usual. I turned it off, and back on, and this time got 5V, and 2 sec to black like normal. The next time I turned it back on, NO backlights... I tried several times, no change. then I unplugged it for a couple of minutes and tried again... still no backlights... I am still getting 4.8V at pin 8, so the on signal has not changed. Still getting almost 15V on 15 & 16, but pins 1 & 3 are showing 0V when they were 2.4V before... Arrrgh.

hi,

can you show me how to get to the capacitor for the monitor? i have the base removed along w/ the front panel. i am stuck now.

there is 4 screws ( looks like it) that are white right behind the lcd. are those screws & how do you remove them?

thanks.

[TV2GO]

Re: Viewsonic VA902b dim lights, then black

First off thanks for all the great info on this forum.
Can anyone tell me what D505 is , and a suitable replacement. mine looks burnt.
Replaced all the Caps , green power light but no back light.

Thanks in advance

[jetadm123]

Re: Viewsonic VA902b dim lights, then black

First off thanks for all the great info on this forum.
Can anyone tell me what D505 is , and a suitable replacement. mine looks burnt.
Replaced all the Caps , green power light but no back light.

Thanks in advance

Match up your power supply with the schematic of the one on post #11 of this thread as there were different model power supplies used for the VA902B monitor. The BAV99 is readily available fron digikey.com.