Proview 3200 no backlights

[smason]

Even though I don't have the time or the space, my wife volunteered me to repair a Proview 3200 LCD TV.

After disassembling it (wow, what a pain) it looks to be fairly good quality.
I was hoping for a few bloated Samxon GFs, but no such luck.

Very few markings on the connectors, though the rear of the inverter board has markings for the connectors.
I have 5v at BL-On 7.6v at VBL. Tracing the connections back, all the VBL red wires go to the 24v out on the power supply. Unplugging the inverter board gives 24v at the power supply connector, and it stays at 24v when powering 2 1157 bulbs in series. So the power supply seems capable.

So, something shorted on the inverter board? Fuses are all good, so it's not drawing much current, and none of the components feel warm to the touch.
Measured 7.6v on one side, 0v on the other of all the obvious bypass caps, so not one of those.
It's a very busy board with lots of small surface mount components. Could be fun finding it...

[tom66]

Re: Proview 3200 no backlights

Plainbill had encountered a PSU before which worked under light load, but dropped out under heavy load.

1157 is only about 2 amps, an inverter draws around 6 amps at 24V. Is there any way to test it at a higher output current?

BTW, if there is a short circuit or high load on that board, and it is 7.6V at say 7A (more than 6A limit), then it would be dissipating 53W in one shorted component, which would be very obvious very quickly (smoke, flames, melted board.)

[retiredcaps]

Re: Proview 3200 no backlights

Even though I don't have the time or the space, my wife volunteered me to repair a Proview 3200 LCD TV.

Wow, Rubycon caps on the power supply.

[budm]

Re: Proview 3200 no backlights

Check that small black fuse that looks like a resistor by the inverter connector to the right, it should have 15 or 20 printed on the top. Common failure for this board.
It looks something like this:
http://s807.photobucket.com/albums/y...ps27a023ac.jpg
You can use 0.22, .33, or 0.47 Ohms 1/4 Watt through hole resistor, one of its leg is connected to the 24VDC input, it is used to supply the power to the PWM IC.

[smason]

Re: Proview 3200 no backlights

Plainbill had encountered a PSU before which worked under light load, but dropped out under heavy load.

1157 is only about 2 amps, an inverter draws around 6 amps at 24V. Is there any way to test it at a higher output current?

Good point. It has been in the back of my mind. I'll see what I can throw together to draw 5 or 6 amps.

I will also measure the current being drawn, I agree you'd think dropping 24v to 7.6 would generate some heat

[smason]

Re: Proview 3200 no backlights

Wow, Rubycon caps on the power supply.

Yeah, I was surprised. And the mains cap is a Nichi.

[smason]

Re: Proview 3200 no backlights

Interesting. I reconfigured the light bulbs to use both filaments on each. Brighter and 24v still stable. Then I switched the meter to measure current, and the bulbs were dim.

THought it was a bad connection or a faulty jumper lead, but nope. Now the 24v output is 7.6 with or without load.
Will troubleshoot the power supply maybe tomorrow.

Check that small black fuse that looks like a resistor by the inverter connector to the right, it should have 15 or 20 printed on the top. Common failure for this board.

Yup, it's open! Once I get the power supply fixed I'll replace it and see what happens.
Thanks for the tip!

[budm]

Re: Proview 3200 no backlights

I think your power supply is OK, you only had the bad fuse.
When you parallel the lamps (Both HI and Low filaments), the cold resistance will be around 2 Ohms, which will have VERY HIGH inrush current for the 24V power supply to handle, I hope the 24V power supply is not damaged.

[mu2]

Re: Proview 3200 no backlights

Hi,

Since you have the unit open you might want to check the caps mentioned in this article on the main board.
http://blog.coppelltvrepair.com/2011...-power-on.html
I worked on one of these sets and wouldn't want to open it up again for another repair.

Mike

[smason]

Re: Proview 3200 no backlights

I think your power supply is OK, you only had the bad fuse.
When you parallel the lamps (Both HI and Low filaments), the cold resistance will be around 2 Ohms, which will have VERY HIGH inrush current for the 24V power supply to handle, I hope the 24V power supply is not damaged.

If it was just the fuse, how would you explain the 7.6v?

It's 2 1157s, each bulb with filaments in parallel, but the bulbs in series, shouldn't be too bad. I really think the PS had issues before I did this.

More tonight hopefully...

[budm]

Re: Proview 3200 no backlights

You were getting 24v before and able to light up the lamps. Please check the cold resistance of that lamps setup you used last and see what it is.

[smason]

Re: Proview 3200 no backlights

You were getting 24v before and able to light up the lamps. Please check the cold resistance of that lamps setup you used last and see what it is.

You're right the cold resistance is around 1 ohm. that's a bit over the top

Current draw on those lamps with an external 24v supply (after initial inrush) is 2.5 amps. Guess I should really come up with a good load for 24v.

Today the supply does output 24v again. More testing later.

[budm]

Re: Proview 3200 no backlights

It may have gone into protection shutdown mode. I used the 1157 low filament for testing power supply all the time, since the cold resistance is not too low, if the power supply cannot handle the cold resistance of the low filament, then I would suspect bad power supply. But using both filaments are too much for cold start, 1 Ohm with 24V, that is a lot of inrush current!
If you want to load the power supply to the max current rating, then use resistor instead.
You need to use meter that has fast capture response time in milli-second range to capture the reading, it will be at least 24A.
24v @ 4A, you will need 6 Ohms resistor rated at least 100 Watts, smaller if you can have 24vdc fan blowing on it.
http://s807.photobucket.com/albums/y...ply%20testing/

[budm]

Re: Proview 3200 no backlights

Or make sure the cold resistance is about 5~6 Ohms.

[tom66]

Re: Proview 3200 no backlights

7.6V across 1 ohm is 7.6A so more than likely it was limiting current at that point.

[smason]

Re: Proview 3200 no backlights

OK this is beginning to tick me off.
I replaced the fuse, same problem. I lifted the + leg of each of the 220UF caps on the inverter board, and checked, all are within a few uf of each other and ESR is fine.

The 24v output of the PS is still flakey, sometimes it's 7.6v even without the inverter board plugged in, I'm leaning back to it as my prime suspect. I'm going to remove the supply, and check the resistors and caps on the 24v section if I get more time later tonight.

[budm]

Re: Proview 3200 no backlights

I think the power supply is damaged, at 7.6V output, sounds like the MOSFETs in the driver side is not driving it correctly. I am sure if you get the 24v back and with that black fuse in place, it should be OK again.

[tom66]

Re: Proview 3200 no backlights

I think it is OK but there is a problem with feedback or startup caps. Failed MOSFETs would cause zero output and blown fuse, usually.

[budm]

Re: Proview 3200 no backlights

Well, it looks like it uses Half -bridge drive setup, if the bottom half of the two Mosfet has blown source resistor, this resistor is used for current sensing feedback, so when he draws too much current the resistor may have gone open, the drive will not be fully function. Also check and see if you still get 400VDC on the large cap just to make sure the PFC booster is still working.

[smason]

Re: Proview 3200 no backlights

Just to clarify, the 12V is fine, and the 24v is intermittent. Sometimes it works, sometimes not. When it does work it can supply a few amps to light 2 1157 bulbs in series. When it doesn't work it's at 7.6V output.