Samsung UA32D4003BM 32" LED NO POWER

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

Check in a dark room, any glow from screen? Also check the LED current as before using your multimeter.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

It wasn't the caps, i lifted the psu board and the logic board up on carton away from the chasis, (everthing still connected) and got 220vdc on the blue wire and 80 on the black wire, and another maybe stupid thing i don't feel breakout voltage when touching the chassis as i felt before.

by the way Q9101 is actually D9101, weird package for a diode

[budm]

Re: Samsung UA32D4003BM 32" LED NO POWER

If you do get over 200VDC on the blue wire, then it sounds like it has open LED on the LED string, need to verify the current draw, if it shows 0 A then.
The rating of the power supply is 203V @165mA (~34 Watts) will have equivalent resistance of about 1230 Ohms. You may be able to make test fixture that has resistor and LED in series to simulate the LED string.
I have a feeling that it has open LED/LEDs.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

no glow from screen, no current with meter :-(
(i never checked current with this meter, so i wanted to make sure that the fuse isn't open,
so i connected it to 1.2v battery and got some reading so im guessing the meter is good.
all the voltages on psu increased: main filter caps, output to led etc.

both of the pcbs were in perfect position before, very strange,

budm, simple ohm testing on both wires can tell me if i have an open string,
am i right?

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

LEDs aren't resistive devices (they're diodes), so checking with an ohmmeter won't show anything sensible

Think it's time to take the panel apart and check the LEDs, truly the Achilles' heel of modern LED LCD TVs.

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

Here's the odd thing. That return wire with 80V on it should have much less on it, I think.

Think about it: If current was 165mA, then power dissipation in return regulation transistor would be over 13 watts, it would melt very quickly (Theta J-A is about 125°C/W for a SOT223 transistor, so the temperature would begin rising towards 1625°C which is quite hot... transistor would probably fail past about 200°C junction temp.)

What you could try: if you feel up to it, build this very simple circuit on a breadboard. Use a 400V sink FET or higher. It regulates sink current:

http://modulatedlight.org/optical_co...ent_sink_2.gif

This will help you safely test the LEDs and it's the most ideal way of doing it, but it's fairly complex. Good learning project, though.

You could also use a resistor, but the risk is you don't know what the forward voltage of the chain is. But you can make a guess. That SOT223 sink transistor shouldn't ever dissipate more than 1 watt. Assume I=165mA. This means V must not exceed about 3 volts.

So the series chain is 200V (e.g. 4V/LED, 50 LEDs in series.) So you need to drop 3V at 165mA so R=18 ohms. Use an 18 ohm, 1W resistor and carefully monitor current. Maybe start with a bigger resistor first, to protect the panel LEDs.

Apply this formula for Vs=220V if you like, though that voltage might drop under load.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

tom66, i would like to open the panel as a last option,
(I never opened an LED panel before, but i replaced an lcd bulb on a SONY VIO laptop a few years ago, (was a pain in the ***) but worked.

About the project you refered to, any recommended parts, as for the comparator, Q1, Cfilter,

The other option as i understand, is less recomended, to use an 18 ohms 1Watt resistor instead of the LED

[budm]

Re: Samsung UA32D4003BM 32" LED NO POWER

I think 80V reading is an error, as you can see the Drian pin of the MOSFET Q9102 is connected to the LED connector, the Source pin is connected to the current sensing resistor R9120 the other end of R9120 goes to ground, the Source pin is also connected to the resistor network to goes to the LED driver IC.
I do not know which ground point you are using to make the measurement.

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

I suspect the transistor is used as a linear regulator for the current, so the voltage drop across it would be an unknown (dependent on the base/gate voltage, the current flowing through it and the parameters of the transistor.) Most likely it is an BJT NPN due to the linear relationship between base and collector currents.

This is different from some sets which regulate current at the boost converter.

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

About the project you refered to, any recommended parts, as for the comparator, Q1, Cfilter,

Comparator: actually it's an opamp, LM358 or LM324 is a good choice, make sure to run it from a 12V or so power supply. Do not attempt to run it from the 200V!! A 100nF capacitor across power and ground of the opamp is a good idea.

Q1: any MOSFET, 250V or more. For example a 7N50.

Cfilter: omit this part it is not necessary for this system.

Maybe size Rsense to 10 ohms, giving you a voltage of 1.65V at 165mA rated current, i.e. 10V/amp. Use a potentiometer for Vin to vary the reference set point voltage. Consider a 100nF capacitor on the wiper.

Test the circuit on a power supply before use and verify its functionality, you do not want to potentially damage the LEDs.

The other option as i understand, is less recomended, to use an 18 ohms 1Watt resistor instead of the LED

Not quite. LEDs still in place, but the resistor will have one end to black wire and one end to the ground, it replaces the on board regulation transistor.

[Ltank]

Re: Samsung UA32D4003BM 32" LED NO POWER

A Samsung I worked on had 100 small Leds on each side of the set. Two wires to each sideparalled to four wire plug. The board was divided into four seperate strings of twenty. five series diodes. So taking a resistance reading would not give accurate results if your unaware of the particular TV led strip layout

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

budm, i will test the voltage again when i get home,
the voltage moved from 60-80vdc, using the screw base on the psu pcb as ground point.

tom, i will try to locate parts and a blank matrix, and update with that as well.

[budm]

Re: Samsung UA32D4003BM 32" LED NO POWER

Can you give us the LCD panel model number? The spec will give us the info about the LED set up.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

Ok retested voltages, at start got 146vdc,
after messing around with it, got 210vdc on blue 70vdc on black

added a picture of panel sticker with all the info.

something about the voltages changing like that really bugs me

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

IMHO that's normal, because the boost converter only switches on once main board sends signal.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

tom sorry didn't understand your explenation,

do you mean that its normal that the voltage, is changing so drasticly?

[tom66]

Re: Samsung UA32D4003BM 32" LED NO POWER

Yes.

They use one SMPS transformer, it powers the 13V & 5V for main board and 145V for boost converter for LEDs.
- 13V is for audio & T-con board
- 5V is for logic & USB port

Now imagine you plug in a hard drive to USB port. Or you turn up the volume. The load on 13V & 5V will naturally increase, right?

The type of power supply uses one transformer for all voltages, so an increase in load on the 5V/13V will require the SMPS to increase the duty-cycle driving the transformer. This will naturally cause the 145V output to also rise, even if its load has remained the same, because the transformer is only regulated at a few common operating points.

This means loud program volume or hard disk spinning up will cause the LED backlight to flicker & that's no good at all!

So they use a post-regulator, which takes this RAW and unregulated dc voltage and boosts it up to 210V, nice and steady, to run the LEDs.

But there's no point in running this in standby, you'd just waste power, so they turn it off.

So that's why it starts low (boost converter off, just passing raw 145V thru the diode and inductor) and then boosts up once turned on... so THAT part is working, but for some reason, there isn't a peep from the LEDs, that's the next step to work out, is there a fault with the regulation circuit on the power board or a fault with the actual LEDs.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

thank you, for the thorough explenation!

i will try to build the test circuit you recommended and update as soon as it will be ready.

[budm]

Re: Samsung UA32D4003BM 32" LED NO POWER

And since you do not see the current spike when using the Ammeter, it is pointing to open circuit in the LED string, but I will wait to see what you get when you make the fixture for testing the LEDs.

[s_henya]

Re: Samsung UA32D4003BM 32" LED NO POWER

tom66, sorry for the delay i had a test last week and couldnt find time to deal with this project.

just to clarify i understood the circuit.

vin is the +12v for lm358?

v+?

and is 2n60b suitable?