Testing LG 47LN5700 backlight ps output w/ resistor

[gliebisch]

As a way to verify proper output of the backlight power supply in LG 47LN5700, is it a legitimate method to simulate the LED load with a resistor? I have the schematic and spec of the two LED outputs, which are each nominally 0.4 A. So I thought about connecting a 250'ohm power resistor to each output, just long enough to get a normal load voltage reading. Since it calculates to about a 40W load, it would be for only a short duration for a measurement on each output, since I don't have a 40W resistor. Access to the LED + and - is possible without bothering the connector, although the connector will be out of course.

I want to do this before I start tearing into screen disassembly chasing potentially bad LEDs.

[budm]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

Well, you need to find out what the Vf of the total LED string is fist.
How did you come up with 250 Ohms?
"I have the schematic and spec of the two LED outputs, which are each nominally 0.4 A." You forget about the Voltage output with loads, 100V 110V? The Vrdops on the LED string is not 100V because you also have MOSFET and current sensing resistor in series with the LED string that being driven. In your case from the other thread, you have 19V drops for the MOSFET and the resistor.
BTW, you also need to understand how the LEDs string are connected in your panel and how set of LED strings it has, 1, 2, 3, 4? You will have to look at the LED driver circuit the board to figure that out.
Your board has two boost converters to drive two sets of LED strings, since I already told you you have open circuit in one of the string, then all you have to do is to swap the two sets of the LED sting wires to see if the problem will follow.

[budm]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

https://www.badcaps.net/forum/showthread.php?t=61174
I already told you you have open circuit in one of the LED string. So what are you trying to do?

LG LN5000 series:
https://www.badcaps.net/forum/showth...ht=LG+47LN5790

[dick_barton]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

There are reasonably cheap led testers on the market.
http://www.ebay.co.uk/itm/SID-GJ2C-L...-/282072043319

[gliebisch]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

then all you have to do is to swap the two sets of the LED sting wires to see if the problem will follow.

OK, thanks , Bud. Since I don't want to butcher the cable going from the LED driver output to the LED strips, suppose I swap the two LED drive circuits by creating a breakout on the PS/driver board from the points noted on the schematic attached here. With pigtail leads from these points I can easily swap LED 1 with LED 2, as you suggest.

I appreciate your diagnosis of a bad LED string earlier. This exercise is purely to convince me that I have no choice but do do surgery on the LED strips, whereas a PS board replacement would be easy.

[budm]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

https://www.badcaps.net/forum/attach...5&d=1509023144
As you can see, all you have to do is to swap just two wires either the L+ wires or the L- wires.
But since you have the diagram (which you should up load for future ref. for other people), do you now see why if you have open circuit in one of the string you will get 0V on the L- return pin?

[Andrew F. Ali]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

From what I have read....you have an open LED on the backlight string. If it was me I wouldn't waste time trying to re-invent the wheel or for that matter invent anew wheel...I would strip down the panel to backlights and determine which strip or strips is the problem and decide if I would use the 'Reecyburger stack' method or buy new strips to repair this TV.

[budm]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

from what i have read....you have an open led on the backlight string. If it was me i wouldn't waste time trying to re-invent the wheel or for that matter invent anew wheel...i would strip down the panel to backlights and determine which strip or strips is the problem and decide if i would use the 'reecyburger stack' method or buy new strips to repair this tv.

+100, already explain how it works and why.

[Davi.p]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

I think the cheaper way to test LEDS are those Boost converters bought from China in conjunction with a ultracheap o free ATX PSU.

[vinceroger69]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

This is the tester i have for testing leds and here is the link for the stacking method led repair

https://www.ebay.co.uk/itm/SID-GJ2C-...75.c100623.m-1

https://www.youtube.com/watch?v=RxwdC7V5Ue0

[gliebisch]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

I would like to thank you, BudM and all the others for the invaluable assistance in getting my 47LN5700 repaired today. It was indeed one entire LED strip (L+R) that was out plus a single LED on another strip. I just replaced them all. Instead of buying an LED tester, I bought a Yescom 30V variable DC supply from Amazon, which is capable of a constant current mode. I was able to then test each pair (L+R) of strips (9 LED's) with about 24 volts and constant 50 ma current. I can use that power supply for general bench work and it was about the same price as the LED testers. In retrospect now, I spent too much time thinking it was the power supply. Upon repowering my TV, the backlight had been set at 100, I turned it down to about 75. Did a software check but it told me I already had the latest software. Not sure if the on screen software check is just for the Smart TV apps, or includes firmware for the system.

For anyone who needs it , attached is the PDF I used to find my way around the power supply LED driver, which I got from another poster who uploaded it in the past.

[budm]

Re: Testing LG 47LN5700 backlight ps output w/ resistor

Thanks for the follow-up, glad to her you got it fixed.
90% of the TV that use LEDs for backlights will have failed backlights due to failed LED/s.