LG 42LE5400 Backlighting & power supply

[dkneyle]

Got this set for next to nothing because it had failed. The owner had replaced the main input fuse but no joy, so sold it for next to nothing. I found a power diode short circuit near the LED driver board. I replaced that and the power supply appears to fire up but nothing on the screen. (The fancy coloured lights beneath the panel come on and flicker in response to the remote control.) It would be easy but expensive to replace the power supply/LED driver board. But I was hoping that someone might have experience with these sets and set me in the direction of what to look for. Fuses blow for a reason; something else would seem to have caused that to happen and that's what I need to track down. Yes, I know I need to start making voltage measurements at the input/output of the LED driver board, but in the first instance I just thought that someone might know this as a common fault.

[selldoor]

Re: LG 42LE5400 Backlighting & power supply

Perhaps post a good clear picture of the power supply front and back using manage attachments - someone may have seen the board in another set.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

This board has various numbers: EAY60803201, YP42LPBL, YP42LPBA, YP47LPBL.

Running my meter over many of the active components I've found one that doesn't seem right. I mentioned in the previous post that there was a power diode 1N5407 that was short circuit. In the same vicinity there is a MOSFET that looks like it provides power to the LED driver board. My in-situ measurements suggested that it didn't seem quite right so I've ordered a replacement for it, and when I remove it from the PCB I'll be in a better position to check whether it measures out the same as the replacement. Without a circuit diagram it's challenging, but I think I'm on the right track. It would appear that many of the power board functions are working OK so I reckon it is on the secondary side, and hopefully only the supply to the LED driver board, and hopefully that hasn't been damaged in the failure. Will advise the outcome.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

I replaced the power MOSFET that was part of the supply driving the LED driver board. Also replaced that power diode that looks like it's a snubber (I've seen some pictures of similar boards where this is not actually used so I don't think it's critical other than it needs to be in working condition if used.) The original MOSFET had blown when compared with the replacement. Power up and set turned on. The picture looked like the backlighting was too bright but I didn't get a chance to explore further. The sound of cracking metalwork and smell of over-heating electronics caused me to shut down before the MOSFET blew again. The heatsink with the power MOSFET was VERY hot. No other signs of any physical damage to the board. I'm working without a circuit diagram at present. Anyone want to suggest what causes the power supply to go into overload mode in these types of situations?

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

My reading of the situation is this: The power regulator that feeds the LED backlighting controller is either a) overloaded, or b) being supplied with too much juice that it needs to dissipate as heat. Anyone with more experience than me want to suggest the more likely scenario?

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

Anyone with the same set that is looking for a repair manual that might help, look here:

Very slightly different model, but I'm sure it will help.

[budm]

Re: LG 42LE5400 Backlighting & power supply

Disconnect those two blue connectors which are for the LED strings, check to see if it show any resistance between chassis and any one of the pins (the female connectors).
The MOSFETs that is getting hot are the four of them on the heat sink close to the LED driver board or the two MOSFETs on the heat sink on the top right that drive the two transformer for B+ for the LED driver board?

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

The MOSFET that is overheating is the one that is close to the LED driver board. (The other 4 three-leg devices that are under that heatsink are actually diodes and check out OK.) The MOSFETS on the LED driver board all seem to check out OK. The MOSFETs that drive the transformer run at normal temperature.
I'll make that measurement looking for LED shorts and will post what I find.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

NO shorts at either the outputs of the LED driver board (actually they are inputs I suppose if you think about it) or at the LED strings. I ran an experiment and powered it up firstly without the LEDs connected. The 120V supply was at 98V, which I assumed was because the power supply detected no LEDs and wasn't actually yet working. I connected one LED plug at a time and all looked good and after 2 seconds shut down (as would be expected with one LED plug disconnected). I plugged both LED plugs into the board and the situation was as it was. The heatsink indicated that the MOSFET was rapidly getting overheated. But here is a clue. The 120 volt supply was only at 109 volts.

[budm]

Re: LG 42LE5400 Backlighting & power supply

Oh, I am looking for shorts in the LED strings due to bad wiring shorted to the chassis that go to the LED strips, not at the output of the LED driver board. The 98V is fed into the Anode of the LED string, the Cathode side is connected to ground through small value resistor and MOSFET for current sensing feedback to control constant current circuits. You can tel how many strings of LEDs are used by look up the spec of the LED DRIVER IC on that driver board, it may also give the application notes schematic to get some ideas how your board is hooked up.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

Yes Bud, I did that measurement as well. It all looks good. No indication of short to ground there either - I checked both the common anode and the cathode of the strings. It's a mystery, and of course I have to be a bit careful not to blow that MOSFET by running the set too long whilst taking measurements. The mystery I think is why we have ~109 volts instead of 120 volts. I reckon that's a clue, but where to from here?

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

I'm beginning to wonder whether the filter capacitors on that 120 volt line are U/S? My thinking is that if the 120 volt rail has exessive high frequency ripple then the MOSFET could suffer from some sort of high frequency switching that causes it to run hot? The caps physically look OK, but it wouldn't be the first time that's caught anyone out. Not too challenging to remove and try a replacement and see if I get a different result. In the meanwhile, if anyone has any thoughts I'd be interested to hear them.

[sordids999]

Re: LG 42LE5400 Backlighting & power supply

Thanks for the manual I have the 47 inch version still sat in the garage will try again soon or to annoy tom

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

I'm trying to reverse engineer the particular section of the SMPS that is causing grief. The 120volt DC supply is a fairly standard high frequency transformer, 4 power diodes to rectify and a couple of parallel electros to filter the supply. I think the MOSFET switches at 50 or 60 Hz to provide the duty cycle for the backlighting LEDs. And I'm assuming the reverse biased diode across the 120 supply is a snubbing diode. The LED driver board switches the LED cathodes to ground via the MOSFET as it provides duty cycle switching.
I didn't take particular note of the type number on the defective snubber diode I removed. I've substituted a power diode IN5408 as they were used elsewhere on the board. I'm wondering if I've got my substitution wrong?
Anyone want to correct my analysis of how these circuits work?

[budm]

Re: LG 42LE5400 Backlighting & power supply

If the snubber is not right, the MOSFET will do run hot and can get damaged. The snubber will usually connected across the winding of the drive side or across the S and D of the MOSFET.
Can you show me the pictures around the areas of this MOSFET circuit?

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

The MOSFET is hidden under the heatsink. It is Q241. The diode in question is next to it, running at an angle. Yes, I think it does go across S & D in actual fact.
Off the top of your head, any thoughts on the correct diode type?
I've seen some images of this board without the diode. Would it be too risky to remove, if only for a short time to check whether the overheating problem goes away?

[budm]

Re: LG 42LE5400 Backlighting & power supply

can I see the bottom side of that area? Trying to see what the D is connected to, Typical Clamping Diode will be around 200V.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

Here you go.

[budm]

Re: LG 42LE5400 Backlighting & power supply

That Q241 is V3+ (connector P205) switching power supply, that diode is the rectifier.
Picture of the whole bottom side?
The snubber is the RC netwowrk R255/C257 for the MOSFET, and C262 snubber for the rectifier D245.

[dkneyle]

Re: LG 42LE5400 Backlighting & power supply

Well, my analysis was way off. But (LOL) I solved it all the same. I replaced the diode in question with a fast recovery diode that I scrounged from an old non-TV SMPS I had lying around for use as parts. I recovered an ER502 which is a 200 volt 5 amp FR diode. Soldered her in, put the board back into place, and the heatsink now runs only warm to the touch. So, whatever the pathway to get here, she's now in soak mode and we'll see if she holds up. All looks good so far. Thanks for your help Bud.