LG 50PC5D YSUS/ZSUS Issues
 

I have acquired this TV from a neighbor, whom said it did the 'pop of death', and now it works without a picture. I looked up info online and apparently the Y-Sustain and/or Z-Sustain boards like to fail in a manner that necessitates replacement of the whole board. I tried feeling the heat sinks to see if they're getting warm, and it appears that perhaps one of the two on the YSUS board isn't as hot as the others. What I'd like to know is if there's a more accurate way to know which of the two boards (hopefully not both!) are failed. The TV has run for several hours now without issues, other than no picture, obviously.

If it is one of the models with IPMs, they often fail but I've successfully replaced just the IPM instead of the whole board. A blown IPM will generally fail shorted so will blow a fuse, usually a 4A ceramic 5x20mm, so check fuses first on the Y-Sus and Z-Sus boards. If one has blown, check for a short on that board, typically Vs to ground is what shorts due to a blown FET/IGBT. If this model uses IPMs, the blown FETs will likely be inside the hybrid module, requiring replacement (though some people have bodged on new external FETs :) ). It's a straightforward task with some solder wick and a decent iron, though a desoldering station is quicker.

If the fuses are all good (look for surface mount ones, too), move on to standard plasma troubleshooting... Check your high voltages... The Vs and Va testpoints are usually on the power supply board and the other voltages on the LGs are usually called things like -Vy on the Y-Sus and and Vzb on the Z-Sus board in LG parlance.

The proper voltages will be on the panel sticker and, IIRC, the second line on the LG stickers below where it calls out the Va (the address voltage for the X boards at the bottom) and Vs (the sustain voltage that does most of the power, comes out the Y-Sus board through the Y-Drives) goes in this order: Set-Up / -Vy / Vsc / Ve / Vzb

Some of these will probably be missing and say N.A. on the sticker as most LG sets do not have all of these voltages but they have maintained the sticker format. Just look for the test points for each and see what voltages you actually have at the testpoints, then I can point you in the right direction for troubleshooting that section.

Vs and Va may only come up for a few seconds at power-on if it senses a fault and goes into protection but you should have time to measure them. A MAX function if your multimeter has one can be helpful.

The service manual is avilable at:
http://www.toms-service-manuals.com/...&mfg=LG-Zenith

Thanks very much, Drussell! I have tested all fuses on all boards (including one each surface mounts I found on the Y and Z boards), and all test good. Now, I am having difficulty locating the test points on the power supply. I'm used to them having 'TP' prefixes, and I'm not seeing anything like that.

I've never worked on this model so I can't say for sure, but taking a quick look at the service manual there are several different possible power supplies and, indeed, it looks like there aren't test points for the Vs and Va. You're supposed to just stick your probe into the back of the output connector. (There should be a silk-screened legend beside the connector.) Your negative multimeter probe can just go to any convenient chassis metal like one of the wallmount holes as the chassis is at circuit ground potential. Vsc/Vscan is usually one of the only ones that you measure across a resistor or capacitor rather than to ground as it is floating.

If you have a decent camera, perhaps you could post a couple good close-up pictures of the various boards and one of the whole back of the TV so I can see what we're dealing with and how everything is interconnected.

I've got the service manual - lemmed klnow if you need it - FREE. Don't pay for it - LG gave it away for free for over 2 years, and then took them off the web. I have a little over 300 LG LCD/plasma/DLP and Zenith CRT sets data, all free from the old LG FTP site.

The manual for the 50PC5D is over 30 megs in size, so a high-speed (DSL/Cable) connection will be needed. I don't have a server, but may be able to upload it to a church website I manage.

PM me if interested.

I'll see what I can do with this iPhone, though unfortunately it only allows me to add one photo per post. My power supply's part number is (I think) 2300KEG006C-F. I've attached a photo of the upper left connector, showing the silkscreened voltages. (The connector is unplugged in the photo, for better visibility.)

Here is a photo of the entire chassis. YSUS at left, power supply in the center, ZSUS at right.

UPDATE: Vs and Va voltages measure 197V (within 2 volts of rated 195) and 60 (right on) respectively.

That will be most helpful! I'll get in touch.

This is with all the board still connected?

If you look at the panel in complete darkness does it have a faint glow when powered up? (IE, is it getting precharge?) Check the rest of the voltages... If you have Vzb (panel bias, on the Z-sus board) then the Z is probably fine and I'm suspecting problems on the Y. Usually if the Z has a problem you will still get a picture it will just be dim and washed out with the panel bias absent.

Something must be missing... If all supplies are there everywhere it is possible that one of the IPMs failed open. They usually short but an open is certainly possible and it won't be generating any Yout signal to drive the panel. Can you post the best shot you can get of the whole Y-sus board? Perhaps I can spot some testpoints for things like the actual Yout signal to the panel.

If it does turn out to be a blown IPM you should check for shorts in the buffers on the Y-Drive boards, you certainly don't want to replace the IPM only to have it blow again due to a failed buffer chip.

Many of the IPMs are potted with a clear potting compound on the bottom and it is easy to see the blown FET/IGBTs from the bottom but it's nice to be sure it's frazzed before you bother to desolder it from the board.

Do you have an oscilliscope?

That link I posted above has the service manual and it's free... Though perhaps you have a better one, this one certainly doesn't cover everything.

Perhaps you guys missed the link in my first post?

I should have mentioned that I couldn't get my device to open that link; it just hung without loading. Sorry about that!

EDIT: The link seems to be working fine now. Odd...

Yes, those two voltages were taken with everything hooked up. I have not checked any other voltages yet. I have verified that the plasma panel has no glow at all; it appears to be completely off. Where would I be checking for VZB on the ZSUS board? I'll try to get photos of both boards for you.

I do technically own a scope...but I don't currently have access to probes for it, and don't know how to operate it properly...yet.

As an aside, I'm not harming anything else by leaving the set running in its current condition, am I?

Here is a photo of the Y-Sustain board:

Next, the Z-Sustain board:

After removing the YSUS board from the chassis, a small black chunk fell out. After judicious inspection, I believe this chunk blew out of a similarly-sized hole I can just barely see on the IPM under the smaller, black heat sink on the YSUS board. I believe we have found the problem...

Heh, that could be a sign of trouble, for sure... :)

I would have expected the IPM to be under the bigger heatsink but perhaps it has modules under both on that model? If it has a hole blown in the side of it, I would definitely desolder it and take a closer look, obviously...

I can't see too well in that photo but in the upper right corner of the photo (so would be the lower right in the TV I assume) there is a <----Vzb---> (Vzb is the Z bias, or panel bias and is really all the Z board does is supply that voltage to an array of elements across the entire panel) so it looks like you're supposed to measure it across that film capacitor, I suspect...

I'm actually somewhat surprised that the panel isn't doing the dull glow but you may have multiple parts/systems failures, of course...

I wouldn't run it except for measurements when you don't know precisely what is wrong yet... Voltages applied to the wrong electrodes when other voltages are missing can permanently damage the panel!

Best to be safe as this still seems like it is likely a fixable situation to make a working TV.

Yeah, there is a large black multi-pin unit (assume IPM) under that black heat sink, which definitely has a chunk missing. I don't know how I'd ever get it unsoldered! I'm wondering if it'd just be easier to replace the entire YSUS board?

There is a 'VZB' mark near those two film caps, but I'm unsure of where exactly to measure. There is a double-ended arrow pointing both ways next to the right-most cap.

I'll check again for panel glow tonight in pitch darkness.

Yeah, that is one of the IPM (Intelligent Power Module) or hybrid module or whatever you want to call it. It is just a bunch of discrete components mounted to one substrate without plastic cases on each device to save on cost. Annoying when one FET/IGBT/diode or whatever blows because you can't just easily replace one $2 part but, whatever... Cost reduction and all that. The advantage, though, is you are replacing most of the drive circuitry within it so you don't have to check a bunch of parts that drive the gate of the blown IGBT or whatever... Everything that is likely to blow is in the IPM.

It's not nearly as difficult as it looks. Using solder wick, you just heat each pin, one at a time, though the wick and it will slurp the solder right out. I find it is easier to melt a little real lead/tin solder on each pin first before trying to wick out the lead free stuff they use during assembly. Solder wick is the key, here. If one pin is stubborn just refill it with lead/tin solder and wick it again. It will come totally loose and just lift right out.

I generally desolder using a solder-slurper vacuum tool but on the IPMs it is WAY easier with solder wick. I've done it on-site with an old 33 watt Ungar iron, so it's not like you need state of the art equipment! :)

Once you desolder all the pins you lift the whole thing out with the heatsink still attached because the screws are from below, mount the new one to the heatsink and solder it back in.

I have been buying my IPMs from a local guy that imports them himself and sells on eBay. I usually pay about $40 each and haven't had a bad one yet or a repeat failure. Many people suggest replacing both of them on the Y-sus board at the same time (one is the drive, one is for energy recovery usually) though I have had good success replacing just the failed one unless I have reason to suspect it was killed by the other or it seems flakey in some way.

Coppel TV repair is well known for supplying thousands of them in the US market with good results. Buying from a reputable dealer that buys quality supplies in bulk means not worrying about them being used pulls from China or something but they are available all over the place, it is a VERY common failure.

EASIER, sure, but where is the fun in that!?? :)

You simply measure the voltage across the cap...
One multimeter probe to one pin and the other probe to the other pin.

That generally always applies to Vscan but Ve/Vzb being across a cap is more unusual, most of the time it is one test point to ground but since this indicates across the cap, across the cap we measure! :)

It is certainly possible to have no glow, it just wasn't what I expected.

It doesn't mean the TV is toast or something, especially if the IPM that is blown is the ERC one, that is usually rigged up to feed recovered power back to the Z board's panel bias supply.

I checked again in total darkness for any sign of panel glow...definitely none. I believe I have located, on the YSUS board, where I am supposed to measure the Vscan, which was the last voltage to measure if I recall. (Photo attached) Just to be safe, I'm asking once again where exactly my probes are supposed to go. :P (Across the cap, or across that string of resistors?)

EDIT: Checked Vscan voltage across resistors and got the correct 133 volts. I also pulled the buffer boards (XL, XR, XC) and checked visually for anything fried/burnt, and all looked good. (Are the buffer ICs the rectangular units embedded in the ribbon cables? There are 16 of them, one for each cable!)

Those resistors are the bleeders for that capacitor and are connected in parallel with it so measuring across either the capacitor or the resistor chain is the same thing.

I can't see any ICs in the ribbon cables in your photos. [Edit: oh, I see you were talking about the X address boards, left, centre and right. They should be fine, I'm talking about the ones along the lefthand side, top to bottom.)

The Y buffer chips (actually they are high voltage shift registers) are the 10 small square chips (they look like they have heatsinks on them in your photo) along the lefthand side of the TV from top to bottom on the two narrow boards (upper and lower Y-Drive boards in LG parlance) that connect to the Y-Sustain board and then connect to the flex cables that go to the panel. Their purpose is to take the Yout signal generated by the Y-Sustain board and connect it in turn to each pixel of each row starting from the top down. The X axis columns are addressed by the boards along the bottom and aren't likely to be involved in this problem. (This is what the Va voltage is for, the "address" voltage that is used by the bottom boards to address each column using the row address electrodes in the panel.) The Yout signal blasts each pixel in turn up to 10 times (at something like 200A peak current!) to light each one up to the required brightness. (This is the 600Hz "subfield drive" that you hear about.)

If there are accessible solder points on all those traces or connector pins on the back of connectors that lead to the flex cables that run down the leftand side of the panel, put your multimeter in continuity beeper mode and put one probe to the Yout signal and run the other probe down all of those points where it runs to the flex cables to the panel in turn. There should be no continuity. (TV OFF, of course!) One trace goes to each row on the panel through the flex cables.

The Yout signal is on most of the connector pins that connect the big Y-Sustain board to the two narrow Y-Drive boards... Look for the pins that are all tied together on big fat traces. Remember, the Yout signal does up to about 200A pulses for each pixel so it needs lots of those connector pins to be able to do that current even in very short pulses so it should be very easy to see which ones connect together and are the actual Yout. There will only be a few pins that are NOT Yout, (basically just a few like 5V and GND, a clock line to shift the shift registers (buffers) and a reset line), so it should be easy to find the Yout since it is most of the pins. :)

If you don't get any shorts on any of those lines that run to the panel all the way down, your upper and lower buffers (the Y-Drive boards in LG-speak) are probably just fine. If there are any shorts (often just the very top row) don't put a repaired Y-Sustain board in there or you'll probably blow up the good Y-Sustain board... :)

LG buffer designs tend to be much less prone to failure than say, many Samsung models, but it is still good to check them anyway when something has fried on the Y-Sustain board just for good measure.

I would always also check any discrete transistor and diode looking devices for shorts also while I'm at it for good measure but in models using the IPMs generally all the faults will be contained within the faulty IPM. To check any for shorts anyway, simply use diode check mode on your meter between each combination of two of three pins forward and backward. (Two pins if it is a diode, obviously.) You should see a diode drop (.5-.7v) one way on most of those... Any that show low or 0 voltage drop are suspect for being shorted and should be removed and tested out of circuit unless there is an obvious circuit design like a low value resistor across it that would make it show 0. Generally if it looks shorted it is probably blown. Again, just good to check for obvious failures while you're at it.

Perhaps germane to the discussion.

Ahh, that is helpful to find all the silly testpoints. They aren't labelled as well in this model as they are in many others.

That explains the Vzb testpoint location, it's the top of that resistor, R7.

That makes much more sense than being across a capacitor.

Thanks again for all your help, guys. Very much appreciated!

So, I have two issues with testing the buffer chips. A) My meter doesn't have a 'beeper' function, and B) many of the traces on the buffer boards are covered by glue, which I don't want to disturb unless necessary. (See photo) If it matters, while the set is powered up, all ten of the buffer chips' heat sinks get at least measurably warm to the touch.

EDIT: I don't see any connections between the YSUS and Y-buffer boards, other than two small cables. Could the large metal plates I'm seeing between the boards actually be the connectors/conductors? :scratch2:

You can just use the ohms function but you have to watch the meter for a low resistance instead of just listening for a beep.

I would probably just test the ones that are exposed, it's at least not worth removing the protective coating. If you really want to completely test them you may be able to disconnect the flex cables and get into the socket connectors that way but I don't know that I'd recommend that, the cables can be very fragile and you certainly don't want to destroy an otherwise good panel by messing up a cable!

Yes, it looks like on that model the top and bottom metal bars are ground and the middle one is the Yout signal. There shouldn't be any continuity between ground or Yout to any of those lines to the panel. Personally, I think I would just test what is exposed for obvious shorts.

If you do repair the Y-Sustain board yourself just be sure to order the correct IPM by the part number on it or it's equivalent. There are different versions by different manufacturers (like an LG number and an STK number) which are functionally the same and interchangable but the two on the board are probably different from each other so make sure you get the correct one if you're not changing both (or make sure you put the right one in the right place if you do change both) as there are some modules that have the same pin locations and functions but are lower power and may work for a short time in the high power position but then blow up.

You certainly don't want to have that happen after all your hard work! :)

Okay, so to test the exposed traces, with the TV unplugged, I put my negative probe on that center metal plate and test each trace on the buffer boards that I can with the positive probe? What should my meter be set to, and what should a correct reading be?

I acquired some solder wick today, too. (RatShack apparently calls it 'desoldering braid'.) Should I go ahead and remove the blown SUS IPM or is there anything else I should do before removing it?

I found Coppell's stuff online and on eBay, but I haven't yet verified which IPM is the correct one. I assume the part number is on the IPM itself and thus will be unreadable until I desolder it?

Yeah, that's the idea. It shouldn't really matter which way around you put the probes either, you are just looking for very low resistance ones that are basically shorted. Probably use your lowest ohms range... They should be essentially infinite to both the Yout and GND if they are good and are usually down right close to 0 ohms if they are blown shorted (under 100 ohms is usually about where most continuity test modes beep). When they short they generally short to Yout but why not check to both for good measure. Don't be surprised if you don't find any shorts, I expect your buffer chips are all just fine.

No, I think so far it looks like everything else is checking out fine for you so I would probably go ahead and desolder that IPM so you can see the part number on it, I don't know what it should be on that model since I've never worked on one. :)

If you're lucky it will be one of the ones that is potted with clear goop on the back and you'll be able to see where it blew itself to smitherines. If there is carbon blast on the board from the mini-explosion, clean that off well once you get the IPM off, too, of course.

A little patience and some desoldering braid should get that sucker right off of there without too much grief. It does take some heat to melt that lead free solder but you shouldn't have too much difficulty. Again, feeding in some lead/tin solder to mix with it really helps lower the melting point and makes the job easier if your iron isn't galactic-temperature-hot. :)

Good luck!

I have been here for two hours trying to get this IPM off and I've only managed to free five of the 33 pins. I don't think I can do this. :no: The solder wick just isn't cooperating. I add leaded solder and it just doesn't want to come up...

Hmm, that's unfortunate... Perhaps Radio Shack's solder wick isn't the best. The stuff I have is a formula for the lead free solder and it sucks it right up.

Is your iron perhaps not getting hot enough to really melt it thoroughly, I wonder?... Do you have a soldering gun or something with some real oomph? It's not like you can hurt the IPM since it is blown so unless you are scorching the board to the burning point you probably can't really hurt anything with some really good heat...

I really think I need a better iron. I have two; one is a Heathkit with three settings from the 1960s, and the other is a cheap Chinese one I paid $6 for...

Invest in a good Weller or Hakko - I've used a WP25 and WP40 for board work in the past, but switched to a Hakko 936 station. MAT Electronics had them on sale in 2009 or 2010, so I took the plunge. Chem-wik is a good brand of solder wick, although I've used others. I also like chip-quik solder for removing multileaded devices soldered with lead-free solder. It lowers the melting point quite a bit. It's a bit pricey, but I still have more than half after 3 or 4 years.

Cheers,

Okay...decision time.

I think what I am going to do is purchase a used YSUS board and install that. I have tested most of the other boards in the TV and see no reason to believe the board will be damaged if I install it. One question, though: If the donor TV's screen was cracked while it was in operation, or it was powered on after being cracked, would the board harm itself or my TV?

Powering on with a cracked panel almost always blows chips on the buffer board and often damages the Y-Sustain board also, yes...

Best to get a known working, tested good, board.

I do know that this seller will take the board back less return shipping if it doesn't work. Now, all I need to worry about is my set damaging his board...

Well, the replacement YSUS board is on its way. Wish me luck in keeping the magic smoke contained! I'll report back once the 'new' board is installed, hopefully with a photo of a happy TV. :yes:

The replacement YSUS board came in today. I checked the fuses and did a visual inspection...all looks good. Upon installation, still no picture! When I turn the set off, there is an ephemeral, barely perceptible flash on the panel. I also noticed that all three heat sinks on both sustain boards don't get nearly as warm as they did before, and in particular, the larger of the two YSUS heat sinks stays stone cold. No fire or smoke, though! Upon closer examination of the new YSUS board, I found the small surface mount fuse near the bottom row of capacitors to be ...odd. With my multimeter, it tests differently than good, but not completely open. I guess this is the problem? I'll probably ask for my refund now, and hope a better board becomes available. :sigh:

Success
 

I got YSUS board #2 in the mail today and checked ALL the fuses. Finding everything good, I put the board into the TV, reconnected everything, plugged in the power, turned the set on, and...was greeted with a nice bright picture! :banana: It's been running for several hours now and still going great.

I do notice a bunch of grain or static in certain parts of the picture, especially when it's dark. Any idea what's causing this? The set also runs really hot. Is this normal? I don't want those damn IPMs popping again! :worried:

It's been a couple years since I worked on one of those models. But if my memory serves me correctly, there are a couple pots on the Y/Z boards that you can tweak to fix the "sparkles". Seems like one of the main ones was labeled "Vscan" or something similar. I am sure there is a "correct" way to do it with a meter, but I always just did it hot and eye-balled it. Maybe somebody who has worked on one more recently can jump in with more up-to-date info.

That's actually pretty much exactly what I did, after some reading up. One of the two pots under the heat sinks on the Y-board was able to clear up the 'red snow', but I still have some 'sparkling' or mostly white flecks in the screen when there's no picture. (i.e. black screen) Not noticeable with a picture on the screen, so far. I didn't adjust any pots very far, and put all but that one back where they were, as I was morbidly afraid of popping my new IPMs or worse! :worried: