Hi another update on this 65 inch LG 4k TV Turns out it's a 3d TV when there's new fangled 3d pieces of rubbish. I should just put it back in the dumpster shouldn't I Anyway, update video on this. Um, people wanted me to do various stuff with it. didn't have time before.

so let's plug in the ethernet ease and check this out with. uh, some search for Eevblog on bing and you end up with some weird looking Yen Dude, he's supposed to be me I Don't get it. fail. Anyway, Um yes, the web interface works.

Everything's hunky-dory and yes, I actually got a remote control for it. so I'm not sure if it's the original one for this, but it seems to work just fine. So as you can see the screen is just fine. I can get into these are app thingies here we got Netflix stand all that sort of stuff now you see.

Look, we can get inside Netflix No problems whatsoever, right? But watch this. Oh, let's watch The Simpsons Movie Well let's watch the trailer for it and see what happens. Look at that. It normally auto plays the trailers and stuff like that, but we get that weird.

my own animated violence, nudity and drug use. Fantastic. But yeah, as soon as it plays something, it will go to pink. So once it's actually displaying certain video content, this is not coming from the HDMI So clearly I was wrong about it.

There were my initial assumption that it was like because I was feeding HDMI and that was causing the problem. but no, it's not that, So it's definitely not the HDMI input decoder and whatever you want to know why you know that part that section of the board. There's obviously something where it renders video that actually causes the problem. Apart from that, the TV is absolute.

The actual screen is absolutely perfect, and this is a bit freaky, but let's watch the video of this. TV Let's go once again, it displays all the apps absolutely perfectly so that's obviously not answer. Look there it goes. Video ads and my money don't.

I Unbelievable. I Just I Don't see heads on any of my youtube videos because I've got YouTube premium. So anyway I recommend YouTube premium. but look there you go.

So there you have it. Obviously when it's displaying any form of video content, it buggers up. so nothing to do for HDMI input. It does it from any like any sort of streaming sauce because it's going to have like H.264 H.265 like hardware decoders, probably in the main chipset or something like that.

But anyway, I Was able to acquire the service manual for this thing, so let's take a look and there's an interesting thing on the back as well which we'll have a look at now. It does actually have component in and that sort of jazz, but I don't readily have the like TRS jak2 component video. I'll probably have one somewhere, but the lab is a mess. let's say the lab.

I'm like standing where the benches used to be. So yeah, it's kind of hard to cobble stuff. Anyway, that one of the interesting things is that we've got a Rs-232 serial interface for like our servicing and remote operation. and as it turns out, the actual manual not the service manual but the consumer manual has all the commands for the Rs-232 interface.

It allows you to remote operator like you can a and change channels and switch the pair or phone and do whatnot. So is it fascinating that there actually got that sort of stuff in a consumer manual? Wow Very impressed. One thing I am going to do is check for updates and spin. Might be out of date, just what.

Make sure you get the latest one. Check for updates. checking Here we go. and if we actually go in here and do a picture test I Don't know what type of picture it's supposed to be, but if we do that, why failed? So obviously this is like internally generating that videos.

so obviously some sort of video decoder is failing. But all the on-screen display stuff and anything that runs fire app and goes through that sort of path instead of a video decoder is a problem. There we go. Let's call them, that's your Z number.

So here we go. We actually have the service manual for this thing. internal use only. Fantastic! This is the exact model that we have and I'm actually very very impressed with this.

There's actually a lot of stuff in here. dates from 24 oh I see look, check leakage hot circuit test. Nice. Alright so we go got specs and all that sort of jazz.

so let's go down here and have a look at what we got now. One of the first things which I didn't know about. See all these like service menus and and stuff like that? Here we go. These are actually accessed via a remote control.

A dedicated service remote control which I Guess you can buy if you're in the service business. If you have this the block diagrams which we'll get into. If you had the Air21 ever and you had the codes for it, you could of course send those codes out yourself. I Don't know, everyone says little, use your phone my phone doesn't have an IR transmitter then yeah and you can do like calibration and all sorts of you know, test the ADCs and DACs and all sorts of stuff.

so it's absolutely fantastic. Now what we've got here the full schematics for this thing, which is absolutely fantastic. All the juicy detail is there. we've got kind of like a troubleshooting process, but it's not.

It's not really the greatest thing, it mostly just says if this board swap. Here's here's the checklist. like repair power board and parts normal, replace inverter or module, but otherwise just a check and replace main board. It's pretty much all you can do is it narrows it down to either it's a panel fault, it's a power supply fault, or it's a mainboard fault.

And that's you know, Pretty much it. So they've got some flowcharts which aren't the best, but hey, at least we've got them about. 20-point white balance ADC Calibration There's the remote. What is it? adjust, adjust, remote or something like that.

So it's basically a service remote control. Neat. Here you go: I CNT is broken. It tells you you know abnormal power section and stuff like that.

So if you're generating this test signal which is kind of a weird test signal dotting across and a faded thing, it's kind of in a weird weird to have that as a test signal. Anyway, it tells you you know, abnormal display, abnormal power section, you know, like stuff like that. So anyway, it doesn't have anything specific to the fault we had. So what we're going to look at here is we know the symptoms now.

so we're going to have a look at our block diagram. The block diagram is very handy cuz that's this is where we're going to be able to narrow the stuff down. And there's basically sort of like you know, two major chipsets here. What? Confusingly, they're both called H13.

So I don't know is that the main boarders caught hastin I Don't get it. Anyway, one is the LG 1150 4d and the LGA 1150 for a in I presume they're different chips I don't have it open at the moment. so this is interesting. We know it's not only like a HDMI receive a problem because even the internal generated test pattern generates the video fault.

Basically, any time you play a video that causes the problem so, but all the on-screen display stuff is fine. So it's got nothing to do with H 13 Over here, this is our AV component video. Scar tuner. All that sort of jazz.

Okay, so it's clearly not any of that. We can rule out all that. Now with the 1154 D Over here, this is like the main processor. It's got everything.

It's got the Ethernet in. So when we do the YouTube video for example, or the Netflix and we're coming in via the ethernet lan, it's going directly into this chip here and it's doing. We'll have a look down here. We'll come back to this in a second.

H13 d Here it is. Ethernet comes straight in. It's doing everything internal might as interest to us. Here is this video encoder: 1080p 30 frames per second.

It's got a dual dual core processor in there and it's got all the sound DSP and it's got the Sakura stuff. It's got a 1 Meg cache and then it's got down here I don't actually I don't know where the 3d we've got to de Graaff except here I Love this GPU Rogue Han: nice he's seen that Han Solo that's that's terrific his on screen display generation for example Oh H 3d is that like the 3d and jitter probably doesn't doesn't seem to do much to do like 3d to add 3d capability or whatever it's got JPEG encoders all that sort of stuff. So you know one possibility is that the video encoder inside this main chip is actually you know there's something wrong with it. But the problem with that is that you know a lot of these faults will be Hardware related in terms of like a large BGA chips like this that they heat up over long term heat up, cooled out.

every time you power them up they cycle. You can get cracks, solder joints, dry joints, all that sort of stuff. That's why often everyone just says I just reflow the BGA you know all that sort of thing or if it's extreme, your reball it. But geez, you have to be pretty desperate to do that anyway.

Yeah, like reflow the chip and stuff like that. The problem with this is that this is all internal on the silicon, so if it was this like it kind of doesn't make sense from that physical failure point of view, which is probably the most likely scenario. So unless the silicon has died like maybe the video encoder section ins in there has died as part of it. but it doesn't like that.

Seems the mancine very likely quite frankly, so it, could be like decoupling for this chip or something as a bit margin. Well then why isn't anything else affected? Why Is it just the video encoding? So yeah, I don't think it's that so I suspect if we reflow that chip that I don't think that's gonna help us much. What I think's possibly more likely its have a look up here. This just shows HDMI coming in over here like this and here's our panel output: 120 Hertz LVDS Combo EP I don't know Vx1 I don't know what any of those are offhand, but of course Ll VDS is the low voltage differential signaling which is goes off - they've got to have a pedal driver there.

And if we sure enough, follow the money down here, look at this, huh? They've got to LVDS paths OS d on-screen display which works fine and F HD full HD I Presume they've got separate on like over a screen overlay paths and separate video path. So aha, it could be that the video path is something wrong with that. so reflow in that chip could help. Who knows.

it could be the penance for that, for example. So obviously the on-screen display everything's working and the apps and things like that. So you're doing web browsing and stuff like that, it's obviously not rendering that to the video engine, so it's it's. probably going via the on-screen display path because all the apps work fine.

You browse the web, you do everything else, you go into your Netflix and all looks good until it plays that video. I Think my money at the moments got to be on this LVDS Full HD part here and then V1x. There's eight Lane thing here and a two lane for the OSD so it could Could Very well. it may not be that path there.

it could very well be these lanes here. These are by lanes physically means differential pairs, high-speed data lanes, paths basically tracers going over, so you know. So we've got potentially three suspect components here. We've got the this path here, which could be sold joints this end.

Perhaps let's assume it's a physical fault solder joints this end this end. So this chip and also this A9 chip over here, whatever that is. And then they've got this one path going off to the panel. so there's obviously nothing wrong with that one path going off to the panel and then go in under the T-con board.

A lot of people said I check the ribbon cables and stuff like that. No, this is happening on the board. This is definitely not anything to do further down the pipe. I Think it's most likely something to do with this sort of interface here.

and this is interesting in that the HDMI input over here comes through the HDMI switch. There's nothing wrong with that. That was my original thought because I was just testing with the haste HDMI So that was a a reasonable first. assumption was that the HDMI input switched Dakotah whenever it is.

That would be an issue. A lot of people said I and the ESD and stuff like that on their boots. No, it's got nothing to do with that because that's what's seen. We've proven that even internally generated video signals.

so you know it goes through a splitter here. and interestingly, it bypasses the main chip here. so if you feed in the HDMI input signal here, it can actually just split right down. Jitter Cleaner and it goes straight to you 14 down here and shoots it straight out so the processor doesn't main processor up here.

H13 doesn't do any of that, it just does all your on-screen display stuff and other things that would you know it has internal video generation because that would be for Netflix and YouTube and stuff like that would is decoding it from Ethernet for example, or decoding it from a USB port or or a hard drive or whatever it is. So oh, by the way, for those who want to know about the Rs-232 serial interface, unfortunately, it doesn't actually give me any useful information. Somebody said I'll control see when you're booting up to try and avoid, like to try break into it I can't get that. So I've got it hooked up here.

Turn the TV on. Okay, TV's on. We get nothing, but if We pair it off there we go. And we got something when we powered it off.

emergency remount ro. Sometimes we get some more data than that, but basically that's it. We just get a message when it shuts down so it's of no help so we can actually try and find. may be an internal debug thing because that is very common.

But yeah, we could potentially maybe get some additional debug stuff out of it. So let's go to this a nine ship which was like the main output driver and we can have a look here. Looks like it's got its own SPI Flash It's got chip configuration, stuff slave boot from SPI flash. There's lots of like configuration jumpers and stuff like that.

Here we go. Debugging it's I squared C Yeah, I'm not that keen to get out the scope and do it. Let's see decoding and stuff like that that you know. No thanks.

I don't see a big value in doing that that. the LG E 7411, the Ursa 9. once again, it's more like option jumpers. is there? The schematics? Probably not going to tell us a huge amount.

Look at all the power pins. That is a metric buttload of they're just the VSS They're just the ground pins. Wow If you lost a couple, lost a few dozen and it's still not going to be a problem. Anyway, this has its own core power supply, so once again, I Don't think it's that because obviously and then it wouldn't.

If the power supplies weren't failed for this chip, you wouldn't be able to get the OSD data coming through for example. So if I went in and measured those local regulators, I'm almost guarantee that they're fine. It just wouldn't make sense for it to be a power supply issue on that chip. So I've got a ground.

where's all the power? Oh over here I didn't see that. There you go. That's all. There's all the power rails.

Sweet look. secret secret squirrel. Um, only for training and service purposes. Its lucky.

I'm doing training and servicing. alright. so I've got the board out I Thought I'd have a least have a go a reflow in a chip now I Know it's not the highest odds, but this person nine chip up here. It's this puppy right here and it's already had the heat sink in pad actually removed from.

it's stuck to the back of the the metal plate on the thing. so I don't know. Maybe that's kind of a you know, a suspect in its own right. Perhaps Maybe it's you know it's gotten too hot or something like that and the adhesive on the back of that's come off I Don't know.

but anyway, this is the of course goes out to the LCD panels here so it's not gonna be anything on this side of it but but certainly something coming in that on screen display bus coming in perhaps? Anyway, it's readily available. the other ones I'm not quite sure where they are, but these seem ridiculously difficult to get off. There's a main chip under here as well and another one up there. They're really just for the pure access sake.

I'm just going to have a go at just re flowing this. I'm not gonna reflow the whole board here you go. It's just a couple of bypass caps there, so nothing really doing there at all. So what I'm gonna do is actually use my pre heater here.

it's a Bowie pooh-pooh II I'm pre heat up this J 8284 those play along at home I got this like for like 10 bucks on eBay it cost more for posties and it did for the actual thing I got at like an auction. nobody bid on it. So it's a preheater so what we're gonna do is just heat the bottom of this board. There's multiple layers of ground planes in here I Don't know how many layer board this is, but basically the idea is just to raise the bottom temperature of that so that we when we hit it with our hot air gun on top.

it's not as big a thermal shock. And there's pros and cons I know some people don't like using pre heaters, other people swear by them, etc. etc. I Just thought I'd give it a go.

maybe just you know, bump it up by a hundred degrees is just help it along. So all the stuff I'm going to do here, please do not take this as my recommendation or best practice or anything like that for Reef lowing or reworking BGA's I Do not do this for a living I don't really have any major feel for it because I don't do it every day or even every month stuff like this. So really I'm just kind of like just sucking it scene if I can make any difference here. So by all means leave comments down below about how you re flow BJ's and all that, but you know, please look just yeah.

keep it nice. Okay, just got a small amount of pre heat under there hundred degrees C that's pretty warm so yeah, that'll maybe reduce the thermal shock, but I don't want to do too much other damage as well. You've got to be careful with any as I've done in previous video. Oh yeah, I've melted the connectors a previous time I tried to do this in controlled in my reflow oven anyway.

I've got my quick 861 DW I said it for a rough from just 400 degrees nominal maximum airflow. so we'll give it a go I could have put some kept on tape around here, but anyway, just looking for any movement maybe help reflow the pins self-center Of course there's an art and science to this. and as I said I don't have a feel for it. This is just me trying to do something, at least just to see if I can get this chip to reflow and make a difference.

If the chip doesn't actually get to 450, there's a lot of loss in the air. It all has to do with air flow and how quickly it can track how efficiently it can transfer that heat and the drop from the element, etc. etc. There's a ton of stuff which goes into this and they didn't see that chip.

you know, expect it to see it. sort of like just maybe slightly wobble a bit and again, you're going to be looking carefully for it. So hopefully those joints get hot enough to reflow and maybe we'll fix something. Well, unfortunately, that didn't work.

I'll spare you the details, but exactly the same problem. it still. we haven't harmed it. It's still hunky-dory so the thermal pad for this one actually came off is the infamous U14, which connects everything else.

You can actually see the traces actually going up here these pairs. They jump down at the bottom and go up to there, so you know we've kinda sort of tried to reflow that. whether or not the balls actually refloat, not 100% sure. Anyway, this little baby here is central to everything.

It's coming from the other main chip itty dude are up here, but this one interconnects the more. so. this one technically has the highest probability of having and interconnect a uie a BGA ball that is suspect on one of those oh not the OSD line. the the actual video line of do the same preheat and reflow on that one and that's the bottle bird chip there.

What's there? Once again, there's a couple of bypass castes, but there's really nothing else. Interestingly, there's a flat flexor cable connector there, so I only put part of the board on there like there's no point heating up the whole thing. It could fit like lengthwise, but then you heat up and it's got those little rubber bumpers on the bottom. I Don't want to melt those or take them off or anything, so just give it a little hand.

That's all alright. Got some pre heater action on there. Let's give this a go. Mac's hundred and twenty liters per minute airflow.

Let's try and heat it up. hopefully I Don't melt this. I should cover those plastic connectors. Should put some tape on those famous last words right? Should I Coulda trying to find the slightest little movement in that just to see that it's should? Hopefully it'll The surface tension of the solder will keep the balls in line.

I Just want to see some resemblance of movement. it could be hard to see. Ah now this is really interesting and if you remember the picture before the squares and it was like pink and it was shimmer in all over the place. it has changed.

It is now fixed. So something has changed and I didn't see any change like this when I reheated that output chip the one that was driving the panel I didn't see that door but heating this one up it's done something. ha and that's the same for Netflix as well. We're definitely onto something so I might take that board out again.

It's just hanging on by a loose thread and just I don't just reheat it again. Maybe I wasn't vigorous enough? Just put some Kapton around there just to protect their connectors there. Hopefully the higher temperature will get the heat through quicker. hopefully reflow the balls quicker without putting less and putting less stress on the on the chip.

But obviously we didn't do any damage last time and this is as high as the Kwik goes by the way, 500 degrees and 200 litres per minute. I could like decrease the airflow, you know? But once again, I said I Don't have a feel for this because unless you extensively do this, like every day every week. yeah, really get a feel for the particular type of gear that you use in then? well you know you're just having a stab like I am all right, let's give it a bill. Yeah, panel still works so we didn't damage it and picture test net same bummer.

and I found my um, wider nozzle and this nozzle is much nicer because your hand doesn't get all the reflected heat coming back. okay I expect a high probability of it not even powering up now not yet completely gone. Dan that chip is gone ski well it's it's moved out of alignment there. I it went in doubt add flux could say well we should use flux to begin with but I've got this liquid flux here I have not used it before so please don't take this as a recommendation.

but it's sort of like it's got a brush and it's liquid so I want to try and get it at least to run down the balls there. So I do all four corners let gravity try and get in there but this so little gap between the balls in there it's going to be hard. Oh Nope. I Do believe we've done our - we've got some lines and they're not sure if you can see that.

Oh now I got lines on this half over here so it seems to be like yep, it's just changing all the time. Well there you go I think we've come a guts are on this side poor little chip. I think it's has made me a bit too much abuse and you know the thing might is shift on there. We're probably like you know, shorted some balls out something like that so that's going to be really hard to get going again unless we actually physically take the whole chip off.

Clean up the pads and and stuff like that. there's I don't know how many you know. Five, six hundred more pins on this thing, clean up the pads and maybe some balls have come off so you might have to reball it and that's not I don't want to go down that rabbit hole really? certainly not in this video and like it's nothing else around it. So I think it is that lane interconnect that we of course reflow this one up here.

It wasn't that what they use different memory here to like? D1 decodes video one, doesn't it? You know it could have potentially have been that, but the fact that we actually got it to change substantially when we reflow this chip shows that that thing was an issue. So that's most likely our diagnosis was right there. so that was a real bummer. But at least we had a go at reflow in this thing to see if it made a difference.

And I hope you enjoyed that troubleshooting procedure in the end. I Believe we guessed the right chip. We, so that you know, looked at the symptoms, narrowed it all down and said the highest probability was that chip. and sure enough, when we refloat that puppy the first time, it actually changed so there was some difference in there.

What that issue? You know it wasn't Did it fix a couple of bad joints that we had on there I Don't know and I've tried several times since to actually try and get this back into a working state, but unfortunately our car might have another go or two because it's like I might as well, right? So I might give it another go. But anyway, I hope you enjoyed that. troubleshooting procedure is valuable to go through and see that chip so ultimately may not. It might be a board swap.

I'm not going to try and find another board like this and take the chip off like a dumped board and take the chip off and all that snow. A few more shots at this. see if I can do if I do I'll put it I may just watch a video on the second channel if I get this thing fixed. if not I Hope you enjoyed the troubleshooting procedure and if you did, give it a big thumbs up.

and as always, discuss down below. Catch you next time.