Hi, Yes, you can see me, It's Dumpster Dive in time again. I found this down in the dumpster room. It's a 50 inch uh, ultra Hd, 4k Hitachi Smart Tv. Um, complete with stand.

Didn't have the remote control. couldn't find it. Pro Dumpster diving Tip: If you find stuff like this, try and find the remote control as well. They'll often throw it like in the dub.

So this wasn't actually in the dumpster, this was just sitting on the floor There, as they often are, Doesn't seem to be any damage on it. I'm not seeing any physical damage anyway. I have actually powered it up and I get absolutely nothing. so it's highly unlikely to be a damaged panel.

like a cracked panel or something like that and failed electronics. that's you know. that's pretty rare, so I suspect the panel's intact. but there's maybe you know input fuse blowing could be dodgy caps again, who knows.

So let's open this bad boy up and check it out, see if we can do a repair. And here's the back of it. It's really hard to manage one of these, uh, like physically handle one of these things Anyway, it is a bit, uh, how are you doing in terms of like scratches and crap on the back so I don't know where they came from. Uh, the legs are a little bit, uh, rusty on it.

discontinued of course, but you can see it is pretty thin. Check that out. Bobby, Desla and the panel for those playing looking home. Technically it's a 49 inch.

I think I looked up the data. she's like 49.5 Or I don't know. Like, you know, 50 inch, 49, 50 inch? Um, ultra high definition 4k. Not sure the exact vintage of it.

Anyway, it's got all the requisite stuff. We've got our Usb 3s. We've got two Hdmis. Uh, we've got optical Usb 2.

we've got the Ethernets. The third Hdmi Don't know what this means though. Is that? Go ahead. Will come pay for part Monday 8th of the 7th.

I don't know. Is it a repair jobby like I refurbed or something? First thing you want to do, of course, is check to see if it's actually drawing anything at all. I could have put my power meter up to it, but just measure the mains input there. I want one.

It's completely open. now. you'd expect something there, whether or not it's a switch. Of course it's a switch mode, but doesn't matter.

Like if it's a switch mode or a linear, you expect to measure at least something not open. Yeah, that could indicate input. Fuse blown. All right.

I think I got out most of the screws. Unfortunately, they aren't exactly the same. I think you can see a difference there in those. That's a little bit annoying, but this.

I'm not sure if I have to take the legs off. Oh no, this one's on top. damn it! Yeah, it looks like I have to get the Uh panel for the legs off as well. and it's all very dusty and rusty and I don't know where this thing's been used.

All right, I think she's gonna come off. Probably should lay this flat, but ta-da we're in Lake Flinn. look at that. You really need a decent sized bench if you're working on stuff like this.

I should put just clean the crap off here and get get it done sideways. anyway. power supply. and right off the bat I spy with my little eye a Pcb mount ceramic fuse there so we'll measure that.

and yep, of course, input fuse blown. Why is it so? so this is where you start your uh, visual inspection. You don't just go replace the fuse Willy nilly. This is, uh, interesting.

Is this the Diode bridge? Um, yeah. they've just got four, uh, regular axial, uh, diodes. One in, you know, five, double o, whatever. Um, and they're just, uh, flapping around in the breeze.

There They, of course, uh, do raise them up off the board like that for uh, ventilation, uh, purposes and expansion of the when they heat up they can. You know, the legs can expand and they don't crack and all the rest of it. Sometimes you might see them with like a little loop in them as well to take out any thermal stress. Okay, it looks like is that a Ptc? It's got an R T.

I'm not looking for anything obviously blowing the uh. main Dc input filter cap. Look at that sausage Jobbie. Hbc.

No, there's nothing. nothing blowing, nothing blown on the secondary, uh, side caps or anything. so magic smoke hasn't been released. I got input chokes, input caps.

um, but not the X and Y class caps down here. They look all intact. Everything looks fine apart from the fuse being blown. Hmm.

So first thing I'm going to do, probably like measure. I don't have to work out the circuit, just measure across the mains. the main Dc cap here. hello, hello, Um, that's a dead short on the output of the Uh bridge rectifier here.

Wow. Okay, we've got to fix that first. Before we, you know, don't just go replace the fuse because it just would have blown again. Okay, so there's no visuals on the top that I can see.

I mean, we've got our switching trenny under there. Um, maybe we can get in a measure, but I've taken the screws off. Let's have a look on the bottom, not seeing anything obvious. You might see a patch around there, but that looks like it's more.

Yeah, that's sort of like almost a conformal coat. It's hard to see this sort of thing on camera, but I can see it anyway. Hot and cold. I love how they call it hot and cold.

There's cold on here. Uh, where is it here? It is. it's cold. There's your update, a couple of feedback.

there's your hot side. I'm not seeing any blowing part or any magic smoke released. Not seen it to give like a dead short. There's a dead shot directly across there.

It's no blowing traces. No, no, that's did that look like. Um, no. that's all on the secondary side.

Anyway, we're looking for a primary side. Uh, short. I love the spark gaps in there. There you go.

I've done a video on spark gaps and demonstrated them. That's really cool. It's always worth looking for any cracks on the input. That also could be a failure mode.

Like if the fuse measured. Uh, right. I'd Next thing I'd probably do is take the board out and just look for cracks on the mains input connector. But no, visually, I'm not seeing it.

We'll measure our diodes. Yep, they're all good. Yep, Yep. no workers.

What's this bad boy down here? Oh hello. We'll move that back to Ohmskee Range. and Bingo. We might have found our culprit because it's more likely that you get a dead shot inside a semiconductor.

Like a diode like that. That is one of their failure modes. They can fail open and short than like inside the cap. Okay, I've pulled him out of circuit.

And yep, that's the culprit. Measure across the mains cap again and that's what we expect. We see, we saw it charge up there. No workers.

So there you go. We have a faulty diode that is an Mur 460.. there you go. It's an ultra fast plastic rectifier according to the data sheets.

Oh, you could put probably any sort of equivalent in there to get it back up and running. So whether or not, that's uh, taking out anything else so I don't know, Maybe. oh, can we take the that's clipped? Is it? Oh no. Is that how is that held in? But uh, yeah.

Maybe you can have a look under there see if there's any blow holes in the um, driving or not? Yeah, there doesn't seem to be any other damage. Usually you'd find something else you can get other, uh, more obscure faults, like breakdown in the insulation of the enamel wire on transformers and stuff like that. But that's you know, that's pretty rare. so I just want to get in there and measure the uh transistor.

It's not a bipolar joby. it's one of those newfangled mosfetty things because it's got g for a gate hey, drain and source shorted. There's another problem. Yep, it's not just the diode.

So there you go. If you just tried to, uh, replace the diode there, and uh, you'd probably become a gutzer and you might, uh, blow it again, because, well, switching transistors shorted as well. Damn and out of curiosity, I just checked. You can actually buy complete replacement uh, boards for this on ebay for 82 bucks? Um, you know, new old stock or uh, ripped from uh junk? Uh T? You know, smash panel, uh Tvs and stuff like that.

So you know if you didn't want to like, get the parts or whatever and try and debug this and spend the time on it. You can, just, um, spend you know, 80 odd bucks at least here in Australia and get a replacement board and probably just work. It turns out, this is a bit of a pain to get out. It's actually, um, screwed in from the bottom side here so you can't even release it from the top.

and then it soldered. Uh, three solder tabs down in the board as so you gotta desolder three major uh tabs for the heatsink and the transistor itself. But it does lift out in one assembly and that is an Mdf-18 in 50. so that would be uh, you can tell by the numbers on these mosfets.

usually 18 will be the current, so that'll be 18 amps. so pretty beefy and 50 won't be 50 volts, it'll actually be 500 volts. Because what I think's going on here is, you can see the architecture here. This is obviously our main switching transformer here, because here's your isolation, right? This is what this, uh, crosshatch indicates.

This is your hot side and this is your cold side as they call it or high voltage, low voltage. And so this is the isolation it goes through. As I said, uh, there's the opto coupler. So this is the main switching transformer here, and the main switching transistors.

for that. They've got to be on the bottom. So what? This is over here. Okay, you've got your mains coming in here.

You've got your input filter, common mode filter section, you've got your bridge rectifier, and then you've got another switching element here, which is this uh, transistor that we just took out here and this little thing which isn't a transformer, It looks like a choke. So when you see a switch in training like this and a choke, and there'll be a chippy on the bottom, there's the controller there, and there's your uh choke over here. So that's interesting. That's actually three terminals there.

But anyway, when you see that switching element after the bridge rectifier before the Uh mains filter cap here, that's obviously doing active power factor correction, and if we were able to pull the part number off that, I'm sure we'll find the conformal coding makes it really hard to see a number on there. But I'll get it. So I'll bet you're a Bitcoin that that's a power factor correction controller. Yep, I get to keep my Bitcoin.

It's a 5590 power factor uh, controller and you can see up here that as we uh said before, it basically is put between. this is your main big ass Dc filter cap, so this is your Ac input. they should bridge rectify. You'll have your uh, you know your common mode chokes and everything else.

and that's this inductor. Is that a big sort of transformer looking uh device. We found the three terminal jobby and uh, q1. There's Q1 right there.

That's our switching transistor and there's our blow-on output diode. So both of these active devices here are blown. So what's caused that? Once again, I don't know like one of them could have died and taken out the other, or whatever. There could be a like an upstream uh fault up here or something like that, but oh geez, I don't know.

Like at this stage you would, just, uh, replace these two active devices and because everything now measures uh, fine across here, it doesn't you know. At least at least with the meter. anyway. So like, there's no dead short up here, that's going to like, you know, blow this diode again and then maybe take out the transistor here.

Who knows. So at this stage, I'd say yeah. The best bet is to get a replacement Diode replacement transistor and see, just power it up and cross your fingers. Although that being said, uh, you probably want to suspect the main Dc filter cap as well and I measured that uh in circuit and it seemed to be okay and there's no damage.

Yes, I desoldered the thing because yeah, I just thought it. It's worth. uh, taking a few seconds to get it out and uh, you know we can just re-gunk it down there. They stick it down, stop it flapping around in the breeze and uh, yeah, let's measure it externally.

Let's put it. Capacitatory Mode: 87 Uh. Mics: It's a nominal uh, 100 mic, so you know that's good enough. That's it.

One Kilohertz. Let's just wake that up to 10. We expect that to drop a bit. Yeah, 79.

Still doing okay, and this, uh, probably won't measure at 100k? No, because it actually has an At 100k. It actually changes ranges and has an issue, so we can't measure that. but that looks okay. And the Dc resistance of the impedance? 0.4 Ohms at 100 kilohertz.

Yeah, that sounds good enough for Australia. Just for completeness, I will get an Lcr meter that can actually measure the Uh capacitance at a hundred kilohertz. and that's at one Kilohertz 10 Kilohertz Once again. 77 Mike: 100 Kilohertz 17 Mike But uh, that's to be expected.

And if you don't believe me, we'll get a low Esr 100 Uh Mic cap. Bingo 18.8 Mike at 100 Kilohertz That's what it drops off at. And of course, that Esr. Again, we can actually measure series resistance.

Yeah, 0.53 This one measures a bit higher at 100k. But yeah, so I deem that cap fine enough to go back in. I can't measure anything else. This diode here is fine.

Uh, everything on the secondary side seems to measure fine. All the caps look fine. Everything's hunky-dory so we're going to replace the transistory and the diode. Now by the way, if you didn't have these replacement parts, you could actually potentially just, uh, bypass this power factor correction system by simply removing the transistor and shorting out the diode there.

or putting a just a regular, uh, you know, replacement high voltage? Um, you know, suitable current a diode in there and then you simply got a a regular, uh, full wave bridge here which then goes into your main Dc filter. Cap your power factor correction. All that does is instead of uh, basically presenting a capacitor across a bridge rectifier and I've done a video showing how this is like a really poor power factor. They of course actively monitor it via the feedback here and then switch the transistor and use this inductor to then, uh, you know, deliver the power to the capacitor at the correct time so that your mains over here.

Um, doesn't think that this is a capacitor anymore. It thinks it's just a resistive load. You could leave the inductor there. It's not going to probably do a huge amount.

Yeah, you could just you know to get it up and running again just to prove that or rather than spend all the time and effort. But anyway, I have done a complete video which I will link in and you've no doubt already seen it which was a, uh, spoiler where I went through ordering a suitable replacement part from Um Rs Components here in Australia. It turned up so we got our transistor. We've got our diodes here.

They had to. You had to buy them in a pack of Uh five. So no worries. And I completely forgot about the Uh ceramic fuse which uh, has blown of course.

Um, this is a five amp Hrc one. I didn't ha. I don't have one of these uh, suitable, especially a leaded one like this. but I do have in stock are some five amp uh, axial ones like this, which should work a treat.

These are 250 volt rated. Uh, these axial ones typically come in, you know, either. like, like low voltage ones designed for like, a secondary side, uh, Dc fusing and stuff like that. Uh, you don't want to use those on primary side domains, but these are, uh, properly, improper 250 volt rated.

They're not as beefy as this big ceramic. Well, these are actually uh, ceramic inside. believe it or not. So yep, that'll be good enough for Australia.

So let's wipe the transistor and dye it in and put the cap back and power it back up. Solder out of these holes. They're annoying. No wackers.

Let's probe up the Clacker 450k. There you go, so you'll notice something's missing in there. the transistor and the heatsink. I've deliberately left that out so that we actually have the circuit like this, I've taken out, I've removed the transistor I put the diode in because you can either short out the diode or leave it in.

It really makes no difference and we'll see if it powers up without the power factor correction circuit because without the transistor in there to do anything, um, with the inductor to, uh, to short it out, then, um, it's just basically just it's going to be Dc through the inductor, through the diode into the cap. like that, It's not going to be the most efficient thing and of course it'll have no zero power factor correction. But who cares? Let's just see if that works well. As I said, um, that will get it actually up and running and it'll probably just work fine forever without the power factor correction.

If you're happy, uh, with the extra load which you're typically in your residential applications and I've done videos on this in your residential household, you're not going to pay for a poor fail power factor devices, although you know it's not good for the environment. Of course you know you want to put the transistor in if you can. Um, but still, you're not going to pay for it. Well, okay Dave, technically you are going to pay for it in terms of like, increased utility costs because you pay a fixed a fee and that pays for all the infrastructure.

Everyone had removed their power factor correction transistors like this from their Tvs. Then the utility company would have to install greater uh, you know, transmission line capacity, bigger lines, and everything else. Um, more power generation? It's it's got to come from somewhere. You know it's not magic, and I know you're curious to see how much this actually draws before and after the transistor.

So this is with no power factor correction. this is in standby. It's drawing 327 milliwatts power factor of 0.02 Yeah, it's pretty bad. So if we check out our Va, it's actually drawing 17 and a half Va.

Or you know, I don't want to say 17.5 watts because it's V8. But you'll only actually pay for 0.3 watts. You won't pay for 17.5 watts. You'll only pay for 0.3 So very poor power factor.

Let's switch it on. and 16 watts? 98? What's 800 watts? That's that's what I expected. Our power factor. There you go.

It's gone up to 0.5 as you'd expect. Um, you? It improves that. It's very poor at low power factors. so you know 0.5 that's not too shabby.

So you know that's 194 Va. But you know you'd only be paying for that 99 watts. Still, let's uh. let's power that on.

Well, it's flapping around in the breeze. She'll be right. Hey, winner winner chicken dinner. I see Hitachi on the screen.

As I said, you really do need like a really lot of bench space to work on these large screen Tvs. There we go. Got red down the bottom? That was the standby. Was it Hitachi Inspire? The next? The next? What? spy? The next repair it is.

Remember this is a 4k, not that. uh, regular Hd. rubbish. Geez that that took some time didn't it? It's got the Netflix's it's got the Youtubes.

Got the Twitteries. Yeah, I was a shadowband on Twitter by the way. if you weren't aware band for like a week and eventually just magically disappeared I could be shadow band again. who knows.

Got all white on the screen. it's looking really good. It looks all washed out in on the camera, but trust me it's good. Oh no.

it's going to have to go straight back to the dumpster. Um, this is how it turned up in the dumpster. Aliens. It's always aliens.

Actually, it turns out once you get to this 404 era here, the soft buttons on the back do absolutely nothing. I can't even power it off. So if we can do a network firmware upgrade, shall we? it does. Connect to the networks by network is lo is log in in is logging in in.

Log in in I'm sure right? that's a There's a t-shirt in that there's a Hitachi t-shirt with is long. Log in in cannot connect the human web. Please check your status. This is hilarious and it's not formatted properly, but there you go.

There's a raspberry Pi Hdmi input. no workers works. A treat. all right, So it works fine without the uh power factor correction.

But now I've added back in the new transistor and let's give it a bow. Hopefully we don't release the magic smoke and 3.2 watts. There you go. So it actually draws more in standby with the power factor correction.

Oh, that's that's a shocker, That's terrible. And the power factor correction is not that great. It's better, but 0.16 But this is typical of uh, low power products like this. that's not great at all, is it? Anyway, let's switch it on.

Wow. Well, that's very interesting. It doesn't work at all with the switching transistor. Are you kidding me? Well, that actually makes no sense at all.

Because if the switching transistor just wasn't let's say the Uh switching. I see. Let's say that was busted and it just wasn't driving that transistor at all. The transistor would just sit there open and it shouldn't do anything.

It'd be just like it's removed. but in any case, it's drawing like a couple of watts. so something's going on. But it seems to be now.

No standby power, presumably going to the logic board. What the? I don't get it, just a shout out to small design touches like this. The standoffs that they've actually used, which are it looks like they're embedded into the frame. Down here.

they have little extended notches on them where it fits into the board like that. So the board, even though it's vertical, can stand there. Brilliant. Someone was thinking.

didn't really want to have to probe around on this thing. I thought I was going to have a winner in a chicken dinner and get this thing fixed. But anyway, let's have a probe. You've got to use your high voltage probe here.

Got the Hvp 70 and by the way, this design. I've got it unplugged right now. But um, the cap is still charged up because they didn't design in a bleed resistor and you'll notice over here that's 50 volts per division, 50 100, 100 and still charged up to 150 volts Dc. It's slowly going down due to the load of the Uh probe, which is four meg, so it'll eventually bleed off, but they've got no bleed resistor on there at all.

so just be careful if you're going to do that. um, use your meter in uh, low z mode if your meter has that, uh, for example and then like short out the main filter cap on these things when you take it off. Anyway, let's power this thing up and see what we get. there.

We go, of course. Yep. 300 and you know, 40 odd volts, Something like that. That's what it expects.

Our Dc is there, but for some reason our soft control isn't switching it on like it did before. Which is very unusual because the power factor correction circuit, that transistor and that Uh. 5590 ic isn't it. That's basically a completely independent circuit.

as you saw, it worked fine. Without it, it's not tied into the switching standby or anything like that. There's nothing this is on the primary side, mains, primary side. There's only those two opto couplers over here.

These two opto couplers coming back and and that's it. So there's nothing coming back onto that, uh, primary side. So I I'm at a loss as to how that is not powering up like it did before. Maybe it's somehow the switching to this is tied into this switching controller of like the main switching controller over here perhaps which is not giving any voltage on the secondary.

Um, that's it's got to be it. There's got to be no voltage on the secondary. If there was, it should just work like it did before. Very puzzling.

You can see that cap, but slowly discharge. That's 10 seconds per division. It'll take a long time to discharge it for me. Okay, just a quick check of the voltage on the secondary.

Don't want to take any chances. Don't accidentally brush anything on the primary side. So we've got our take away container protection here and let's measure okay. Powered on.

Now this is the V power going over to the board. so luckily they've got the pin outs on here. Yes, Yep, nothing. So yeah, I didn't expect that to come on unless the standby was on the one next to it.

that should be V Stand by. Don't short it Dave. Don't short it. No, we're getting nothing on V standby, which is why it is not powering up.

Hmm, we'll just discharge that cap again. There we go. Boom. Drops it down to 98 volts yet.

no wackers. So that's very curious. Why if we get in our full voltage on our main filter cap, that should be powering the switching circuitry which is under here, which drives the transformer primary, which then, um, you know, should switch that and we should get voltage on the secondary. And then of course this has, uh, the standby signal comes back on this connector and then tells this, uh, board to provide, uh, full power out.

but there's no standby voltage there unless I'm measuring it wrong. I don't have a schematic for this thing if you do, leave it in the comments down below. So all I can think of is that as I said, this switching controller for the power factor correction is somehow tied into this, the main switching controller over here. Which makes no sense because this is supposed to be a completely independent circuit.

It's not tied in at all, yet it works without the transistor. It just passes a standard bridge rectifier, Um, passes through the diode into the main filter cap and everything's hunky-dory Everything powers on just fine, so that is. That is really puzzling now. Unfortunately, this video has gone on long enough and in show business they always tell you, you know, leave people wanting more So I'm not going to go any further with with this in this video.

Maybe that's for another video. and people can speculate what's actually going on here. So I'm just going to remove that transistor again. get the Tv going, and we might have to revisit this one because that's very interesting.

Hmm. leave your thoughts in the comments down down below and here you go. I have actually lifted the pins on this, so I'm going to power this up again. So transistor is now missing.

didn't have to get the whole heatsink out. Well, what do you know? Damn it. Look at this three watts again. I bet you it's not going to soft power up.

Something has gone wrong with it since. Uh, after I've installed the that transistor and then I've removed it again and it's all right. I've physically removed the heatsink again, so it's exactly the same thing. I haven't done anything to it except put that transistor in, power it up, and take the transistor back out.

and you saw it was working before and now it's no. We've still got the three Watts. So this is some sort of new fault. Unbelievable.

Yep. Confirmed. I just tried the standby button and sure enough, it doesn't work. So by reinstalling the transistor, have I damaged something else? Wow.

Okay, this was a repair and then I killed it by repairing it more I guess. Catch you next time you.