EEVblog 1388 - Dumpster Diving 4K TV Murphy's "Repair"

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 [Music] you

Channel: EEVblog

Views: 83,167

Rating: 4.9327493 out of 5

Keywords: eevblog, video, murphy, murphy's law, repair, dumpster diving, dumpster diving tv, hitachi tv, power factor correction, pfc

Id: SrKLVzChy9M

Length: 29min 41sec (1781 seconds)

Published: Fri Apr 23 2021