First impression is that the original shunt resistors were tiny, so they weren't carrying much current. But here is a more accurate calculation on the wattage of the shunt resistors... The front label states that output current is 4 amps, which means that the 12v output of the power supply provides 48 watts. Therefore the input stage of the power supply sinks approx 48 watts (ignoring losses) with an input voltage of 340 volts DC. This means that the pair of shunt resistors is carrying approx 0.15 amps maximum. Let's say 0.2 amps to allow for losses. Each new shunt resistor of 0.68ohm rated at 0.25W can carry 0.6 amps because I = sqrt(P/R). Therefore the new resistors are well within specification for this circuit. By the way, those scorch marks look like a solder blob rolled between the pins of the MOSFET, pulled up the base and dumped a huge current into the shunt resistors. This current draw took out the rectifier at the same time.
This is an insulating bead, not a magnetic bead. And it looks like something joined the traces from gate to drain, not actually touching the legs themselves, because metal was melted from the traces.
The blob doesn't need to touch the legs of the MOSFET themselves, the burnt traces imply that something joined the Drain trace to the Source trace (not base as I said above).
Excellent! Thank you, brilliant, thanks for sharing your knowledge. So after reading the comments, I'm wondering if it was possible that too much solder was applied to 2 of the MOSFET pads which flowed through the through holes and rested on the underneath of ferrite bead causing a short (solder bridge) Once turned on it shorted causing the scorching and blew the resistors, bridge and fuses. I'm not sure when these things are machine soldered if it is possible for too much solder to be applied or could the same 2 joints have been done twice perhaps????
Well done. From yesterday's video, I took apart my failed IR soap dispenser, found out what the problem(s) are, the main which is the pump is like a car oil pump with tiny nylon gears, one of which is cracked, the screws are not SS and corroded, so I 'could' superglue the gear and replace the screws, but all in all, for $10, it is not worth it, but I may use the IR sensor for something later. Finding the two 0.6 resistors faulty is why I watch, I learn every time. Thank you, Capt. Porcupine Head 🙂
When working out what has failed, think of the current path. Probably a small fruit fly or other insect was vapourised as it touched both the drain and source of the MOSFET. As the MOSFET was presumably shorted across drain to source - the drain of the MOSFET gets it's voltage from the primary transformer which in turn is fed by the rectified mains from the bridge rectifier. The current flow goes through the fuse, then the bridge rectifier then the primary of the switching transformer then the drain of the MOSFET to the source and then to ground through the resistors that attach the source of the MOSFET to ground (which I failed to notice until you actually saw they looked burnt). It's amazing how tolerant these circuits are when you think. The new bridge rectifier was well within spec to deal with the current and voltage it has to handle. But the new MOSFET you used was 100v lower than the original. Although you might think the voltage will get nowhere near this level, when looking at the output (load) of a switching circuit you would be surprised at what the voltage can peak to on the drain of the MOSFET. You will probably have around 300v or more going down to almost zero at whatever the switching frequency is. That's a lot of stress. If the MOSFET was left permanently conducting at the maximum current when it is switched on it would blow apart in no time. In fact that is what usually happens when the drive circuit fails. So I think they design a circuit to be running well inside the tolerances of the components used (or at least they should do). Those resistors you have used measure 0.5 Ohms so should be ok to leave in. Your only slightly limiting the current by creating a slightly larger source to ground voltage drop. If anything the MOSFET will have a slightly easier time of it than before - but really there's no discernible difference. The resistors are there to act as a fuse when the current exceeds a certain level (which the original resistors proved).
I think Vince is always thinking of future content, "How to clean/replace carpet, How to treat acid burns, How to fix up the shed to live in when you are banned from the house, etc."
As the MOSFET solder pads showed damage and device as a whole looked quite new, maybe there was a solder bridge across two of the MOSFET pads (manufacturing defect) which when power was applied blew the resistors, diode bridge and fuses? The MOSFET itself was OK because the pads, solder bridge and other components soaked up all the energy before the fuses blew.
Good thinking, thank you Green Jim. Maybe the solder came through the through holes and sat on the ferrite bead which shorted the legs!!!!! I think you have it, well done! 👍👍👍👍
that feeling when you finally finish a difficult fix love to see you still get massively excited after all this time you can see your enthusiasm in every video you make vince and thats why i watch them all keep enjoying yourself mate because we enjoy watching
Thanks Craig, I wasn't sure if it was a little long with a lot of back and forths, but so far the viewers have appeared to have liked it at 98% thumbs up. Cheers mate 👍👍👍
44:03. Bridge rectifiers usually dont cause any Components on the output to fail and also that big smoothing cap at the output was still intact which tells me it was something at the output of the br and that blew the br causing the other fuses to blow. If the br was to fail, that mains cap would probably also have exploded and there would be even more damage due to the fact that you feed unfiltered ac into a dc circuit. It was either a solder bridge or a tin splash or something shorting the mosfet. And due to the unknown thing shorting the mosfets pins, it did not take out the mosfet itself but rather the resistors pulling down the pin of the mosfet to gnd and everything at the output of the bridge rectifier and the fuses
A POWERful video today vince. I found that ELECTRIFYING and very ENERGIZING. I am impressed that you don't CHARGE for these repairs and simply keep your mind TOPPED UP. I'll grab my coat.................
Regarding you mentioning the 240v into the bridge rectifier, remember thats RMS not peak voltage. So the bridge must be able to take the peak of 340v. You did use a replacement that is well in spec but worth a mention. Back in the day I saw someone use a 300v rated bridge on mains and it blew instantly.
Already thinking and forgetting what i did with my life before a daily 30+ minute MMV video came out every evening. Gonna miss this daily routine when christmas is over
Hahaha, I won't Lee, I'm already done in😂 By day 12 I'll be fixing loose screws in the door handles and by day 20 I'll be that desperate I'll be changing the smoke alarm batteries for easy fixes😂 Only kidding, glad you're enjoying them, I am too at the moment👍
YEEEEEASSSSSSSSS! Nice fix!!!! If I were a manufacturer and I could get away with 1/4 watt vs 1 watt resistor and save a 1 penny each.... That's all I'm going to say. CONGRATULATIONS! What an awesome, satisfying fix!
Good fault finding, congrats! 2 Points: I am sure the Fault came from an Residue or something on the Legs of the switching Mosfet. Be aware to remove all of the metal which gots evaporated over the Mosfet and the isolating Ferrule! (think about applying Soldermask on the Ferrule and the lower Part of the Mosfet). Second: Never connect a Charger to the Batt at first. Save Way is to connect it to Mains and the Battery afterwards. All the Best, and thank you for your Videos. 😊
For the resistor power rating... The resistance value will set how hot it will get for a given application (voltage/current). The power rating is how it can sustain that heat. 0.5W can take heat for longer without changing its characteristics. However... you can ameliorate this by wicking the heat away into the copper more effectively. When you solder the 0.25W parts in, be generous with the solder and inspect how much copper is around them. The power rating must be used in conjunction with the copper area to calculate the temperature and, therefore, the stress on the part. It's very difficult to do that, so just lean heavily on the solder. If you really want to go to town, solder in stubs of copper wire sticking up out of the pads to act as a small heatsink. Make sure the wire stubs are stiff enough to not bend over and touch something else. I have done something similar when upgrading a Wii Remote Sensor Bar to run off USB and shine brighter for use further away from the TV using ultra bright IR LEDs diven at 200mA each.
Very interesting, they are connected to a via that the heatsink is soldered onto so hopefully that alone with wick away lots of heat. I will happily put massive blobs of solder on though to help with the mass. Thank you! 👍👍👍
loooved this repair. I watched it while walking on my walking pad :D. And its so funny how you said in the beginning " it probably cant be fixed ". Thats so funny. You say that like 80% of the time and then proceed to fixing it anyways :). Well done Vince. I appreciate your efforts in making these videos. I know its quite some work.
Brilliant fix on this one!!! If those resistors cannot handle the current, they would burrn up open circuit very quickly! Just test it for a few hours, measure them - if they look OK and still measure OK, they will be fine.
Dear Vince, It is never a bad idea to use thermal paste when reassembling the transistor, it can only improve the thermal transfer. The black tube that was over the middle pin of the MOSFET is not an insulator but a ferrite bead. Usually their primary use is it to kill RF oscillations... A ferrite bead is a passive device that filters high frequency noise energy over a broad frequency range. It becomes resistive over its intended frequency range and dissipates the noise energy in the form of heat.
Nice fix Vince, that burning on the mosfet was definitely arc flash, but not from any component failure it could be a bug was in there and went across the legs of the mosfet, spiders and alike can cause such damage.
It seems the short was going through the drain to source which means absolutely something bypassed the mosfet damaging the bridge rectifier and the shunts probably an ant or something that got evaporated considering the burn mark!
Faulty irons, broken kettles, soap dispenser that doesn't dispense and now a dirty old battery on the living room carpet. I'll bet your wife loves you some days Vince🤣🤣
That battery looks clean and new. Ur the third person to shout about the battery on the floor when that battery looks cleaner than most people's shoe soles. Calm down, drink some tea or something. That battery looks like it just left the shop brand new. If Vince's wife was going to find a reason to flay him alive it would have happened by now already.
Nice job Vince 🙂 When connecting a battery it's a positive thing (+ first) / when disconnecting it's a negative thing (- first) on a modern vehicle (with a -ve ground) is how I remember it.
Isn't that because of the chance of touching the car body with your spanner? With the negative disconnected you are safe to connect the positive and touch the spanner to the chassis, because at that point the chassis is floating, but the other way round, chassis already connected to battery negative means spanner touching from battery positive to chassis is bad. I don't think the battery cares which connection first, it's just about where you could short and get into trouble.
Vince - I think you best start thinking how to celebrate your 1 million subscribers! 2025 will be the year I think and this advent marathon has to boost it no end.
Thank you Christopher, it would be nice, but unfortunately my channel has sort of stalled the last couple of years, I think I get around 30 subs a day so realistically the next 100k will take an eternity to realise, but you never know with RUclips, strange things can sometimes happen. As long as I get to continue for as long as I can, I'll be as happy as I can be 💪😎👍
Persistence rewarded 😀👍 Regards the meter REL button … according to the EEVBlog page the majority said it did not work (some suggested it only worked on some ranges but didn’t specify). I think meters are meant to come with a few missing functions or you wouldn’t need to buy another one.
37:25 Clean the banana contacts on your meter with some Q-tips and IPA, and do the same with the plugs on your testleads. That should get you ohms readings down well below 1/2 an ohm where they should be.
Actually, I would bet that there was a foil bridge on the top side of the board that burned off.. That's a manufacturing fault in the PCB at the start. that would hang the 330V right across the "switch" and while the foil fault was corrected. it took out the rectifier and fuses as well.. along with the sensing resistors down stream, that saw a LOT more than the 12 or so volts the sense network was designed to take. This was a real good fix! Thanks.
Whenever you find very low value, large size resistors, they are usually current sense resistors. If you double the resistance, the controller will think the current is twice what it actually is and probably reduce the max output current from 4A to 2A. So they should never be replaced with smaller ones as that could cause excessive current. In regards to the power rating, it is hard to know the required power of these. We know the 4A output current but these resistors are on other side of the transformer and we don't know its turn ratio. Anyway, just get the highest power ones with the given footprint.
I am one of them today, this series is good really enjoying it so thought it was about time I showed some support been watching for quite sometime now.
Can you try turning your meter off and then holding the leads together and turn it back on to the ohms position and see if that zeros it? Glad you stuck with it and fixed that. Good job.
“Positive first, then negative. When disconnecting the cables from the old battery, disconnect the negative first, then the positive. Connect the new battery in the reverse order, positive then negative.” That way if you slip and your wrench hits a negative surface there is no short.
41:55 The resistors are on the other side of the board so it cant be them leaving that mark on the MosFet. I suspect there could have been a small solderbridge there left from the manufacturing that caused it to short, and blow up the two resistors. They are supposed to act like fusible resistors anyway.
when working on anything live i would suggest using leads that are GS38 compliant. even if you have a heat shrink for the tips to slide over. It would save a few unwanted issues if you accidentally touched leads whilst testing.
Could they be working as shunt resistors to measure the current going to the battery? You can use a big external resistor. Sometimes they use SMD cause it is easy to manufacture bu robots but external resistors can work better and are way easy to find locally.
yes shunt resistors - but not for the current going to the battery but for the current going to the transformer. the chip needs to know this to get the switching timing right and/or detect overload conditions
Agree. Rectifier failure caused all the problems. My Aneng 870 looks the same as your meter and pressing the REL button should zero the display. After that all readings are relative to that stored value, i.e. from the new zero point. Don't know why yours didn't work, maybe need to get your probes to show a stable value first (you leave it on REL after, by the way, don't press the button again). I bought some gold plated, really sharp probes from Ali Ex for 1.70 and they are good enough. In fact, I bought 2 pairs. Something to add to the order next time you order from China.
For the resistors, measure the voltage drop over the resistors under load (while charging). If the charging current is 4A the load should be at it's maximum. *I'm assuming here that the resistor load is directly related to the charging load, this may not be the case.* From ohm's law & definition of electrical power, we get >> v * v/r = w For 1 watt per resistor, meaning 2 watts in total, the max voltage drop accros the resistors should be ≈ 0.825 v. If you're not charging at maximum current multiply the wattage value accordingly, for instance charging at 1A while max is 4A multiply by 4. If the resistors are measuring the full charge current directly (simple solution, but bad circuit design) they should have a rating of at least 2.5 watt or higher.
I think that ferrite bead on the centre pin of the MosFet trapped some moisture, and that Moisture took out those sacrificial resistors and oh~wow! they actually saved the MosFet, ergo saving a buck..
The small tube around the drain pin of the mosfet is actually a ferrite bead, and they are conductive, it’s too big for that application, it should have been much thinner, the thing is, it caused a short and sparked across drain and source, all the current of the collapsing magnetic field in the primary side of the transformer returned to the rectifier and shorted at least one of its diodes, and blew the current sensing resistors for the driver. Really low quality manufacturing.
If you can get 1206 size resistors I would go with the highest wattage you can, the design engineer did it this way (using 2) to spread the current between the 2 resistors instead off just using one so half the power through the resistors. This is difficult to advise without looking but you will probably be ok but perhaps the first time you use it proper put your hand on the case every now and then although you have to allow for the MOSFET. I think 0.5 watt is as big as it gets anyway in 1206.
The ic is probably a SG6848T or an OB2263. (Sot23-6 housing) Pin 3 of this ic is probably connected to ground via a 100k resistor which sets the switching frequency at around 65-70kHz The philips ORT7500 (red retro radio) uses a nearly identical primary side of the power supply, and in these radios the 6 pin ic fail ALOT!!. Don’t use mosfets with a lower Vds than the original, the DC voltage on the drain of the mosfet might ’only’ be 325V, but because it is switching an inductive load (the transformer) the pulses are way higher, around 600Vp-p, something to consider! I think the cause has got something to do with the ferrite bead around the drain, it’s to big and probably they fixated it with some glue or other crap and with high frequency and high voltage pulses on it likes to spark to the source pin, which is virtually ground, current sense resitors go up in smoke and the current peak also took out the bridge rectifier and the fuses.
I didn't Standish!!! I did show that in the video and also ask the question if it was put there for noise reasons to stop interference between the legs, BUT then it got cut out because of the length of the video. I only realised I didn't put it in after soldering in the last anchor point and I couldn't face unsoldering it all again. I still have it so I can put it in when I solder in the resistors as the pain of unsoldering the MOSFET would have eased by then 😂 Is it for interference reasons do you know? Thank you, I haven't heard from you in a while, hope you are doing good 👍
@@Mymatevince The use of ferrite beads is quite a complex issue and can have a number of effects (beyond my knowledge pay grade😊) - but I trust the designers of this unit knew what they were doing and aimed to reduce either the effects of noise, interference or Mosfet turn-on speed. Thanks for asking about me Vince. I'm OK, just been quite busy and having a few health issues.
the wattage depends on the use case. it looks to me like its in series with the load leg of the mosfet - plus theres 2 in parallel already by design - i would go with same wattage or higher - definitely not lower.
I can't believe the factory forgot the thermal paste. Here's my best guess on what happened: someone tried to use it, it turned on, after a couple of minutes (or less) the mosfet cooked itself, it took the resistors and the bridge rectifier, because it took the rectifier, the fuses blew open (maybe even literally, would not be the first time I saw it). That's by far the funniest of faults I've seen in a while. P.S.: The leads thing, try to change the battery, if the problem persists, either your leads are broken, or worse the connectors on your multimeter has issues. Edit: I am surprised that the mosfet still works, I still would not trust it for long term use, also, the bridge rectifier is surely at fault for the fuses, not the blown resistors
Well they had 2 x 0.68 ohm resistors in parallel. Why? You can get 0.33 ohm parts, so the only reason for them to use two chunky resistors in parallel is for the watts rating. Remember, these resistors are going to carry the full switched current on the primary side.
Y'all are flipping your sh*t over the battery. It's brand new and if your brand new battery that says "BMW" on it leaks, then you have the worst luck in the world, like 1 in 5,000,000 bad luck. I've never had a car battery leak out of the box, and one that's never been in service is less likely than ever to leak.
One of my favorite mysterious foults :) my guess on the original foult is dodgy electric outlet. Is the negative in a garage is a bit hit and miss you get 400V and things like battery chargers and dc adapters stops working, ask me how I know 😉 brilliant work ❤❤❤
Bourns is a reputable manufacturer for resistors and trimmers, good choice on the brand. If you want to be scientific about it you can install the new resistors (R680) and then measure the voltage across. That way you can do the math for the current and then you can get the wattage. If the calculated wattage is within the specs of the resistors they'll be fine. If it exceeds it you'll need to get bigger resistors. I don't know the 1206 limitations, probably 0.5Watts is the maximum for that size? You can check a catalogue on one of the large parts suppliers (DigiKey, Mouser etc).
I think Watts matter for resistors. You need to know how much energy will be dissipated before replacing it. If the rated Watts are too low, the resistor will overheat and break.
Im curuous about the stacking resitors on top of each other. Wouldnt power just flow through the path of less resistance. So a stack of them would net the one resistors value ? Not add. Unless in series?
I actually suspect the charger maintainer originally had a higher spec bridge rectifier, MosFets, etc, etc but after some research the manufacturer realised that cheaper components would still actually work for a while, Yes they probably tested them to see how long they lasted and because they lasted at around the end of warranty they chose to use the cheaper components which may need a smaller foot print or different footprint, ergo what looks like a clever design on the PCB that would accommodate various sizes of components, it was Profit grab or at best cost cutting.. Yeah I know~I know I am a sinical little Far Err-Mudder..😁 But I am sticking to why a PCB has this accommodation of different components.. It's all about duh Dosh.. Almost fur gutz.. yup no thermal Paste saves a buck for every 1000 units 🤣😂😅
You should really get your self a Variac transformer. That way you be more saft when plugt in to the mains.. Stay safe. Love your videos. Well safer ;)
Would a crossed polarity on the battery take those resisters out ? I know these things normally have crossed polarity circuit protection built into them, but if that was faulty ??
no as they are on the primary side. it was a solderbridge on the mosfet legs. the short got vaporised in an instant but the arc can carry a lot of current for a long time (in milli seconds) and so the fault propageted until the damage was done. the factory QC failed
how are your and your family do. too day on this Tuesday evening i hope your have Very nice wonderful day Thank your for do this amazing Video that you make 👍 Thank you so much God-bless your and your family
A brave man - car battery on the carpet! Does the wife know? 🤣
He had an iron sitting flat on the carpet a day or two ago!
She'll certainly know now,,, I hope he knows how to remove blood stains from the carpet 😂
@@peterferguson2344 LOL
Service free battery is not often leaking, I would not worry to mutch, but i mabe wrong.
That battery looks brand new
I love the attitude that it's about the diagnosis and not the fix.
Got to hand it to Vince, putting the potentially explodey thing on a massive battery to save the carpet. 😂😂
I am sure it is fine...The acid from the battery base mark in the carpet will disquise the explosion marks.
actually it's really not dangerous at all in the worst case it can only makes a little burn on the top of the battery
😂
😂😂
@@Mymatevince I can't comment direct. I dunno why. It could be that there was a false polarity on the battery clamps.
First impression is that the original shunt resistors were tiny, so they weren't carrying much current.
But here is a more accurate calculation on the wattage of the shunt resistors...
The front label states that output current is 4 amps, which means that the 12v output of the power supply provides 48 watts.
Therefore the input stage of the power supply sinks approx 48 watts (ignoring losses) with an input voltage of 340 volts DC. This means that the pair of shunt resistors is carrying approx 0.15 amps maximum. Let's say 0.2 amps to allow for losses.
Each new shunt resistor of 0.68ohm rated at 0.25W can carry 0.6 amps because I = sqrt(P/R).
Therefore the new resistors are well within specification for this circuit.
By the way, those scorch marks look like a solder blob rolled between the pins of the MOSFET, pulled up the base and dumped a huge current into the shunt resistors. This current draw took out the rectifier at the same time.
But how can it roll between the legs of the mosfet if the middle leg hast a magnetic bead ...
This is an insulating bead, not a magnetic bead. And it looks like something joined the traces from gate to drain, not actually touching the legs themselves, because metal was melted from the traces.
The blob doesn't need to touch the legs of the MOSFET themselves, the burnt traces imply that something joined the Drain trace to the Source trace (not base as I said above).
@@brendan_hassett Never seen an insulation in such an application, seen ferrite beeds multiple times (for EMF puproses).
Excellent! Thank you, brilliant, thanks for sharing your knowledge. So after reading the comments, I'm wondering if it was possible that too much solder was applied to 2 of the MOSFET pads which flowed through the through holes and rested on the underneath of ferrite bead causing a short (solder bridge) Once turned on it shorted causing the scorching and blew the resistors, bridge and fuses. I'm not sure when these things are machine soldered if it is possible for too much solder to be applied or could the same 2 joints have been done twice perhaps????
I thoroughly enjoyed this video. It is like the work of a detective narrowing it down to the culprit. Well done. And very clever. Great video.
Thank you very much 👍👍👍👍
Well done. From yesterday's video, I took apart my failed IR soap dispenser, found out what the problem(s) are, the main which is the pump is like a car oil pump with tiny nylon gears, one of which is cracked, the screws are not SS and corroded, so I 'could' superglue the gear and replace the screws, but all in all, for $10, it is not worth it, but I may use the IR sensor for something later.
Finding the two 0.6 resistors faulty is why I watch, I learn every time. Thank you, Capt. Porcupine Head 🙂
😂Thank you. I'm from a long line of Porcupine Heads!!
When working out what has failed, think of the current path. Probably a small fruit fly or other insect was vapourised as it touched both the drain and source of the MOSFET. As the MOSFET was presumably shorted across drain to source - the drain of the MOSFET gets it's voltage from the primary transformer which in turn is fed by the rectified mains from the bridge rectifier. The current flow goes through the fuse, then the bridge rectifier then the primary of the switching transformer then the drain of the MOSFET to the source and then to ground through the resistors that attach the source of the MOSFET to ground (which I failed to notice until you actually saw they looked burnt).
It's amazing how tolerant these circuits are when you think. The new bridge rectifier was well within spec to deal with the current and voltage it has to handle. But the new MOSFET you used was 100v lower than the original. Although you might think the voltage will get nowhere near this level, when looking at the output (load) of a switching circuit you would be surprised at what the voltage can peak to on the drain of the MOSFET. You will probably have around 300v or more going down to almost zero at whatever the switching frequency is. That's a lot of stress. If the MOSFET was left permanently conducting at the maximum current when it is switched on it would blow apart in no time. In fact that is what usually happens when the drive circuit fails. So I think they design a circuit to be running well inside the tolerances of the components used (or at least they should do). Those resistors you have used measure 0.5 Ohms so should be ok to leave in. Your only slightly limiting the current by creating a slightly larger source to ground voltage drop. If anything the MOSFET will have a slightly easier time of it than before - but really there's no discernible difference. The resistors are there to act as a fuse when the current exceeds a certain level (which the original resistors proved).
I think Vince is always thinking of future content, "How to clean/replace carpet, How to treat acid burns, How to fix up the shed to live in when you are banned from the house, etc."
😂😂😂😂😂
As the MOSFET solder pads showed damage and device as a whole looked quite new, maybe there was a solder bridge across two of the MOSFET pads (manufacturing defect) which when power was applied blew the resistors, diode bridge and fuses? The MOSFET itself was OK because the pads, solder bridge and other components soaked up all the energy before the fuses blew.
Good thinking, thank you Green Jim. Maybe the solder came through the through holes and sat on the ferrite bead which shorted the legs!!!!! I think you have it, well done! 👍👍👍👍
Yes, that is what I thought. A solder blob bridging the mosfet pads would cause that kind of damage
that feeling when you finally finish a difficult fix love to see you still get massively excited after all this time you can see your enthusiasm in every video you make vince and thats why i watch them all keep enjoying yourself mate because we enjoy watching
Probably one of my favourite MMV videos here, really enjoyable with the right amount of tension, putting the thing that went bang on a car battery XD!
Thanks Craig, I wasn't sure if it was a little long with a lot of back and forths, but so far the viewers have appeared to have liked it at 98% thumbs up. Cheers mate 👍👍👍
44:03. Bridge rectifiers usually dont cause any Components on the output to fail and also that big smoothing cap at the output was still intact which tells me it was something at the output of the br and that blew the br causing the other fuses to blow. If the br was to fail, that mains cap would probably also have exploded and there would be even more damage due to the fact that you feed unfiltered ac into a dc circuit.
It was either a solder bridge or a tin splash or something shorting the mosfet. And due to the unknown thing shorting the mosfets pins, it did not take out the mosfet itself but rather the resistors pulling down the pin of the mosfet to gnd and everything at the output of the bridge rectifier and the fuses
Agreed. For the resistors the higher the wattage the better (lower chance of overheating and self destructing ) .
13:52 i wasn't really paying close attention and when i looked back i thought he propped the board up with a banana 🤣
banana for scale
LTT screwdriver?
😂😂😂
A POWERful video today vince. I found that ELECTRIFYING and very ENERGIZING. I am impressed that you don't CHARGE for these repairs and simply keep your mind TOPPED UP.
I'll grab my coat.................
Your comment is lightning fast and you are so posetive thank you ❤
Ohm-y God that was awful...
@@SWATStrachan Thank you. It was shocking I agree.
@@andersmmvfc.8376 You are too kind. I was just being a bright spark
Factory solder bridge on the mosfet looks fav to me....Great series Vince, looking forward to every evening this month🥳🥳🥳
Excellent, thank you. I now think you are spot on with the cause of failure 👍
Regarding you mentioning the 240v into the bridge rectifier, remember thats RMS not peak voltage. So the bridge must be able to take the peak of 340v. You did use a replacement that is well in spec but worth a mention. Back in the day I saw someone use a 300v rated bridge on mains and it blew instantly.
Already thinking and forgetting what i did with my life before a daily 30+ minute MMV video came out every evening. Gonna miss this daily routine when christmas is over
Hahaha, I won't Lee, I'm already done in😂 By day 12 I'll be fixing loose screws in the door handles and by day 20 I'll be that desperate I'll be changing the smoke alarm batteries for easy fixes😂
Only kidding, glad you're enjoying them, I am too at the moment👍
YEEEEEASSSSSSSSS!
Nice fix!!!!
If I were a manufacturer and I could get away with 1/4 watt vs 1 watt resistor and save a 1 penny each....
That's all I'm going to say.
CONGRATULATIONS! What an awesome, satisfying fix!
Very interesting and instructional. Thanks a lot for sharing this. Happy First Advent.
Good fault finding, congrats!
2 Points:
I am sure the Fault came from an Residue or something on the Legs of the switching Mosfet.
Be aware to remove all of the metal which gots evaporated over the Mosfet and the isolating Ferrule! (think about applying Soldermask on the Ferrule and the lower Part of the Mosfet).
Second: Never connect a Charger to the Batt at first.
Save Way is to connect it to Mains and the Battery afterwards.
All the Best, and thank you for your Videos. 😊
For the resistor power rating...
The resistance value will set how hot it will get for a given application (voltage/current).
The power rating is how it can sustain that heat. 0.5W can take heat for longer without changing its characteristics.
However... you can ameliorate this by wicking the heat away into the copper more effectively. When you solder the 0.25W parts in, be generous with the solder and inspect how much copper is around them.
The power rating must be used in conjunction with the copper area to calculate the temperature and, therefore, the stress on the part. It's very difficult to do that, so just lean heavily on the solder.
If you really want to go to town, solder in stubs of copper wire sticking up out of the pads to act as a small heatsink. Make sure the wire stubs are stiff enough to not bend over and touch something else.
I have done something similar when upgrading a Wii Remote Sensor Bar to run off USB and shine brighter for use further away from the TV using ultra bright IR LEDs diven at 200mA each.
Very interesting, they are connected to a via that the heatsink is soldered onto so hopefully that alone with wick away lots of heat. I will happily put massive blobs of solder on though to help with the mass. Thank you! 👍👍👍
I knew that replacing the bridge wouldn't fix it. There was still 30 mins left of the video. 😀
Don't let him know that, or he'll go all sneaky with multi-part videos or something!
Vince, this was great entertainment. Just love the do or die tests.😂
Thank you Wayne 😎👍
loooved this repair. I watched it while walking on my walking pad :D. And its so funny how you said in the beginning " it probably cant be fixed ".
Thats so funny. You say that like 80% of the time and then proceed to fixing it anyways :). Well done Vince. I appreciate your efforts in making these videos. I know its quite some work.
Brilliant fix on this one!!! If those resistors cannot handle the current, they would burrn up open circuit very quickly! Just test it for a few hours, measure them - if they look OK and still measure OK, they will be fine.
Great job, Vince, as usual both entertaining and informative!!!! Happy Holidays to you all and I look forward the the next Advent episode 🎄🎆
Dear Vince,
It is never a bad idea to use thermal paste when reassembling the transistor, it can only improve the thermal transfer. The black tube that was over the middle pin of the MOSFET is not an insulator but a ferrite bead. Usually their primary use is it to kill RF oscillations... A ferrite bead is a passive device that filters high frequency noise energy over a broad frequency range. It becomes resistive over its intended frequency range and dissipates the noise energy in the form of heat.
Thank you Patrick, I thought it might have been for unwanted noise when I was editing the video, but I wasn't sure. Thanks for clearing that up 👍
Nice fix Vince, that burning on the mosfet was definitely arc flash, but not from any component failure it could be a bug was in there and went across the legs of the mosfet, spiders and alike can cause such damage.
The next video will be Vince fixing his acid burnt carpet
😂
Another stellar video, happy holidays 👍
Thank you Paul, Happy Holidays to you and your family too 👍😎
Nice one mate, well done on fixing that one 🙂👍
Red to Red, Black to Black, flick the switch and stand well back 😀
It seems the short was going through the drain to source which means absolutely something bypassed the mosfet damaging the bridge rectifier and the shunts probably an ant or something that got evaporated considering the burn mark!
Worth watching just to see how excited you got, very entertaining Vince.
Vince, Try using some of your DeOxit on your meter lead connectors. That could be the source of your high ohm reading when you short out your probes.
Faulty irons, broken kettles, soap dispenser that doesn't dispense and now a dirty old battery on the living room carpet. I'll bet your wife loves you some days Vince🤣🤣
Dirty old battery ?? It looks brand new
That battery looks clean and new.
Ur the third person to shout about the battery on the floor when that battery looks cleaner than most people's shoe soles.
Calm down, drink some tea or something.
That battery looks like it just left the shop brand new.
If Vince's wife was going to find a reason to flay him alive it would have happened by now already.
Nice job Vince 🙂 When connecting a battery it's a positive thing (+ first) / when disconnecting it's a negative thing (- first) on a modern vehicle (with a -ve ground) is how I remember it.
Isn't that because of the chance of touching the car body with your spanner? With the negative disconnected you are safe to connect the positive and touch the spanner to the chassis, because at that point the chassis is floating, but the other way round, chassis already connected to battery negative means spanner touching from battery positive to chassis is bad. I don't think the battery cares which connection first, it's just about where you could short and get into trouble.
Enjoying these fixes, look forward to the next 21 episodes 👍
I am looking forward to the next 21,000 episodes 🙂
Vince - I think you best start thinking how to celebrate your 1 million subscribers! 2025 will be the year I think and this advent marathon has to boost it no end.
Thank you Christopher, it would be nice, but unfortunately my channel has sort of stalled the last couple of years, I think I get around 30 subs a day so realistically the next 100k will take an eternity to realise, but you never know with RUclips, strange things can sometimes happen. As long as I get to continue for as long as I can, I'll be as happy as I can be 💪😎👍
Persistence rewarded 😀👍
Regards the meter REL button … according to the EEVBlog page the majority said it did not work (some suggested it only worked on some ranges but didn’t specify). I think meters are meant to come with a few missing functions or you wouldn’t need to buy another one.
You almost gave up and finally yess!! Great result Vince. A video every day is a lot of work, thanks for that. and keep up the good work! 👌
Thank you Marcel 😎
Excellent fix Vince.. i really enjoyed it. A couple of faults and a victory 🎉
Thanking you kindly 👍👍
37:25 Clean the banana contacts on your meter with some Q-tips and IPA, and do the same with the plugs on your testleads.
That should get you ohms readings down well below 1/2 an ohm where they should be.
Love it vince looking forward to all your content
Actually, I would bet that there was a foil bridge on the top side of the board that burned off.. That's a manufacturing fault in the PCB at the start. that would hang the 330V right across the "switch" and while the foil fault was corrected. it took out the rectifier and fuses as well.. along with the sensing resistors down stream, that saw a LOT more than the 12 or so volts the sense network was designed to take.
This was a real good fix! Thanks.
Whenever you find very low value, large size resistors, they are usually current sense resistors. If you double the resistance, the controller will think the current is twice what it actually is and probably reduce the max output current from 4A to 2A. So they should never be replaced with smaller ones as that could cause excessive current.
In regards to the power rating, it is hard to know the required power of these. We know the 4A output current but these resistors are on other side of the transformer and we don't know its turn ratio.
Anyway, just get the highest power ones with the given footprint.
Thank you Martin 👍👍
Enjoying this series, nice fix.
This is the best chanel on youtube! Evereybody wants a mate like Vince!! 🥰🥰
Excellent video Vince 👍
I only just noticed that your mat has a perfect spot for your eye loop.
Nice one Vincey boy, keep em coming
Seen you've ticked over 900k now? Well done 👏
I am one of them today, this series is good really enjoying it so thought it was about time I showed some support been watching for quite sometime now.
Its a dim bulb tester you need
Hey, congrats on the 900K 💪
Can you try turning your meter off and then holding the leads together and turn it back on to the ohms position and see if that zeros it?
Glad you stuck with it and fixed that. Good job.
“Positive first, then negative. When disconnecting the cables from the old battery, disconnect the negative first, then the positive. Connect the new battery in the reverse order, positive then negative.” That way if you slip and your wrench hits a negative surface there is no short.
41:55 The resistors are on the other side of the board so it cant be them leaving that mark on the MosFet.
I suspect there could have been a small solderbridge there left from the manufacturing that caused it to short,
and blow up the two resistors. They are supposed to act like fusible resistors anyway.
Well spotted 👍
Epic - great work Vince mate
Solid gold tweezers, Rolls Royce... 👑
Your Fixathon is great way to end the evening 🔥
when working on anything live i would suggest using leads that are GS38 compliant. even if you have a heat shrink for the tips to slide over. It would save a few unwanted issues if you accidentally touched leads whilst testing.
Happy Christmas to you and your loved ones Vince mate i hope you all have a good one.
Nicely done Vincent.
Another great video! I suspect there was a bit of conducting crud shorted across the Triac legs that blew out the components.
Another nice repair Vince.😊
Nice repair! About connecting positive terminal first, does that only apply to cars or any circuitry?
Ring the cheap automotive stuff, not to be confused with Ring, the expensive doorbell 😂
Could they be working as shunt resistors to measure the current going to the battery? You can use a big external resistor. Sometimes they use SMD cause it is easy to manufacture bu robots but external resistors can work better and are way easy to find locally.
yes shunt resistors - but not for the current going to the battery but for the current going to the transformer. the chip needs to know this to get the switching timing right and/or detect overload conditions
Great Work. Thank you for that Video.
Agree. Rectifier failure caused all the problems.
My Aneng 870 looks the same as your meter and pressing the REL button should zero the display. After that all readings are relative to that stored value, i.e. from the new zero point. Don't know why yours didn't work, maybe need to get your probes to show a stable value first (you leave it on REL after, by the way, don't press the button again).
I bought some gold plated, really sharp probes from Ali Ex for 1.70 and they are good enough. In fact, I bought 2 pairs. Something to add to the order next time you order from China.
For the resistors, measure the voltage drop over the resistors under load (while charging). If the charging current is 4A the load should be at it's maximum. *I'm assuming here that the resistor load is directly related to the charging load, this may not be the case.*
From ohm's law & definition of electrical power, we get >> v * v/r = w
For 1 watt per resistor, meaning 2 watts in total, the max voltage drop accros the resistors should be ≈ 0.825 v.
If you're not charging at maximum current multiply the wattage value accordingly, for instance charging at 1A while max is 4A multiply by 4.
If the resistors are measuring the full charge current directly (simple solution, but bad circuit design) they should have a rating of at least 2.5 watt or higher.
I've worked on a couple CRT TVs and none of the MOSFETS (or voltage regulators) had thermal paste on them, yet they worked well. So you're okay there.
I think that ferrite bead on the centre pin of the MosFet trapped some moisture,
and that Moisture took out those sacrificial resistors and oh~wow! they actually
saved the MosFet, ergo saving a buck..
The small tube around the drain pin of the mosfet is actually a ferrite bead, and they are conductive, it’s too big for that application, it should have been much thinner, the thing is, it caused a short and sparked across drain and source, all the current of the collapsing magnetic field in the primary side of the transformer returned to the rectifier and shorted at least one of its diodes, and blew the current sensing resistors for the driver. Really low quality manufacturing.
Thanks for the break down of the fault/info John👍👍👍
Youre amazing Vince...
If you can get 1206 size resistors I would go with the highest wattage you can, the design engineer did it this way (using 2) to spread the current between the 2 resistors instead off just using one so half the power through the resistors. This is difficult to advise without looking but you will probably be ok but perhaps the first time you use it proper put your hand on the case every now and then although you have to allow for the MOSFET. I think 0.5 watt is as big as it gets anyway in 1206.
The ic is probably a SG6848T or an OB2263. (Sot23-6 housing)
Pin 3 of this ic is probably connected to ground via a 100k resistor which sets the switching frequency at around 65-70kHz
The philips ORT7500 (red retro radio) uses a nearly identical primary side of the power supply, and in these radios the 6 pin ic fail ALOT!!.
Don’t use mosfets with a lower Vds than the original, the DC voltage on the drain of the mosfet might ’only’ be 325V, but because it is switching an inductive load (the transformer) the pulses are way higher, around 600Vp-p, something to consider!
I think the cause has got something to do with the ferrite bead around the drain, it’s to big and probably they fixated it with some glue or other crap and with high frequency and high voltage pulses on it likes to spark to the source pin, which is virtually ground, current sense resitors go up in smoke and the current peak also took out the bridge rectifier and the fuses.
Hope you put the Ferrite Bead (noise filtering) back on the Mosfet? Good find on the resistors BTW
I didn't Standish!!! I did show that in the video and also ask the question if it was put there for noise reasons to stop interference between the legs, BUT then it got cut out because of the length of the video. I only realised I didn't put it in after soldering in the last anchor point and I couldn't face unsoldering it all again. I still have it so I can put it in when I solder in the resistors as the pain of unsoldering the MOSFET would have eased by then 😂 Is it for interference reasons do you know? Thank you, I haven't heard from you in a while, hope you are doing good 👍
@@Mymatevince The use of ferrite beads is quite a complex issue and can have a number of effects (beyond my knowledge pay grade😊) - but I trust the designers of this unit knew what they were doing and aimed to reduce either the effects of noise, interference or Mosfet turn-on speed. Thanks for asking about me Vince. I'm OK, just been quite busy and having a few health issues.
Does UK got fuse on the plug and also breaker box? If yes, that's nice. Double protection
the wattage depends on the use case. it looks to me like its in series with the load leg of the mosfet - plus theres 2 in parallel already by design - i would go with same wattage or higher - definitely not lower.
They will drop 120-150mW under full load as they are - they will be fine.
I can't believe the factory forgot the thermal paste. Here's my best guess on what happened: someone tried to use it, it turned on, after a couple of minutes (or less) the mosfet cooked itself, it took the resistors and the bridge rectifier, because it took the rectifier, the fuses blew open (maybe even literally, would not be the first time I saw it).
That's by far the funniest of faults I've seen in a while.
P.S.: The leads thing, try to change the battery, if the problem persists, either your leads are broken, or worse the connectors on your multimeter has issues.
Edit: I am surprised that the mosfet still works, I still would not trust it for long term use, also, the bridge rectifier is surely at fault for the fuses, not the blown resistors
It is not uncommon to have no Paste - it depends on Design and Cost Calculation
Thermal paste is not needed in this case - likely not even the heat sink. It is only a 40W charger - the moseft will only have to dissipate ~2W.
Well they had 2 x 0.68 ohm resistors in parallel. Why? You can get 0.33 ohm parts, so the only reason for them to use two chunky resistors in parallel is for the watts rating. Remember, these resistors are going to carry the full switched current on the primary side.
Well done vince 👏👏👏👏👍
have you ever considered doing a photo search check for items. i find this actually works for me whenever nothing matches by basic text search vince?
Quick tip - dont put an acid battery on a surface without a tray as acid seepage coukd mark the surface. I know from bitter experience
If the battery has no cracks, it should not leak acid, and it is even less likely with an AGM battery.
Y'all are flipping your sh*t over the battery.
It's brand new and if your brand new battery that says "BMW" on it leaks, then you have the worst luck in the world, like 1 in 5,000,000 bad luck.
I've never had a car battery leak out of the box, and one that's never been in service is less likely than ever to leak.
@@Boogie_the_catThe battery is not brand new. Quite the opposite, it is exactly 12 years old. Since it is an AGM battery, lt won't spill anyway.
One of my favorite mysterious foults :) my guess on the original foult is dodgy electric outlet. Is the negative in a garage is a bit hit and miss you get 400V and things like battery chargers and dc adapters stops working, ask me how I know 😉 brilliant work ❤❤❤
Bourns is a reputable manufacturer for resistors and trimmers, good choice on the brand.
If you want to be scientific about it you can install the new resistors (R680) and then measure the voltage across. That way you can do the math for the current and then you can get the wattage. If the calculated wattage is within the specs of the resistors they'll be fine. If it exceeds it you'll need to get bigger resistors.
I don't know the 1206 limitations, probably 0.5Watts is the maximum for that size? You can check a catalogue on one of the large parts suppliers (DigiKey, Mouser etc).
Vince, After the spark did you check that both the internal and plug fuses were still good?
If you heated the Heat sink with the hot air station, would it be easier to remove the MOSFET?
Hey Vince did you check out the latest EEVBlog Video 1655 about how to safely discharge mains voltage capacitors using your multimeter?
something poetic about the "QC passed" sticker plastered right on the board lol
I think Watts matter for resistors. You need to know how much energy will be dissipated before replacing it. If the rated Watts are too low, the resistor will overheat and break.
Im curuous about the stacking resitors on top of each other. Wouldnt power just flow through the path of less resistance. So a stack of them would net the one resistors value ? Not add. Unless in series?
yes .. and this was the goal. bundle up 4 resistors with a higher value to create one with a lesser resistance
Could the leads have been connected the wrong way round? If the charger has no protection it would cause damage.
Well done fella. Down to Chinese crapp again😊😊😊
I actually suspect the charger maintainer originally had a higher spec bridge rectifier,
MosFets, etc, etc but after some research the manufacturer realised that cheaper
components would still actually work for a while, Yes they probably tested them to
see how long they lasted and because they lasted at around the end of warranty they
chose to use the cheaper components which may need a smaller foot print or
different footprint, ergo what looks like a clever design on the PCB that would
accommodate various sizes of components, it was Profit grab or at best cost cutting..
Yeah I know~I know I am a sinical little Far Err-Mudder..😁 But I am sticking to why
a PCB has this accommodation of different components..
It's all about duh Dosh..
Almost fur gutz.. yup no thermal Paste saves a buck for every 1000 units 🤣😂😅
Cheers Koala 👍
Hi Vince. What is the make and model of the phone you use to look things up? Thanks.
You should really get your self a Variac transformer. That way you be more saft when plugt in to the mains.. Stay safe. Love your videos.
Well safer ;)
Would a crossed polarity on the battery take those resisters out ? I know these things normally have crossed polarity circuit protection built into them, but if that was faulty ??
no as they are on the primary side. it was a solderbridge on the mosfet legs. the short got vaporised in an instant but the arc can carry a lot of current for a long time (in milli seconds) and so the fault propageted until the damage was done. the factory QC failed
how are your and your family do. too day on this Tuesday evening i hope your have Very nice wonderful day Thank your for do this amazing Video that you make 👍 Thank you so much God-bless your and your family