I love these types of videos, helps people get an idea of how legitimate electronics repair goes. Will also help right to repair a teensy bit. Thank you for this Clive!
I consider Mitsumi a quality manufacturer, had lots of 3,5" floppy drives, never failed. Could be dirty with dust bunnies, still wouldn't misalign heads. This PSU looks to be quite tidy, too.
The best keyboard I ever had was a Mitsumi. Cost £3, only bought it to pad an order out for free p&p. Nice and clicky despite not being a proper key switch. Reliable, light, nice typing angle.
@@Zadster I think i used one of those for 10 years, until repairs kept piling on and it no longer felt great. But nowadays, i'm exceptionally happy with a generic Chinese mechanical that i found used years ago, this is the best i've had so far. But they're also the manufacturing company behind Nintendo's products, also contributing some of their engineering. Gamecube, Wii era stuff was so good! Then again the word is they're also the ones manufacturing Nintendo Switch, which retains its glaring controller drift issue, though ultimately it's an engineering issue and not materials or assembly quality issue, and ultimately Nintendo isn't doing anything about it, so that's pretty telling.
The choke / capacitor filter at the front end is more to protect the mains from radiated HF interference. Those HF switching PSUs can be pretty electrically "noisy" - and most of the Compliance Authorities (e.g. FCC) now mandate such suppression.
That's why the Alibaba $19 gaming computer power supplies are the best. One or two of them can corrupt an entire MRI machine in the same building with their electrical interference.
@6GDC You don't mean computer PSU surely? It bears reading the specs carefully, if it's a programmable power supply, they will have voltage dependent maximum current. 40A at 3V output maybe? Still a massive massive accomplishment, IF they did it, but it can be done, but it will be rare to see anything with more than 10A outputs. I don't know the product you're speaking of, but by far not all impressively specced and aggressively priced Ali products are scam junk. Ali computer PSUs that the dollar guy speaks of, are just salvage junk or copies of old ATX1.3 computer PSUs, with maybe a power handling of 200W or so that rough order of magnitude, just made to look like modern PSUs on the outside.
@6GDC How feasible a pocket-size 40A switch-mode power supply is depends on a few things, such as: how large "pocket size" is; the input & output voltages/voltage ranges; whether they bothered with EMC (needing space for filtering components); etc. Modern MOSFETs have extremely low on-state resistances (the "R_DS(ON)" value¹, i.e. the resistance, in Ω, between the Drain and Source pins at a specified Gate voltage). Even very cheap MOSFETs can have R_DS(ON) ratings well below 10 mΩ, and this results in very high current ratings, especially for the size. Lower resistance means a lower voltage drop, less resistive heating, and thus more current before things get out of hand (I'm kinda guessing here, but it makes sense). Clive's shown a couple of devices on this channel containing SOT-23 MOSFETs capable of several amps of continuous current - and SOT-23 devices are only 2.9×1.3 mm (top surface; Wikipedia doesn't list the height of SOT-23). The "USB power bank that [intentionally] gets very hot" video comes to mind. It used one of those SOT-23 MOSFETs to switch its heaters, IIRC. There's also the recent emergence and popularity of gallium nitride MOSFETs, which claim higher voltage ratings, higher efficiency, smaller overall device size, and other benefits. They've driven the recent influx of "GaN" SMPSs in the consumer space - e.g. USB power supplies and phone chargers, especially high-power USB-PD supplies. Those devices can supposedly be made smaller and lighter with GaN MOSFETs, as well as being more energy efficient, in theory. (No doubt some of those devices are just slapping "GaN" on the packaging/product description as a marketing lie, but some are probably genuine!) Further reading: en.wikipedia.org/wiki/Gallium_nitride#Transistors_and_Power_ICs (and the rest of the article for more general info about GaN tech) So your "40A" SMPS _could_ be genuine! It's not totally unrealistic, IMO. ...but it could also very well be a wild exaggeration! 😁 ¹ R_DS(ON) is typically written as "R" followed by "DS(ON)" in subscript (no "_"), but I can't use subscript on TH-cam - at least not without obscure Unicode characters that I don't have easy access to on my phone - so I'm using "_" to indicate subscript.
@6GDC NOT 40A at 170V, almost 7kVA, you can be sure. Read between the lines - maybe 40A at the lower end and use that calculated wattage to work out what you will get at the higher end. That's how Shenzhen do. (mebee one or two amps if you are lucky?) Still respectable but not 7kVA.
Had the engineering manager where I used to work try to discharge the caps on a 600Kw vsd, with the "Short it out" technique he had heard about. Luckily he was not injured, but it "Brightened" his day.
Thank you Clive! This looks like a fault I had recently in a Peavey mixer. Another chip had visibly blown on the PSU, which I replaced. Now the desk pulses on and off just as you described. I’ll dig it back out and have another go at fixing it now!
I have a bass amp that has dual secondaries +-15 and +-45 the 15v neg rail failed and was doing what you describe but after replacing caps there was some huge charge then a bang across the -15v rail the windings are now open on the secondary, smps are great but if you lose the transformer it's game over
Nippon Chemicon capacitors are generally a sign of someone not cheaping out, so reinforces the "common sense and safety" comment in the description :) (That big, painful 450V capacitor prominently showed the Nippon Chemicon logo - they only include the company name inside the logo in _really big_ capacitors - and the large capacitor on the secondary side was in the same style, even though the logo was hidden.)
I re-capped my DBX 286A microphone processor with all new Nichicon caps. It wasn't a cheap prospect. I don't regret it. I got it faulty from Ebay for $49. It's fully functional now and quite impressive. The only two I did NOT replace were 2 bi-polar caps that I had not prepared for.
this kind of repair is the god of all repairs, these days, electronic technicians are not the purest kind.. i remember a friend of mine who works with radar system in middle east, he travels even during wee hours for sometimes 1000 km just to fix a radar. i saw him fix my laptop by cutting some IC connections and jumping it from somewhere with several wires, insulate it and bingo, it works. people like clive and my friend are few among millions who does a work with electronics by just smelling it. salute to you clive. nice video and thank you for sharing it with all walks of life.
A long time ago, someone thought, "Let's take the AC mains, rectify it into DC, connect s coil with a DCR of 0.05 ohms to it, and then short it to ground with a fragile mosfet." That man ended up in an asylum, but danged if it didn't work.... It has always been my theory that switching supplies work because of voodoo.
Actually, this exact thing took place over a hundred years ago when Henry Ford put the ignition coils in the earliest cars. Then along came Kettering with the coil, 'condenser' and points to repeat the same thing. Now we have chips that do it in microseconds. 👍👍
@@acmefixer1 That's true. And the "flyback" circuits in CRT TV's and video monitors work similarly. Still, it's kind of amazing that these circuits are so reliable if built correctly. On paper, it seems like they'd be doomed to a fiery death in the first second of operation.
You forgot the HF chopper. In the 80's I fixed dozens of SMPS' in commercial equipment (third party IBM network muxes). They used all discrete components back then. It was always the output electrolytics that bore the brunt. They ran HOT. Times sure have changed.
@@ScottGrammer That's because you're thinking of this like it's a DC circuit. It isn't. You're not thinking about the inductance involved. When the mosfet first turns on, there's not a dead short across the supply. The coil is nearly open at that point due to its inductance. It gradually starts conducting more and more current until the mosfet turns off, sometime before it becomes a dead short across the supply.
@@joshuabest100 In a test like this it is much better to replace your fuse with say a 40w or 60w incandescent light bulb, if it comes on the device is still shorted, but will not explode because there is not enough power available for blowing stuff up. For a TH-cam channel though it is better without it: because you will have sparks and flying shrapnel which is always cool! :D
Would love to see more general troubleshooting and repair videos. They are few and far in between on your channel and you’re a very good teacher and wonderful at explaining things!
I have no electric or electronic background,but I am fascinated by your videos and pretend to understand everything you explain. I wish I had studied electronics and a teacher like you would make the complicated circuit boards easy to understand. But thanks anyway, your understanding and hardworking in your videos is greatly appreciated.
If you want to remove a component and need to pull it gently to remove it there is an easy trick I use. Grab a pair of forceps and clip them onto the resistor, mosfet etc and then hold the board so they hang, once you heat the pads up the weight will pull gently for you. An added bonus for exposed metal components is it helps reduce the heat in the parts. Sorta like the opposite to using bluetack to hold parts in place while you solder them in place. Forceps are an excellent tool for the toolkit for all sorts of little tricks. Very handy and usually a great cheap addition to the toolkit.
I work on industrial battery chargers, we have dc capacitors we run around 680vdc. The day I found a bad bleed down resister was very fun. Yes I reached in like a dumb a$$ to change the igbt.... That was the day I learned to ALWAYS double check the capacitors .. Be safe Clive .
A few decades ago, I was working on (diagnosing) a single-triode power ultrasonic oscillator at the base of an ultrasonic cleaning machine. Went to measure the anode volts, should've been between 5 and 10kV. They were, and my high-voltage probe failed, and punched a hole through my thumb. Very thin hole, but a hole, nonetheless!
I would add to the "proper repair" while in there I would also replace the small bulk cap for the chip VCC. Those little things tend to just wear out over time and cause all kinds of problems, not worth saving 10 cents and a minute of time to find out there are more problems about to go off. A pretty nicely designed unit this one or perhaps not much used, often there is distinct heat discoloration and dirt on the boards.
Yes! So often I find the psu pulsing or not starting up because of that little chip/switcher start cap has dropped from say 10u to 4 or 5u. The worse the power quality or noisier the environment (industrial) the shorter the life span of the caps seems to be.
I always love your videos, but have to admit it was good to see one that was not an ozone / ion generator related one thrown in. Always good to see a bit of diagnostics and repair :).
Wow! Thank you. This is the exact problem I have with a similar power supply that powers a set of mirrored led lights. PSU led onboard is pulsing, witch causes the lights in the unit to pulse. The caps look fine, and I was suspecting the diode, but wasn't sure how to test. Guess what? I followed your video and yes definitely the diode! I'm going to hit up our sparky at work to see if I can pick up a freebie. Hopefully I will have another working DMX fixture to add to my program 😃
Much rather too much Flux then not enough.. I'm the same way when doing electronics specially pcbs cause I'm always worrying that I may burn up the paths
@@plageran For the battery, smoke detector batteries. Energizer lithium for the win. They're not the cheapest (tho they've come down a bit) but worth every penny.
@@plageran Haha We don't have Poundland here but I'm gonna go out on a limb and say it's similar to "Dollar tree" here... If I'm right, you'd be absolutely correct! 🤣
This will be helpful next time I repair a PSU that I actually need to diagnose - I've fixed a few PC monitor PSU's recently, and they all died the same way: cheap no-name caps on the secondary side failing (in very visibly obvious ways). It's funny because the board that deals with inputs, driving the LCD, etc usually has good caps on it, but the PSU sitting right next to it is full of garbage (and in the last PSU I fixed, also full of vacant spots a cap was *supposed* to occupy, and now does).
Very big coincidence! I got into something identical today... my LED magnifier desk lamp died several years ago and I finally got fed up with using a flashlight ('torch') to trim my nails. Hauled the thing to work (extender arm etc) and took it apart... found a board almost identical to the Mitsumi, albeit pure Chinesium. The chip was a 1202 (EDIT: Innuovo XN1203) using the same setup... the opto bridged a slot in the circuit board to give extra isolation. I tested the electrolytics, they were iffy as ChongX caps normally are, but the failure was in the input filter stage (just before the big cap)... the 1.5mH pass inductor was open. New parts on the way from Mouser, soon I will have my magnifier lamp again! Very glad to be able to get the PDF of the Chinese chip, the example circuit diagrams didn't need translation thank goodness.
Even just AM radio is basically unusable in my apartment block because at least one household uses these shitty PowerLAN/DLAN ethernet-over-powerlines adapters, turning all the electrical wiring into a giant antenna transmitting a f*ckton of noise/interference (think cheap chinese switchmode PSU with barely any filtering, but on steroids)
@@Knaeckebrotsaege I'm using these every now and then for easy LAN connection across my apartment. Say, someone would like to catch the noise with their SW defined radio, at which frequency range would they have to look? I'm too lazy to do my own research. :-/ Asking for a friend.
This is a flyback converter. They are notoriously hard on the capacitors and the output diode because of the trapezoidal current waveforms they produce. Since the output of the flyback only receives energy for part of each switching cycle it means that the currents must be quite large so that the average current (the DC load current) can be met. So if the output is, say, 1A DC, chances are pretty good that the peak output currents are on the order of 3-4 amps. That is tough on the diode and the output filter capacitor. Usually the component that fails on a flyback is the blocking diode, as happened here.
Your caution with the capacitor reminds me of my college days in the 70s. It was all valves with lots of volts available on the bench. We used to get big electrolytic caps, extend the connections with bare wire and tape it down each side of the cap. We would then charge it to 200 volts and toss it across the lab shouting 'catch' . The recipient got quite a surprise 😂 Oh the folly of youth.
I have similar stories of that period, in the testing department of a PABX manufacturer and students at the electronics lab of a local school kept doing that all the time, leaving charged capacitors over the benches, waiting for a "sucker".
These are used on a high end Song blu ray player and fail on a regular basis. The biggest offender is the double diode on the secondary. It’s more than adequate for the job especially as the diodes are wired in parallel so I suspect a bad batch.
So here's a weird thing that happened in my mind: 9:27 I noticed you said live, but not the requisite "at mains voltage". So in my mind I imagined a radio broadcast where an announcer with a really bassy voice said "Big Clive. Live - at mains voltage!"
An interesting and very practical display of skill made understandable with the voice over. This guy should be selling'EASY ELECTRONIC REPAIRS,lessons by post.He is really great,showing and telling. Thank you. Terry Offord
Flux on the de-soldering braid, awesome piece of knowledge, you may have just saved my oscilloscope kit that I ordered to build for practice soldering.
I pronounce it sodder. Actually it depends, sometimes the L finds its way in but won’t be complete and will sound like sowder. But yeah, if guys like clive and paul Carlton weren’t around, I would really have to say it like them exactly to get my dose of soLder
@kasperabigail9129 some have notice that I pronounce it differently depending on time of day. I'll say solder in the morning, sowder in after noon more and turns into sodder at night.
For 10 years ago, my mom by new DVD player. It broke few days after she used in first time. The capacitor was dry out. The store there DVD come from told her by new because it will be too expensive to fix. I take the player with me too my work and change this capacitor. DVD player is fully active 10 years later. I fixed player for 100 isl. Krona. New DVD player cost about 30.000 isl. Krona.
I had an IBM PS/1 PSU with exactly the same fault. Gave it a brand new diode, a complete recapping for good measure. Now just have to finish repairing the floppy drive bezel, which is apparently made of tortilla chips with too much dip on them, because it cracks every time I try to reassemble it. Pretty soon it'll be made of all-new plastic, though!
I hear the words "dead short" and I immediately think "capacitor" too. Next, I postulate, what else has this capacitor, if found, (alternative, shorted tracks?), taken out? A short (or even a "medium"?) does amusing things, elsewhere on board, sometimes. Diodes are a natural, but it can tickle other parts too, transformers, for one. Sometimes all! Also, have you ever noticed on odd (very?) occasions where restoring with a new component somewhere places stress, renewed or new, on another component, causing it to fail afterwards? An "It tested okay before, isolated, why is it shot now", kind'a thing?
On page 12 of the datasheet there's a demonstration board schematic, which might be much closer to the board you're looking at. There's an interesting detail about the diode on the suppression network on the primary, the datasheet calls for a 1N4007 and not a fast recovery diode as used elsewhere on the design. One possible explanation may be that using a "slow" diode allows some of the spike current to flow back, marginally improving efficiency.
4:49 I prefer inertial desoldering - heat the pads and whack the PCB edge-on on the bench top. most of the solder will splash off. Small pcbs can just be dropped on the floor
I owned an commercial radio communications business for 40 years (think police/fire/taxi radios) and sold/serviced alot of power supplies over the years. We regularly repaired linear power supplies, but switching PS's are a different beast. We soon learned it was much better to replace a faulty SMPS than to repair it. Once you repair a piece of electronics the customer believes any further failures are your fault and should be covered by your repair warranty. They are also much cheaper to replace than repair these days... unless this is for your own equipment.... but still might not be cost effective to self repair!
13 people wanted you to discharge that cap with your fingers I guess, great vid! I have no clue where to start looking for faults if it's not black or crispy thanks Clive!
I love watching your videos, but I’m not very knowledgeable about this stuff. The other day, I got brave and tried to fix a switch mode PSU for a Nintendo 64. I now know why we idiots should not be doing that with the unit plugged into mains. I made a big scary pop and immediately threw the whole thing in the bin and announced my retirement from PSU repair.
lol we all have pop accidents. don't just give up yet. PSUs are very easy to fix once you understand its basic fundamentals. safety precautions comes automatically once you nail down its working principle. you'll be surprised how many electronics fail because of single faulty component on PSU and how it can easily be repaired, saving you tons of $$. be patient and learn your basics. dont give up yet
Also you can use one of those cheap “in circuit” ESR meters to test components without removing them….. because it uses a high frequency AC measurement. “Diodegonewild”made a good video about it on his channel…. where he’s troubleshooting and repairing an ATX power supply this past year. It’s about a 25 minute video and he explains the use of an ESR meter to measure all types of components without removing them from circuit. Just an idea and would love to see you tinker with some stuff like that and see more troubleshooting videos on your channel. There’s a huge void on TH-cam of nice to the point troubleshooting videos like this. You’re such a good explainer!
Thanks BigClive for going through a bit of switch mode diagnostics, I'm encouraged to whip the PSU out of a misbehaving plasma TV that does have a puffy cap in it; I wouldn't have thought about the diode though. I'll have to dig my Hameg out of the loft check a few components.
Great video. Just the right amount of troubleshooting and detail, at least for me. Of course, hardly anyone bothers with this we just buy a new power supply. Of course, I usually have something suitable salvaged from when the load device fails or falls into disuse.
understanding cannot be conveyed. you can only have it.. but you, my friend, help a lot in that case! thanks for motivation, inspiration and teaching! 😌👌
Ah if only I had known some of this before. My monitor died recently with zero power, wish I’d sent you my power board in for diagnostics! Well done and thank you for this video Clive :)
I worked in test of equipment so I preferred analog power supplies especially the Coutant Lambda ones. They still exist in Ilfracombe as TDK Lambda and make very good smps. Thanks for the great explanation, diagnosis and video.
Who are the 4 morons who have disliked that ? -_-' Mitsumi is a pretty reputable and historic supplier. They're famed for flooding the world with their floppy drives amongst other components. Their name is all over the place in mostly high end consumer devices.
Now 14. That's probably a pretty good average from 2.7k likes. About 0.5% hit the wrong button. I swear at least 3% of the population can't find their own arse with both hands and a search light, so this is way ahead of the curve.
It's 38 dislikes now....however its not good to call someone a moron just because they have a valid opinion in opposition to the majority. Its also quite possible to accidentally swipe the dislike on a mobile device.... unfortunately Clive even went down the route of calling people stupid when he decided to way in on the 5g debate a while back.....this was very frustrating because he was talking bolllocks and didn't understand the modus operandi of the effects of high frequency radiation on cell generation and replication....
That little repair reminds me of a friend fixing a very expensive bit of kit by replacing one component. A lot cheaper, and environmentally friendly than ditching the whole thing. Although I think I missed something from the video as I couldn't see which bit the ozone was supposed to come out of. :)
This happens all the time, one example is I repaired a 24" computer monitor by replacing one broken resistor inside for a few cents of cost at most. Been in use for 6-7 years after that and still going strong.
When I fix something like this, I like to use a diode with a voltage rating one step higher (as you said, 2*5A should be more than enough for the BD player, so current handling capability likely isn't the issue), and possibly add a heatsink to the diode. I got a cheap +5V/+12V supply with an USB/IDE adapter that developed the same kind of fault (one of the secondary rectifiers shorted), but it had a simple 2-transistor primary-side oscillator that kept running into current limit each pulse, so no kind of pulsing or short circuit protection, which heated the transformer. The supply started smelling "hot" after like 10 seconds of operation in that condition. The broken diode was a single DO-201 diode, but the PCB and heatsink in this supply were designed to take a single TO-220AC diode (two legs only), and after replacing the shorted diode with a heatsink-mounted TO-220AC, the supply works fine again. When the supply failed, I was testing stuff with a dead HDD, so I power-cycled it a lot. Maybe the diode got killed by inrush current when it was still warm. The specs seemed to be OK, 40V reverse and some amps forward. A scope showed that the actual reverse voltage during operation was far from being 40V.
Mitsumi was a popular brand of CD-ROM drives in the early 1990s, glad they are still in the business. That double diode makes me think of the classic two-winding full rectification circuit, with one half winding powering the output during opposing cycles (diode concic A-C-A). The other possibility is a single winding providing opposing DC voltages from opposing cycles (diode config A-C A-C).
You can usually find the location of a short without removing any parts if you use a decent bench meter and a 4 wire resistance measurement. You just probe all the components on the 12V rail and wherever the resistance is lowest is where your short is.
Sorry for the dumb question - I’m a noob: so, you would place the red probe on the power rail and the black probe on a ground point, to do what you suggest? Please clarify, I want to learn. Thank you kindly
HI Clive, I've been enjoying your videos for a few years now. I'm a heating engineer and fix a lot of boilers... I have to replace quite a lot of fans on Worcester Greenstar boilers. The "mechanicals" of the fan are invariably fine, but the fan control boards seem to fail quite frequently after a power cut. I'd love to send you a failed one (and one that works also if that would be helpful) for you to work your magic on. Ditching a £170 fan for what is probably a single component failure irks me!
@@bigclivedotcom Thanks Clive. I've just mailed back and front photos. There is a good chance the email will have gone into your spam. About 50% of my emails to gmail addresses seem to end up in spam.
Actually, sensitive analog circuits or other high-impedance circuits can behave differently if flux still remains after power on. Cleaning the board can be a good preventitive measure: Or, a corrective measure if you're testing it after resoldering/replacing something and it's not 100% working AND there is visible flux residue.
After buying a couple of PCB's at an electronics rally, I was offered a 10 to 15 Kg box of the same with 50p refund!!! So I now have a LOT of these type of PSU's taken from failed TVs' and even more of the Hi-V plasma driver coils to play with. I was interested in the HF properties of the larger ferrite transformers but will be looking for pulse driver IC's thanks to BigClive and the simplicity of the circuitry. Got loads of the double diodes, FET's and 450V caps to have fun with as well.
Thanks so much for this Clive, I have some smart bulbs that I've flashed to Tasmota to de-cloudify them. The PSU has failed on some of them, which I'd like to revive, as manufacturers are making it increasingly hard to flash firmware on newer smart things. This video will make fixing them a more reasonable proposition.
3.15 is part of the ISO R10 series of preferred numbers. It is approximately the tenth root of ten taken to the fifth power. A different preferred number series (the E series) is used for resistors, capacitors, inductors, and zener diodes. Preferred series are why you see numbers like 4.7, 63, and 220 more often than 5, 60, and 200 for resistors, etc. See en.wikipedia.org/wiki/Renard_series for a better explanation.
@@whitcwa I figured it was something like that. I always got a kick out of those for some reason though. I'm used to seeing fuses in whole or half-numbers with most of the stuff I work on.
Many years ago, I worked for a company that repaired PlayStations. The PS1 PSU was a similar design, just with a range of different outputs voltages. These PSUs suffered from 3 common faults, most common was the solder joints failing on the input socket, 2nd was again a failed solder joint, this time on the 400V 400uF primary capacitor, 3rd was a failed duel diode on the 12V output. Oh and yes a 400V 400uF capacitor is very spicy, having been bitten by them on more than one occasion!!
We used to have piles of fax machine PSU's with the fault written on the PCB. This particular fault was named "Pâc Pâc" and we knew that it could be one of two things, the shotky or driver FET. Good old days !
Another good troubleshooting video! I've never worked on such circuits that have a high voltage capacitor but I can see where a multimeter is a must when working on those to avoid any nasty surprises. If you touch a capacitor that's charged up to 220+ volts, that's gonna hurt!
Even when you have a meter it's a smart move to build a little discharge device with a suitable resistor and some insulated wires so you can discharge any caps safely before working on equipment. The meter could be broken and show zero volts when there is still 300 left in the cap you know. Just jamming a screw driver or whatever metal between the cap's legs will discharge it but that causes all kinds of damage.
I love love love your videos! I love your schematics ! I took electronics engineering in school many years ago! I still tinker! Only thing I remember from digital was and or and/or gates lol! I have a DC brushed motor controller I've gotten out of a treadmill ! I have the motor and speed controller and for 2 years been trying to figure out ware to inject a pwm to controll the motor speed. It takes 110ac drops it to 90vdc and then to the motor! Been relearning about how to drive higher voltages with mosfets with low voltage controls with octo couplers ! It's been so long and I over think things. I was going to just use a servo tester for the pwm but that's max 8.4v and if I'm not mistaken that will only drive the mosfet to dump 8vish and all the current it wants. I know I'm wrong! I want to MacGyver a small go-kart with a inverter a 12v batt and a 90v 1/2 hp motor ! I can bump the 12v to 90v with my buck booster/regulator from drok ! Was 900w but I added a shunt and beefed up the tracks and have pulled 2kw out of it but the 3s 4400mah 75c batt wasn't happy! Lol was wondering if I sent you pics in 8k /108mp both sides could you spot ware I would inject a 5v pwm ? Think it would make a great video! Appreciate all your videos! You have entertained me through out the years and the geek in me thanks you!
I had a diode AND the cap fail in a dvd/vhs combo psu one time. didn't blow the cap , but it had an ever so slight bulge , and both parts measured 0 ohms. 2 euro's in parts fixed it.
@@martinlelle6863 from painful experience I can say that DC kicks harder than AC volt for volt, 240V AC is annoying and a reminder that you are probably being stupid! but DC, especially when there is a nice fat cap backing it up will make you reconsider your life choices!
Years ago I purchased a "flat" CRT computer monitor on eBay, it was almost brand new good but the VGA connector was smashed, since I had used cables with the connectors still in good shape I opted for replacing the cable, I took off the back cover and simply out of habit I checked if the electrolytic of the power supply was discharged and still had *about 290V* and it had taken a week to arrive and I still hadn't plugged it! So definitely the "I already unplugged it yesterday" does not guarantee any security when working with high voltage capacitors
The dual diode might have lasted longer if it had been fastened to the PCB as a semi heat sink. Heat kills. I think that these electromechanical devices get to be several years old and the lubricants get gummy and that puts a heavier load on the motors. Then the heavier load draws more current from the power supply and the results are what Clive found in this power supply. A bit of lubricant here and there can make the problems go away. It's kind of a shame that an otherwise good piece of equipment should end up in the landfill for such a simple problem. Thanks, Clive.
Some capacitors do read as shorted but that's because they have a charge, or are building up a charge, I've found you have to wait a little while to make sure properly👀
![ลองสุ่มไอโฟน 990 โกงมั้ย? [ โกงมั้ยครับ ep.101 ] | DOM](http://i.ytimg.com/vi/bfsWe0j492k/mqdefault.jpg)
I love these types of videos, helps people get an idea of how legitimate electronics repair goes. Will also help right to repair a teensy bit. Thank you for this Clive!
I like them as well, this also is a very good way of teaching people, especially as Clive gives them a little fun in the mix too.
It's amazing how quickly you are able to fault trace this problem. Top marks for you and awoo for the fix!
Finally, a classic uncut repair video of something other than a LED lamp. ;-)
Still a diode !
seriously it was kinda getting monotonous here with all that LEDs
Diode was obviously defective, no light output!
@@GreeceUranusPutin Gold star !
@@GreeceUranusPutin any diode can be a light emitting diode if you stuff enough power through it.
I consider Mitsumi a quality manufacturer, had lots of 3,5" floppy drives, never failed. Could be dirty with dust bunnies, still wouldn't misalign heads. This PSU looks to be quite tidy, too.
The best keyboard I ever had was a Mitsumi. Cost £3, only bought it to pad an order out for free p&p. Nice and clicky despite not being a proper key switch. Reliable, light, nice typing angle.
Yep, good old Japanese brand name. Nice spacious design here, and repairable.
@@Zadster I think i used one of those for 10 years, until repairs kept piling on and it no longer felt great. But nowadays, i'm exceptionally happy with a generic Chinese mechanical that i found used years ago, this is the best i've had so far.
But they're also the manufacturing company behind Nintendo's products, also contributing some of their engineering. Gamecube, Wii era stuff was so good!
Then again the word is they're also the ones manufacturing Nintendo Switch, which retains its glaring controller drift issue, though ultimately it's an engineering issue and not materials or assembly quality issue, and ultimately Nintendo isn't doing anything about it, so that's pretty telling.
Indeed. Mitsumi is an old established brand. I would assume the product is good. Which is much different than the usual random maker.
i got them confused with Matsushita
I’m so glad you covered the charged up input capacitor danger.
The choke / capacitor filter at the front end is more to protect the mains from radiated HF interference. Those HF switching PSUs can be pretty electrically "noisy" - and most of the Compliance Authorities (e.g. FCC) now mandate such suppression.
That's why the Alibaba $19 gaming computer power supplies are the best. One or two of them can corrupt an entire MRI machine in the same building with their electrical interference.
I'd still buy one for a dollar.
@6GDC You don't mean computer PSU surely? It bears reading the specs carefully, if it's a programmable power supply, they will have voltage dependent maximum current. 40A at 3V output maybe? Still a massive massive accomplishment, IF they did it, but it can be done, but it will be rare to see anything with more than 10A outputs. I don't know the product you're speaking of, but by far not all impressively specced and aggressively priced Ali products are scam junk.
Ali computer PSUs that the dollar guy speaks of, are just salvage junk or copies of old ATX1.3 computer PSUs, with maybe a power handling of 200W or so that rough order of magnitude, just made to look like modern PSUs on the outside.
@6GDC How feasible a pocket-size 40A switch-mode power supply is depends on a few things, such as: how large "pocket size" is; the input & output voltages/voltage ranges; whether they bothered with EMC (needing space for filtering components); etc.
Modern MOSFETs have extremely low on-state resistances (the "R_DS(ON)" value¹, i.e. the resistance, in Ω, between the Drain and Source pins at a specified Gate voltage). Even very cheap MOSFETs can have R_DS(ON) ratings well below 10 mΩ, and this results in very high current ratings, especially for the size. Lower resistance means a lower voltage drop, less resistive heating, and thus more current before things get out of hand (I'm kinda guessing here, but it makes sense).
Clive's shown a couple of devices on this channel containing SOT-23 MOSFETs capable of several amps of continuous current - and SOT-23 devices are only 2.9×1.3 mm (top surface; Wikipedia doesn't list the height of SOT-23).
The "USB power bank that [intentionally] gets very hot" video comes to mind. It used one of those SOT-23 MOSFETs to switch its heaters, IIRC.
There's also the recent emergence and popularity of gallium nitride MOSFETs, which claim higher voltage ratings, higher efficiency, smaller overall device size, and other benefits. They've driven the recent influx of "GaN" SMPSs in the consumer space - e.g. USB power supplies and phone chargers, especially high-power USB-PD supplies. Those devices can supposedly be made smaller and lighter with GaN MOSFETs, as well as being more energy efficient, in theory.
(No doubt some of those devices are just slapping "GaN" on the packaging/product description as a marketing lie, but some are probably genuine!)
Further reading: en.wikipedia.org/wiki/Gallium_nitride#Transistors_and_Power_ICs (and the rest of the article for more general info about GaN tech)
So your "40A" SMPS _could_ be genuine! It's not totally unrealistic, IMO.
...but it could also very well be a wild exaggeration! 😁
¹ R_DS(ON) is typically written as "R" followed by "DS(ON)" in subscript (no "_"), but I can't use subscript on TH-cam - at least not without obscure Unicode characters that I don't have easy access to on my phone - so I'm using "_" to indicate subscript.
@6GDC NOT 40A at 170V, almost 7kVA, you can be sure. Read between the lines - maybe 40A at the lower end and use that calculated wattage to work out what you will get at the higher end. That's how Shenzhen do. (mebee one or two amps if you are lucky?) Still respectable but not 7kVA.
Had the engineering manager where I used to work try to discharge the caps on a 600Kw vsd, with the "Short it out" technique he had heard about.
Luckily he was not injured, but it "Brightened" his day.
and... that's why he was the manager
This Clyde could turn children into electronic tech's! He makes the most complicated, easy too understand. GREAT JOB CLYDE! CHEERS!!
Thank you Clive! This looks like a fault I had recently in a Peavey mixer. Another chip had visibly blown on the PSU, which I replaced. Now the desk pulses on and off just as you described. I’ll dig it back out and have another go at fixing it now!
I have a bass amp that has dual secondaries +-15 and +-45 the 15v neg rail failed and was doing what you describe but after replacing caps there was some huge charge then a bang across the -15v rail the windings are now open on the secondary, smps are great but if you lose the transformer it's game over
Nippon Chemicon capacitors are generally a sign of someone not cheaping out, so reinforces the "common sense and safety" comment in the description :)
(That big, painful 450V capacitor prominently showed the Nippon Chemicon logo - they only include the company name inside the logo in _really big_ capacitors - and the large capacitor on the secondary side was in the same style, even though the logo was hidden.)
I re-capped my DBX 286A microphone processor with all new Nichicon caps. It wasn't a cheap prospect. I don't regret it. I got it faulty from Ebay for $49. It's fully functional now and quite impressive. The only two I did NOT replace were 2 bi-polar caps that I had not prepared for.
@@TechGorilla1987 The Nichicon Muse ES series are the ones you want for the bipolar caps. Lovely shiny green sleeves!
this kind of repair is the god of all repairs, these days, electronic technicians are not the purest kind.. i remember a friend of mine who works with radar system in middle east, he travels even during wee hours for sometimes 1000 km just to fix a radar. i saw him fix my laptop by cutting some IC connections and jumping it from somewhere with several wires, insulate it and bingo, it works. people like clive and my friend are few among millions who does a work with electronics by just smelling it. salute to you clive. nice video and thank you for sharing it with all walks of life.
A long time ago, someone thought, "Let's take the AC mains, rectify it into DC, connect s coil with a DCR of 0.05 ohms to it, and then short it to ground with a fragile mosfet." That man ended up in an asylum, but danged if it didn't work.... It has always been my theory that switching supplies work because of voodoo.
In the early days they were outright scary.
Actually, this exact thing took place over a hundred years ago when Henry Ford put the ignition coils in the earliest cars. Then along came Kettering with the coil, 'condenser' and points to repeat the same thing. Now we have chips that do it in microseconds. 👍👍
@@acmefixer1 That's true. And the "flyback" circuits in CRT TV's and video monitors work similarly. Still, it's kind of amazing that these circuits are so reliable if built correctly. On paper, it seems like they'd be doomed to a fiery death in the first second of operation.
You forgot the HF chopper. In the 80's I fixed dozens of SMPS' in commercial equipment (third party IBM network muxes). They used all discrete components back then. It was always the output electrolytics that bore the brunt. They ran HOT.
Times sure have changed.
@@ScottGrammer That's because you're thinking of this like it's a DC circuit. It isn't. You're not thinking about the inductance involved. When the mosfet first turns on, there's not a dead short across the supply. The coil is nearly open at that point due to its inductance. It gradually starts conducting more and more current until the mosfet turns off, sometime before it becomes a dead short across the supply.
"This is good", I know that feeling when you're expecting a pop or smoke and nothing happens :-)
Anything to do with mains I always use a 3a fuse in the plug
@@joshuabest100 In a test like this it is much better to replace your fuse with say a 40w or 60w incandescent light bulb, if it comes on the device is still shorted, but will not explode because there is not enough power available for blowing stuff up. For a TH-cam channel though it is better without it: because you will have sparks and flying shrapnel which is always cool! :D
@@perhansson6718 that's a pretty good idea tbh I might use this
@@perhansson6718 that explains why my TV repairman grampa always used a light bulb for various kinds of testing
@@perhansson6718 haha indeed😂😂😂😂
I'm not sure why but I always find PSU's to be some of the more interesting things to watch.
More componet sparse so it's easier to follow along maybe?
Yeah me to, Diode Gone Wild does a lot of 'em.
You're drawn to power electronics, it's Just how it is
@@masteryoda394 A powerful draw, it is
@@andymouse Freaking love that guy of couuuuuurse. And here is my cat, of couuuuurrrse...
My favourite videos are the repair videos because they are done by Clive so they are not boring and I learn stuff from them! Thanks for that clive
Would love to see more general troubleshooting and repair videos. They are few and far in between on your channel and you’re a very good teacher and wonderful at explaining things!
I have no electric or electronic background,but I am fascinated by your videos and pretend to understand everything you explain. I wish I had studied electronics and a teacher like you would make the complicated circuit boards easy to understand. But thanks anyway, your understanding and hardworking in your videos is greatly appreciated.
If you want to remove a component and need to pull it gently to remove it there is an easy trick I use. Grab a pair of forceps and clip them onto the resistor, mosfet etc and then hold the board so they hang, once you heat the pads up the weight will pull gently for you. An added bonus for exposed metal components is it helps reduce the heat in the parts. Sorta like the opposite to using bluetack to hold parts in place while you solder them in place.
Forceps are an excellent tool for the toolkit for all sorts of little tricks. Very handy and usually a great cheap addition to the toolkit.
I work on industrial battery chargers, we have dc capacitors we run around 680vdc.
The day I found a bad bleed down resister was very fun.
Yes I reached in like a dumb a$$ to change the igbt....
That was the day I learned to ALWAYS double check the capacitors ..
Be safe Clive .
That's ac input 3phase?
A few decades ago, I was working on (diagnosing) a single-triode power ultrasonic oscillator at the base of an ultrasonic cleaning machine. Went to measure the anode volts, should've been between 5 and 10kV. They were, and my high-voltage probe failed, and punched a hole through my thumb. Very thin hole, but a hole, nonetheless!
I use a cable from a dead DVD player as a test mains lead for SMPS boards. The almost all use the same connector for the mains input.
I would add to the "proper repair" while in there I would also replace the small bulk cap for the chip VCC. Those little things tend to just wear out over time and cause all kinds of problems, not worth saving 10 cents and a minute of time to find out there are more problems about to go off. A pretty nicely designed unit this one or perhaps not much used, often there is distinct heat discoloration and dirt on the boards.
Yes! So often I find the psu pulsing or not starting up because of that little chip/switcher start cap has dropped from say 10u to 4 or 5u. The worse the power quality or noisier the environment (industrial) the shorter the life span of the caps seems to be.
I love the gigantic primary/secondary labeling. There's no missing that
I always love your videos, but have to admit it was good to see one that was not an ozone / ion generator related one thrown in.
Always good to see a bit of diagnostics and repair :).
"Yeah, that's good enough."
No truer words have been spoken.
Wow! Thank you. This is the exact problem I have with a similar power supply that powers a set of mirrored led lights. PSU led onboard is pulsing, witch causes the lights in the unit to pulse. The caps look fine, and I was suspecting the diode, but wasn't sure how to test. Guess what? I followed your video and yes definitely the diode! I'm going to hit up our sparky at work to see if I can pick up a freebie. Hopefully I will have another working DMX fixture to add to my program 😃
This is so educational to a hobbyist electronics person, awesome sauce! Thanks Clive.
You can tell Big Clive is a good guy because he gives too much flux.
Ha ha ha 😂 ❤️ the flux is the spirit of the electrons
Much rather too much Flux then not enough.. I'm the same way when doing electronics specially pcbs cause I'm always worrying that I may burn up the paths
Yeahhh..more flux with leadbased solder😁😁😁😁
XD😊
@@johngarritzcx6733 always hate non lead based solder always have to use 80-100 watt stations. Can't use my 40 watt gun or my 60 watt station.
This is really useful, thanks! I would be interested in seeing more repair vids.
I really need to get into fault finding on this level, but never seem to have the time... Thanks for this encouraging video! :-)
my router broke over a year ago and i still didn't fix it. best tip: document what you do with pen an paper. this is my mulitmeter's 2nd battery.....
@@plageran you only need 2 things for a easy fix , a trash can and Amazon
@@plageran For the battery, smoke detector batteries. Energizer lithium for the win. They're not the cheapest (tho they've come down a bit) but worth every penny.
@Hill lmao and @Mad that's abit of a mod if you're referring to Poundland batteries😂😂
@@plageran Haha We don't have Poundland here but I'm gonna go out on a limb and say it's similar to "Dollar tree" here... If I'm right, you'd be absolutely correct! 🤣
This will be helpful next time I repair a PSU that I actually need to diagnose - I've fixed a few PC monitor PSU's recently, and they all died the same way: cheap no-name caps on the secondary side failing (in very visibly obvious ways).
It's funny because the board that deals with inputs, driving the LCD, etc usually has good caps on it, but the PSU sitting right next to it is full of garbage (and in the last PSU I fixed, also full of vacant spots a cap was *supposed* to occupy, and now does).
Also funny where the arrangement of heatsinks and power diodes is such, that the caps get baked real good, entirely avoidably so.
PSU repairs are very interesting to me. Perhaps I will view this. One moment please....
Usually when people do a false accents i find it unbearable but you’re so naturally entertaining I actually enjoy it.
Very big coincidence! I got into something identical today... my LED magnifier desk lamp died several years ago and I finally got fed up with using a flashlight ('torch') to trim my nails. Hauled the thing to work (extender arm etc) and took it apart... found a board almost identical to the Mitsumi, albeit pure Chinesium. The chip was a 1202 (EDIT: Innuovo XN1203) using the same setup... the opto bridged a slot in the circuit board to give extra isolation. I tested the electrolytics, they were iffy as ChongX caps normally are, but the failure was in the input filter stage (just before the big cap)... the 1.5mH pass inductor was open. New parts on the way from Mouser, soon I will have my magnifier lamp again! Very glad to be able to get the PDF of the Chinese chip, the example circuit diagrams didn't need translation thank goodness.
Enjoyed that, really clear run through. That cap didn't want to discharge in a hurry.
Noise is becoming a big problem, especially our amateur radio.
73 de N2NLQ active
Even just AM radio is basically unusable in my apartment block because at least one household uses these shitty PowerLAN/DLAN ethernet-over-powerlines adapters, turning all the electrical wiring into a giant antenna transmitting a f*ckton of noise/interference (think cheap chinese switchmode PSU with barely any filtering, but on steroids)
@@Knaeckebrotsaege I'm using these every now and then for easy LAN connection across my apartment.
Say, someone would like to catch the noise with their SW defined radio, at which frequency range would they have to look? I'm too lazy to do my own research. :-/
Asking for a friend.
@@froedlmetallmann4643 Well the AM broadcast band is from 550kHz to 1720kHz.
Taking a look around, the frequency range seems to be quite broad, being between 2MHz, and around 70MHz.
This is a flyback converter. They are notoriously hard on the capacitors and the output diode because of the trapezoidal current waveforms they produce. Since the output of the flyback only receives energy for part of each switching cycle it means that the currents must be quite large so that the average current (the DC load current) can be met. So if the output is, say, 1A DC, chances are pretty good that the peak output currents are on the order of 3-4 amps. That is tough on the diode and the output filter capacitor. Usually the component that fails on a flyback is the blocking diode, as happened here.
Your caution with the capacitor reminds me of my college days in the 70s. It was all valves with lots of volts available on the bench. We used to get big electrolytic caps, extend the connections with bare wire and tape it down each side of the cap. We would then charge it to 200 volts and toss it across the lab shouting 'catch' . The recipient got quite a surprise 😂 Oh the folly of youth.
I have similar stories of that period, in the testing department of a PABX manufacturer and students at the electronics lab of a local school kept doing that all the time, leaving charged capacitors over the benches, waiting for a "sucker".
I was almost waiting for a solder sucker! But I guess the Cu braid does the job. :)
Thank you so much. You addressed the cause of flipping LED. I like the way you explained the circuit with a single output setup.
These are used on a high end Song blu ray player and fail on a regular basis. The biggest offender is the double diode on the secondary. It’s more than adequate for the job especially as the diodes are wired in parallel so I suspect a bad batch.
Could the lack of heatsinking be to blame?
So here's a weird thing that happened in my mind: 9:27 I noticed you said live, but not the requisite "at mains voltage". So in my mind I imagined a radio broadcast where an announcer with a really bassy voice said "Big Clive. Live - at mains voltage!"
You are like a God. I would never have found this problem. I still don't get how you solve these things soo fast.
An interesting and very practical display of skill made understandable with the voice over. This guy should be selling'EASY ELECTRONIC REPAIRS,lessons by post.He is really great,showing and telling. Thank you. Terry Offord
This was a big help to me. I’ve been trying to figure what was wrong with my PSU and it turns out it was a bad resistor. Thanks for the process!
Flux on the de-soldering braid, awesome piece of knowledge, you may have just saved my oscilloscope kit that I ordered to build for practice soldering.
I love these. Still trying to get the basics and this is so useful.
You have no idea how nice it is to hear you say SOLDER. All the USA content people say "SOW-DER" and it eventually sets my teeth on edge
I pronounce it sodder. Actually it depends, sometimes the L finds its way in but won’t be complete and will sound like sowder. But yeah, if guys like clive and paul Carlton weren’t around, I would really have to say it like them exactly to get my dose of soLder
@kasperabigail9129 some have notice that I pronounce it differently depending on time of day. I'll say solder in the morning, sowder in after noon more and turns into sodder at night.
Sawder
For 10 years ago, my mom by new DVD player. It broke few days after she used in first time. The capacitor was dry out. The store there DVD come from told her by new because it will be too expensive to fix. I take the player with me too my work and change this capacitor. DVD player is fully active 10 years later. I fixed player for 100 isl. Krona. New DVD player cost about 30.000 isl. Krona.
I had an IBM PS/1 PSU with exactly the same fault. Gave it a brand new diode, a complete recapping for good measure. Now just have to finish repairing the floppy drive bezel, which is apparently made of tortilla chips with too much dip on them, because it cracks every time I try to reassemble it. Pretty soon it'll be made of all-new plastic, though!
I hope you don't have the yellow rifa caps of death in there..
I hear the words "dead short" and I immediately think "capacitor" too. Next, I postulate, what else has this capacitor, if found, (alternative, shorted tracks?), taken out? A short (or even a "medium"?) does amusing things, elsewhere on board, sometimes. Diodes are a natural, but it can tickle other parts too, transformers, for one. Sometimes all! Also, have you ever noticed on odd (very?) occasions where restoring with a new component somewhere places stress, renewed or new, on another component, causing it to fail afterwards? An "It tested okay before, isolated, why is it shot now", kind'a thing?
love your method of taking pictures and enlarge it, great video as always!
Always a pleasure when Big Clive gets his probe out and goes to work! Sometimes the components bulge up excitedly…;)
I like the Fault finding vids when they come along.
Thanks Clive. Can we have more repair videos please? You explain how the circuits work.
The exact quote is: "The bigger the gob, the better the job" - Louis Rossmann, indeed. Cheers, Clive.
Not sure if that went as intended... :-)
Small dab’ll do ya, big glob,ll screw ya.
"If you’re not Fluxing, you’re not Soldering" - Justin, The Art Of Repair. 👌
On page 12 of the datasheet there's a demonstration board schematic, which might be much closer to the board you're looking at. There's an interesting detail about the diode on the suppression network on the primary, the datasheet calls for a 1N4007 and not a fast recovery diode as used elsewhere on the design.
One possible explanation may be that using a "slow" diode allows some of the spike current to flow back, marginally improving efficiency.
4:49 I prefer inertial desoldering - heat the pads and whack the PCB edge-on on the bench top. most of the solder will splash off.
Small pcbs can just be dropped on the floor
I owned an commercial radio communications business for 40 years (think police/fire/taxi radios) and sold/serviced alot of power supplies over the years. We regularly repaired linear power supplies, but switching PS's are a different beast. We soon learned it was much better to replace a faulty SMPS than to repair it. Once you repair a piece of electronics the customer believes any further failures are your fault and should be covered by your repair warranty. They are also much cheaper to replace than repair these days... unless this is for your own equipment.... but still might not be cost effective to self repair!
13 people wanted you to discharge that cap with your fingers I guess, great vid! I have no clue where to start looking for faults if it's not black or crispy thanks Clive!
I love watching your videos, but I’m not very knowledgeable about this stuff. The other day, I got brave and tried to fix a switch mode PSU for a Nintendo 64. I now know why we idiots should not be doing that with the unit plugged into mains. I made a big scary pop and immediately threw the whole thing in the bin and announced my retirement from PSU repair.
That's quite scary if it was you that made the pop.
@@brancheternal Well, not directly, I guess. I’m about as good at wording things as I am at PSU repair…
lol we all have pop accidents. don't just give up yet. PSUs are very easy to fix once you understand its basic fundamentals. safety precautions comes automatically once you nail down its working principle. you'll be surprised how many electronics fail because of single faulty component on PSU and how it can easily be repaired, saving you tons of $$. be patient and learn your basics. dont give up yet
Are you sure the pop came from the board and not from inside your pants? 😂
that's the way a lot of us started out, then learned wisdom
That was interesting indeed🧐
You already know that I can't help but say something about the quality of those pictures. 😝😂
Yes Yes Yes !
I have a very similar no good PSU with the same symptoms and now I have 'a clue' about 'how to attempt a repair'
Thank you Clive !
Also you can use one of those cheap “in circuit” ESR meters to test components without removing them….. because it uses a high frequency AC measurement.
“Diodegonewild”made a good video about it on his channel…. where he’s troubleshooting and repairing an ATX power supply this past year. It’s about a 25 minute video and he explains the use of an ESR meter to measure all types of components without removing them from circuit.
Just an idea and would love to see you tinker with some stuff like that and see more troubleshooting videos on your channel. There’s a huge void on TH-cam of nice to the point troubleshooting videos like this.
You’re such a good explainer!
Thanks BigClive for going through a bit of switch mode diagnostics, I'm encouraged to whip the PSU out of a misbehaving plasma TV that does have a puffy cap in it; I wouldn't have thought about the diode though. I'll have to dig my Hameg out of the loft check a few components.
Great video. Just the right amount of troubleshooting and detail, at least for me. Of course, hardly anyone bothers with this we just buy a new power supply. Of course, I usually have something suitable salvaged from when the load device fails or falls into disuse.
I find a stiff nylon toothbrush to be a great desoldering aid. Simultaneously apply iron and scrub connection
How would you use it? Would you use the shaft, or the brushes, or am i completely lost..?
So you just need a third hand to end up with a molten brush? somehow I don't see how this would work at all...
but with braid you know where the solder is going, a brush might flick solder everywhere
understanding cannot be conveyed. you can only have it.. but you, my friend, help a lot in that case! thanks for motivation, inspiration and teaching! 😌👌
Ah if only I had known some of this before. My monitor died recently with zero power, wish I’d sent you my power board in for diagnostics! Well done and thank you for this video Clive :)
I worked in test of equipment so I preferred analog power supplies especially the Coutant Lambda ones.
They still exist in Ilfracombe as TDK Lambda and make very good smps.
Thanks for the great explanation, diagnosis and video.
Very nice, Clive. Your videos are often magical.
I have a failing PS in an undercounter TV/RADIO/DVD player that needs you.
Who are the 4 morons who have disliked that ? -_-'
Mitsumi is a pretty reputable and historic supplier. They're famed for flooding the world with their floppy drives amongst other components. Their name is all over the place in mostly high end consumer devices.
The PM, Home Secretary, Foreign Secretary & Health Secretary 🤷🏻♂️
Probably someone that has 4 accounts that doesn't know anything about electronics and has a TH-cam channel.
Now 14. That's probably a pretty good average from 2.7k likes. About 0.5% hit the wrong button. I swear at least 3% of the population can't find their own arse with both hands and a search light, so this is way ahead of the curve.
The only moron is the person that thinks everyone has to like what they like on the internet. I liked it but I understand others may not and let them.
It's 38 dislikes now....however its not good to call someone a moron just because they have a valid opinion in opposition to the majority. Its also quite possible to accidentally swipe the dislike on a mobile device.... unfortunately Clive even went down the route of calling people stupid when he decided to way in on the 5g debate a while back.....this was very frustrating because he was talking bolllocks and didn't understand the modus operandi of the effects of high frequency radiation on cell generation and replication....
This is great! A simple project which was relatively easy to follow along. (Obv, power supplies are seductively simple, yet dangerous.)
That little repair reminds me of a friend fixing a very expensive bit of kit by replacing one component. A lot cheaper, and environmentally friendly than ditching the whole thing.
Although I think I missed something from the video as I couldn't see which bit the ozone was supposed to come out of. :)
It could power the Russian Nuclear Corporation unit.
This happens all the time, one example is I repaired a 24" computer monitor by replacing one broken resistor inside for a few cents of cost at most. Been in use for 6-7 years after that and still going strong.
When I fix something like this, I like to use a diode with a voltage rating one step higher (as you said, 2*5A should be more than enough for the BD player, so current handling capability likely isn't the issue), and possibly add a heatsink to the diode. I got a cheap +5V/+12V supply with an USB/IDE adapter that developed the same kind of fault (one of the secondary rectifiers shorted), but it had a simple 2-transistor primary-side oscillator that kept running into current limit each pulse, so no kind of pulsing or short circuit protection, which heated the transformer. The supply started smelling "hot" after like 10 seconds of operation in that condition.
The broken diode was a single DO-201 diode, but the PCB and heatsink in this supply were designed to take a single TO-220AC diode (two legs only), and after replacing the shorted diode with a heatsink-mounted TO-220AC, the supply works fine again. When the supply failed, I was testing stuff with a dead HDD, so I power-cycled it a lot. Maybe the diode got killed by inrush current when it was still warm. The specs seemed to be OK, 40V reverse and some amps forward. A scope showed that the actual reverse voltage during operation was far from being 40V.
Spicy Voltage is my new favorite word!
great video. Sure, a lot of people can build circuits from scratch, but troubleshooting and making your own diagram is what cuts the cake
Mitsumi was a popular brand of CD-ROM drives in the early 1990s, glad they are still in the business.
That double diode makes me think of the classic two-winding full rectification circuit, with one half winding powering the output during opposing cycles (diode concic A-C-A). The other possibility is a single winding providing opposing DC voltages from opposing cycles (diode config A-C A-C).
I remember an early mitsumi CD drive totally bombing win98 consistently.. the drivers just never worked.
@@benbaselet2026 Yeah, their unusual interface required drivers that weren't really made for later Windows versions.
@@benbaselet2026 and now they are spitting cherry MX clones
Thank you.
To me it is a marvel what these small components can handle.
Thank you, I always curious to know about switching ps but there's not enough good explanation like what you did, I really appreciate.
Clive "i don't want to touch the capacitor... it is spicy"
Internet "damnit Clive just do it!!"
This is not ElectroBOOM, you know! ;)
It's good for cardiac arrest stimulation.. probably not a great idea Ryan lol
@@NaterNorris charge the capacitor to 450v then tell your buddy to catch
A zap to the fingers will have you hopping ...done that many a time..
Im glad im not the only one that gets surprised by hot components! ;-)
You can usually find the location of a short without removing any parts if you use a decent bench meter and a 4 wire resistance measurement. You just probe all the components on the 12V rail and wherever the resistance is lowest is where your short is.
Sorry for the dumb question - I’m a noob: so, you would place the red probe on the power rail and the black probe on a ground point, to do what you suggest? Please clarify, I want to learn. Thank you kindly
HI Clive, I've been enjoying your videos for a few years now. I'm a heating engineer and fix a lot of boilers... I have to replace quite a lot of fans on Worcester Greenstar boilers. The "mechanicals" of the fan are invariably fine, but the fan control boards seem to fail quite frequently after a power cut. I'd love to send you a failed one (and one that works also if that would be helpful) for you to work your magic on. Ditching a £170 fan for what is probably a single component failure irks me!
It would be interesting to see a picture of one of the PCBs. Can you email one to bigclive1@gmail.com
@@bigclivedotcom Thanks Clive. I've just mailed back and front photos. There is a good chance the email will have gone into your spam. About 50% of my emails to gmail addresses seem to end up in spam.
@@rainbm Could you resend the email and images. I don't see it in my inbox and had to bulk delete spam due to a build up.
Hi Clive, did you ever receive my email with the photos? I sent them again after your spam purge.
"There's no problem with having too much flux." It flux it up sometimes, though ;-)
Actually, sensitive analog circuits or other high-impedance circuits can behave differently if flux still remains after power on. Cleaning the board can be a good preventitive measure: Or, a corrective measure if you're testing it after resoldering/replacing something and it's not 100% working AND there is visible flux residue.
After buying a couple of PCB's at an electronics rally, I was offered a 10 to 15 Kg box of the same with 50p refund!!! So I now have a LOT of these type of PSU's taken from failed TVs' and even more of the Hi-V plasma driver coils to play with. I was interested in the HF properties of the larger ferrite transformers but will be looking for pulse driver IC's thanks to BigClive and the simplicity of the circuitry. Got loads of the double diodes, FET's and 450V caps to have fun with as well.
I just repaired PSU thanks to this video. Thanks greatClive!
Thanks so much for this Clive, I have some smart bulbs that I've flashed to Tasmota to de-cloudify them. The PSU has failed on some of them, which I'd like to revive, as manufacturers are making it increasingly hard to flash firmware on newer smart things. This video will make fixing them a more reasonable proposition.
3.15 Amp fuses seem to be really common for some reason. I see them in all types of power supplies and other devices.
3.15 is part of the ISO R10 series of preferred numbers. It is approximately the tenth root of ten taken to the fifth power. A different preferred number series (the E series) is used for resistors, capacitors, inductors, and zener diodes. Preferred series are why you see numbers like 4.7, 63, and 220 more often than 5, 60, and 200 for resistors, etc. See en.wikipedia.org/wiki/Renard_series for a better explanation.
@@whitcwa I figured it was something like that. I always got a kick out of those for some reason though. I'm used to seeing fuses in whole or half-numbers with most of the stuff I work on.
Many years ago, I worked for a company that repaired PlayStations. The PS1 PSU was a similar design, just with a range of different outputs voltages. These PSUs suffered from 3 common faults, most common was the solder joints failing on the input socket, 2nd was again a failed solder joint, this time on the 400V 400uF primary capacitor, 3rd was a failed duel diode on the 12V output.
Oh and yes a 400V 400uF capacitor is very spicy, having been bitten by them on more than one occasion!!
We used to have piles of fax machine PSU's with the fault written on the PCB. This particular fault was named "Pâc Pâc" and we knew that it could be one of two things, the shotky or driver FET. Good old days !
Good explanation switching power supply
Another good troubleshooting video! I've never worked on such circuits that have a high voltage capacitor but I can see where a multimeter is a must when working on those to avoid any nasty surprises. If you touch a capacitor that's charged up to 220+ volts, that's gonna hurt!
Even when you have a meter it's a smart move to build a little discharge device with a suitable resistor and some insulated wires so you can discharge any caps safely before working on equipment. The meter could be broken and show zero volts when there is still 300 left in the cap you know. Just jamming a screw driver or whatever metal between the cap's legs will discharge it but that causes all kinds of damage.
I always go for the tripple: discharge with a resistor, measure AND jam a screw driver between the pins for final safety check.
I love love love your videos! I love your schematics ! I took electronics engineering in school many years ago! I still tinker! Only thing I remember from digital was and or and/or gates lol! I have a DC brushed motor controller I've gotten out of a treadmill ! I have the motor and speed controller and for 2 years been trying to figure out ware to inject a pwm to controll the motor speed. It takes 110ac drops it to 90vdc and then to the motor! Been relearning about how to drive higher voltages with mosfets with low voltage controls with octo couplers ! It's been so long and I over think things. I was going to just use a servo tester for the pwm but that's max 8.4v and if I'm not mistaken that will only drive the mosfet to dump 8vish and all the current it wants. I know I'm wrong! I want to MacGyver a small go-kart with a inverter a 12v batt and a 90v 1/2 hp motor ! I can bump the 12v to 90v with my buck booster/regulator from drok ! Was 900w but I added a shunt and beefed up the tracks and have pulled 2kw out of it but the 3s 4400mah 75c batt wasn't happy! Lol was wondering if I sent you pics in 8k /108mp both sides could you spot ware I would inject a 5v pwm ? Think it would make a great video! Appreciate all your videos! You have entertained me through out the years and the geek in me thanks you!
I had a diode AND the cap fail in a dvd/vhs combo psu one time.
didn't blow the cap , but it had an ever so slight bulge , and both parts measured 0 ohms.
2 euro's in parts fixed it.
I could tell Clive was scared of the big capacitor!😂
The capacitor bite hurts a lot though, it can be quite furocious._
@@martinlelle6863 from painful experience I can say that DC kicks harder than AC volt for volt, 240V AC is annoying and a reminder that you are probably being stupid! but DC, especially when there is a nice fat cap backing it up will make you reconsider your life choices!
This sort of content is GOLD.
Years ago I purchased a "flat" CRT computer monitor on eBay, it was almost brand new good but the VGA connector was smashed, since I had used cables with the connectors still in good shape I opted for replacing the cable, I took off the back cover and simply out of habit I checked if the electrolytic of the power supply was discharged and still had *about 290V* and it had taken a week to arrive and I still hadn't plugged it!
So definitely the "I already unplugged it yesterday" does not guarantee any security when working with high voltage capacitors
The dual diode might have lasted longer if it had been fastened to the PCB as a semi heat sink. Heat kills.
I think that these electromechanical devices get to be several years old and the lubricants get gummy and that puts a heavier load on the motors. Then the heavier load draws more current from the power supply and the results are what Clive found in this power supply. A bit of lubricant here and there can make the problems go away. It's kind of a shame that an otherwise good piece of equipment should end up in the landfill for such a simple problem. Thanks, Clive.
BigClive, BigTamer of unruly power supplies ❤️❤️❤️
Some capacitors do read as shorted but that's because they have a charge, or are building up a charge, I've found you have to wait a little while to make sure properly👀