A couple notes: Indeed, when using a DMM diode test on MOSFETs, I always make the verification twice - altering the polarity of the test leads. The diode test on most DMMs provides about 2.5 V and the gates are very low leakage, so a modern N-channel device (like IRF450) gate “remembers” the test voltage (+2.5 V or -2.5 V) for several seconds, maybe minutes. When you next measure the Drain to Source, you find either a short after positive, or open after negative charge, respectively. P-channel devices, like IRF9450 of course behave in the opposite way. I have used this for years in my MOSFET testing, but just recently bought one of those component identifying/testing devices that you used. Good to keep that in my mind! Another thing I noticed is that your board evidently does not have conformal coating, as you make your probing so readily. A conformal coating is a real pain for any probing. Using sharp pin extenders on the probes may help a little. On the other hand, I have seen, how an unprotected board looked when coming back from a Gulf Coast site after a hurricane. Talk about salt mist exposure! Keep on producing your educational videos….

After watching this it encouraged me to have a go at fixing my 12 volt to 240 volt inverter that had been lying around in my cupboard. When I connected the battery after I had finished it exploded spectacularly, like a firework, but at least I learnt a few things. Enjoy all you videos

How much simpler was an old school linear battery charger. Just a transformer and rectifier. So much less to go wrong.

Small tip - when testing fuses always check it by way of the contacts of the holder that the fuse is fitted into - I say this as sometimes the fuse may not be making good contact with the holder at one or even both ends, usually because of some corrosion that can't be seen.

lol you needed to run that disclaimer longer. I had to pause it to read it all to the end. I do just happen to be over 206 yrs old. That whole thing reminds me of Monty Python and the Holy Grail lol

Good save. I was staring to feel like Mrs. Doubtfire there for a while. You have been awarded 🎓🎓🎓🎓🎓 Smarty I.T. awards.

Wish I knew more about how to diagnose and repair these boards. Thanks for sharing👍

Blimey Mick, I've not heard the term Schottky Diode for 30 years! 😃 Nor seen a solder sucker for nearly as long. 😊

Well done

Loved the Disclaimer. Great video and I like the way you guide us through the process from start to finish, keep up the good work and looking forward to seeing more

8:03 gated, meter provides enough to gate them , slow avalanche

Good fault finding very logical

As soon as you said generator, my internal warning siren was going Danger Will Robinson Danger........ Tip, those heavy resistors, i'd rework those joints. Another tip, one recommended to me after working on very very expensive power supplies, put heat shrink on the pins of your probes, right down to about 2mm from the tip. We used to make our own hi precision probes, soldering a pin into a 4mm banana socket, wrap it back with heatshrink, so the tips mega sharp. Great work bud.

The best fine print that I have read in it's entirety! Of course it's the only one for good reason

The 2 series filter capacitors form a doubler on 110v input by jumpering the center tap to ac rail going to the bridge rectifier

cool description of SMPS.

That area of burnt PC board, just above the output area, and corresponding to the two green resistors on the other side was screaming at me! What was that all about? (if you can still remember) Thank you for a very nice explanatory repair!

😂😂😂 best disclaimer off all time 👍👌👏

Someone else has mentioned I think on one of your other videos, that, with a Fluke meter that has gone into standby, just press the yellow 2nd function button (and possibly any of the other buttons too) and the meter will spring to life again. You don't have to turn off and turn on again..

Well done on the repair. The resistors may also be for balancing (voltages across capacitors effectively in series), so may be worth rechecking capacitor voltages / resistor values (though with no oscillator running there are likely to be strange readings). Also well done on volunteering at a repair cafe, especially if that's the type where the customer can watch your every move.

I noticed you don't mention the charcoal briquettes that seem to be two power resistors, and I have to say I was more than a little bit worried about how many times those AC power leads got twisted around. Other than that very healthy fault finding techniques used

Cracking fault finding 😊👍👍

Once again a great video, very informative. I repair PSU's for a living and come across many with those SG3525 fitted, they seem to be a common failure

wonderful repair!

With no Vref I thought there must be something like 431 ref ! clearly not and I thought it was an input..great repair, lots of people get scared when you mention SMPS and they used to bother me but if you take a bit of time the mystery out of it I bet you've fixed hundreds !...cheers.

Great job , mate.

Another great helpful video. thank you

Great fault finding, not quite as great as your disclaimer 😂 I was sure it was going to be the big burnt looking resistors on the output! So how did you know the Vref is generated internal to the chip? Was it stated in the spec sheet or is this just common knowledge. Thanks Mick 👌

The schematic shown at 9:57 is a half bridge, not push-pull as it is labeled. The converter on the bench is clearly a half-bridge. (Half bridge uses the "centre tapped" DC supply; push-pull uses a centre tapped primary on the transformer)

CHEERS Another top video pal keep it up

Awesome 👍🏻👍🏻👍🏻✌🏻✌🏻✌🏻

Brill video nice simple explanation of a switch mode p/s.

I know this comment is very late, but I recommend using a series lamp, if it just lights up for second , u know that the input circuit works, if it shorts out and fully turns on then its at fault, or if its not taking power. On the other hand its the circuit after the transformer which is at fault. It just greatly reduces the time to diagnose which part of the board is faulty

Those poor output resistors look like they have been pretty cooked. Also some evidence of excessive heat right by the chip the failed, possibly a contributing factor. Good vid, I don't normally get so deep into an issue like that before I just give up. I might take a look at my Peter Chou arcade SMPS that failed about 20 years ago, but I've kept around just in case.

Please let me ask,the sg3525 is a pwm control circuits,could you check the output signal with a scope?

Was interested in this as I have a HP monitor 2009v that is dead and suspect the power supply. Tried to get one on line but no luck. It has ran for 14 years then just suddenly stopped. Tried to get Circuit. I have been aware of SMPS for many years but never knew how they worked. At 73 I do now. Thank you. In the meantime I got an LG monitor from CEX. The boards in the HP do not show the heating marks yours did and it looks like the switching chip has got warm on yours. I can now have another play. 😍😍

Looking at the state of the board that looks like it got really hot you can see parts of it scorched😮😮😮

I've got an expensive tagan 550w pc psu that's just turned a hard drive to a burn on my desk and now all rails read high on my psu tester, any interest to you?

You say the discharge resistors prevent you getting a shock from touching the board but would I be right in thinking their actual purpose is to prevent a shock from unplugging the device and touching the prongs of the plug because of energised caps?

wheres the local repair cafe ive never heard of one before, are you in the north east i live in sunderland and i wonder if theres one near me. how does it work can you take something in to see if it can be fixed and do they have set charges. sorry for all the questions but the repair cafe sounds like a great idea

Switch mode power supply battery. Charger repair. KT2SM.

Too much time since , but any ideas how plugging into generator - would make it fail ? Thanks agian

My father owns a vacuum cleaner Polti (model Lecologico) and it powers on but nothing happens when pressing the Start button. The motor works when applying power, the switches and cables are OK, replaced all caps and triac, don't see any damage on the pcb nor cracked joints. The main chip has nothing written on it, zero. It's an old model and sent an email to Polti telling my troubleshooting and asking for schematics, which they replied "No". Now what do I do next? Thx

dont you ever use a scope in xy mode+a iv curve tracker(ie huntron tracker type thing),very handy things,spot faults at a glance.

How many amps can this handle?

whats a local repair cafe? never heard of something like that

An active low Shutdown pin. Doesn't that mean that 0V on that pin means that it is in shutdown?

Did you notice the burnt out compont?

I told you at the start of the video "it's the controller", and now at 16:45 I may end up with egg by the end? :)

I have very good eyesight apparently, and you are missing a few words in your disclaimer ;)

Plug a switching power supply into a dirty generator will blow something. THD needs to be 5% or less.

How can I send a device to you ? No matter the cost to me.

Mosfet/igbt/power switching Semiconductor heatshink not sufficient, design doesn't have thermal shutdown system

You should explain what you are doing better

Sorry to be a pooh pooh, but leaving your solder wire so close to the work area is inviting being ZAPPED!!!!!!!!!!!!!!!