How To Repair ATX PSU. The Full Tutorial. Computer Power Supply Repair

How To Repair Any ATX PSU
Learn Electronics Repair #085
In this 1 hour video we take an in depth look at ATX PSU repair. Here you should learn everything you need to know to get started.The topics we cover are...
1. How the ATX PSU Works
2. How to recognize the various sections and components that make up the PSU
3. What Tools and Equipment you need to start your repair work.
4. PSU Repair Safety Issues
5. What Goes Wrong - most common faults
6. Real life fault examples
I have a large box full of faulty ATX Power Supplies. At the end of the video I go through all the faulty ones that I have, describing the faults and inviting you to vote for which one you want me to fix first. I intend to fix all of them of them and upload the videos. By that time you should have all the knowledge you need to repair ATX PSU. Feel free to vote for the ones you would like to challenge me to repair first, or the one with the most similar fault to your own PSU
Place your votes in the comments below, everyone welcome, the more the merrier!
Here are some links I mention in this video
PSU Repairs Playlist
• Power Supply Repairs
PFC Circuit in Detail (from 48mins 50secs)
• LER #047 ATX 600W PSU ...
ATX PSU tester review
• ATX PSU Tester Analyze...
I work in collaboration with:
The Electronics Channel (with Carlos and Detlef)
/ @theelectronicschannel
Gran Canaria Uncovered
/ @grancanariauncovered (with Detlef and Julie. Not electronics related)
Det Builds Stuff - / @detbuildsstuff8128 (Detlef)
Retro Upgrade - / @retroupgrade (Carlos)
For All Your PCB needs: free $5 discount coupon
www.pcbway.com/setinvite.aspx...
Equipment used in my videos. These are affiliate links, you pay the normal price and I make a small commission.
TEST METERS
Aneng AN8009
s.click.aliexpress.com/e/_DdW...
amzn.to/469ZgOS
www.banggood.com/custlink/mD3...
KM601
s.click.aliexpress.com/e/_DBw...
www.banggood.com/custlink/KD3...
VC480C+
s.click.aliexpress.com/e/_DEY...
amzn.to/3Q3LEz7
amzn.to/46tmNdo
MESR-100 ESR METER
amzn.to/3tiy8hK
amzn.to/3RwvejK
s.click.aliexpress.com/e/_Dmi...
XC6013L CAPACITOR METER
s.click.aliexpress.com/e/_DmX...
amzn.to/3t71Nul
amzn.to/3LDdcbO
TM-902C TEMPERATURE METER
s.click.aliexpress.com/e/_Dm9...
amzn.to/3LFzERv
amzn.to/3PTwuvW
FNB58 USB ANALYZER
s.click.aliexpress.com/e/_Dmd...
amzn.to/3tb1QW8
amzn.to/46oMn30
www.banggood.com/custlink/mKm...
PCI POST ANALYZER
s.click.aliexpress.com/e/_DBs...
amzn.to/46m9Gep
TL460S Plus PCI_E ANALYZER
s.click.aliexpress.com/e/_DnB...
amzn.to/3ZERvOs
MULTIMETER PROBE KIT KET05
s.click.aliexpress.com/e/_DE7...
OSCILLOSCOPES
FNIRSI 1014D
s.click.aliexpress.com/e/_DEz...
amzn.to/3RxMFk1
www.banggood.com/custlink/KKG...
FNIRSI DSO-TC3
s.click.aliexpress.com/e/_DC8...
amzn.to/3RBsgdS
www.banggood.com/custlink/Gm3...
FNIRSI DPOX180H
s.click.aliexpress.com/e/_DBX...
amzn.to/468oooT
www.banggood.com/custlink/GGG...
BENCH PSU
NPS3010W
s.click.aliexpress.com/e/_DDP...
amzn.to/3rxsuIf
www.banggood.com/custlink/3DD...
SOLDERING
T12 Station with M8 9501 Handle
s.click.aliexpress.com/e/_DFV...
amzn.to/3FdG8Ul
SUGON T26D
s.click.aliexpress.com/e/_DFx...
QUICK 861DW
s.click.aliexpress.com/e/_DDo...
amzn.to/3PBOKZH
amzn.to/3RCSqNh
PROS'KIT SS-331H
s.click.aliexpress.com/e/_DDo...
THERMAL CAMERA
Infiray P2 Pro
s.click.aliexpress.com/e/_Dmu...
amzn.to/457PgEb
www.banggood.com/custlink/DDG...
MICROSCOPES
Amscope Optical Microscope (copy)
s.click.aliexpress.com/e/_Dlh...
uk.banggood.com/custlink/mKGy...
amzn.to/4675Qp0
amzn.to/459uvrY
ANDONSTAR AD407
s.click.aliexpress.com/e/_Dl4...
amzn.to/3PSQKxJ
amzn.to/48sxej4
EEPROM Programming
TL866 II+
s.click.aliexpress.com/e/_Dls...
amzn.to/3RTCRB9
amzn.to/3LDu3Lu
CH341A
s.click.aliexpress.com/e/_Dk0...
amzn.to/48IZiir
If you would like to support this channel
You can send donations
www.paypal.com/paypalme/youtu...
You can subscribe to Patreon
/ learnelectronicsrepair
You can click Join to become a channel member
Thank you
Richard

[@brianterry8512]

Thanks for this! I am converting an old ATX into a benchtop supply for general mad-scientist type stuff and thought it was shot. I knew there was a 5v Standby but didn't realize it actually needed to be grounded in order for the rest of the supply to fire up. Thanks for all the good info, keep up the good work!

[@inwerp]

this is hands down one of best repair channels if not the best. ladies and gentlemen, subscribe, like and share.

[@LearnElectronicsRepair]

Thank You!!!

[@peddersoldchap]

What Tools and Equipment you need to start your repair work 34:20 PSU Repair Safety Issues 39:56 Common faults 49:37

[@markgoogolplex2572]

Thanks very much!

[@lucaskiai2159]

These videos are very informative and helpful. They are helping me improve on my diagnostic techniques and approach to electronic repairs.

[@messageobliquespe100]

Great outline of the supply components - & no doubt you’ve saved some lives too - I’ve obviously been lucky that I’m still here - so Thank-you! 👍

[@bozz3058]

i tried to do this by my self with no knowledge or idea of what i was doing so i sqrew up my corsair 500w psu then i got electric schoked so yeah... i just ordered a new one xd. thanks for the vid man!

[@LearnElectronicsRepair]

Really that is surprising. I hope you did not get an electric shock after watching my videos because that means I am not doing my job properly

[@mrrace7971]

@@LearnElectronicsRepair no i did not get it when i watched the video. I got it before and then i decided it probaly is best to watch a yt vid on how to repair it!

[@mariopic]

crosair supplys are built as one massive 12v power supply and use small dc to dc converters to step down 12v for ur 5v and ur 3.3v rails one trick is meter each rail pulse the pon wire to ground the bad rail wont out put any thing most of time its 3.3v rail if u isolate it on the supply u can still power the psu with the dc to dc card removed and u inject 3.3v to the monitor rail for the dc to dc converter this is the tx and rm and ax models do this

[@Olivia-W]

Um. All future readers- charged capacitors can be dangerous! No touchy until you discharge it safely.

[@delatroy]

@@Olivia-W yes you’re right. Can hold charge even after a long time. There’s a reason they’re hard to get into

[@DD-jk3nf]

Thanks for the video buddy! The supply you were using as an example is almost exactly the same as the one I'm going to attempt to repair, close enough that I could identify all the components you were talking about 😊 I'll be following your tutorial. I had already checked the 5v stb before finding your video, now on to the common faults 😁

[@edward20049]

Thank you very much Sir for a very comprehensive detailed instruction on how the power supply repair should be carried out.

[@hullinstruments]

I would stick up for my buddy diode gone wild and his KZbin channel and videos on repairing ATX power supplies. Maybe not as comprehensive but just as valuable. And you were both excellent teachers and very skilled at explaining electronics engineering and repair type stuff. Much love from the USA!

[@martinda7446]

Diode G W is special. Being in the industry I know he has real qualities. Check his website to confirm.

[@giuseppemangeruca6898]

Hi there, I'm not a technical expert or a monster but I love the electronic thing, and I will try to repair my power supply following your guidance,.hopefully, I hope to do it, of course, of course I don't have the perfect tools but it does matter the most important is to be careful and patient. Thank you for everything,👍ciao!👋

[@jasonlarnach6770]

Yeah your drawing was rough, but I'm a complete noob and it taught me a lot. Ty.

[@LearnElectronicsRepair]

You're welcome

[@knightwar3]

Another interesting channel about electronics, one of the best explanation about atx smps and the diagram is very useful

[@LearnElectronicsRepair]

Many thanks! And welcome to #LER

[@personaldronerepair6141]

I just purchased an HX1200i that is supposed to defective (still in shipping). This video may come in very handy. Thank You Learn Electronics Repair. Subbed ! Belled ! Yeaahh :)

[@kevinmagsy0312]

Thank u master, thanks sharing of the knowledge for atleast understanding how psu works.. so we have an idea to repair ours

[@DevilbyMoonlight]

Well presented, well explained, what more could you want - thankyou, if it were up to me I would choose to fix PSU's that had more on the 12v rail than the lower ones, more modern boards have power delivery issues with the old style atx psu's and can make the psu appear faulty when they arent- but they still have their uses even if not connected to a pc :) keep up the good work

[@darrenbird2526]

Great explanation and a perfect drawing example. Thanks again!

[@German_byte]

Many thanks for an informative and entertaining explanation.

[@todornikolov7286]

One of the best explanation! Thank you.

[@masterAilan777]

Soo cool and in depth. Thanks. Very fascinating.

[@japhetobandajunior1943]

Well planned lesson. Very informative 😊 thank you.

[@asmodeuso6]

Thank you for sharing your knowledge, sir!

[@cybergarri]

I discover your chanel few days ago and thats videos are gold, well for the future PSU repair I would found more interesting a partial failure like less output voltage

[@LearnElectronicsRepair]

Have a look at the 'hall of shame' at the end of this video - if you find one that has the sort of fault you would like to see repaired next then let me know and i will make a video for you. So far I looked at Number 3 (video to upload tomorrow and it was a tricky one!) and next will be number 4 for another one of my subscribers

[@Nick_R_]

I've been trying to get my old Olivetti PCS 11 back to life. After a period of not being used, it fired up fine, then in the middle of a game of Elite it stopped and won't power back up. Your excellent video has taught me that I know a bit and now a bit more, but still not enough to fix the PSU. It's propietary and tiny, so I have no replace options either. Are there people who repair 1992 ish PSUs?

[@prmayner]

I love you man. NOW I GET IT, Such a clear explanation even a caveman can understand it. Thank you,

[@patrickmegan8224]

Hi, this was a cool video. I learned a lot from it, especially the safety section, these things can be dangerous! I'll vote for 1 and 6 which had no output and #13, the unknown soldier.... I'm keen to see how you diagnose this and how to get at some of those hidden parts.

[@LearnElectronicsRepair]

Hi Patrick. Number 1 is fixed kzbin.info/www/bejne/mGWTmGaEed5so68 I'll get a look at number 6 next for you

[@FirstLast-tx3yj]

@@LearnElectronicsRepair what is there other than capacitors after the conversion is made to 12v DC in order to have all the atx connections? Because I can get a power brick that gives me 12V and satisfies my 100w need but I need to convert 12V positive and negative to a full ATX plug Like the pico PSU but tjose are expensive so I want to adapt my own and know what is in the circuit other than the capacitors!

[@blackhorserepairs]

Excellent video, thank you!

[@markojovanovic6021]

Awesome video! Great content on the channel.

[@johnvanwinkle4351]

Thank you for the great information! I never knew how they worked. I have a couple of nice ones I need to repair....grin

[@smada36]

This is excellent. I built my sister a PC about five years ago and a year ago it stopped working. To fix the PC I exchanged the PSU. Just for fun, I thought that I would open up the broken PSU to see if I could learn anything and maybe fix it. What are hobbies for? Firstly, the power supply was full of dead cockroaches. I noticed a MOSFET with a large chunk out of it, and beneath the board, a black mark with a cockroach blown to bits. I'm assuming that this little guy was looking for somewhere warm to sleep and got more than he wanted. Effectively the high side and the low side MOSFET's were closed at the same time? On a component tester that MOSFET shows up as one resistor, and the one next, and connected to it shows up as two resistors, so I have replaced both. However, the PSU is still dead. Further investigation shows that two pink, resistor looking components, with a matt coating, have open circuit level of impedance (Beyond what my meter can measure). One next to the MOSFET, and a smaller one in the middle of the board that can be traced back to the MOSFET. I have assumed that these are fuse resistors and have ordered what I can find to replace these. Not many people seems to talk about fuse resistors in PSU's, but this has six, four of which have very low resistance and seem to tally with the coloured bands. It is an EVGA 650 GQ, but sadly schematics are not easy to find.

[@mohamadhegazy-ef2pz]

Thanks for all the good info

[@bradleycooper3523]

Great video, would love to see the number 8 psu repair video. My one has a similar fault id love to get to the bottom of, the fan spins for a moment and then nothing but the 5v is working great.

[@JCboss29]

Incredible, thank you so much!

[@bobflyer4346]

Excellent set of video's and thanks. On the safety issue, what do you use for Isolation transformer? Recommendations....TNX again

[@cactusstudiossouth]

HELLO LER! Am an avid Subscriber, you are my first 'Go to' when diagnosing PC issues! I recently built a new Computer in my old Tower with a new Raidmax Vampire 1000W 80 Plus Gold Power Supply, an MSI Z390-A PRO Motherboard with an Intel Pentium Gold Processor G5400 CPU @ 3.7GHz! Next added 64.0 GB DDR4 3200 Ram! Also a ZOTAC GAMING GeForce GTX 1 Video Card! Recently upgraded all my HHD's with an Intel 660p Series M.2 1 TB SSD Drive C and two WD I TB internal SSD's! My Computer now boots up completely for a cold start in 30 seconds!!!! (Once in a while when I turn it on, it Boots into the Bios Screen)!!!(But when I wont to boot into Bios, no matter what key combination I press it wont go into Bios on Command)! Windows will not display on my Digital Monitor but I can see the Bios Screen on another HDMI Output to my 32 inch TV! Am running XMP Mode and Gaming Mode on the MSI along with allocating 3/4 of my Ram to Process my Adobe Premiere 2022! My Question is: How do I get the computer to Boot into Windows EVERY TIME? And secondly, How do I get my computer to boot into Bios on command?

[@marcuswilliamobrien]

No 7 or 15 if I'm not too late. Your too quick, I'm watching it now. Thanks fot the great info.

[@LearnElectronicsRepair]

Haha I do my best LOL

[@gameofquantity96]

Sir please make more videos They’re so helpful

[@learngamecheating]

nice explanation Richard

[@LearnElectronicsRepair]

Glad you liked it

[@notesanom9546]

Good tutorial fixing psu pc thanks sir

[@LearnElectronicsRepair]

Most welcome

[@mr.fixcamer]

So good, thank you

[@LearnElectronicsRepair]

Glad you like it!

[@enriconapolitano4321]

Hi, Good tutorial, Thank you a Lot

[@LearnElectronicsRepair]

You are welcome!

[@morto360]

Great video!!!! It was really well explained and it looks like they are much easier to repair then I thought. You did not mention anything about isolation transformers... because some people say they mandatory when working with power supply's. Can you comment on that? I assume your lightbulb is "kinda" doing a similar job by limiting the current?

[@freda5344]

This guy is suggesting anyone can fix a psu, what about when you need to check if its switching properly with and oscilloscope. Have a shot with your best (or not) drink, and a drinking game with every "yeah", you'll be dead flat if you even reach the end.

[@windward2818]

If you look at electronic repair taught at community college level in the USA, the PC switching power supply repair falls under consumer goods repair main powered with a hot chassis design. Which is very dangerous to both the technician and test equipment. It is usually prudent to where eye protection when you testing and have the cover removed during initial power on. I have seen supplies blow up on the bench, both a capacitor and a transistor which actually launched metal and plastic shrapnel. Generally, on a consumer repair test bench you have an isolated variable AC power supply CB (circuit breaker) protected on the output and a current limit function, there are usually two AC Ammeter current ranges and an AC voltage meter, I prefer traditional analog meter movements for this type of supply. If the bench and bench equipment is earthed (which is usually the case for personal safety) then the earth ground is also isolated. In the USA we have a version of the mains receptacle which has the ground isolated (orange type), so if it is attached to a chassis the chassis is not strapped to ground. Two resettable circuit breaker (hydraulic magnetic type, I like Carling Technologies) ranges are generally supported (like 2A and 10A switched MBB with a surge characteristic), and current limit function is provided. You can use a filament light bulb to give the current limit function or you can just use a large low inductance power resistor on a heat sink. Having a variable AC supply using a boosting variac is very handy because you can ramp up the AC voltage to see how the DUT reacts, and test at high line (135 VAC in the USA). Unfortunately, repairing AC consumer goods safely is not cheap. My approach is to use a battery powered DMM (which is very common) and I also use a Fluke Scope-Meter for initial testing, which is also battery operated and both channels are isolated from each other and from earth because the instrument is floating. Once I narrow down the fault I can use a more expensive bench meter with differential high voltage probe if I need more bandwidth or I want to get a high resolution capture to send to somebody or use in training material. I like using the PicoTech 5000 Series (again not cheap) over a expensive scope from Tektronix or Keysight. The other go to testing device is a thermal camera (also not cheap). If you apply load(s) to a repaired PC power supply and let it stabilize you can understand a lot about the design by looking at its thermal performance under load. You can also spot poorly selected high stress capacitors because the increase in ESR will product a lot of heat, and if so you can replace with a better quality or higher value to reduce the heating due to ripple current. If you actually work on a lot of these types of supplies it seems, in my opinion, that the bridge+input capacitor(s) are always under designed. Other hot spots would include semiconductor power devices and diodes. You should also have a capacitor discharge load with clip leads or probes to discharge high voltage capacitors before de-soldering.

[@johnmcgiv1]

Hi excellent video. What would cause no 5v stby voltage. Many thanks again.

[@allenyoung9092]

Number #4 outstanding...

[@LearnElectronicsRepair]

Hi Allen - I've started the video on No #4 then I started to do some motherboards as I felt I had done a lot of PSU recently and try to keep the video subjects varied. I'll get back to that one on Monday and see if I can complete it (or at least diagnose the fault and order parts if needed, in which case I will upload a Part 1 video

[@allenyoung9092]

@@LearnElectronicsRepair Thank you I really like your chanel and how you diagnos the electronics before repair again Excellent!

[@jjdigitalvideosolutionsllc5343]

Your videos are a great learning aid. I'm still trying to digest it all. I have a micro-ATX with a -12v line that reads -11.5v. That doesn't seem too far off, but it won't boot my old Amiga 4000 anymore. I compared levels to those from an original Commodore PSU that will boot the Amiga 4000, and the only significant difference is that -12v line. The original PSU is reading almost -12v on the nose. The problem with the original PSU is it's only 145W and when I load up the Amiga with all the expansion cards and drives that's underpowered. The micro-ATX is 350W, plus the smaller form factor makes cable management in the cramped Amiga case easier. Can I bring that -11.5v closer to -12v? is this an aging capacitor issue? Or should I be looking for a different problem/solution? Thanks.

[@whoisntwhoisit2126]

This channel is all you need to know about Autoplay, once you are on this channel, you will never see autoplay work again! (At least this is my experience, 90% of the time I end up on this channel and it never leaves once there, just plays this channels video's forever. Really infuriating!!!) Went from enjoying the channel to cursing it every time it comes on.

[@dev639]

Awesome video! I am trying to fix a EVGA Supernova 850 P2, it has a fried ceramic capacitor that burnt of it's identification on the board and the transformer seem to be shorted. My issue is I can't find any info anywhere as to which components compose the board and I also can't find schematics for it. How do you guys find these things out? I mean measuring those things isn't helpful because they are toast. What's my next step?

[@lukedavis436]

I might attempt this on two cheapo Power supplies (one of which blew a capacitor giving me the fright of my life) at least to bring them back to life for a short while, one is Zoostorm branded whereas another is unbranded

[@caine0729]

youre wxplaining this to someone who after 5 years of learning electronics , you would give him one last lesson

[@allen5976]

Thanks for the very informative video. I've tried to repair a psu that gives out 3,7 and 12vdc. I measured the outputs and all seem fine, but the tech comes back telling me that it is still faulty. It seems to be unable to maintain the voltages under load. What would be the component or area that I'd need to investigate further? Thanking you in advance and a merry Christmas from Galway, Ireland

[@mohammadkhlaifat1651]

many thanx what the component which indicate the power 300 w or any value else?

[@LeCommeiLFaut]

Awesome Man, Awesome! Best video on KZbin! Btw, could you advise, I have Thermaltake Toughpower 1350w 80 plus silver. Big psu, 12 years old, and works like clock. But 3.3v rail is lowered to 2.9 , other rails are good (12v 12.280 , 5v 4.60 , 5aux 5). Only 3.3v is lowered to 2.9. Could you advise could it be false capacitors or something else, and overall worth it to repair? Btw it works about 4 years at 2.9v (at 3.3v rail) 😄, and I think maybe repair can be not so expensive or..

[@peterfindlay5820]

My vote is 12 as I have a ps with both 12 v + and - low @ ~ 11.5 and -10.5 approx. it’s a Shaw 680w Already checked all the electrolytics all ok

[@Wolf-yo1gx]

No. 12. Will be interesting.

[@LearnElectronicsRepair]

Hi Wulfrano, I will do that one next

[@mesterak]

16, 2, 3, 5, 7 are my picks

[@bobyk87]

If the input thermistor burns, I replace it not putting in a shrink insulation on it, and especially if it may sit right on the middle of the cooling fan "luft", may it not want to lower its 10R and cause trouble? In any case 10R is almost a short, but @1 amp (1A*220V=220W, light PC load) there is I²*R=10W or 2 amp (440W) gives I²*10=40W, so actually 10R is a lot on the primary side under load! My thought: make sure that NTC keeps warm at steady operation.

[@ronelconci817]

its clear

[@dababroad]

Hello Sir, very informative video, I am currently modifying a power supply and noticed that one of the 3,3 V wires is connected to a brown wire, which senses the v output. If I want to use the ATX for a 3,3/5/12 volt supply should I als connect this brown wire or can I cut it loose? reg Dirk Bosman

[@shokdj1]

I've managed to locate the problem thanks to this video. Only problem is I can't find the part anywhere online that's shorted I think it's the 12v transistor on the high voltage side of the board. P13009a

[@shokdj1]

I found a pretty close equivalent if anyone comes across the same problem D13009K

[@iyaramonk]

I'm curious about the capacitors that can hold a charge big enough to kill you. You would have to touch the actual circuit right, it wouldn't be enough to touch the capacitor? I'm asking because I want to clean my psu fan, but the connector is quite close to the capacitors so I'm not sure I could reach it without rubbing against them.

[@peddersoldchap]

Hi Richard, would a 40W "old technology" light bulb be enough for a current limiter?

[@shokdj1]

I'm glad I watched the health and safety part I didn't realize the heat sync is live

[@mobiusm7440]

Great tutorial. I am trying to solve a problem. My power supply was subjected to a short circuit on one of the two 12 volt outputs, 12V2. Something has failed, 0 volts output. 12V1 displays 12.1 to 12.3 volts. I have no schematic. Is there a most probable failure point under short circuit conditions?(the sc resulted in a small fire in the computer, filled the living room with smoke).

[@LearnElectronicsRepair]

Can you see which parts have burnt? If so email me a pic to [e-mail]

[@miloszboher7431]

Hi, great tutorial. Can you advise me on my problem. I had a blown fuse. I replaced the mos fet in the pfc part. I now have a voltage on the main capacitor of 350v. I have 5v sb voltage. When i connect the green wire to ground the fan wants to start but calms down. I have no output voltage 12, 5 or 3.3v. the output diodes are good. Where to look for error. Thanks in advance.

[@LearnElectronicsRepair]

You need to look at whatever PWM controller is driving the main switching transistors. The fist thing, is the controller on the low voltage side or the high voltage side? If it is on the low voltage side the PSU will probably have 3 transformers in a line (one larger two smaller) and one opt-iisolaor Check out his video it may help you to diagnose your fault kzbin.info/www/bejne/eZ3Ti5qBe9CNkKc

[@marshallcalder247]

No. 12 & No.14 ?? Please?? If/When you come to these? Okay then Whichever Comes up?? They"re great videos to watch - by the by, so... Thanks. Those bits still in the ether begin to make sense. Started out in electronics as a hobby at around age 12 but never built any circuits with components. The Old Man brought home a kit box for beginners - the old springs terminal component box with connector wires and circuits to build and try out but but keen as I was , couldnt make any work and eventually lost interest. Now I'm 60 and want to make good on my interest in electronics. I will in time.

[@alainvidmar1277]

Hello friend. Nice video for troubleshooting. I have one issue at my ATX PS. I did not realize at your hall of shame. ;-) I switch on PS and working, but if I conect any load on 12V rail it turns off. What could go wrong with this?? Where can I find the problem. First look inside everything looks fine. Can you help me about that??

[@TheMono25]

Well I soldered a trace on the pcb that had burned and broke connection and tried the green and black wire and the fan came on this time I just need to test the pinouts and make shure there right before I connect to the motherboard

[@technoperu]

Greetings, thank you for your knowledge, I have a problem with a power supply, I hope you can help me... I have a broken ax 1500i power supply, the plug melted due to too much load... the fuse looks like a black resistor wrapped in plastic, it is open, it no longer has continuity. I took it out but there is no replacement for this fuse.... what fuse and how many amps can I put in when I finish repairing it? I hope for your wise advice, friend... knowing that this source delivers 120 amps in total with 100% work, what do you recommend? teacher ....

[@celicdusan]

can I change output capacitors with higher capacitance caps ? (For example on +12 I have 1000uF/16V , but replacement is 2200uF/25V )

[@miosz8531]

When PSU turning off during load have i check all capacitors? Whad do you think, how to find issue?

[@pablosoto5407]

Did you talked about resonant stages of some atx psus in some of your videos? seems like this relative new technology give us a lot of tricky to find issues in this psu designs. Regards.

[@VainasDominicanas809]

My corsair cx750 fountain unfortunately flooded by accident when it was turned off, but connected to the power strip, that is, when the water flooded the room, the power strip activated while everything was off, I put the fountain to dry for several weeks, then when I wanted to turn it on, it no longer I lifted it up, I checked the secondary part of the source, and it doesn't have a short, the purple cable measures 4.5 volts, not 5, the green cable measures 2.4 5.6 6.6. and 3.3 volts, bone varies in the measurements, and when I make the bridge the source does not start, I did the oscillation test and if it is oscillating, the transformer hums when it wants to start, but it does not happen, what could it be? Thank you

[@farben_]

Another great video, I have a Corsair hx620w that only works for one day and it works great but then over the next days it doesn't work anymore, almost certain it's the big cap and one of the smaller caps looks a bit inflated, debating if it's worth to repair it since it's not very easy to find another 470uf 400v Hitachi capacitor.

[@LearnElectronicsRepair]

The large capacitor is not so important to have a low ESR, also they don't tend to fail unless they have visibly leaked or corroded. The smaller caps on the output (low voltage side) are more likely to cause this sort of problem because they are dealing with high frequency on the output of the switching transformer. Is the problem that the PSU comes on some days and will not power up on other days? When it does not power up do you have 5V STB or not?

[@farben_]

@@LearnElectronicsRepair My multimeter is missing the probes, they broke when travelling, it's on my parts list to buy. How do you go about testing a 400v cap from a PSU? It's one thing to test those regular 2000uf 16v caps but 400v is too dangerous for a novice to be handling, can you test it indirectly like probing the 24 pin connector with the multimeter? The other components at least visually look normal, no burned marks. When I get the probes I'll check the voltage values on the ATX connector to see if I get some clues.

[@LearnElectronicsRepair]

@@farben_ Hi Farben - you test the 400V capacitor same as you would test 16V capacitors - with a capacitor meter, and ESR meter if you like. You don't need to apply high voltage to test it, unless you wanted to test it can handle it's rated working voltage. Just make sure it is discharged first.

[@farben_]

@@LearnElectronicsRepair Thanks that was what I wondering! Guess I really need to get an ESR meter, too many issues I'm having with my old computer parts are around the capacitors, seems like a good investment in the long run. I don't want to damage the capacitor so I'll try to get a resistor and discharge it that way. The PSU has been unplugged for months now.

[@LearnElectronicsRepair]

@@farben_ Actually you can check the capacitor using a power supply, a resistor and a multimeter. A resistor and capacitor in series have what is called a Time Constant (TC), which is the time it takes the capacitor to charge to 63% of the supply voltage. The power supply could be a 9V battery for example (measure the actual battery voltage to get an accurate value) The resistor should be something reasonably high - 10K or more The formula is TC (seconds) = R (ohms) x C (farads) but you don't need to work it out yourself. You can use this website and enter the capacitor, resistor and voltage values then it will tell you the TC www.mouser.co.uk/technical-resources/conversion-calculators/time-constant-calculator Connect the capacitor and resistor like this (making note of the correct polarity, power supply +ve to capacitor +ve 9V--------/\/\/\/--------||------------0V Now connect your multimeter across the capacitor, make sure the capacitor is fully discharged. Now turn on your power supply and measure the time it takes to charge to 63% of the supply voltage For example with a 470uf capacitor and a 22k resistor the time constant is 10.3 seconds (so 10 seconds is close enough using a clock or watch) For a 9V supply the target voltage on your multimeter is 5.67V You can easily calculate this target voltage for any supply voltage, the formula is Target Voltage = Supply Voltage / 100 * 63 You should be able to do this in circuit for your main smoothing capacitor Have fun

[@gloubiboulgazeblob]

Hi LER, I'm reading "The Art of Electronics" (3rd Edition from from Paul Horowitz). Does the "booster" you talk about in the beginning of this video (PFC part after the bridge rectifier) works like a "synchronous boost converter" seen in the book in the first chapter (Figure 1.52B, if you have the the book) ? In the book, there's the DC in, then a capacitor (directly to ground), followed by an inductor, then a "switch" (50% duty cycle, to ground), again followed by a second capacitor (to ground as well), which in turn provides Vout = 2 * Vin. Is that the kind of "booster" we found in ATX PSU's primary stage ? I assume the duty cycle is dynamically tuned by some feedback controlled by an IC (for instance, a CM6805BG in my ATX), right ?

[@bobbyy8711]

Hi Richard I pick #12 and 14 on the Hall of S, that is what I experience from the one I have. Oh and also if you can include one those 5V unstable ones, thanks so much.

[@LearnElectronicsRepair]

OK Bobby - I just finished a dead laptop (video later today) so I'll look at some more PSU now - No 12 is next (already voted) then I'll get on to yours.

[@bobbyy8711]

@@LearnElectronicsRepair Thank you so much. I am now watching "LER #044 Laptop LCD LED Screen Repair - Back Lights OK but no picture" I need to learn great things from you, I am mostly at home now after losing a job last year due to the pandemic. My child's Lenovo G40 has white and black lines on the display, hope to have it fixed someday.

[@LearnElectronicsRepair]

@@bobbyy8711 Hi Bobby, with the lenovo laptop, go to www.badcaps.net/forum/index.php register for free and post your question plus some pics in the 'troubleshooting laptops' section. I am sure you will find some specific answers to your problem. I wish you luck with your search for work, I truly hpoe things improve soon. I'm also affected by the pandemic as is the whole island where 80% of our GDP is directly or indirectly related to tourism. And In the south of Gran Canaria where I live that is almost 100% From google: What is the main industry in the Canary Islands? tourism industry The main bulk of the Canarian economy centers on the tourism industry, accounting for up to 80 per cent of the gross income of the archipelago. With one of the most favourable climates in the world, the Canary Islands attract millions of tourists every year. So you can imagine how tough things have been here.

[@tanchienhao]

Due to the PFC, my current limiter bulb keeps blinking. Is it safe to bypass the PFC for testing? I intend to desolder the PFC inductor

[@bobbyy8711]

Hi, great tutorial here, so happy I found this. Can you help me? I replaced three bulging caps in my psu board, after replacing them the tester reads only 11.7 to 11.9 volts and not 12 as before. Is this just in the beginning or should I continue using it? I'm afraid it may affect other parts. First time I see a very interactive youtuber, you answer most of the comments which is very healthy. Thank you and greetings from the Philippines.

[@LearnElectronicsRepair]

HI from the Canary Islands. :-) The voltages on your PSU are within tolerance of the ATX Specification which is +/- 5% on a voltage rail. I wouldn't worry too much about them. www.lifewire.com/power-supply-voltage-tolerances-2624583 Do you have a load connected to the PSU or just the tester? If you have no load try connecting a couple of HDD and then see what the voltage is, or better still connect it to a PC and use your multi meter to check the voltages. Almost every part of your computer has buck regulator circuits (VRM) to reduce the 5V and 12V to low working voltages such as 1.8V, 1.5V, 0.9V etc and these circuits will produce a very stable output voltage regardless of the supply voltages from the ATX PSU.

[@bobbyy8711]

@@LearnElectronicsRepair thank you so much, this is great news, stay safe sir..

[@omarjaysultan6293]

I am experiencing the same problem in my Corsair VS650, it has an output of 11.8v and not reaching 12v. And I notice that the fan is not spinning fast, i replaced it with a new fan which i had tested b4 replacing but still it spins slow. Should i continue to use my psu?

[@theoneohmresistor]

@@omarjaysultan6293 it has a temperature control circuit for the fan

[@SuspiciousPixels]

Comprehensive video even though you say yeah a million times.

[@timmarkham522]

I’m pretty sure it’s a Tourette’s tick or something of the sort, it seems subconscious and uncontrollable

[@louf7178]

Why do power rails (5V, 12V) get weak under load? My guess is the feedback to the PWM (besides any piece will effect almost everything).

[@chau7825]

How did you fix number 4? Thanks.

[@RandomTorok]

I have a very different psu problem. Brand new, Corsair, 1200 HDX. Turns on fine, but blows every motherboard I connect it to. Corsair refuses to respond to my emails. I have one of the power supply testers that you showed but it shows the power supply as being fine.

[@CollosalTrollge]

I have a cooler master and inside are 2 huge capacitors, one of which is leaking electrolytic fluid, swollen at the top. PC does not turn on. When i bridge Green and Black and switch on PSU the fan does spin ok, but when i try again the fan does not spin, it is intermittent. Can a capacitor cause this?

[@DojuMelody]

hi Sir! It's a great and instructive video! I have a question! If you have time, please answer me. My PC power supply (Mod. H305P-02). came on an unfamiliar over power power supply burned it. So when i turned the power supply, some 3 things burned when i saw it, Fuse, 16v 2200uf & DVR. And those things i replaced it(i changed things with his own behavior rightly. Then power supply it's worked even the green light is ON. & i checked all the output voltage & it's all output volt is ok (exact)! But when i put (plugged) it with a PC it's the pc doesn't start-up, even that the green light is lost. Then i pulled it out and checked it again it's ok it's work! when? without pc. so what can i do! really i confused so much! Please Sir, if u can & u have a time pls help me or give me same advice 🙏 Thanks!

[@LearnElectronicsRepair]

Check to see if you still have 5V standby on the purple wire when the ATX is connected to the motherboard

[@carterwheat7644]

my vote is on number six

[@DIYerGuy]

Great video on the ATX PSU! There's a lot going on inside these PSUs we take for granted. Here's a question for you that has me wondering: I have a Version 2 ATX PSU that uses a 24 PIN connector (by the way, the V2 ATX does not use a -12v). When it's connected to the motherboard (an Asus), it will not turn on. Removing all the power connectors and testing the PSU with a Thermaltake Dr. Power II PSU tester (v2) all power such as +12v1, +12v2, +5v, +3.3v indicate OK on the tester except for the +5v standby which indicates Fail. So on the face of it, it would appear the +5V standby is the problem. One can either try and repair the PSU of course, or simply replace it. However, I also did the "paper-clip test" which allows you to start the PSU to see if it operates. Using the paper-clip test the PSU does, in fact, operate, and after I got the PSU running I used a voltmeter and tested all voltages. All voltages, including the +5v standby are present and are within spec: This is in direct contrast to testing the PSU using the Thermaltake ATX PSU tester (which gets good ratings). So, the bottom line is that using the PSU tester indicates a failure on the +5v standby - which would tend to indicate the reason the PSU and PC will not start, but using the paper-clip test which gets the PSU running, all voltages, including the +5v standby are present. Now one thing I should mention is that the Thermaltake Dr Power II PSU tester does not put a load on any of the PSU voltage rails - it just tests for voltage and if they are present it checks that they are within spec. Can you comment on these observations and perhaps suggest another test? By the way I do have a new PSU (an exact match) on order but have not yet received it. I'd be very curious to your opinion on the discrepancy of the Thermaltake ATX PSU tester saying there's failure with the +5v standby power, and the fact that when I measure voltages with a digital voltmeter, all voltages, including the +5v standby, are coming up present and within spec.

[@LearnElectronicsRepair]

Hi Peter - I think this just proves you can't always trust your test equipment - this has caught me out a few times as you will see if you watch my videos. It would be interesting to see how your ATX PSU tester grounds the green wire internally, on mine it is just soldered directly to ground. It sounds like yours is maybe using a slightly more complex method that for some reason it is not starting your PSU - o something stupid wrong with the 24 ways ATX socket on the tester?

[@DIYerGuy]

Yes indeed. I had to get the PC back to operating, so I could not spend the time trouble-shooting the PSU. I previously ordered a new replacement which I was fortunately able to obtain locally. For comparison, before installing the new PSU I tested it with the Thernaltake Dr. Power II tester and as expected, all voltages, including the critical +5vsb, tested perfectly OK and there were no failure indicators whatsoever. And again, I tested the original (which showed a good +5Vsb using a voltmeter) and the Thermaltake Dr. Power II still showed an immediate FAIL on the +5Vsb. To my relief, after installing the new PSU the PC came back to its former glory. I suspect the issue with the discrepancy between the failing +5Vsb uning the PSU tester and the "good" reading using a voltmeter might have to do with loading: It might be that the original PSU which tested OK on the +5Vsb using a voltmeter had to be loaded down more to bring it to a failure condition because, as you know, a voltmeter will provide hardly any load. In fact a really good voltmeter is designed to have an internal resistance much greater than the minimum 20,000 ohms/per volt, so there's really no load on the source when testing its voltage with most voltmeters. I suspect the Thermaltake Dr. Power II is putting more of a load onto the +5Vsb power rail sufficiently to load it and that load - what ever it is in ohms - brings it down sufficiently to cause it to FAIL. I am just guessing about that, and there's no straightforward way to test for that since there's no schematic available for the tester to see what's going on internally (though I could measure the resistance to ground on the +5Vsb pin on the PSU tester and see what the resistance is). The Intel ATX Specs, unless I missed it, do not specify minimal load values for the +5Vsb rail either. In any case, I found your video quite interesting and worth the time to watch, so I have subscribed to your channel.

[@sl45time]

Thanks for the explanation! 1) I once had a power supply few years back. When I power on for the first time, it will start up, and then shut down. When I power on for the second time, it will start normally. Is it possible for you advise where is the problematic area? 2) For my current computer (using for ~9-10 years Seasonic 650W PSU), it is complaining that it cannot boot up with my m.2 SSD. After removing the SSD, it can function for a short period of time, and then start to feedback that 1 of my HDD is faulty. So, I replaced with a good HDD. However, also after a short period of time, it is showing problem as well. I have replaced the PSU since then. It is possible to advise where is the possible fault? Or how shall I proceed with the trouble shooting?

[@LearnElectronicsRepair]

Not sure about either of those problems - I think it is something I would need to see in front of me to have a clue what it could be

[@mohamedkikia7338]

Check the capacitors as a start. Check if there are any with bulging tops and replace those.

[@coooool221]

Hey there, I have a NPS 875BB the problem is that sometimes it turn on. i don't know what is the problem. the first time i have this problem i just clean it with some air pressure and it works fine now it randomly turn off and don't want to turn on.

[@waelelkholi3406]

Hello, Thank you for the effort you put in teaching us how things work :). I have a 1600 watt PSU that does not output PGO. I thought is is a bad chip , I mNged to get an equivalent, But when i started measuring the inputs the power supervisory chip i found that PGI signal is also zero . couls that still be a bad chip? if not what supplies the chip with PGI. where can i get schematics for this power supply. I checked badcaps but i got lost :) there are so many types of power supplies :) the chip is ( 1. HY510N WT7510 ST9S313 PS113 RT7501 ) your help would be appreciated . Thank you in advance. The brand of the psu is SURMA, 90 plus gold

[@LearnElectronicsRepair]

Hi Wael If the PGI signal is also low (and there is no short or low resistance from PGI to ground) then I would suspect the supervisor chip is doing exactly what it is supposed to do. Try tracing PGI t osee where it comes from. Regard the schematic, I don't have it. I would suggest you actually post in badcaps and ask if anyone has it (in the PSU repair section) www.badcaps.net/forum/forumdisplay.php?f=32 They are a very friendly bunch there so with a bit of luck someone can help you.

[@waelelkholi3406]

@@LearnElectronicsRepair ُ Thanks, I will do that. I actually searched badcaps could no luck. Without a schematic it is very difficult.

[@robbievandewetering8636]

number 3

[@sikandarbaloch7086]

Yeah 👍

[@riversedge64]

Hello! Awesome tutorial. I have a Fractal FD-PSU-INTEGRA-500 power supply which is stopping to work after minute or two and next he doesn't want to work for some time (he needs to cool out?). For sure there is not a problem about capacitors (I've replaced them). So do You suggests that one of transistors are bad and goes warm and makes the short circuit? Or something else? Would be cool to diagnose the problem because I'd like to repair this PSU instead of replacing it. Thanx in advance! All the best :)

[@LearnElectronicsRepair]

Get some freezer spray - that should help you pin down the problem to one area of the board or even the faulty component www.aliexpress.com/item/1005001317666429.html?src=google%2Chttps%3A%2F%2Fwww.aliexpress.com%2Fitem%2F1005001317666429.html%3F_randl_currency%3DEUR&src=google&src=google&albch=shopping&acnt=439-079-4345&slnk=&plac=&mtctp=&albbt=Google_7_shopping&gclsrc=aw.ds&albagn=888888&ds_e_adid=483620149093&ds_e_matchtype=&ds_e_device=c&ds_e_network=u&ds_e_product_group_id=539263010115&ds_e_product_id=es1005001317666429&ds_e_product_merchant_id=109055844&ds_e_product_country=ES&ds_e_product_language=es&ds_e_product_channel=online&ds_e_product_store_id=&ds_url_v=2&ds_dest_url=https%3A%2F%2Fs.click.aliexpress.com%2Fdeep_link.htm%3Faff_short_key%3DUneMJZVf&albcp=11758679187&albag=116841987329&isSmbAutoCall=false&needSmbHouyi=false&gclid=Cj0KCQiA-K2MBhC-ARIsAMtLKRtb4HgzO6-JFM-OOJG-WXre-K9L-4KYgkjIiPnGp2zj4ugPtm0pXZMaAmxVEALw_wcB&aff_fcid=a2b61f4fd05b4b6185ad0a6fc7f77850-1636577057355-05248-UneMJZVf&aff_fsk=UneMJZVf&aff_platform=aaf&sk=UneMJZVf&aff_trace_key=a2b61f4fd05b4b6185ad0a6fc7f77850-1636577057355-05248-UneMJZVf&terminal_id=73afdb6de87c46a39ebe66cc27dbf49d

[@riversedge64]

@@LearnElectronicsRepair Thank You for Your answer. Actually the PSU stopped to work completely and I gave up with repairing but I'm working now on another repair of old computer where machine gets bugging after 1-2 minute of work, so it's definetly one of chips getting warm, so the freeze spray may will help me with finding which chip exactly is getting wrong.

[@mariopic]

wondering do u offer a repair service ?

[@godfree2canada]

my PSU's 5v[red] is drooping with a ripple and 1.1khz according to my multimeter. itll droop to 4.4VDC any fix ideas?

[@LearnElectronicsRepair]

Here is the Hall of Shame vote for whatever you would like me to repair next *** Updated 28/09.2021 ATX PSU Hall of Shame No 1 - 300W - Dead Voted by me: Video Here: kzbin.info/www/bejne/mGWTmGaEed5so68 No 2 - 300W - 12V high, 5V low No 3 -500W - No Output Voted by me: Video here: kzbin.info/www/bejne/bH-ymKWdlpqrZpY No 4 - 400W - Powers on for 2-3 seconds then shuts down Voted by Farben: video here: Part 1. kzbin.info/www/bejne/i3bSc6Ompap1npo Part 2. kzbin.info/www/bejne/en7TYaetj8-ahK8 No 5 - 250W - 5V & 3.3V unstable No 6 - 400W - Dead kzbin.info/www/bejne/gnLJfXaVjbVqp5Y kzbin.info/www/bejne/sGjPnoZ_rrqLn9E No 7 - ???W - 5V Standby Unstable kzbin.info/www/bejne/maCplWSLpdyobs0 No 8 - 300W - powers on and straight back off again No 9 - 350W - No Output kzbin.info/www/bejne/e6Cch3ucjNFlpJI No 10 - 400W - 5V Unstable kzbin.info/www/bejne/haOtZ5Jpi7uBlZY No 11 -400W - 3.3V too high kzbin.info/www/bejne/eXPXmnegaraMptk No 12 - 240W - 12V too low - voted by Bobby Y PART 1: kzbin.info/www/bejne/ZqfKaIp-btZ6aqs PART 2: kzbin.info/www/bejne/mXalmolvaNxjaKc PART 3. kzbin.info/www/bejne/moPYfnp9gtycfc0 No 13 - 450W - ??? Mystery Fault ??? No 14 - 420W - 12V too low kzbin.info/www/bejne/oXfPkHyZjbx6b7M No 15 - 450W -5V Standby Unstable kzbin.info/www/bejne/maCplWSLpdyobs0 No 16 - 470W - 5V too low - same PSU as No 4 - so I am repairing these together. See No 4 Videos

[@jamesmackie6641]

Where would you recommend for purchasing components

[@LearnElectronicsRepair]

I get a lot of my components from AliExpress and they work fine. I also salvage as much as I can. As long as you keep our components very well organised, a large stock of salvaged parts is very useful. I think you may find fake or bad components if you are trying to buy something rare or valuable, but for generic spare parts and component collections it all seems to work just fine to me - as in 'things stay fixed when I fix them'.

[@kinkykane0607]

I don't suppose you could do the same video but for "AT" PSU if you ever have the time ?? :) I have been trying to follow along but ATX PSU and AT PSU are quite different.

[@mr.fixcamer]

Number 3

[@LearnElectronicsRepair]

Hi, Number three is here No 3 -500W - No Output Voted by me: Video here: kzbin.info/www/bejne/bH-ymKWdlpqrZpY

[@SteveMcQueen-zp7uu]

power supply #3

[@gasparreis303]

I bought a Nox Urano II 630w atx power supply, but it never turns off, with the fan always running without being connected to any motherboard, I think the green Power On wire is short circuited with the black wires? How can I solve this problem ?

[@LearnElectronicsRepair]

58:20 - The hall of shame... I have 16 faulty ATX PSU to repair. Watch the parade of dodgy power supplies and let me know which ones you would like to see repaired! The one with most votes gets repaired first!!

[@farben_]

Number 4 for sure! Since it's similar to the issues of my Corsair PSU and wanted to check what the problem might be, I have a feeling it's the main capacitor but I'm not an expert.

[@LearnElectronicsRepair]

@@farben_ Hi Farben. OK Number 4 next. I've already done number 3 (video to upload tomorrow) it turned out to be a really indepth repair and a serious look into ATX PSU design - 2 hours of video and no apologies for that

[@farben_]

@@LearnElectronicsRepairthe info is very advanced for a novice but I'm starting to understand the role of each component section, up until a few days ago I had no idea what capacitors did. This channel is exactly what I was trying to find.

[@LearnElectronicsRepair]

@@farben_ I made two beginner videos about capacitors - the first one explaining what they do kzbin.info/www/bejne/aIe8Xq2qptmho6M and the second one explaining what goes wrong with them (how they fail) and how to test them kzbin.info/www/bejne/aIe8Xq2qptmho6M Check them out if you like.

[@farben_]

@@LearnElectronicsRepair Going to watch it, found a page with more info on electronicshub as well, I need a multimeter with a capacitance setting, the one I currently have is too basic, I only used this one to check the potentiometers on some console's laser drives to check if they needed to be calibrated.