How to repair a switching "switch mode" power supply. See what's involved. Also a brief explanation about the difference between Linear and switch mode power supplies.
You started out with a SMPS made by Power Systems, Inc. I was the Product Safety Engineer for PSI for nearly 30 Years. I think the model shown was in the series of late PS-1600's or PS-1700's. Enjoyed watching that work again! LOVED that company and my old job! I traveled all over the world for them. BTW; that green module in the primary has a saw tooth generator and a low voltage dc reference voltage that slides up and down across the saw tooth. thereby tell the switcher to stay on for a longer or shorter period of time based on the load conditions, as felt thru the feedback loop from the sec to primary using that opto you spoke about. Wow, I am retired and its been over 10 yrs since I left America. (retired in Philippines) PSI went out of business about 8 years before that! So, the memories are wonderful! Also, the reason we used the RTV between all the caps was to add ruggedness to the supply. We used a HALT test chamber and a random vibration test at the same time we took temps up and down to find the failure points. We brought them up to about 140C before they failed, then down to -60C before they failed and found that most of our supplies lived at 30G's but would fail by 60G's. I ran that area. My people monitored about 30-50 components during the test for temperature and we monitored the input and output voltages on each channel. I also ran the RFI/EMI Testing and got them approved thru UL/CSA/TUV, etc. (I believe the standards that one met was UL/CSA60950) and often made sure they would pass the conducted and radiated emissions testing also to ISO Standards. I also sat on two technical committees in Canada helping to write the International Safety Standards for the IT industry. Later, I worked for the German TUV Agency as a Product Test Engineer. LOVED my work!
Absolutely the finest teaching around. You have a gift, I wish my instructors would have had even 50 percent of your ability to transfer knowledge in such seamless way. Best on youtube for sure. Thank You!
@Jeremy Mettler eh, there's something about teaching that you can't really buy. There's some merit to your point, but at the same time the big-salad-earning folk... while probably great at what they do, they're not necessarily also great at teaching. There's a number of vital components that have to be there to make a good teacher. That's why it's a rare breed. Be that as it may, we're still underappreciating the average teacher. The lot of us just want to be taught, but they can't do all the work for us. We have to learn... to learn.
Watching these SMPS videos reminds me of my time working at a power supply manufacturers. Testing units from the production line to working with design engineers building their prototypes and developing them to production status. One challenging job I had was developing a 30kV pulse supply to be used on a magnetron. I wasn't privileged enough to know what the final product was (secret stuff). Another excellent video Mr C. and I've watched quite a few of them so far.
Your shop brings back memories of my father's shop when I was a little boy. My father was an sound engineer for the RCA corporation. Repairing sound systems all across NYC it was great fun being in the projector booths and seeing racks of AMPS in the winter you never needed heat the equipment provided it. My dad passed away just as the transistor came on the market every thing was tube operated. I miss his great wealth of know how from radio and tv repair which I saw from a very young age. He started his repair business in 1918 when he was 14 in his fathers basement. I became an electrical contractor installing alarm systems and generators building my own controls as much was not commercially available. Customers learned very quickly never say to me it can't be done I always proved them wrong. Much to there dismay and amazement. Today every one in my field is an expert or certified my father would have called them tube pullers putting wheels on miscarriages. Artificial stupidity is what I am waiting for.
One more important comment, to regulate the output voltage, the control circuit senses the output current and increases the conduction time of the switching transistor to regulate the voltage. Love your work!!
The best explanation on how SMPS works with an in-depth explanation of where these are unsuitable and/or what additional devices are needed for these when running sensitive audiovisual equipment.
Paul, great video! I've been servicing SW mode PSs for many years and have found anomalies that don't make any sense and today was one of those days. Working for a college that has very limited budgets, I have taken it upon myself to repair as many electronic items as possible. The majority of these failed devices are due to a bad SW PS. With a few exceptions, all they needed were new electrolytic caps, and most of the time it is a small 10µf to 47µf cap near the control/regulator chip that needs to be replaced, although, I usually also replace the filter caps at the output for good measure. Anyway, the sole purpose of that small cap seems to be one of obsolescence, or a timer cap. When it starts to lose its capacitance, the output amperage starts to drop and when enough capacitance is lost, the PS may make a ticking noise due to the PS starting and stopping on a continual basis, even with no load. Now for the weird part... I pulled the "timer" cap from this no-name PS (20V @ 2A) since it was "ticking" on and off at about 2 Hz, and my cap tester showed it was well within tolerance of the 47µf value. Normally, this B&K, model 830, cap meter has always showed a greatly diminished value but not this time. I replaced the cap anyway and the PS came back up to full voltage and power output. I then analyzed the bad cap using a sensitive analog VOM and both the old and the new cap exhibited the same readings for charging time and leakage, so I'm at a complete loss as to why the PS circuit likes the new cap but not the old one. And this is not the first time this has happened. The only possibility I can think of is that breakdown is occurring in the cap, such as its leakage is exceedingly high at its rated voltage and causing a shut down. So I put the bad cap on my lab PS and set it to 20 V, and the leakage current reading is 0.2 µA. Then ran the same test with a new cap and it had a current leakage of 1.0 µA, five times higher than the old cap. So, if the leakage current goes to low, these PS's shut down? Could this get any weirder? If you have any insight or info about this anomaly I'd appreciate hearing from you. Thank you.
If I might add a little bit in here, the problem might be that the ESR of the capacitor had greatly increased. The capacitor with increased ESR might test just fine at higher frequencies, but this cap needs to filter and hold up the voltage to the regulator @ 120Hz and if the ESR is too high and the capacitor Xc' diminishes in relation to the ESR. I had it happen before on guitar amplifiers that I've repaired. I check the filter caps and they show within 5% of correct value @ 1KHz or higher, but when I check it at 100Hz it registers anywhere between 40 - 60% low. Replace the caps and Bob's your uncle. Again, this is just something I've experienced in the past, so I thought I would pass it along
Best electronics teaching channel I found hands down. I've been repairing electronic devices nearly 40 years full time and there is always something new to learn!! Thank you. 😉
Very good explanation of the switch mode power supply, especially with your diagram you drew. If not already, you would make an excellent electronics professor. You have a very good teaching method.
Mr Carlson's Lab I am no electronics expert by a long shot. I have learned so much from RU-vid. I had an old friend several years ago that you remind me of. He could fix anything electronic. If any of the area cities electronics repair shops could not figure out a problem, they would bring it to Jack. I think he really enjoyed them bringing him stuff they couldn't fix! Almost always, next thing you know, the apparatus was working again. You sir are so very talented, and it is a pleasure to watch you work, even if it is above my head. Thanks for the videos.
Carlson, you are really a good teacher! I learnt lot of electronics theory and trouble shooting skill from your video. Your speaking was so clear and very understandable, even to me a non-English speaker. Thank you very much!
If I need to replace a through-hole IC, my preferred method is to cut the pins as close to the IC body as possible in order to leave as much of each individual pin in the board as I can, then I simply go round and remove the remains of each pin with tweezers and my iron. Then remove the solder with wick. Using this method lessens the chances of damaging the PCB significantly, it's also a lot easier and faster.
+Mr Carlson's Lab Mathematically, switching power supplies are very complex to understand. I don't think type 2 and 3 compensation feedback is easy to understand.
I never attempted to try and understand the engineering of of SMPS and for me it was enough that I could comprehend them enough to effect repairs to high replacement cost OEM units that made sense to repair like your TEK supply. I cut my teeth on IBM PC supplies back in the 80's to learn about SMPS technology.
Many thanks for your excellent tutorial. I was a TV technician 50 years ago, repairing tvs to component level. Then I moved up to Avionics tech job and later on to an aircraft engineer's job. I retired after 40 years without board level repairs. After retirement I find a pensioner's life extremely hard. I am refreshing my electronics knowledge to be able to repair TVs or Computers.Please keep up the good work. aftab ahmed. London
@ANDREWAUSFA hi mate Just wanted to say, i have not seen the PYE logo since i was a kid thanks for the memory i had a PYE Tv later i had a Rank Arena hehe
Ah the good ole SMPS the most common failure in modern junk! I spend hours working out faults on that kind of crap. Televisions are notorious for SMPS failures, crusty caps and short circuit diodes being the most common failures, but they can go right down to the darn right obscure and hard to find component failures that make you question the real meaning of life! (But hey, you get a warm feeling when you find the fault or touch the 400v LOL). Keep up the good work, good clear explanations and gripping videos. 5*
Thanks so much for a well scripted film.If it's not scripted,you have an incredible sense of what to show us.I'm working on a Behringer PMP1680s Europower powered mixer that just blinks the leds on the front panel.I'll now be able to check some of the SMPS, key components. Thanks again!
Excellent video. I'm just getting back into analog electronics after a couple of decades focused on digital and programming and find these videos both an excellent learning tool and refresher. Thanks for doing these.
"Tittally""...I haven't heard that one before :) You mention the one of the greatest tips that I figured out years ago for component removal/replacement...apply fresh solder to any joint to facilitate easy and much quicker part removal, hence, less heat to a possibly fragile board. Also, on an IC that I'm removing and not going to reuse, I recommend snipping the legs off of the body of the IC and then removing the legs one by one. Once they're out, use a small stick-pin and apply the solder tip to the hole to melt the solder and push the stick-pin through the hole to clean it. This process saves a lot of solder wick and/or solder-sucking on most repairs and saves the consumables for the most difficult situations that need those tools. Cheers Paul, Dan G.
When replacing polarized parts which are not marked (I.E. caps), the easiest way to identify the negative is to put your meter on continuity and put one probe on (chassis) ground and the other on each pin of where the cap goes until you get a beep.
Almost 10 years ago, I repaired one of these supplys that wouldnt turn on. In the process, I shorted something and blew up the switching transistor and the hybrid. I ended up doing exactly as you said. I scraped the green stuff off the hybrid and replaced the damaged components. That was a real pain in the butt! Turned out that the no start condition was from a bad cap under the heat sink. The smallest one on the board. Love your channel!
Absolutely fascinating; the world of electronics and the way YOU present it to us. You are an inspiration at best. Thank you so much. I feel pepped up and positive. Good luck and much love. Cheers! 😊👍
Great demonstration Paul; switch mode supply are very efficient, the only downside is they need about 90 volts to start making variable increase very tricky and in some cases blow the power supply when it pulls to much current. As I recall the switch mode operates within +-1% making regulation very tight.
Excellent switch mode PSU description. It really filled in the limited knowledge that I previously had. Especially regarding the safety requirements with the AC line voltage. Thank You!!!
Hi Paul, Couple of suggestions that might make it easier to remove the components from the board without damaging the traces or PCB. 1. If you are just replacing an IC then cut the legs off the IC before attempting to desolder the pins that way you only need to remove 1 pin at a time. It also means that you can grab the loose pin from the IC side whilst touching the solder side with the iron and gently pull the pin out. It really leaves a very clean hole that way. 2. When applying heat to the joint try and not press the iron into the board. I.E. try and float the iron on the molten solder. That way there is no physical force transferred directly onto the board. As you are no doubt aware the traces are bonded to the board and that bonding comes loose when you apply heat. If you can minimise any physical force during this time then the traces should stay in place and not get damaged. Regards, Graeme
Hi Paul, MCL, Excellent tutorial on Linear and SMPS supplies. Good job on the fix. MCL is a goto channel that delivers credible, accurate, and, depending on the subject matter, easy to understand information, when your a newbie, it makes a difference. Thankyou. Take care, C.
Paul, Another fine video. I'm working on bringing back into service a Yaesu FT-901DM transceiver that has not been used in 25 years. It does power up and receiver is OK. The plate current and PO meter gives no readings, however there is output into the dummy load. Just started working on the project and hope there in nothing too seriously wrong. The radio gave me many hours of good service.
+Mr Carlson's Lab Thanks Paul. Peter TRXBench thinks I should save up for another radio. :-) I'm going give it a try and see what I can do. Found a Canadian ham on Ebay with a recap kit for all the electrolytics in the radio. It's not expensive and I think I will buy it. I have a Flex1500 to use, but no antennas. That will take precedence, getting my triband beam down for maintenance.
+Mr Carlson's Lab Hi Paul, My luck keeps getting worse. I bought a Flex1500 off Ebay last year to have something to use and I just tried it with a 10m dipole and no output. I click the MOX button in PowerSDR 2.7.2 and hear the relay click, and nothing. :-(
Yeah, back in the 1970's we cut them off at the chip-case, then pull the pins off the board 1 at a time...solder sucker....then you got nice place for the new ones......I worked on a lot of TTL back then.....those 7400 series, ran hot and failed often....a gate goes bad...change the whole chip..
Always enjoy your “speciality” work. I’ve got a cheap switcher that’s dead and I look forward to attacking mine. I wanted a cheap high current 12volt PSU but it died a whimpering death. Always enjoy your programs to glean all I can to get back to getting my vintage equipment back to use Thank you as always and Happy Thanksgiving, Merry Christmas and hope we both have great 2024! Bill Martin, KA1TIU
If I haven't said this before I'll say it again.. you are a quality teacher. Some people honestly don't have the ability to teach no matter how much training they receive but you have the natural ability. Please keep up the good work!
The best investment I ever made was buying a hakko desoldering gun. Man that made my life so much easier. I can't believe I was throwing away PCB's because I couldn't be bothered with desoldering when it is so much easier with a desoldering gun. You don't have to get an expensive one like hakko either, I just like to buy tools to last.
yeah! Mr Carlson is the ultimate sen-say.. And his Kung Fu is strong!!!! what i really love about him is he never say's a fix is impossible.. the man's a real gem!!!
Very cool, I have some electronics that over stated their "120-240v" capabilities in NZ. I'm going to have to re-watch all your troubleshooting videos before I go after them. Thanks for posting!
+Mr Carlson's Lab I'll be back on Saturday, 21hrs of travel from when I leave until I get to Toronto. No more super glue and rubber tape repairs for me!
Excellent instructional videos, sir. Thanks for sharing your knowledge. I think I can now attempt to fix the power supply in my Cisco 500 series POE switch. Greatly appreciate your time.
excellent n fruitfull video for those who r interested in SMPS power supply.Good that u have shown the importance of SMPS over linear power supply.thanks for video.
Best tutor in the world,You would Love Electronics after Bad things Happen..........Thumps up for this guy!Now thinking on building a Power Supply that turns the whole world on!!!!!
Hi Paul, thanks for this video. Very professional as always. If you dont have such a good desoldering tool, you can cut off the legs of the IC's as close as possibe on the chip. After the IC is removed, you can easy remove pin by pin. Simply attatch a alligatorclip on the pin (from top side) turn the pcb upside down and the leg falls off by gravitation during soldering/heating. I do this every time if possible, when its clear that i will never use or need this part again. Maybe i would use sockets fore the IC's in case that i have some unexpected issues on that board during testing and measuring. Great respect ! CS
xing jiang He was not sure if those IC’s where good, they are cheap and difficult to replace after soldering the board back in. So he replaced them just to be sure. As a precautionary method.
Worth mentioning a current limiting mains supply can be improvised by using an incandescent lamp in series with the supply. A 150watt mains tungsten or halogen lamp will do. If it lights up brightly you know the power supply is shorted and can disconnect the supply. Another useful side effect of this is that the lamp will flash indicating the working of an active PFC circuit where present. Still you definitely want an isolation transformer if your testing power supplies on the bench.
Man, I love this channel. Thanks for showing your technique for removing I.C.'s. They have proved to be a real pain for me in the past. Excellent stuff. Now, where is that old CB power supply I have that does not work?
The easiest way to remove a chip from a multiple layer board is to cut the chip body away from the legs, remove the legs from the board - then clear the plated holes. Vcc & Vdd always require more heat. This method has served me well for over 40-years.
Thank you for pointing out how noisy the switch mode power supply is. As an amateur radio operator I have heard what happens when a radio uses one of those supplies.
Finally got around to checking this video out since the p.s. in my Tek TDS460 is the same and found the caps good, but what appears to be quite a few solder joints on the output pins looking dry. Pulled the caps and all mine tested good with no physical leakage. Thanks for the video!
Awesome video Paul. I always love a good tutorial on switchers. Was very happy to see you do this. My main pc in the house just went south. Unfortunately, will not repair the smps due to output filter coil has a dead short on all pins. Take care my friend, enjoyed that.
Hello ! Once again a nice video and good explanation on how both power supplies work, but I think you could have told that the switching power supply has a much higher efficiency than the linear power supply which will dissipate a lot of energy on the power transistors that stand the voltage drop between the rectified voltage and the output voltage. 73 and CUL.
Dear sir, Great video and very clear information! I did the measurements as you described: I found an unstable 4.6v instead of a stable 5.0 volts.I hope replacing the four capacitors (as in my board version) will solve the problem of my TDS 420A which doesn't power-up, just flashing for a second if pressing on "on/stby" button.Thanks again ! Yaron (from Israel)
There are some resistors in front of the vertical potted board that will corrode at the board level (in the PCB hole.) This vertical potted board is kind of a green color and is located by the switching transistor. Try to lift the ends of the resistors carefully to see if they break off where they go through the PCB. Replace all the capacitors.
Thanks Mr. Carlson.After a very long time of waiting the new capacitors they have been arrived.I ordered the same/original Nichicon 1500uF 35V. I replaced them all (6 capacitors).All voltage values measurements are now unstable. The capacitors are 1450uF which is still "in the range" so this is not the reason I believe...the replacement was carefully done, so i can't guess what causes to this behavior/fault.If "listening" to this board I can hear "tick tick" like an analog watch- I don't know if this was before or it's a new behavior...I hope I will be able to figure out/fix this PS. BTW, the vertical board is o.k. and I didn't find any breaks there.I hope you will have some tips how to move forward...thanks!! Yaron
Thanks for the tip on adding solder to the IC before desoldering. I wish I'd heard of it sooner as it would surely have helped. I remember desoldering 72-pin SIMM sockets on a motherboard a long time ago with nothing but a low wattage iron, solderwick, patience, and a mop to swab my face with between pins to keep from sweating all over the thing. Talk about a Zen-mode inspiring time. Yikes.
This is one of the very best electronic channels I have come across. You have an awesome talent and knowledge base my friend. Here is a question on smps I have wanted to know the answer to for some time. Q) Is it possible/safe to parallel the outputs of two identical smps's? eg. two 36 volt, 10 amp supplies to achieve one 36 volt 20 amp supply? The problem I can see is how to exactly share the load between the two supplies.
Some SMPS will work in that configuration, others will object when substantial load is drawn. Some SMPS have provisions to "daisy chain" It all depends on the design... So no real straight forward answer. Thanks for your kind comment Jim!
Hello again. Very many thanks for your answer in reply. I have just discovered that some of the MEAN WELL range of smps' can be paralleled: see: Industrial Enclosed Power Supplies - Meanwell RSP-1000 spec. As you have already said, it depends on the design. All of the voltage outputs of these supplies must be within 0.2 volts of each other and the total current output is calculated thus - # of psu's x rated current per unit x 0.9.Thanks again.
I did a (UK) ONC in electronics back in the early 90's. (I subsequently moved onto computing.) I wish my electronics lecturers and tutors had been as good and as easy to follow as Mr C. (One lecturer however did say that the then current syllabus took the fun out of electronics.) Regarding SWPS some even caused problems with certain computers at one time and it was not unknown to resolve by moving onto another make of power supply. I would have thought bench variable power supplies are best if of linear type to get over the RF issue as you don't always know what you will be testing, though many do use variable SMPS for this.
Tnx sir, you shared a very helpful and informative knowledge especially to us who repairs smps powersupply. This very helpful to me sir as electronics technician. Tnx sir for sharing knowledge sir🙂
Lol, on rare occasion I have been totally stumped on the problem section of circuit responsible for a fuse popping. Jumping the popping fuse, and looking for a pillar of smoke lead me directly to the shorted component. While I strongly suggest against this practice NOT to be done for safety reasons, well, umm, lets just say that one sometimes might feel they have no other choice other than scrapping the unit. Certainly not the correct, professional or safe way to go about troubleshooting, and would strongly suggest against it esp for safety reasons. One might start a fire, have components explode with fragments into one’s eyes, etc!
Nice informative video. Thanks for all the good technique. I only wish you had shown where to attach the mains supply. I'm always intimidated by switching supplies.
Hi there , I don't know where to start , so I will start by say Thank You for the video . I never knew which is which and way are the difference between them !!! . This is one aspect , the other is , and here I will Quote TOM "Very good explanation of the switch mode power supply, especially with your diagram you drew. If not already, you would make an excellent electronics professor. You have a very good teaching method..... Mr Carlson's Lab I am no electronics expert by a long shot. I have learned so much from RU-vid...... You sir are so very talented, and it is a pleasure to watch you work, even if it is above my head. Thanks for the videos." and "RadioHamGuy Excellent video as always, great information and you make it easy to understand." I can say no more than Thank You and keep up us on post . P.S. I subscribed of course . LOL.
I wish I had seen this video sooner. I have thrown out many SMPS in the past just because I didn't know how to work on them. Thanks for sharing this video. Maybe you can do a video sometime on how to use an isolation transformer?
Wow Paul !!!. I only just found your 'Chanel', and am impressed with the diversity & clarity. I noticed you are from Canada, and one of the BIGGEST things that was impressing me, when you speak, is hearing the correct pronunciation of 'SOLDER', and NOT 'SODDER' !!!!! Excellent work. I'll be back very often !!! Thanks, from Western Australia.
