Q-TEC 500W P/C PSU repair

By on April 28, 2016
Q-TEC 500W P/C PSU repair

 

 

 

 

 

 

 

 

 

My friend George bought recently this (used) PSU shown below, at a very cheap price, in order for him to use it as a spare PSU for his own P/C.

Q-TEC 500W P/C PSU repairing

Unfortunately, when he tested it, the unit was not functioning at all.

When I visited him some days later he asked me if I could fix it. I used his own bench, tools and instruments for that purpose and I opened the unit.

You can see below its internal view.

atx power supply repair

In the photo above, you can distinguish in yellow color the P.F.C correction coil beneath the small PCB which contains the input filter of the unit, which looks like a small ordinary low frequency power transformer.

This technology, using simply the addition of a coil after the rectifier bridge, was the first applied method for “cos phi” correction in SMPSes. It looks like the filter stage of the good old vacuum tube power supplies, which were using a huge iron cored coil similar to this one in shape along with two electrolytic caps, one at each end of it, forming a “pi” type low pass filter at the output.

The modern ways make use of sophisticated ICs which do this task with much better efficiency than the simple coil right after the rectifier stage. Moreover the inclusion of an active P.F.C circuit in every SMPS, no matter what its use is, is nowadays compulsory. This refers especially to high end units and it is common practice on the manufacturers’ part during the recent years. Directives of relevant institutions, which are in effect internationally, make this use of active PFC circuits compulsory.

The first thing I noticed when I opened this PSU was that the big fan mounted at its top cover, beneath the protective metallic grid, was stuck. This is always bad news… Right afterwards I dismantled completely the unit in order to test its semiconductors which control the primary winding of the power transformer.

Below you can see the noise suppressor filter assembly along with the (empty) fuse holder of the A.C input fuse.

ac input fuse

The fuse was not only burned with all its metallic material spread around all over the internal glass surface of it, giving it a shiny black colored hue, but its glass case was also broken in pieces! I removed them one by one, both those still standing at the fuse holder and the ones which had fallen inside the unit…The appearance of the fuse in such a condition, made me think that this bad news feeling I had in the first place seemed to be even more intensive now…

I kept on with static tests of the components at the primary side. The rectifier bridge was (remarkably) intact although I was detecting a short circuit right after the burned fuse. Then I removed the heatsink which holds the power transistors which control the primary side of the power transformer. After testing, one of them was completely shorted. Zero Ohms between all of its terminals.

The transistor was a 2SC2625. Because George had no new spare transistors of this type but he has an entire junk-yard there, I searched around and found a similar salvaged PSU using the same transistors which had its fuse intact and apparently had a “no start” problem. I removed both of them from it and after testing them I put them to the unit under repair.

By further testing, I found a fast diode (FR254) short-circuited as well. It was connected in anti-parallel to the collector-emitter terminals of the defective transistor I found before, used as freewheeling diode, for protecting this transistor. Finally I removed both of the protection diodes from the salvaged PSU and replaced both of the similar diodes of the unit under repair.

The first one found defective is the diode you see in the photo below, just below blue disk capacitor at the middle of the two electrolytic caps. The second one is that shown just below the left electrolytic cap, near the heat-sink. The defective power transistor which I replaced first is located exactly behind the first aforementioned diode, mounted on the front-right side of the heatsink shown. You can distinguish the end of its holding screw at that place on the heat-sink’s body. For space economy I didn’t take many and detailed photos…

atx parts replace

In the driver circuits of these transistors were also used two small 10µF/50V electrolytic capacitors. Since George doesn’t own an ESR meter up to now and me thinking over again the very likely overheating which the circuit had suffered because of the blocked fan, I decided to replace them both without previous test of them. But before disposing them off I kept them in my pocket in order to test them at home…I was thinking of the blocked fan and didn’t want to expose the switching transistors again in any likely further risk of inefficient driving, in case these two caps were dry…

When I was back at home this fear I had about those caps proved to be true. Both of them were measuring open circuit (meaning more than 100Ω ESR for my own meter)…

By the way here is another tip for you:

Whenever you see a power output driver circuitry which makes use of those small electrolytic capacitors, it is certain that the power semiconductors of the output stage are ordinary power BJTs (=Bipolar Junction Transistors, i.e. not MOSFETs). So if you find them exploded, at least you will know what kind of semiconductors they were.

Below you can see all the defective parts I found (except the broken 5A fuse of course).

power components replaced

After finishing the repair of the electronics, I dismantled its fan and made maintenance on it cleaning the petrified grease debris from its sleeve type bearing and the rotor axle and re-lubricating it. Next, I checked the bearing for excessive wear and found it intact. I finally confirmed this by applying externally 12V to it. It was spinning smoothly without making any mechanical noise or vibrating.

Then I reassembled the entire unit and ran a lamp test on it. Everything was in order. The unit started normally without any surprises. I removed the lamp, connected it directly to the mains and loaded it up with a load that George made using many halogen lamps connected in parallel. A real stove for the winter! The unit was functioning normally with the load at its output drawing a lot of current.

Because I could not estimate accurately the condition of the electrolytic capacitors of the output filters, due to unavailability of an ESR meter, I measured the A.C component in all its D.C outputs using his oscilloscope, while they were heavily loaded. I had no other option about that needing to confirm that it would function properly and reliably when put in ordinary use. All outputs presented normal level of ripple and noise components, meaning that their filter caps were functioning properly.

The unit was already fully operational, ready to serve George for some years to come.

paris aziz

This article was prepared for you by Paris Azis from Athens-Greece. He is 59 years old and has more than 30 years’ experience in electronics repairs, both in consumer and industrial electronics. He started as a hobbyist at the age of 12 years and ended his professional carrier as a senior electronics technician.  He has been a specialist in the entire range of consumer electronics repairs (: valve radio and BW TV receivers, transistorized color CRT TV, audio amps, reel and cassette tape recorders, telephone answering and telefax devices, electric irons, MW cooking devices e.t.c) working in his early stages at the official service departments of National-Panasonic first and JVC afterwards, at their premises in Athens.

Then he joined the telecoms industry, working for 20 years as field supporting technician in the sector of DMRs (: Digital Microwave Radio transmission stations), ending his carrier with this subject. Now he is a hobbyist again!

Please give a support by clicking  on the social buttons below. Your feedback on the post is welcome. Please leave it in the comments.

P.S-  If you enjoyed reading this, click here to subscribe to my blog (free subscription). That way, you’ll never miss a post. You can also forward this website link to your friends and colleagues-thanks!

 

Note: Please check his previous repair article in the below link:

http://www.jestineyong.com/atx-psu-modified-into-a-car-battery-charger-part-3/

 

Likes(110)Dislikes(1)

32 Comments

  1. Albert van Bemmelen

    April 28, 2016 at 7:55 pm

    Great and again very informative article Paris! The P.F.C correction coil
    you explained as only used in older type of PSU's was new to me. And
    knowing that they now use electronic circuits for better P.F.C correction
    was something that I expected because I found indeed special PWM controller chips in the newer SMPS units. Also the power output driver circuitry Tip
    you gave can be very useful if the item description on the BJT/Mosfet
    components, depending if small ecaps are used, is unreadable.
    The info on repairing these Units is very useful.
    Thanks!

    Likes(5)Dislikes(1)
    • Paris Azis

      April 29, 2016 at 3:03 am

      Thank you, Albert. Yes, the PFC methods have been much improved in modern designs. Their controllers now are usually split internally in two main parts. One is the already known PWM controller used for the regulation of the output voltages of the PSU and the other is again a PWM controller which usually works using the same timing signal of the previous one, but exploits the opposite phase of it. This controller drives a pre-stage of a classical boost regulator configuration, consisting (usually) of a small toroidal coil and a diode. This circuit is connected in series to the output of the mains rectifier. In this way the DC voltage at the terminals of the mains filter cap is elevated and stabilized to a value of 385-390V by the PFC controller. This voltage goes then to the primary of the switching transformer for further processing. The controller checks at every instant the current curve and holds the input current draw of the PSU in phase with and proportional to the input voltage.
      If you are interested to this process in detail, I suggest you to read carefully the preliminary literature of the CM6800 chip, included in its data sheet. I met this IC many times in modern PSUs and became familiar with it. Among others, some excellent designs of Tagan are using it. This is only an example of course. There are too many ICs for this purpose, from many manufacturers, working in various different modes. However this is the base the stand on.

      Best Regards

      Likes(1)Dislikes(0)
      • Albert van Bemmelen

        April 30, 2016 at 4:11 pm

        I certainly will look into the CM6800 datasheet. Thanks for the very usefull new theoretical background on how these PFC operated SMPS units work, Paris!
        You are, next to a Transformer expert, obviously also a Master in everything concerning Power Supplies!
        Maybe this new information finally also will help me to fix the still not working AN4800 with TNY277 controlled PSU.

        Likes(1)Dislikes(0)
        • Paris Azis

          May 1, 2016 at 12:01 am

          Hey Albert,

          I also thank you for your kind word. Well, it is true that I spent a lot of my (ex) professional time repairing switching PSUs, right since they appeared in the market. I was working for Panasonic Service at that time…
          Remember that I would be pleased to help you repair that PSU you have defective.
          That TNY277, drives the little STBY transformer which produces the STBY 5V along with the PWM chip supply. If you measure this voltage in the violet coloured cable located on the multi-pin connector to the motherboard, while you are in STBY, this tiny chip is OK. This should be your starting point. If there is no voltage there, do not expect the PSU to work!
          Only if this voltage is present you should move forward, searching for other bad components.
          Is your fuse intact or not? This is key question. If the fuse is intact, you rather miss the auxiliary 5V supply, along with the higher (18 to 22V) voltage for powering the (F)AN4800.
          This chip is very robust and fails rarely. If your STBY 5V are present (which I doubt) check next the Vdd voltage on pin 13, especially when you short the green with a black (ground) cable. It should be above 11V. If this is also OK, check pin 12 for pulses there (Output, Power Factor Correction). If this works normally the high voltage on the bulk capacitor should be very near to 400V. And finally if this is also OK, check pin 11 for pulses (Output, Pulse Width Modulator). That would be all Albert.
          Good luck with it. Keep me informed!

          Likes(1)Dislikes(1)
          • Albert van Bemmelen

            May 1, 2016 at 2:27 pm

            Great additional information dear Paris! I noticed that the CM6800 you had mentioned is completely pincompatible with the (F)AN4800 chip.
            Mayby this will help to fix my 500Watt Power Supply. I have an ESR tester so I won't use my oscilloscopes to check on the ripple component in all D.C outputs while the PSU is heavily loaded. The Fuse is still intact. But I noticed a small extra board inside this unit I haven't checked yet. Maybe this is a P.F.C. related circuit the now prevents my unit from starting. I previously already replaced both AN4800 and the TNY277 without any luck. I hope to have some time to have another look into it. (because although I am going 60, I have a lot of problems with our local authorities relating to finding work and being unemployed).

            Likes(1)Dislikes(0)
            • Paris Azis

              May 1, 2016 at 11:48 pm

              Hey Albert

              I am very sorry to hear about the unemployment. I passed through this channel before and I know very well what it means. I wish you the best of luck about that my dear.
              As for your PSU, given that the fuse is intact, follow the instructions I gave you. Pay special attention to those very small electrolytic caps around the controllers. Your case is tricky. I bet you will find a completely insignificant component which causes you that trouble…
              If that extra board you refer to is located at the secondary side, I think it’s irrelevant with the trouble…Instead if it is separately connected to the main board with four thick cables ending in a plugged connector, this board is your PFC correction circuit. A simple voltage measurement across the bulk cap will show you if it works (390V) or not (320V or less).
              Good luck with it once again. I am anxious about the result!

              Greetings

              Likes(1)Dislikes(0)
              • albert van bemmelen

                May 2, 2016 at 2:30 pm

                Paris I just fixed today one of my P.F.C. PSU's with the (F)AN4800 and TNY277 controlled Supplies. Thanks to you I noticed this 5Volt standby circuit around the TNY277 and a Transformer. It wasn't active, there was no 5V that came indirect from the BIG 400-450V Capacitor through a defect resistor. After I replaced the 4.7 Ohm 1/4 Watt (yellow purple 2x gold black) and the shrinksocket around it the supply is working like new.
                A new repair article has born. I will show all photos and the schematic I had drawn to find the 5V Standby circuit.
                Cheers!

                Likes(2)Dislikes(0)
                • Paris Azis

                  May 3, 2016 at 1:33 am

                  I am glad to hear that Albert. Also I am looking forward for your article. “Big troubles have (almost always) insignificant causes”… Remember? When you informed me that the fuse was intact I was sure about that. Once again this proved to be true…

                  Likes(1)Dislikes(0)
  2. Parasuraman S

    April 28, 2016 at 9:33 pm

    Excellent article on PSU repairs and very informative!

    Likes(1)Dislikes(0)
    • Paris Azis

      April 29, 2016 at 3:54 pm

      Thank you, Parasuraman.

      Best Regards

      Likes(0)Dislikes(0)
  3. Dennis

    April 28, 2016 at 9:56 pm

    Paris, Thank you for your informative and well presented article. I always learn from your experiences.

    Likes(2)Dislikes(1)
    • Paris Azis

      April 29, 2016 at 3:56 pm

      Hi Dennis

      Thank you too for your kind words. Keep on learning!

      Best Regards

      Likes(1)Dislikes(0)
  4. Humberto

    April 29, 2016 at 12:26 am

    Hi Paris, you have done a nice repair. If the cooling fan was stuck, that's a negative sign, because the internal components can overheat and dead, mainly the semiconductors. Congrats. another device saved from dump

    Likes(1)Dislikes(0)
    • Paris Azis

      April 29, 2016 at 4:04 pm

      Hi Humberto,

      Thank you for your supporting comment. And yes, this is the common mechanism of thermal destruction with these PSUs. When the fan is stuck the components start overheating, the electrolytic caps vent or suffer a rupture of their aluminium case or even explode depending on their internal pressure and finally the power semiconductors are exposed in high transients which along with their overheating leads to their destruction in very short time...Always the same story...

      Best Regards

      Likes(1)Dislikes(0)
  5. beh

    April 29, 2016 at 1:01 pm

    Paris: Indeed your articles are great and the subjects that you are explaining are the first time in the history of electronics in the world
    and/ or i have not seen any where else i want just to say thank you behalf of the all the members of this web log and say keep it up this good job.
    beh

    Likes(2)Dislikes(1)
    • Paris Azis

      April 29, 2016 at 4:22 pm

      Hey Beh

      Thank you for very kind words. However the expression "the first time in the history of electronics in the world", although very sincere and touching as I feel about it, it is also exaggerating! Thank you for your supporting enthousiasm anyway!

      Best Regards

      Likes(1)Dislikes(0)
  6. Anthony

    April 29, 2016 at 2:44 pm

    A fantastic repair and an excellent explanation as always when you present one of your very enjoyable articles Mr Paris sir ! And always accompanied by very nice clear photographs that lets us hobbyists see and learn from your electronics repair experience. I appreciate how you take us through the step by step analysis and troubleshooting process and why you took them. Thank you for sharing these articles here !

    Regards

    Likes(1)Dislikes(0)
    • Paris Azis

      April 29, 2016 at 4:39 pm

      Hey Anthony

      Thanks a lot for your positive and supporting comment.
      A good documentation of a repair case is self-evidently necessary in order to articulate a comprehensive presentation. And of course successfully taken photos add more value to the text. “A picture is a thousand of words” as a wise saying says…

      Best Regards

      Likes(1)Dislikes(1)
  7. suranga bandara, Suranga Electronics

    April 29, 2016 at 3:36 pm

    Mr-Paris Azis,

    Good Repair work ! and thanks
    for sharing your Article.

    Likes(1)Dislikes(0)
    • Paris Azis

      April 29, 2016 at 4:41 pm

      Thanks a lot, Suranga.

      Best Regards

      Likes(0)Dislikes(0)
  8. Robert Calk

    April 29, 2016 at 11:10 pm

    Thanks for sharing the repair with us Paris. Your friend is lucky to have someone like you to learn from. If he is going to learn electronics repair, then an ESR meter is a must, or just direct replace all e-caps.

    Likes(1)Dislikes(0)
    • Paris Azis

      April 30, 2016 at 3:10 pm

      You are (always) welcome Robert. Yes, George is aware about the necessity of owning an ESR meter. No doubt about that. He has programmed to buy a good one. Recently he received two different cheap units, but he was completely disappointed with their performance and quality and sold them immediately…
      Direct replacement of all e-caps of a device is really disastrous and I would not recommend this method to anyone. I did this only once when I decided just to rejuvenate my older CRT TV, a 21” Goldstar I had then, and I regretted it immediately. The set revealed three different failure symptoms after the replacement! (I didn’t have an ESR meter at that time and took me many hours to resolve problems not existing before the…rejuvenation, by replacing again some… brand new e-caps). So the best method in my humble opinion is: “test immediately after a single replacement and keep on this way”…This is finally the shortest and most effective way to go when it comes to wholesale replacements.

      Best Regards

      Likes(2)Dislikes(0)
      • Robert Calk

        May 1, 2016 at 5:47 pm

        I don't see how replacing old e-caps with new ones can be bad unless the new e-cap is bad. He should get a Blue ESR meter. It can also check the ESR of batteries.

        Likes(2)Dislikes(0)
        • Paris Azis

          May 2, 2016 at 12:26 am

          Hey Robert

          If you own an ESR meter and test the new replacement caps in advance (which I would strongly suggest or at least this is what I would do in a case of wholesale replacement) yes, you are right, this cannot happen. And this is an additional benefit of owning an ESR meter.
          I said that because in cases the wholesale replacement is to be done “in blind”, without prior verification of the quality of all the caps to be replaced, this is quite likely. When I did that replacement the ESR meter was not even existing here…
          Moreover, I have many times rejected brand new caps right after buying them and performing an “incoming inspection” on them before putting them in storage selves. Especially when they are sold in “tape” form, many sellers pull them hardly holding their body against the paper “tape” in order to remove it and this has deteriorating effect on their internal bonding with their terminals. If they don’t reveal the problem at once, this will come very soon in the future. Unfortunately I have seen that many times as well…
          If the sellers intend to do that in my presence, I always stop them and buy the caps “taped”. When I am back at home I simply use my cutter to remove the “tape”. The problem is when the caps are already free from the “tape” in the storage selves of the shop waiting for us to buy them…

          Greetings

          Likes(2)Dislikes(0)
          • Robert Calk

            May 2, 2016 at 10:52 pm

            Hi Paris,
            Yes, that is true. I always make sure new caps are good also. I have never found a brand new e-cap to be bad, but I am new to electronics and don't repair as a business. It's hard to believe that anyone would be stupid enough to go yanking on the e-caps like that, though I know that there are people like that. I cut them also like you do.
            I would recommend that his first ESR meter to be a Blue ESR meter. Plus he can get the kit and gain more experience soldering. After looking inside my Blue ESR meter that I bought complete, I would recommend the kit. The soldering inside mine looked atrocious, so I re-soldered everything and added some IC sockets.
            I also have a EDS-88A Series II DCR/ESR meter from Electronic Design Specialists Inc.. It is expensive, but very nice. I usually only use it for SMD caps.

            Likes(2)Dislikes(0)
            • Paris Azis

              May 3, 2016 at 8:58 pm

              Hi Robert

              Well, components’ sellers are (at least not always) electronics’ technicians and they do that…
              I recommended the same meter to George. I own it as well and I am satisfied with it. I am also quite satisfied with the analog one I built about fifteen years ago. It has an amazing accuracy for an analog circuit that is. Just wonderful. And the DC blocking capacitor in its input allows me to measure caps with the unit under repair powered from mains. It can reject effectively the 120Hz component after the rectification…I bought the Blue ESR only to make comparisons with that one!
              I know the EDS-88A. In fact this is the one I intended to buy in the first place but I preferred the Blue ESR because of its numerical display, which renders it more flexible because one can measure sub-ohm resistors with higher accuracy than that of any ordinary multimeter.
              Recently I bought a LCR bridge which is just amazing! Among its other characteristics it is capable of measuring sub-ohm resistances with an accuracy down to one mΩ level (!!), for the cost of 120$ only (including its basic accessories), plus the transporting expenses (sent from Japan). I also classified all the unknown coils and transformers I had in stock, spares from destroyed PC PSUs, for any likely use of them in repairs or in hobby constructions…

              Greetings

              Likes(0)Dislikes(0)
              • Robert Calk

                May 5, 2016 at 10:01 am

                The Blue ESR meter is good to measure resistors for people that don't have a good DMM. I trust my Brymen 869s more for resistors. The Blue ESR meter can test batteries also.

                Likes(0)Dislikes(0)
  9. Ulises Aguilar Pazzani

    May 2, 2016 at 9:02 am

    Mr Azis, grait job Sir very analytical on your jobs,
    alWaYS READ Your articles I was reading the george don't have a ESR tester on line You can fine a schametic for a ESR meter I make couple of them they were pritty good

    Likes(1)Dislikes(0)
    • Paris Azis

      May 2, 2016 at 2:29 pm

      Hey Ulises

      Thank you for your kind words and your support. George, after testing some cheap ESR meters, decided to buy a good one. I have also built my own meter, long ago, implementing a schematic I found in the web and works fine. Its display is analog, but nevertheless it meets the requirements for electrolytic capacitor testing. Thank you again.

      Greetings

      Likes(1)Dislikes(0)
  10. Yogesh Panchal

    May 12, 2016 at 11:03 pm

    Paris,

    Thanks for sharing your experience and for good explanations.

    Likes(0)Dislikes(0)
    • Paris Azis

      May 24, 2016 at 1:39 am

      Thank you too Yogesh.

      Best Regards

      Likes(0)Dislikes(0)
  11. Ulises Aguilar Pazzani

    December 26, 2016 at 11:41 am

    Mr Azis , ok I was going to give a web site , I buy ESR meter and make one , I use both and the one I make its pretty accurate

    Likes(0)Dislikes(0)

Leave a Reply

Your email address will not be published. Required fields are marked *