Don't Miss

About Conceptronic 90W Power Supply (Part 2)

By on May 8, 2015
power supply repairing







The first part can be accessed in the below link:

I replaced yesterday the defect components by all new parts from China after almost 4 weeks of waiting. And I noticed that to my shame the Conceptronic 90W Power Supply didn’t behave as I had hoped for. Although my article in waiting for the parts arriving proved to be a good way to analyse exploded controllers I now come to the conclusion that the transformer sadly must be partly defect. Somehow the fuse should have protected the transformer but didn’t. And this is how I came to that conclusion.

power supply repairPhoto of loose wire primary side under now removed (was glued) 400V Buffer Elco.

I found a loose primary wire that was hidden under the glued 400 Volt Buffer Elco, and after soldering it back to the pin on the board the Supply finally came to life.


From the left to right are the first 2 wires. The DC Run Vcc voltage for the LD7575 controller.

power supply schematic

Above : Principle schematic of the Conceptronic 90 Watt universal PSU.

But on the controller OUT was a shorted diode (SD1 above schematic) without any marking besides its Cathode marking and so I removed it without replacing. (without any consequences so far. Possibly only for some kind of protection).

Also the low output impedance measured on pin OUT of the LD7575 is a direct cause of the internal pull-low resistor to prevent any floating conditions.

A 5Volt 0.11A bright Led light worked but it burned about one second, went Off for a moment and then gave light again for a second etc. But my 60 Watt light Bulb only briefly went on when the PSU was switched on. And the secondary 5 Volt voltage was with 5.4 Volt a bit too high. So first I checked the transformer again with my Blue Ring tester.

The transformer had one coil for DC run controller Vcc purposes. And it had only 1 or 2 flickering leds with the Blue Ring test, which is perfectly normal! And measured with my Digital meter on diode test it beeped for low resistance check. And the second and last coil wire (also connected to the gnd of the Buffer elco) of this first coil had one or two extra wire(s) that seemed to go into the transformer but I was unable to identify where to?

The other and second primary coil was not connected to the first coil and had one twisted wire pair coming out of the transformer top, and 3 bottom wires of which one only was connected to the transformer bottom itself. The Twisted wire pair of this second primary coil went to the Anode of special MURF840 diode. The most left bottom coil wire was plus via the diode coming from the + of the diode bridge, and the most right wire was connected to the Collector of IGBfet 20N60C3. And all Blue coil tests were okay except the most left coil bottom pin (+ in 300Volt) if checked against the Twisted wire pair.

But one can easily oversee this because the second coil has 4 connections of which one wire is only going to a bottom Transformer solder spot. And from the 5 possible Blue ring tests only the one from left pin against the twisted wire pair fails!

The next thing I assumed was that Vcc was oscillating. And measuring this with my digital meter proved that the voltage went fluctuating from 12 to about 22 Volt. So I checked the bottom components again and noticed that diode SR11 in serie with the Vcc pin of the LD7575 controller was somehow high impedance in 2 directions what shouldn’t be the case. No problem for the Vcc of the LD7575 controller because it handles up to max 30V DC.

After replacing it by a 1N4007 the power supply indeed worked uninterrupted. Without fluctuations continuously but with an secondary output USB voltage of about 5.4 Volt still a bit to high.But worse the 60 Watt light bulb was now burning about 30 Watt or so. But the secondary display and the configuration switch worked fine because I could set the output voltage to any voltage from 12 Volt to 24 Volt. So in conclusion I have to confess that the Conceptronic transformer is kaput beyond repair. Too bad that fuse (5A !) and other protective parts didn’t save the transformer what one should have expected.

Else it would have been a fantastic power supply repair. And sure it cost me some time and some new parts. But I now can use this partly functioning Power supply to test any LD7575 controller components and I am also new knowledge richer! And maybe I’ll find one of these days another Conceptronic 90 Watt PSU with a good functioning transformer for replacement. Because the other parts are functioning perfectly. But I hope this article explains how and what is, and what is not repairable. But one only can be 100% sure afterwards when every error is fixed and only if every bad component can be replaced. And old ways that prove to be wrong, lead ultimately to new and better ways of doing things.

Albert van Bemmelen, Weert, The Netherlands.

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!





  1. Albert Hoekman

    May 8, 2015 at 10:21 pm

    Nice article and job Albert!

    • Albert

      May 9, 2015 at 12:34 am

      Thanks! My name and country sharer- correct me if i'm wrong? Haha.

      Of course I should have tested the Transformer better with the great Blue Ring tester.
      But since I did check it from the pcb bottom side it proved afterwards to be a bad way. Because the second primary coil also had 4 wires of which one was not visible on the solder side of the pcb. And one should only expect at least 2 or mostly 3 wires of this coil of which one wire afterwards wasn't even connected. So that's why it didn't show up in the first measurements.

      And not many Engineers would dare to try to repair this almost completely exploded PSU.
      (giving Exploded View a new meaning Haha). Which shows that I am a confident Electronics repair engineer, not afraid to repair the almost impossible.

      And after all, if the transformer wouldn't be partly shorted afterwards, everything would have been completely working now!
      Because just that last part isn't replaceable easily, even doesn't mean I wouldn't even try that in future!

      After all it only did cost me one MURF480 diode, a 20N60C3 IGBfet, a 5A fuse, 2 resistors (1 x 620 ohm plus 1x 0.22 ohm), a LD7575 controller, a 1N4007 which I all still can use in other repairs again. And a hobby may cost some time.
      (and Nothing exploded after replacing all parts which means I did it correctly).


  2. Bill Eitner

    May 9, 2015 at 2:54 am

    It's somewhat impractical, but the transformer can be rewound.
    Anyone who claims to be an electronics tech should have repaired a transformer or coil via rewinding at least once. That's the advantage coils/transformers have over other components; they can be repaired.

    It's like a right of passage and is common in antique radio repair.

    Old vacuum tube amplifier stages are transformer coupled.
    The DC plate voltage goes through a transformer primary winding.

    As we all know, tubes are controlled the opposite of transistors.
    Grid voltage turns off a tube, whereas base/gate voltage turns on a transistor.
    Lose the base/gate voltage the transistor turns off (no damage).
    Lose the grid voltage the tube runs full on eventually destroying the tube and often overheating/burning/damaging the transformer.

    When you can't find a replacement transformer the only option is to rewind the old one. Enameled wire of the right size is all it takes.
    Although not really recommended, in a pinch the bad spot can be spliced depending on the level of damage. Sometimes the failure is just a break at a weak point. What you can get away with depends on the condition of the enamel coating the wire and the design of the device you're working on. For example, with the common flyback style SMPS the transformer turns ratio isn't critical, so the removal of a bad spot or even a short length probably won't effect performance.

    So, there you have it.
    Albert, if you want to become even more knowledge/experience rich, you still have an option that can result in a complete repair.
    Good Luck.

    • Paris Azis

      May 12, 2015 at 3:03 am

      Hi Bill,

      Nice and encouraging your comment. I fully agree, except for the transformer rewinding advice. Here is why (Motorola,Linear/Switchmode Voltage Regulator Manual,1983, page 11-1).
      ...It takes quite a bit of time to develop a "feel" for this craft and be able to use both experience and intuition to find solutions to second and third order problems...
      And what's more:
      ...For flyback designs, a gap is necessary...and a test circuit is needed to evaluate the effect of the gap...
      If this task is scary for design engineers, is not possible for repairs technicians to succeed with rewindings I humbly think.


      • Albert

        May 16, 2015 at 4:35 pm

        Paris, thank you for mentioning the Motorola book and your usefull advise. I sadly haven't got the book. So I haven't read it either. I take it that the Gap part you're mentioning is only applicable to Flyback transformers in TV's and sorts? And in this case the defect PSU transformer isn't intended to create High Voltage output. So I guess in this case there is no Gap construction in this Universal Conceptronic 90 Watt PSU? I hope to find the book you mentioned to be able to understand what the Gap part is about. Thanks for this information.

        • Albert

          May 16, 2015 at 4:57 pm

          For anyone that wants to know more about SMPS.... just found a very usefull Guide on Switching Power Supplies. Only 73 pages and very concise from ON semiconductors !

          And Free. See:


  3. Albert

    May 9, 2015 at 3:30 pm

    Hi Bill, thank you for your possible new repair solution but I don't think I even know how to just open a transformer core without destroying it.
    It probably is a lot easier then to dismantle the 50/60 transformer types, which are quite heavy laminated iron type cores. To reduce any unwanted flux or something it was for. And I read somewhere in the past that a guy knew how to dissolve these cores to be able to open them in some kind of chemicle solution but I forgot what he used. And it seems to be a very uncertain way to repair these. Sure these coils are no Tesla type of coils which are a very precise made type of coils. But aren't these PSU transformers not also made on precise automated winding machines?
    If I could see someones report with photo's on such a project on how to repair these transformers, I probably knew what I was up against?
    For now I won't go this far yet.

  4. Anthony

    May 10, 2015 at 4:40 am

    Thank you Albert for taking the time to share the very valuable information from this article with us. Even though it
    wasn't the desired outcome the experience has enriched yours and everyone else's knowledge on this blog !
    We have all gained from this electronic repair article and that can never be a waste of time.

    Thanks again !

    • Albert

      May 10, 2015 at 9:56 pm

      Thanks Anthony, Nice of you to say but I do hope to present a better ending repair article next time. Also because of the a bit disappointing result. The more we learn the less a gamble it is. But it must be as they say in English speaking countries: No pain, no gain. Or the he who never tries, never wins stuff.
      Bye bye.

  5. Humberto

    May 11, 2015 at 9:16 pm

    Albert, good explanations very scientific. Congratulations.

  6. Paris Azis

    May 12, 2015 at 3:08 am

    Hi Albert,

    Nice job no matter the result. Knowledge has always a price. It was never for free.


  7. Yogesh Panchal

    May 14, 2015 at 2:55 pm

    Good article thanks for sharing.


Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.