Connect with us

Computer Switching Power supply

Discussion in 'Power Electronics' started by Y2KEDDIE, Mar 20, 2013.

Scroll to continue with content
  1. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    I' have a switching power supply, and it works intermitently. I'm thinking a component is out of tolerance because it works some of the time. I measure the Vs (start up) and it measures only 4.2 Volts ( when its working , and when its not). I'm thinking this is the culprit. Any ideas?

    Are then any good (practical) books on switching supplies?
     
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    First thing to do is to check the capacitors.

    Then replace them.

    The voltage is the symptom, not the cause. Measuring the output voltage may be misleading. You really need to look at it with a scope (and see what it does under load)

    Don't know any good books. If you find one, let me know.
     
  3. Electrobrains

    Electrobrains

    259
    5
    Jan 2, 2012
    If your power supply works "intermittently", it could be a sign that you have a bad contact or solder place somewhere, or that you overload the supply (or it overheats because something else is bad) and it automatically switches off for a while to cool down.

    If you google around a bit you will find a lot of information and books about SMPS's (Switched Mode Power Supplies), even on-line free ones.
    But the theory of such technique would demand that you have a basic general understanding of electricity and electronics.
     
    Last edited: Mar 20, 2013
  4. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    Computer smps

    Actually, my computer runs on 24 VDC , and the supply is a dc to dc converter. It has been in operation 24/7 , 8 years. I think its just tired. It still has useful life. I've replaced the supply and have it working. I'm trying to troubleshoot the old tired supply, because they are no longer available in the same package.

    The odd thing is, while bench testing, I apply power from a cold start and maybe 5- 10 minutes later it fires up. Then, after a while it will shut off. If I cycle the power a few times, sometimes it will start. It never gets warm, and as I said even on a cold start it's intermittent.

    I've tried tapping on the components, flexing the board with no reaction.

    Spec says 5Volts for Sb (start up) TTL available in quiet mode, but I have 4.2VDC on and off.

    Eddie
     
  5. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    If it only works reliable after a few starts (i.e. after it warms up a little) then look at the capacitors. As they warm up ESR drops. If they are failing and their ESR is getting high then "taking a while to start" is a common warning sign.

    And as I said before, check the output with a scope. You may find that there are spikes or ripple on the output.
     
    Last edited: Mar 21, 2013
  6. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    I'll get the scope fired up this weekend. There are only 8 or 10 electrolytics, Ill check them all.
     
  7. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    Well, I didn't scope yet, but did one better. I removed every capacitor (15 pcs), and checked leakage, value, and ESR with my Sencore LC-101 Analyzer.

    I found (8 pcs) of 1000ufd @16 to have no leakage, value of 100ufd, and ESR of 5 ohms (almost 10 times higher than spec).

    I want to replace the capacitors but the new ones I have, are physically bigger.
    The bad ones are 1/4" DIA, the new ones 3/8"DIA. I don't think I can physically fit them all on the PC board. (but I'm going to try).

    Eddie
     
  8. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    5 ohms is quite high.

    If the capacitor has 1A of ripple current, you'll get 5V of ripple!

    I also assumed you measured 1000uF, not 100uF (typo?)
     
  9. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    No, measured 100 ufd and High ESR (5 ohms). Definitely all bad. I checked a new 1000ufd and got good results.with 1,8 ESR.
     
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    Interesting. Often the ESR deteriorates way before there is any significant change in capacitance.

    It would be great if this is all you have to replace.
     
  11. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    Unless these capacitors were labeled wrong. I would think 1000ufd would be typical. Most were in the output side (load side), but not all. I don't have a schematic. It's a 24VDC input, and then the typical ATX supply voltages on the output. 200Watt supply.

    The ESR is 10X what Sencore chart says for a 1000ufd cap. at 16V, and the measured value is 100ufd. I couldn't measure any leakage, my instrument just flashed all 888, indicating a fault.

    I will start repopulating the board with new capacitors. Wish me luck.
     
  12. Y2KEDDIE

    Y2KEDDIE

    259
    15
    Sep 23, 2012
    I didn't want to leave this hanging : I gave up. I replaced all the electrolytics and it no longer works in any fashion. I believe I have a solder bridge somewhere on the board. With out a schematic it seems hopeless. Theres a lot of surface mount chips on the foil side. I'm shelving it for now; maybe some day when I'm bored I'll pick it up again. I hate to quit but I have other projects.

    I did find a lot of capacitors with High ESR and low value, checking them was a good tip.

    Thanks for your support.

    eddie
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-