Connect with us

PSU Silverstone ET550-G dead after thunderstorm (turned itself off)

Discussion in 'Power Electronics' started by hansmuller, May 4, 2019.

  1. hansmuller

    hansmuller

    4
    0
    May 4, 2019
    Hello members ,

    The PSU was running when a lightning bolt struck far away. turned off and cannot power on again.
    so probably overvoltage exposed.
    I opened the PSU to replace the fuse. it is a ceramic one, there I cannot see on first sight if it is blown. Also removal seems unnecessarily difficult as it is soldered and no holders that may clamp it.
    Did a sight inspection and did not see anything apparent.
    As i do not have good soldering equipment, do you recommend to bring it to technician to solder replace the T10A fuse?
    What else maybe to check before that?

    with best regards
    Hans
    IMG_20190503_225704131_1.jpg IMG_20190503_225522646_1.jpg IMG_20190426_181433784_1.jpg IMG_20190426_175347479_1.jpg

    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
     
  2. Martaine2005

    Martaine2005

    2,309
    616
    May 12, 2015
    Hello Hans,
    There are many things that can be checked. First would be the fuse. Do you own or have access to a simple multimeter?. They can be bought for as little as £10.
    Or a diy continuity tester, a piece of wire, torch bulb and battery.
    upload_2019-5-4_19-28-9.gif
    Where the switch is, simply connect across your fuse. If the bulb lights, the fuse is fine.
    Let us know how you get on.

    Martin
     
  3. Bluejets

    Bluejets

    3,422
    690
    Oct 5, 2014
    Might be the photo but top of cap beside D22 looks corroded also.
     
  4. dave9

    dave9

    704
    158
    Mar 5, 2017
    Even if the fuse is blown, don't replace it yet. Odds are your primary side switching transistor(s) (mounted on the longer heatsink) are blown and will need replaced first, or possibly a bridge rectifier diode pack (or both).

    As far as "bring it to technician" to do anything, you will probably find the labor rate is higher than (a used PSU) it's worth. If you don't even have a soldering iron sufficient for replacing a fuse, it would seem like you're putting the cart before the horse in trying to repair it in this particular case.
     
  5. dave9

    dave9

    704
    158
    Mar 5, 2017
    Then again this is an electronics forum. If you're a gambler and a DIY type of individual who would rather buy tools than pay someone else the labor, with the idea that you might not do much better than break even (??) but have the tools to use again next time, then there exists the possibility that you could educate yourself about troubleshooting transistors and diodes, invest in a sufficient soldering iron, and dive in! A very frugal person could even calculate that you could still save a lot of money, a budget grade soldering iron of 40W or more (might not need that much but being conservative) could be had for $10. A multimeter could be had for $10. Replacement transistors or diode bridge could be had for $5, roughly for all these figures.

    It could easily be cost effective for you to repair it but the investment is your time. What you learn could be reapplied later in life, but there is a reasonable limit to the "learn everything there is" philosophy of life rather than spending time on things that are more repetitive.

    I did not venture into electronics based on an event like this, but everyone has to get from point A to point B somehow. It's up to you...

    I would look at whether I had decent power surge protection and consider upgrading that.
     
    Last edited: May 5, 2019
  6. hansmuller

    hansmuller

    4
    0
    May 4, 2019
    thank you @Martaine2005 . it turned out that my very basic multimeter was not doing any more. had to wait for delivery of a new chinese. arrived today and checked the fuse right away. The fuse is not blown.
    [​IMG]

    @Bluejets. yes I had checked all capacitors at sight and did not see any corroding capacitors. here a photo of capacitor tops of both C34 and C60. Seems i forgot to mention this PSU is just 1.5 years old.
    [​IMG]


    thank you @dave9
    yes exactly i like to learn stuff like this. ordered a few cob leds last week as well to upgrade low lumen led chips in e27 bulbs. I do have a basic gun like solder iron and solder and paste.

    as you assumed already the fuse is ok, no resistance measured.

    Well I could bring it to technicians here in my low wage country of residence, however i would like to see first how large the damage is and learn something on this way doing it. Maybe it is something i feel confident to do myself and i do not like to throw a rather nice, young and not so cheap piece of electronics to the garbage.

    attached a photo of your mentioned heat-sinked transistors. I am not sure - which one/ where is the bridge rectifier diode pack ?
    Another day I will look into a power surge protection yes.


    [​IMG]

    IMG_20190618_181647870.jpg IMG_20190618_183015158.jpg IMG_20190618_183234406.jpg
     
    Last edited: Jun 18, 2019
  7. Bluejets

    Bluejets

    3,422
    690
    Oct 5, 2014
    With almost zero electronics knowledge you may have set your starting point just a little to high.
     
  8. hansmuller

    hansmuller

    4
    0
    May 4, 2019
    would not call it zero, yes may be in over my head, but I can tell if a capacitor is corroded. once had such a mainboard in times of the capacitor plague.
     
  9. Bluejets

    Bluejets

    3,422
    690
    Oct 5, 2014
    So can my 2 year old grandson but I very much doubt he could go any further.
    Instances of corroded or bulged capacitors do happen but not as frequently as other smps problems especially those caused by power surges.

    There are some here that will "walk you through" any diagnostics but in this case, a circuit diagram would be necessary.
    Not so much for the professionals here, they can navigate almost with their eyes shut.
    More-so so they can point to components and refer to them in an alpha numeric form so you can follow.
     
  10. hansmuller

    hansmuller

    4
    0
    May 4, 2019
    so just tested the bridge rectifier
    with the multi meter in diode mode

    IMG_20190619_221209885.jpg

    all diodes only allow the current in the right direction giving a reading of about 540 mV each
    as shown by example in the shot
    so i can confidently say that the BR is okay as it is working as expected
    will try the transistors tomorrow
     
  11. 73's de Edd

    73's de Edd

    2,609
    1,063
    Aug 21, 2015
    Sir hansmuller . . . . .



    Well . . . .since you now do have it all apart , I will assist you a bit. . . . .BUT . . . . . that lil' puppy is just BRISTLING with uber circuit complexity.

    I am referring to the last and very bottom photo, of post 5's clustered series of 5.

    The AC power enters at the top right and passes down thru components until it reaches your just tested Full Wave Bridge Rectifie block.
    It stops being AC there and is converted to pulsating DC which must be passed on to the left to the large black E-filter *** which then stores up
    a reserve of that DC voltage that is now no longer pulsating , but being a pure steady DC power.

    (****Dat's being the one with the top label, that sez that inspector C plugged the unit into power for 5 minutes and that it ran and he passed it . . . .
    so ? I don't know why you are having trouble with it already ? )

    You should find your FWB rect negative connection making direct connection to that E-caps 1 bar /stripe down the side negative terminal .
    If your pic of the probes across the FWB . . . AND . . . . if being in OHMs mode instead of Dode Test mode, you may ahve already passed the next test.
    BUT, to be sure, you want metering in low ohms range and short the probes together to confirm how the display reads on a dead short.
    Now look at the heat sink mounted, central, Q2 POWER transistor position and you see that its center lead has to be bent a bit forward to mate into its hole in the PCB while the two side ones pass directly down into their holes.
    That central pin is being our one of interest on Q2 and on Q1 ? beside it and also Q3-Q4 at the far end.
    To test and see if possibly Q1-2-3 and 4 might have a dead CRUNCH short in it /, as their failure mode(s).
    You test from a center transistor pin to the ground terminal of the main E-filter cap. Do a probe polarity reversal each testing time also.

    If no hard shorted transistors . . .GOOD.
    Also if wanting to do some further diode testing go back to diode test set up and check the large diode in front of Q2.
    Then you may need to pass ID of the D11 diode numbereing, as it seems as a single diode between its two pins BUT it also could be a
    full wave rectifier with the diodes ends sharing the common heat sink tab at the top and each diode free lead at the exposed pins at the PCB holes.
    Expect a much lower forward voltage reading than the other FWM tested units on this /these D-11 types of diode(s).

    Just Relative Info . . .
    The SMALLER green conformal coated power transformer behind Q1-Q2 is the standby transformer and it provides continual low power for >48hrs a day!
    The LARGE green conformal coated power transformer behind Q3-Q4 supplies all the main power, when switched on.
    Pass on the ID of the 8? pin IC, mounted between the Sby tran and the finned heat sink. Also D11 if you like.

    Thaaaaaaasssssssssit . . .for now. . . . . .

    73's de Edd . . . . .


    The average man's judgement is so poor, he runs a severe risk every time he uses it.


     
  12. Bluejets

    Bluejets

    3,422
    690
    Oct 5, 2014
    Word of warning...discharge those HV caps before sticking your fingers in there.
    They will hold quite a wack for a long time after being isolated from the supply.
    I use a 1K5 2w resistor with a couple of probes soldered to the end of 150mm extension leads.
     
  13. TCSC47

    TCSC47

    31
    9
    Mar 7, 2016
    I think you have gone as far as is practical with such a fault. You have shown it is not a blown fuse and that the rectifier seems ok. Without a circuit diagram and more knowledge than we can supply simply here I would seriously suggest you bite the bullet and buy a new PSU.
    Two things would stop even me going further. The first is the critical nature of the PSU. It could fail with over voltage and blow up the equipment it is powering and you will be even worse off. Damage has been caused to one or more components but some may have only been stressed and may fail in the near future.

    The other is the way the lightening strike may have caused damage to the PSU. It could have been a temporary over voltage down the mains line but it could have also been the electromagnetic pulse (EMP) from the lightening. I have seen this happen. Absolutely any component may have been damaged and you will require a detailed knowledge of what you are doing to find them.

    If you are interested in electronics, then soldering is one of the necessary skills you need. I would suggest buying a decent soldering iron and solder sucker. Practice de-soldering the components and start a junk box collection. Maybe you could make some experimental circuits with the bits you take out or you could learn about them by testing them with your multimeter. You can google how to do this.

    I do not wish to sound condescending, but the thing to me that indicates your are at the beginning of learning about electronics is your picture of the multimeter reading when testing your rectifier diodes. You say the reading is 540 mV but it is actually 0.540 ohms.

    Anyway welcome to the world of electronics. We are all learning all the time.
     
  14. 73's de Edd

    73's de Edd

    2,609
    1,063
    Aug 21, 2015
    Sir hansmuller . . . . .

    Where he go o o o o o o o o .o o o o ? . . . .Why you no come-a- back to us ?

    et

    but . . . . . . .but . . . . . .BUT... BUT... BUT.

    " your picture of the multimeter reading when testing your rectifier diodes. You say the reading is 540 mV but it is actually 0.540 ohms."

    Au contraire . . . .

    Look vewy,vewy, VEWY *** closely at . . . .

    The upper and lower lettering of the different range scales, with the diode / ohms scale being the one of interest.
    The yellow lettering aspect is indicative of the right orangish yellow SEL button having been pressed, and the unit is now being in biased, diode test mode , with the diode symbol being displayed at the display top and the V at the display bottom . . . . . . . . ergo . . . . . with a measurement of 0.540 Vf across the tested, forward conducting semiconductors junction.

    fin

    *** Copywight 2019 . . . . .Elmer Fudd

    73's de Edd . . . . .

    Good example of a Brain Study:
    If you can read this you have a strong mind:
    *********************************************

    7H15 M3554G3
    53RV35 7O PR0V3
    H0W 0UR M1ND5 C4N
    D0 4M4Z1NG 7H1NG5!
    1MPR3551V3 7H1NG5!
    1N 7H3 B3G1NN1NG
    17 WA5 H4RD BU7
    N0W, 0N 7H15 LIN3
    Y0UR M1ND 1S
    R34D1NG 17
    4U70M471C4LLY
    W17H 0U7 3V3N
    7H1NK1NG 4B0U7 17,
    B3 PROUD! 0NLY
    C3R741N P30PL3 C4N
    R3AD 7H15.
    PL3453 F0RW4RD 1F
    U C4N R34D 7H15


     
    Last edited: Jun 27, 2019
  15. TCSC47

    TCSC47

    31
    9
    Mar 7, 2016
    Well, there you are. I have learnt something.
     
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

-