Drawing too many amps from SMPS, still works - but not that well?

Discussion in 'Power Electronics' started by Nicholas, Aug 21, 2016.

  1. Nicholas

    Thread Starter AAC Fanatic!

    Mar 24, 2005
    121
    1
    Sorry for the strange description, but here is my issue:

    I have been using a 36V 10A 360W SMPS to power a coil in bursts. It worked well for some time. Now it doesn't work
    that well; I find out that the coil has a resistance of 2 ohm, which means it draws 18A. Now the power supply cannot
    energize the coil, although it is trying to move (solenoid). I switch the coil to a 4ohm one (drawing 9 amps), still cannot
    energize it. Only tries to do it.

    So, my question is, can the power supply be made 'somewhat defective' in this capacity? Not to be able to fully drive
    something...but not dying? If so, can I fix it?

    Thanks!

    Nicholas
     
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,791
    1,103
    That would obviously depend on the design of the particular power supply, but I would say yes. For example, the supply might have a current-sense resistor which has been 'cooked' through excess current and has changed its value, consequently causing the supply to limit the output current. Or the main switching transistor could have been damaged and be operating in a crippled way. Or a solder joint may have over-heated and gone high resistance. Or .......
     
  3. Nicholas

    Thread Starter AAC Fanatic!

    Mar 24, 2005
    121
    1
    Thanks for the reply, it sounds reasonable!

    Well, maybe I should open it up...or order another one.
     
  4. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    783
    The electrolytics on the secondary side of an SMPSU work hard anyway - harder if you've been overloading it.

    If they run hot; the electrolyte can be dissipated until the capacitor ESR rises.

    Usually; failing electrolytics build up a head of steam and bulge the top of the can. But occasionally they lose enough electrolyte so there's a vacuum inside the can when they cool down. The sign of that is a slightly sunken top.

    Quite predictably; a capacitor sans electrolyte feels lighter than a healthy one of similar size.

    Failing electrolytics in a SMPSU also get hot when its running - but don't poke around in the primary side, the cans aren't isolated.
     
  5. Nicholas

    Thread Starter AAC Fanatic!

    Mar 24, 2005
    121
    1
    Hi guys

    I just opened the SMPS. It has two C3320 transitors, which is rated at a max of 15A each (if I understand it correctly). Could I assume that the 18A drawn by the coil has damaged these transistors?

    Thanks!
     
  6. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,791
    1,103
    You can; but your assumption may be incorrect :). Transistors are among the most likely components to fail. As per Ian's post, do any of the electrolytic caps look suspiciously deformed/leaky?
     
  7. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    783
    If those transistors are on the primary side of a SMPSU; there is a transformer turns ratio between them and the current you were drawing.

    The current rating of those transistors isn't an exact direct relationship to that ratio - they have to have some safety margin for spikes etc. You also have to allow for the fact that PWM means the transistors aren't conducting 100% of the time - but generally speaking, the collector current will be less than in a stepped down output.
     
  8. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    783
    If it doesn't have good enough overcurrent shutdown to protect the transistors - I wouldn't bother repairing it.

    Other than severe mains transients - the 2 most common causes are dry joints and high ESR capacitors.

    Usually there is a small electrolytic that samples the chopper pulses for regulation - high ESR there can cause a fairly entertaining sequence of events........

    A dry joint in any of the circuitry that charges or samples that cap, will have similar results.
     
Loading...