Diode oscillates at low currents?

Discussion in 'General Electronics Chat' started by Maverik, Feb 20, 2014.

  1. Maverik

    Thread Starter New Member

    Jul 24, 2012
    5
    0
    So we're using a diode that is something like this except a little older so I can't find a datasheet.

    It's rated at really high currents and it works really well when we put large amounts of current through it, but we're noticing that when we put lower currents (<15amps) we're getting a ~20 Hz oscillation on it. I'm attaching pictures for reference.

    Has anyone ever encountered this? Do you know what might cause this?
     
  2. #12

    Expert

    Nov 30, 2010
    16,354
    6,852
    Are you saying the whole circuit is a battery with a diode across its terminals?
     
  3. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    it might be a slight problem with the diode. if the doping were wrong, it might be acting like a tunnel or Esaki (sp?) diode. those exhibit negative resistance and will oscilate. I have never seen one do this though.
     
  4. crutschow

    Expert

    Mar 14, 2008
    13,056
    3,245
    Post the circuit diagram.
     
  5. Maverik

    Thread Starter New Member

    Jul 24, 2012
    5
    0
    Lol, sorry I totally forgot about the circuit around it.

    We're boosting up a voltage and the bucking it back down. The reason I thought it was the diode was because we bypassed it and we got the smooth output.

    Today I hooked up the diode across a power supply and limited to 10 amps and under and got no oscillations, so I'm thinking it has something to do with the combination of the output filter and the power diode to give us a weird control loop.

    Anyone experience anything like this before?
     
  6. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Put a large cap (with LOW ESR) directly on the output of the Buck module.

    Your circuit as it stands allows the diode characteristics to interfere with the closed loop voltage regulation of the buck module, causing it to oscillate.
     
  7. Maverik

    Thread Starter New Member

    Jul 24, 2012
    5
    0
    I'll try that on Monday, thanks.
     
  8. tindel

    Active Member

    Sep 16, 2012
    568
    193
    Constructive critisism, please don't take it the wrong way:

    Why are you boosting, bucking, and then using a crappy (2V!) diode to produce 43V? You'll end up with horrible efficiency and poor regulation... Why not just boost your 24V supply to 43V and be done with it?

    If you're using a diode to 'diode-or' 2 supplies then I get what you're doing, but even so, there are much better power diodes that provide less voltage drop than 2V at heavy loads, and it would be better to just have a single converter stage still.
     
    Last edited: Feb 22, 2014
  9. Maverik

    Thread Starter New Member

    Jul 24, 2012
    5
    0
    I'm using a boost and then buck because the buck offers current limiting. The voltage drop is estimated, I'm not concerned with efficiency as much as I am about proof of concept and this is injected into a larger system.
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, those supplies are a couple of grand each; why not consult with the manufacturer and get ONE boost-type supply that will meet your requirements? Also, you could replace that diode with (a) MOSFET(s) used as ideal diodes to eliminate the majority of that 2v drop across the diode.
     
  11. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    I have seen supplies like this before, but they were used at line levels to boost the line voltage to around 350 volts then had multiple output modules running off them for whatever output voltages you need. sola makes some, and astec makes them also.
     
  12. Maverik

    Thread Starter New Member

    Jul 24, 2012
    5
    0
    I'd love to have the manufacturer build a specific converter for us, but he was not interested. We're not ordering a lot (University Research), so he isn't really inclined to spend much time with us. I will look into replacing the diode with MOSFETs, it's just going to be slightly annoying to add additional driver circuits and the sort as we run around 100-300 amps through here.

    Thanks everyone for the input, it has been very helpful.
     
Loading...