Simple HV power supply with ripple problems

Discussion in 'Power Electronics' started by Sjöholm, Mar 15, 2019.

  1. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    Hejsan all,

    I'm building a simple HV power supply, but no matter what smoothing cap I throw at it, the ripple is mostly unchanged - I guess I need a special capacitor, but which?

    I tried MKP and foil capacitors ranging from 1nF to 200nF to no avail.

    This is the LTSpice sim. I'm using CL01-12 diodes (didn't have the spice model).
    Screenshot from 2019-03-15 20-55-11.png
    And this is how it looks on the scope (multiply voltage with 1000):
    IMG_20190315_203358.jpg
     
  2. crutschow

    Expert

    Mar 14, 2008
    23,081
    6,848
    So how much ripple can you tolerate?
     
  3. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    So that I can put a IRG4PC50W across the output and switch it, I tried with the ripple as is, but the darn thing wouldn't turn on.

    Also I'm a bit baffled by the fact that even with a 1nF cap the sim shows a few volts ripple, with a 900nF foil cap in real life I can't get even close to what the sim shows.
     
  4. wayneh

    Expert

    Sep 9, 2010
    16,102
    6,219
    Do you really want R1 to be 1G? That's a fantasy in my world. What happens in the sim if you take that down to 1M or less?
     
  5. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    I use a 1G glass resistor in series with a 1M film resistor (not in the diagram) to make a voltage divider so I can measure various HV's. The 1G will obviously not be the final load.

    As per your suggestion I tried with various values of R1 in the sim, and only at values below 4K you start to see some ripple about 10V, the current being around 200mA which the transformer can supply.

    So with 1G I have about 200V ripple, why? I'm perplexed.
     
  6. sghioto

    Well-Known Member

    Dec 31, 2017
    1,039
    157
    What is the DC voltage output?
    SG
     
  7. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    As you can see on the scope picture, it is 600V peak with 200V ripple.
     
  8. sghioto

    Well-Known Member

    Dec 31, 2017
    1,039
    157
    Those CL01-12 diodes are for microwave power supplies. Will they operate at 50KHz?
    SG
     
  9. thedoc8

    Active Member

    Nov 28, 2012
    126
    11

    I see 200mv ripple not 200volt and what do you expect to filter with 100n. What am I missing.
     
  10. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    You missed "And this is how it looks on the scope (multiply voltage with 1000)" i.e. 200mV * 1000 = 200V
     
  11. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    Yes, and I tried with some 2CL77 Fast Recovery, the ripple problem persists.
     
  12. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    With BY203-20 diodes

    IMG_20190317_151513.jpg
     
  13. wayneh

    Expert

    Sep 9, 2010
    16,102
    6,219
    Can you experiment with turning down the frequency? And goosing C1 up to 1µF or more.
     
  14. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    It's a fixed zero voltage crossing transformer exciter so I can't adjust freq.

    But I think I have figured it out now, apparently the ~600nA though the 1G resistor wasn't enough to turn on the diodes properly, so I switched that to a 1M resistor and it now looks like below. I will keep the 1M resistor as ballast. Case closed. Thank you all for your suggestions.

    IMG_20190317_155836.jpg
     
    sghioto likes this.
  15. wayneh

    Expert

    Sep 9, 2010
    16,102
    6,219
    Suspected as much.
     
    Sjöholm likes this.
  16. Sjöholm

    Thread Starter Member

    Jun 13, 2018
    35
    2
    Yes your suspicion was right, here is your prize ,-) And I gave you a thumbs-up.

    first_prize.jpg
     
    wayneh likes this.
Loading...