Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Need ideas on how to induce spikes on 600VDC signal

  1. Jul 24, 2012 #1
    I have an issue I hope you guys will help me solve.

    I need to re-create a problem seen on Rail equipment where high voltage spikes are affecting equipment (resistive load) normally running on 600VDC.

    I thought about using a simple RC circuit to produce the spikes over the 600VDC nominal voltage but it would not work because:

    - Spikes need to be of a certain length (100 - 300 us duration) and a certain amplitude (1.5 up to 5 kv ) a simple RC circuit could not modulate and ramp up the spike amplitude accurately.

    - Spikes will probably feed back to the power supply and damage it.

    I am looking for ideas on how to inject voltage spikes on a 600VDC signal. Thoughts and ideas are more than welcome.

    Thank you

  2. jcsd
  3. Jul 24, 2012 #2


    User Avatar

    Staff: Mentor

    Welcome to the PF.

    Are you familiar with the EN 61000-4-x series of immunity tests? They are done on electrical/electronic equipment in order to get a CE Mark, to be able to sell those products in the European Union (and other places).

    The EN 61000-4-5 Surge test, or the EN 61000-4-4 Electrical Fast Transient tests may be what you are looking for... The equipment to run these tests is fairly expensive, but test time at local test labs is not too bad.
  4. Jul 24, 2012 #3
    Thank you for the quick reply berkeman.

    I am somewhat familiar with the fast transients tests but I doubt that the same equipment could be used in this case.

    For my experiment, I need to be able to generate the spikes "on top" of the 600VDC nominal voltage. The fast transients created by the surge equipment used in a test lab would in my opinion either be absorbed by the power supply filters or eventually damage the power supply.

    I made spikes tests on my equipment using what is commonly called a Hi-Pot tester but the results were not conclusive enough. That is why I am trying to replicate the same scenario seen by the equipment when used on a rail car.

    Thanks again,

  5. Jul 24, 2012 #4
    The CE shock gun is a good idea, You can also try the stun gun for cheap. You don't need to even tough the rail, just let it jump.

    Arc is RF, your rail might be very low impedance at low frequency, it is quite high impedance at RF. That's how they transmit data through wall plug AC lines for control in the house. My first guess is they are much higher than 50Ω in RF world. Get a stun gun and try it. Either it work or not, cheap to try than to talk!!! I think you can get it in a lot of places for cheap.

    Don't the power station have arc protection? If a stun gun can bring down the whole power station, that's sad!!!
  6. Jul 24, 2012 #5
    I am currently building a test setup to replicate what the rail equipment would see when powered by a sub station. I will be using an AC-DC power supply to feed the 600VDC to the test unit, not a sub-station.

    I also need to be able to modulate the amplitude of the spikes which I could not do with a stun gun.

    I like the idea of using a stun gun but I will have to keep that as a last resort for the moment.

    Let me know if you have other ideas.

    Thanks anyways.

  7. Jul 24, 2012 #6
    What is the impedance of your 600 V line? In other words, if you generate a 1.5k or 5k spike, how much current will that require? How much current is being drawn at 600 V when the spike occurs?

    What are the rise and fall times of your voltage spike?

    Do you have an adequate source to create the spike at the required voltage and current?
  8. Jul 24, 2012 #7
    Current for the spikes should be fairly low. At 600VDC, current consumption is around 10-12 Amps.

    Rise and fall time should be fairly quick, I would say around 10% of the spike duration. It is not the rise and fall time I am concerned about, it is really how to produce the spikes on top of the 600VDC.

    The main voltage source for the experiment will be coming from a shop disconnect box either 480VAC or 575 VAC limited by 100Amps fuses. AC-DC power supply can accept either.

  9. Jul 24, 2012 #8


    User Avatar
    Gold Member

    yes +1
  10. Jul 24, 2012 #9


    User Avatar
    Gold Member

    Well yes that's what the power supply filters are supposed to do, absorb the transients, and frequently what that surge/transient equipment is designed to test. You have proposed a contradiction: you want impose 1.5kv spikes but are afraid the spikes may damage connected equipment. So which is it?
  11. Jul 24, 2012 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    there used to be something called IEE SWC acronym for "Surge Withstand Capability" test. It's now ANSI C37.90.1, and SWC test generators are about.

    The test applies spikes of shorter duration but similar voltage to yours.

    It came about 1968-ish when electronic relays began to replace electromechanical ones in substations. Its purpose was to simulate the voltage induced in sensing leads during rapid collapse of fields related to substation switching, which were eating the newfangled solid state relays.

    A search on that phrase might lead you to some test generator schematics. I can remember seeing some ca 1972, those'd probably be simple enough for your guys to fabricate one out of the local junkpile.....

    old jm
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook