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

Hi-pot test of a circuit with a diode

  1. Apr 24, 2014 #1
    We manufacture crank case heaters and they are designed with an 800V 1A STD Recovery diode. We do not Hi-pot the heater with the diode in it we only test to make sure we have an ohm reading in one direction and an OL in the other.

    However, our customer installs our heater into a system and Hi-pot the entire system. They do a 1 second test at 1.8 kV and then a ramped test maxing out at 3.25kV for one minute. I claim they are damaging if not destroying the diode by doing that test with it in the circuit which is leading to hi-amp readings at their final QC station if not total failures.

    Not being an expert, I am just looking for some verification that their test could be or is damaging the diode thus causing failures. Now how and if I can design around their test is another issue.

    Thanks
     
  2. jcsd
  3. Apr 24, 2014 #2

    berkeman

    User Avatar

    Staff: Mentor

    Can you post a schematic of the system and how your heater is hooked up? Generally Hi-Pot testing is done with protection components removed from the circuit...
     
  4. Apr 24, 2014 #3
    Schematic for heater diode installation

    The schematic is rough but the circuit is simple. We integrate the diode into one of the leads of the heater after we test it and before we ship it. I think I attached the schematic.

    Thanks
     

    Attached Files:

  5. Apr 24, 2014 #4

    berkeman

    User Avatar

    Staff: Mentor

    Is the Hi-Pot voltage applied differentially across the heater wires (unlikely), or in common-mode with both wires with respect to the overall ground of the system (case of the heater?)?
     
  6. Apr 24, 2014 #5
    It is the common-mode.
     
  7. Apr 24, 2014 #6

    berkeman

    User Avatar

    Staff: Mentor

    Then the diode should not be affected.
     
  8. May 13, 2014 #7
    Mr. Berkeman,
    I am still not convinced that the hi-pot test does nothing to the diode. When we receive rejects back, the diodes are bad. The heaters make it past the hi-pot test but are rejected at final QC for pulling high amps. They pull the high amps because the diode no longer is functioning as it should. Here is their test procedure, do you see any potential here? I am still not comfortable with an 800V diode being tested at such high voltages. Thanks for your help.

    Typically a 1 second test is performed.
    Eg 460 volt unit
    Test would be performed on main input line of unit L1,L2,L3 to Frame Ground. It would pass through a set of fuses and a side switch on the contacor
    Test would start and voltage would ramp up to 3.25KV in 5 seconds then hold that value for 1 second.
    There is also an Arc Sense feature that is turned on. It is not a requirement but used to make sure that the operator does not move the test wand for the 6 seconds while the test is performed
    If the test indicates a failure the test would be performed after the heater fuses ( load Side) at 2.71 KV where the ramp up time is still 5 seconds however it would have to hold that voltage for 60 seconds.
    If there is a failure here the heater would be disconnected from all circuits and the minute test would be performed directly to the component wiring.
     
  9. May 13, 2014 #8

    berkeman

    User Avatar

    Staff: Mentor

    If the Hi-Pot voltage is applied to both wires, then there is no voltage across the diode. It seems unlikely that would be able to do any damage to the diode.

    How often do you get failures? What percentage of the devices fail final QC for bad diodes? If it's a high percentage, then I'd recommend testing the diode by hand before and after the Hi-Pot test, to see if it really is failing there. What other tests are run on the assembly?
     
  10. May 13, 2014 #9
    We get enough failures to be on the radar of our customer. And it is always for high amp draw with the occasional hi-pot. Less than 5% but as of recently there were 3 failures in a short time frame. The diode is enclosed in SS armor and sealed so it is difficult to test on the final product without destroying the heater. We hi-pot instantly when we get to the 3.25 kV at the plant so I have no idea how any of them pass.

    The heater assembly is ohms tested and hi-potted at 1.8 kV. Then the diode is installed and final ohms to verify diode installation is taken. The heater is installed into the system at the customer locale and hi-potted in accordance to the previous standard. Assuming it passes, it is then tested for amp draw in final QC which is where the majority of rejects occur.

    I appreciate your help, it is hard to find anyone who knows about diodes let alone hi-pot testing.
     
  11. May 13, 2014 #10

    berkeman

    User Avatar

    Staff: Mentor

    Are you sure the Hi-Pot with the diode is done by shorting the two wires together and applying a high voltage to the two wires together with respect to the grounded metal of the enclosure? Do you have a picture of this test?
     
  12. May 13, 2014 #11
    Schematics

    This is what they gave me.
     

    Attached Files:

  13. May 13, 2014 #12

    berkeman

    User Avatar

    Staff: Mentor

    Where does the diode go? How is the Hi-Pot test performed (L1+L2+L3 to Earth Ground?)
     
  14. May 13, 2014 #13
    The heater is wired to L1 and L2 after some fuses. The diode is either in L1 or L2 since there is only one in the heater and I am sure they do not know which lead contains the diode for it is encased. The only thing they explained was that the hi-pot test is hooked into L1, L2, and L3. They do not specify if they tie those together. But they are hooked to frame ground on the other side of the test.
     
  15. May 13, 2014 #14
    Sorry, print says it is hooked up to L2 and L3. But only two legs.
     
  16. May 13, 2014 #15

    berkeman

    User Avatar

    Staff: Mentor

    Is the unit powered by L2 and L3 during the test? Or is the connection to AC Mains open circuited during the test? Usually it is open circuited, but I'm suspicious that the test is putting a differential voltage across the diode, which could certainly damage it...
     
  17. May 13, 2014 #16
    The unit is a 480V unit and our heater is either a 240 or a 120 depending on the style called for. Are you asking if the heater is powered up? They state that there is an auxiliary contact that needs to be closed for the heater to turn on. What if it is not closed during the hi-pot test? Could that create a potential difference? I guess what question should I ask the plant in terms of their test so I can bring value added info back to this chat?
     
  18. May 13, 2014 #17

    berkeman

    User Avatar

    Staff: Mentor

    So your heater is powered by the differential AC Mains voltage difference between L2 and L3, when a contactor is closed, right? The question to ask is how the Hi-Pot voltage is applied to L2 and L3 during the test, and whether the contactor is closed during the test.

    In my experience with Hi-Pot testing, both of the input wires would be tied together (in this case L2 and L3, so there could not be any AC Mains voltage connection for L2 and L3 during the test), and that pair of wires has the Hi-Pot voltage impressed on it with respect to the chassis/Earth ground of the device. You could ask whether this is how they are doing the Hi-Pot test of your unit.

    If on the other hand, they are keeping the AC Mains connection to L1, L2, L3 during the test, and only applying the Hi-Pot voltage to one line at a time with respect to chassis/Earth ground, that could potentially cause some issues, I would think.
     
  19. May 13, 2014 #18
    I have shot them some questions and will keep you posted. Thanks again for the help.
     
  20. May 13, 2014 #19

    berkeman

    User Avatar

    Staff: Mentor

    Sounds good. BTW, what is the reason for the diode? Are you just wanting half of the heating effect from the AC heating element?
     
  21. May 14, 2014 #20
    To the best of my knowledge that is the reason. This is more of a warming element than heating.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Hi-pot test of a circuit with a diode
  1. Diode circuit (Replies: 2)

Loading...