Ask ADUM4136 DESAT fault under ringing condition

Dear  Sir / Madam :

I use ACPL-36JT in my Gate drive board design to drive IGBT model; there is DESAT fault problem on switch ringing.

For example: 

my High voltage is 270V;

when switching, The switcher collector Pin can go to -10V; then ACPL-36 make a fake DESAT fault.

Could you tested ADUM4136 under this ringing condition?

If there is no fake DESAT fault; I'll replace gate drive chip with ADUM4136.

Thanks

Wei Liu

Electrical Design Engineer

GE Aviation Systems

 

T +1 937 204 1357

wei2.liu@ge.com

 

111 River Park Drive

Dayton, OH 45409

  • 0
    •  Analog Employees 
    on Mar 17, 2017 2:29 AM over 3 years ago

    Hello Wei,

    I have a few questions about your conditions:

    1) Are you running in a unipolar or bipolar setup? Do you have a voltage source between VE and VEE on the ACPL36JT, or are they connected together?

    2) Is there any protection on the DESAT pin like a schottky diode from DESAT to VEE?

    3) What frequency is the ringing that goes to the -10 V?

    The ADuM4136 uses internal transformer coils to communicate across the isolation barrier, unlike the ACPL-36JT, which uses LEDs. The DESAT pin on the ADuM4136 can not go below -0.3 V referenced to GND2, or the part risks turning on an internal body diode. This can be prevented in the situation you are showing by placing a schottky diode from DESAT to GND2 to clamp the negative voltage at the DESAT pin to the Vf of the schottky diode. R12 provides current limiting, and should be sized to allow for an approximately 10 V drop for the duration of the ringing you are seeing.

    RSchnell

  • 1:I'm running  on IGBT module, voltage source for IGBT is 270V; current is 120A.

       ACPL36 is powered by +16V/-8V and +5V for logic.

    2: we have added  schottky diode protection for DESAT when we found this issue. 

    3: Our IGBT switching frequency is 13Khz, the ringing is narrow, I think it's created by big current and inductor on the trace and circuit.

    by the way, I try to test it on EVB so  I ordered EVAL-ADuM4135EBZ and tested:

    VDD1 to Vss1 = 5V;

    VDD2=15V; VSS2=-8V;

    (I connected D1 negative side to GND2 for DESAT detect) 

    when I power on the EVB; Output is -8V, nFault is high. Ready is high. nReset is high. VI+=0, VI-=0; 

     

    1: I connected VI+ to signal generator for 10Khz, 0 to 4V square wave; nFault is Low, why?

     

    2: I disconnected  signal generator, use jump wire connect VI+ to VDD1; nFault is Low, (Output is -8V).

     

    any VI+ input high then the nFalut goes to low?

    This is new EVB I order from Digikey

     

  • 0
    •  Analog Employees 
    on Apr 5, 2017 1:37 AM over 3 years ago

    Hello Wei,

    Are R12 and D1 populated? It sounds like the DESAT pin is getting to the 9 V threshold, and reporting a fault. You can try shorting the DESAT pin to GND2. This will disable the DESAT detection. In normal operation, the IGBT provides a path for the DESAT pin to be grounded, but if not IGBT is in place, we needto hold DESAT below 9 V. The easiest way to do this is place a 0 Ω resistor in the C6 placement.

  • 0
    •  Analog Employees 
    on Aug 2, 2018 3:46 PM over 2 years ago
    This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin