This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

What does ISO5852S Split Outputs to Provide 2.5-A Peak Source and 5-A Peak Sink Currents

Other Parts Discussed in Thread: ISO5852S, UCC256302

Hi All,

I used ISO5852 as IGBT driver for PMSM. In the datasheet, it said: "Split Outputs to Provide 2.5-A Peak Source and 5-A Peak Sink Currents". Does it mean each ISO8252 need provide 2.5-A peak source and 5-A peak sink currents to each IGBT. If I used 24VDC power to provide to ISO5852, and I have 6 ISO5852s and 6 IGBTs, I need prepare 6x5=30A maximum for 3 phase PMSM. If I use UCC256302 to generate 24VDC power for the driver and IGBT, the output current Iout = 30A for maximum. Is my calculation for the current for my driver to IGBTs correct? Or can you let me know the different calculation for the current of driver for IGBTs?

Best,

Hao

  • Hao, 

    Gate drivers can only source / sink their Peak rated current for short periods, they cannot be used like load switches, and for driving IGBT/FET, they dont need to be. 

    The average current of the driver is any quiescent/operating power plus the driver switching power: Qg*fsw

    or Cgate*(VDD-VEE)*fsw. Where Qg is the gate charge of the fet and in device datasheet. essentially it all comes down to the device youre driving and how fast to switch. 

    Even if you design for peak current at 2A or something this happens only for a very short time. Your average current is usually in 10-100s mA range. For example,

    lets say youre driving a IGBT with gate charge 500nC, and switching at 50kHz. That ISO5852S driver stage's average current is only 25mA (plus quiescent current for logics). 

    So yeah you dont need to have a 50A capable supply. The power would be ridiculous. Each gate driver need at most just a few watts each. 

    If this answers your question, please press the green button. 

    Best

    Dimitri

  • Dimitri,

    Thank you for reply. I understand your consideration and calculation are correct. I also refer to an other resource. If my driver uses the power are 15v and -8v. I also add an serier gate resistor which the value is 10Ohm. I receive the gate current is (15-(-8))/10 = 2.3A. So it is pretty high. Can you let me know if my calculation is wrong?

    Hao

  • You must also consider the Drain-source resistance of the internal FET. Note that this also changes over the entire turn-on/turn-off period, since Ron changes with FET Vds, and the fact that our drivers have hybrid pullup structure with N/Pch FETs in parallel .

    You can use the values of "ron-max and roff-max" located in page 30 of the datasheet, but these values are max, for the purpose of calculating the worst-case power dissipation in the driver. 

    I would suggest using the following values for calculating peak source/sink current. : Roh_eq (internal peak on resistance) -> 3.2Ohm, 

    Rol -> 0.65Ohm . Note that these are not datasheet specifications and used for estimate purposes only. 

    Add your gate resistance to these values and recalculate, your peak current calculates to roughly 1.7A, below threshold. 

    These values will be close, for the purpose of peak current calc, If this answers your question, please press the green button. 

    Please keep in mind these are estimates, the actual peak source current may vary

    If this answers your question, please press the green button. 

    Best

    DImitri

  • Dimitri,

    As you calculate, the peak current was roughly 1.7A. If I have 6 ISO5852 to drive 6 IGBTs for a 3 phase PMSM, I need prepare the current around 10A (1.7Ax6). So the power I need prepare is 10^2*13.85 = 1385W. Please let me know if I am wrong.

    Thanks,

    Hao

  • Hao,

    Please see my first response to your post. This question is fully answered in that reply.  

    The average power is based on the gate charge of fet/IGBT.

    you will find it will be on the order of 10-100mA per driver. This is NOT based on peak current. Peak current happens for a very short time. The average current is an order or 2 orders  of magnitude lower than peak current .

    If we think about it, 1.7kW is a massive amount of power. The ICs would melt. 

    Best 

    Dimitri