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.

UCC21521: Dual low side driver configuration

Part Number: UCC21521
Other Parts Discussed in Thread: UCC5390

Hello,

I am using the UCC21521 in a dual low side configuration to drive parallel SiC MOSFETs in a non-synchronous boost converter application.  Is there any guidance or reference designs for this configuration?  It is mentioned in the data sheet but not shown as an example anywhere.  My specific questions is related to VDDA and VDDB.  I am deciding between using separate 2W isolated bipolar DCDCs to power them individually or one larger isolated bipolar DCDC to power them both.  I do not anticipate issues either way, but always like to see a reference design or application guide for a specific configuration. 

One other question - as this configuration defeats by design the channel to channel isolation - is there a similar dual gate driver that is cheaper that is not isolated channel to channel?  Is there a single channel equivalent?  THis seems like a newer product and I cannot find the single channel equivalent. 

Thanks!

  • Hi Tim,

    I am an applications engineer supporting UCC21521. At this time, I think we do not have any reference designs with UCC21521 outputs in parallel driving SiC MOSFETs. I will double-check this and update you if I find such a design. For now, I recommend you look at the white paper UCC21520: A Universal Isolated Gate Driver with Fast Dynamic Response, section 4.2, regarding using two output channels in parallel.

    Since the channels are tied together, I recommend using a single bipolar supply for both. With the dual configuration, any asymmetric noise on the supplies could result in output contention.

    While at this time there is no dual-channel IC with no channel-to-channel isolation, there are single-channel isolators which may be less expensive. Take a look at the UCC5390 family, specifically the "E" variant for bipolar supplies.

    Regards,

  • Hey Derek,

    Thanks for the fast and detailed response. I will check out UCC5390. I have read that white paper, but let me clarify my question. I would like for each channel of the UCC21521 to drive a separate SiC MOSFET. The FETs themselves will then be in parallel, not the gate drive. I shouldn't need more gate drive strength than one channel can provide. Do you have anything for that setup?

    Thanks!

    -Tim
  • Hey Derek,

    Sorry if this appears twice, looks like my reply got eaten or is maybe still in an approval queue.

    Thanks for the fast and detailed response. I am actually trying to drive 2 SiC FETs per UCC21521, so I will be paralleling the FET, not the gate drive. I should not need more drive strength for my FETs. Do you have any ref designs or advice for that setup?

    I think single supply or dual supply will work. It may just come down to trading off cost, reliability, and ease of layout.

    The UCC5390 looks promising and may ease layout more.

    -Tim
  • Hi Tim,

    I checked through our reference designs, but I didn't find any with two SiC FETs in parallel on a single output. My advice for paralleling SiC FETs , or any MOSFETs in general, is as follows:

    • The outputs should share as much trace length as possible, and the length of the split segments should be equalized to minimize layout-related propagation delay mismatch.
    • The connections to transistor gate and source should be run very close together to minimize loop inductance. Keep this loop as short as possible.
    • Keep the split supply bypass capacitors close to the IC supply pins.

    Regards,