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.

TIDA-050063: Active Precharge Circuit ref design, usable at low capacitance values?

Part Number: TIDA-050063
Other Parts Discussed in Thread: TPSI31P1-Q1, TPSI3100-Q1, TPSI31PXQ1EVM

Tool/software:

Hi.
My name is Tom and I have recently taken over a PDU design where a prior coworker has used this reference design and accompanying calculator to design a precharge for a 220uf system. this is a much smaller amount then the reference design. when I input the values into the component selection tool it seems to suggest a switch frequency over 1mhz which seems unreasonable.

I was hoping someone could confirm is this topology could be used for smaller capacitances like 220uf and possibly even recommend some acceptable starting values.

  • Hello Tomas,

    Thanks for reaching out to our team on E2E! This active precharge circuit can definitely be used for charging smaller capacitances.Typically the switching frequency is impacted by voltage, inductance, and current delta, but not capacitance. What values have you selected in the design/calculator for the voltage, precharge time, current delta, and inductance? If there is a schematic, we can determine the current delta based on shunt resistance and hysteresis circuit resistances. 

    Best regards,
    Tilden Chen


    Solid State Relays | Applications Engineer

  • Thanks for your reply.

    I have attached the existing schematics below.
    I believe the intention was to use a powerful 1ED3124MC12H 14A gate drive IC fed from a dedicated isolated 18V supply. in conjunction with a SiC MOSFET with very low gate charge to make the faster switching possible. to make up for the under sized inductor.

    Upon review of the supporting documentation provided I discovered the 18V isolated power supply selected wont be able to source nearly enough current.

    With a re-spin of the board guaranteed I can focus on a slightly more reasonable implementation of an active precharge. 

    Working on understanding the TI example and calculator a bit better. the switching frequency is dependent on the available power and the Qc of the FET. a smaller inductor will saturate quicker and thus needs switched quicker. So I suppose its a balance of the 2 to get the desired function.

    with a Qg of 10nC, the calculator suggests a switching frequency of 470khz is achievable using the example isolated supply. is this something you think is reasonable to implement?

  • Hello Tomas,

    Thanks for your reply. I think 470 kHz is going to be a tough switching frequency to hit when you account for delay in the circuit. The delays will result in the current overshooting the peak current target and the inductor will saturate. 

    Perhaps you can consider our refreshed active precharge design, TPSI31PXQ1EVM. Solution cost should be comparable to what is in your schematic, while charging faster, generating less EMI, and needing less board space. This would be able to charge 220 uF to 800-V in 83 ms (unclear what your precharge requirement is).

    Best regards,
    Tilden Chen


    Solid State Relays | Applications Engineer

  • Thanks for getting back to me. That sounds like an absolute win. less board space is the key one. the design is going into a PDU unit that needs to precharge 5 separate HV outputs independent of each other. 

    I hadn't realized there was a new active precharge design.
    Looking at it, a single chip solution is definitely preferable.

    Is there any more information available on the TPSI31PXQ1 chip used in the evaluation module?
    I had a search both internal and external to the TI site and drew a blank on datasheets or design guides. or indeed any way to order the chip separate to the evaluation module.
    Is this because the module is a new development?

    Cheers
    Tom.

  • Just to add a note I have put an order in for one of the TPSI31PXQ1EVM boards so that we can evaluate a baseline.

  • Hello Tomas,

    Thanks for your reply. Please see attached for TPSI31P1-Q1 datasheet. Yes, the device variant is new. The architecture builds off the TPSI3100-Q1 device family which we have been sampling for almost a year now. Apologies that the TPSI31P1-Q1 device product page is still unavailable, should go live any minute now Blush. Once the device product page goes live, there should be an option to order samples.

    Thanks for ordering the TPSI31PXQ1EVM. What is currently in-stock is rev B, which targets a 10-A average charging current. This is the newest board version, we made these changes since many customers seemed to want faster charging. But the EVM product page and user's guide do not reflect the updates yet at this time, they still show information for rev A (2-A average charging current). It seems like rev A is a better fit based on your schematic, I can ship you a rev A board free of charge if that helps with your evaluation, sending a private message to collect shipping information.

    TPSI31P1-Q1 Datasheet - Oct 2024.pdf

    Best regards,
    Tilden Chen


    Solid State Relays | Applications Engineer

  • Hi
    That's awesome news. I will have a look through the datasheet.

    I didn't realize there were two versions of the evaluation board, for this particular project 2A is definitely beneficial but as we are a design consultancy haveing the option of a 10A precharge is no bad thing.