• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 183 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ76952: Driving SiC or GaN FETs with BQ parts

Harry McNally
Intellectual 375 points
Part Number: BQ76952

Hello

I haven't found any discussion, white papers, or forum questions about using SiC or GaN main switch FETs for BQ battery protection devices.

Is this discounted because of the drive requirements and reduced Vgs that have to be observed or is price sensitivity limiting their adoption ? I assume manufacturing volumes will drive device cost down so are they being considered now for future BMS designs ?

For surge robustness and low Rds(on) energy losses, these newer parts are attractive.

This TI discussion promoted the surge robustness of TI GaN devices:

e2e.ti.com/.../no-avalanche-no-problem-gan-fets-are-surge-robust

Do TI support people have comments or can you direct me to white papers; apologies if I failed to find these ?

All the best
Harry

  • 0
    TI__Mastermind 27440 points

    Hello Harry,

    We have not tested our devices with SiC/GaN FETs, so we cannot say how they would behave with our battery AFEs. Something to note for our devices is that it takes into account the body diode of conventional MOSFETs. From my limited understanding of GaN FETs, these do not have body-diodes. 

    Most BMS use the body-diode to their advantage (e.g. if a battery is overcharged, one FET will be ON, the other OFF. You can still discharge through the body-diode of the OFF FET and once the BMS detects a discharge current the FET would be turned on). If the GaN FET has no body diode, then this would not work.

    Something I read quickly about the SiC FETs is that they have a high body-diode voltage drop, which sounds like it could be inefficient. 

    I would recommend that you post the question to the FET team. They may know the specifics better for the FET considerations as I have little understanding.

    For reference, our device's FET drivers will drive the gate-source voltage of the FETs to ~10-V to 12V. 

    Best Regards,

    Luis Hernandez Salomon

  • 0 in reply to Luis H. S.
    Intellectual 375 points

    Thanks Luis

    I was looking at raising the FET to 200V for 13S to 16S and this example agrees with your analysis:

    https://epc-co.com/epc/Portals/0/epc/documents/datasheets/EPC2304_datasheet.pdf

    There is a body diode but Vf is 1.5V minimum. An external diode could be added rated for modest current before the FET conducts.

    Related to the BQ itself, the main concern is that FET Vgs(max) is 6V and a forward leakage of 0.8mA @ 125degC. A limiting circuit may be possible adjacent to the gate [Even BQ low overdrive lists V (FETON_LO) as 7V (Max)] but I didn't know if the BQ charge pump can sustain 0.8mA gate leakage current (per FET) and any consumption by the limiting circuit.

    Reading between the datasheet lines Page 75 "Note: If the charge pump is enabled with CP1 connected to BAT, the device will consume an additional
    ≈200 μA." Does this imply the charge pump can't deliver 1.6mA for back-to-back FETs ?

    All the best
    Harry

  • 0 in reply to Harry McNally
    TI__Mastermind 27440 points

    Hello Harry,

    You likely would have to add back-to-back Zeners at the gate-source to protect the FET. 6-V max and -4-V min is a really narrow range. Most power MOSFETs have a +/- 20-V max gate-source and our AFE FET drivers will always attempt to drive to ~10-V to ~12-V. So the Zeners would always be conducting to protect the FET. This would increase the consumption and/or overload the charge-pump.

    You are correct, the charge-pump would not be able to drive 1.6-mA of current. The charge pump can only maintain a 10-V drive voltage at a max of ~40-uA load current. It can provide 200-uA maximum when shorted. Theere is some discussion about this topic in Section 7 CHG Driver Current Paths of the Multiple FETs with the BQ769x2 Battery Monitors application note

    Best Regards,

    Luis Hernandez Salomon

  • 0 in reply to Luis H. S.
    Intellectual 375 points

    Thanks again Luis

    It is fair and I thought it was too much to expect that current out of the charge pump. Who would have anticipated a FET with such a leaky gate ?

    It might be possible to have an external charge pump which is real estate and power. But if the GaN can be made to work it has an impressive RdsON and, like SiC Rds only rises around 50% at the higher junction temperatures on a BMS encapsulated in a pack or cabinet. (And the robustness against surge from that TI link above.)

    I can close the thread but with an idea. Could revisions of any 12S to 16S BQ parts consider a charge pump with lower gate voltage and higher current to accommodate GaN ?

    All the best
    Harry

  • 0 in reply to Harry McNally
    TI__Mastermind 27440 points

    Hello Herry,

    No problem!

    I will bring it up with the team so it is considered in the future!

    Best Regards,

    Luis Hernandez Salomon