• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 1003 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.

Original question:

BQ76940EVM: Schematic TIDA-00792

CSD19505KTT: Issues while using TIDA-00792 schematic with CSD19505KTT mosfets (paralleling MOSFETs)

Max Peters
Prodigy 70 points
Part Number: CSD19505KTT
Other Parts Discussed in Thread: TIDA-00792, CSD19536KTT

Hello everybody,

I wanted to build a battery management system. Because I am not very expierenced in building these systems I used the TIDA-00792 (sheet 1) reference design and changed the CSD19536KTT into CSD19505KTT. The other three sheets of TIDA-00792 I did not respected. After finishing the schematic I designed the pcb of my own.

While testing the discharge direction I have the problem that one of the two DSG mosfet is not working properly while leading 10A over all current. In the low currents tests I did not have any problems leading. After having this issue, I measured the transit resistance. I observed that one of these mosfet had an low-ohmic source-drain resistance. Therefore I replaced the damaged mosfet and started the measurement again, but now with a thermal camera. I noticed that one of the source-connecting tracks had a lot more thermal burden than the other. I took a look at the layout and saw that the track width is noticeable different. Can this difference cause the problem? While reasearch I often read about oscillation problems coused through incorrect routing. But I also placed the 200R ferrite bead infront of every mosfet corresponding to the reference design.
Finally I made an measurement using only one MOSFET each side, and I had no errors (only high track warming).

It would by glad, if you would share your ideas with me!

Thanks,
Max

  • 0
    TI__Mastermind 19775 points
    Hi Max,

    The difference in track width can definitely cause issues similar to what you are seeing! The thinner the trace, the higher the resistance of that trace. I would highly recommend you change the board to have the same track width for each of the source-connecting tracks.

    Let me know if that fixes the thermal issue for you!

    Thanks,
    Vince
  • 0 in reply to Vince
    Prodigy 70 points

    Hello Vince,

    at first I want to thank you for your fast reply. What effect will couse the damage of the mosfet? Is it an measureble oscillation? I don't want to order a new pcb without knowing what the problem most likely is. Maybe I can fix it with an replacement of the ferrtit bead?

    Best regards,

    Max

  • 0 in reply to Vince
    Prodigy 70 points
    It's not only a thermal problem of the track, also the mosfet had permanent damage (low ohmic). Did I missunderstand your answer?
  • 0 in reply to Max Peters
    TI__Mastermind 19775 points
    Hi Max,

    I apologize for the confusion. The mosfet damage is likely from one of two areas: thermal burn (sustained high current) or a large transient current.

    For the thermal burn, I would suggest that you make sure that the MOSFET chosen does not need a heat-sink at the 100A being used (ensure the thermal junction to ambient of the mosfet is sufficient for the watts being used). For transient, I would suggest, if possible, to replicate the setup used previously and measure the current between drain and source. Be careful, as most ammeters cannot measure at 100A, so it may be best to do the test with a high power resistor and voltmeter instead.

    Let me know if you have any additional questions!

    Regards,
    Vince
  • 0 in reply to Vince
    Prodigy 70 points

    Hello Vince,

    thanks for your reply. I'm using a heatsink for each mosfet. Additionally I would like to point out, that I talked about 10A.
    The first point you mentioned I can exclude, because I have no problems using only one mosfet.
    At what frequency would the transients occur? Can they had been caused by the wrong layout? I have to say, that I noticed transients in the past with a repetition rate of about 100 ms in the source current. But I thought I will by out of question, because the one mosfet works pretty well. Although later they didn't occur anymore. Furthermore they didn't occur anymore later and the measurement looked fine.

    Best regards,
    Max

  • 0 in reply to Max Peters
    TI__Mastermind 19775 points
    Hi Max,

    The transients could be at various frequencies depending on the source. Did you notice these transients shortly after hot-plug of the battery stack? Do you happen to know the amplitude of these transients?

    Thanks!
    Vince
  • 0 in reply to Vince
    Prodigy 70 points
    Hello Vince,

    I got these transients after attaching a ohmic load (0R33) on the load side. The voltage source I mounted instead of the battery stack on battery side. The amplitudes reached currents up to 50 A. They were gone, after reaching 10 A (3 V output voltage). So I think it was caused by the low output voltage drop.

    Can we conclude with saying, that a optimized track routing (first answer from you) maybe solve the problem?

    Best regards
    Max
  • 0 in reply to Max Peters
    TI__Expert 6085 points
    Since the thermal IR-images are giving you the hotspots, it will be relatively quick for you get 2oz Cu instead of the 1 oz for the high-current path.
    Adding thermal vias under the FET is a easy way to pull the heat out!