• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 471 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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:

BQ769142: Basic test setup for BQ769142

BQ769142: BQ769142: short circuit test

hamzeh ahangari
Prodigy 111 points
Part Number: BQ769142

Hello,

In my BMS card, I wanted to have a short circuit test on my load. My BMS card is built almost similar to BQ769142 evaluation design. After short-circuit, I saw the spark on load connection, and later I noticed that FETs were damaged (I saw smoke!), although short-circuit protection was activated (probably a bit late). An interesting thing I saw is that after the damage, charg-pump value (+10V above bat voltage) has been seen on load line. I guess may FET has been damaged and it passed its gate pin voltage to drain-source in a way.... Anyway I replaced all FETs. I want to manage another test, by 1)reducing SCD delay from default 30us, to minimum possible 15us to activate protection faster, and 2) assemble the spare (parallel) FET to divide high current between two FETs (I have two charge FETs and two discharge FETs).
The problem is now it seems that the BMS is damaged slightly. Before replacing it I want to be sure. Now, there is I2C connection, and cell voltages, battery/load voltages, temperature sensors all are readable and valid. However, it seems there is a problem on FET driver pins and load can not be connected.

Now, every time that I re-plug the battery, the BMS, CP1 pin (charge pump) is ok, it is around 10V above my battery (45V+10V=55V). However, when I try to have a load connected, I notice that CP1 pin is no longer around 55V. it drops to battery level and remains in that state. Later, while BMS says that discharge FET is ON (read FET status using I2C command), but since charge-pump is not ok, FET is not active in practice and load is not connected. What can be the problem? are charge-pump circuit, or FET enable pins damaged, while rest of BMS is ok?

Thanks

  • 0
    TI__Guru 54715 points

    Hi Hamzeh,

    The BQ769142 could be damaged or there may be an unexpected short on the board.  The CHG and DSG pins can be pulled to VSS without damage. When other components failed the pins may have been exposed to a transient above abs max which may have been damaging. You might remove the resistors between CHG and the FET gate node and the DSG pin to discharge FET gate node.  With the resistor removed you should be able to switch the CHG and DSG to see if they move the full range.  Use an external source to drive the board gate nets about 10V above the source and measure the required current. 

    Remember that SCD is a threshold for the part, the SCD mechanism will operate if current is above the threshold.  You might begin tests with a current above that threshold and see if things switch as expected before continuing in higher steps to the full current capability of your cells.

  • 0 in reply to WM5295
    Prodigy 111 points

    thanks, actually what I concern is about reaction time of BMS to short-circuit. For example,

    - if  there is no load, and suddenly a short circuit happens on load side, will DSG turn-off quickly, or CHG transistor turns-on quickly, before damaging the body diode of CHG? if not, is there any solution?

    I had another observation. During one of my discharge tests, once I noted a high-temperature on FETs, even with low current (~1AM). When I slowly increased the current to around 2A, I saw an smoke and later understood that CHG FET is burnt (damaged body diode). This happened only once, but I am thinking what can be the reason. The only answer I found is CHG did not turn ON and then body diode is damaged. I replaced the FET and didn't see this again at least in a few run. if this is true, what can be the reason for CHG to not turn on? these are my guesses:
    - I may have configured the FETs in test (none-autonomous mode), and in that mode you can turn a FET off, and this can be dangerous.
    - I used two FETs in parallel both for DSG and CHG. maybe that is a not a good configuration and cause one FET to not turn ON? I don't know. But I've seen they working in many tests without any heat.

    Best regards,

  • 0 in reply to hamzeh ahangari
    TI__Guru 54715 points

    Hi Hamzeh,

    The body diode protection is expected to be slow, in the ms or 10's of ms range, so it is not likely to react to turn on the charge FET for short circuit even at the maximum delay setting.

    The FETs must be enabled (FET_EN set) or the part can't turn off FETs.  If using the FET test commands the software must respond to control FETs, SCD is not likely to work well.

    Check for body diode protection operation at low current as you indicate.  I don't know why your FET would smoke with low current, most power FETs should take a few watts when mounted, check your FET's specifications.

  • 0 in reply to WM5295
    Prodigy 111 points

    Thanks, after replying by Math and having more test and study, I explained the issue with more details in below thread:. As far as I understand, there is no good solution for short circuit....

    e2e.ti.com/.../bq769142-short-circuit-and-damaged-bms