0
\$\begingroup\$

I'd like to design a Battery Managements System board (based on TI bq40z50-R2) for my own small (2-4s) lithium battery pack, but I am still unsure how can I test it safely.

I am worried that an error in my board design or in the configuration of the bq40z50 might cause a short circuit or some other unsafe condition to the batteries.

The tests would also include batteries short circuit, undervoltage, overvoltage, overcurrent.

I was thinking of buying for the first tests a 2s or 3s commercial Lithium battery pack with protection (short circuit, overvoltage, undervoltage per cell,…) already integrated (unspecified IC).

I am unsure whether that protection circuit might interfere with the normal operation (measurement accuracy,…) of the bq40z50, invalidating the test.

I could also use stricter protection thresholds for the bq40z50: for example I'd set 4.23V for overvoltage on the bq40z50, whereas the protection IC in the commercial battery pack is set for 4.35V. This way I'd see if the bq40z50 sets the appropriate flag during an overvoltage test. During such a test, can I also use a multimeter to monitor the batteries voltage or would this also interfere with the protection ICs?

Once these first tests are done, I'd proceed using the designated batteries for my battery pack (branded 18650 Li-Ion or RC LiPo without protection IC already in them).

Could this work or do you suggest an other procedure/hardware?

Is there another way to simulate a Lithium battery without actually using one?

Thanks in advance for you help

Sincerely

\$\endgroup\$
1
\$\begingroup\$

If you wish to test just the BMS, there are battery simulators out there. The generic tool for such a job is a 4 quadrant power supply, it can both source and sink current in both polarities. This will allow you to test all possible output states with one device.

\$\endgroup\$
  • \$\begingroup\$ Thanks Lior Bilia. Wow, a bipolar power supply or a battery simulator would be great. After a quick search I find only industrial devices, out of reach (price and availability) for an hobbyist like me. Is there a solution in the middle between your answer and Boxer's? Such as using a 4 channel power supplies, even if they cannot sink? At least they are more affordable and available. \$\endgroup\$ – haku15 Nov 11 '19 at 20:27
0
\$\begingroup\$

To mimic a li ion cell is going to be difficult due to there discharge characteristics. Every chart will show you what i mean, but speaking of lifepo4, they have a very stable charge/discharge curve. For a simple bms for this kind of battery i would suggest some cut of relay set at 3.0 V, and 4.0 V to stop charging. It depends of course on your total cells what you connected in series. A resistor board for balancing above 4.0 V per cell. With all this added you can safely test your bms board. But keep in mind, that my numbers are pretty standard, are very safe and for a lot of cycles, 4x more than factory standard cycles.

If you keep all the margins 0.1 V less then the margins of the TI bq40z50-R2, you can test whatever you want.

The C.I.D. chip in a protected cell, does nothing more to protect the cell from being overcharged and short circuit, thus prevent a thermal runaway. When it is popped, your cell is useless.

\$\endgroup\$
  • \$\begingroup\$ Thanks Boxer. Yes, for some tests I could use resistors but that would not be enough: I wouldn't be able to simulate different voltages of the batteries, to test the passive balancing of the IC, for example. I also could not test overcurrent conditions from the batteries. Or I am missing something? \$\endgroup\$ – haku15 Nov 11 '19 at 20:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.