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.

BQ78350 host communication motor noise/interference issue

Other Parts Discussed in Thread: BQ76940, BQ78350, BQ76940EVM

We're using a BQ78350 to manage a BQ76940 AFE in an e-bike application. We lose SMBus connectivity with the BQ78350 when the system is connected to the bike's electric motor under load. Communications errors increase as the motor load increases and eventually the SMBus fails completely. The motor draws a maximum of 12A @ 36V. There are no problems when we operate the BMS using an equivalent [noiseless] resistive load. 

I've pasted an oscilloscope trace of the SMBus showing successful and failed communications. The trace on the left shows a successful SMBus transfer, the trace on the right shows an unsuccessful transfer. In both cases, you can see the BQ78350 ACKs the host's 8-bit data (highlighted by red arrows). In both cases the clock is temporarily suspended following the ACK. On the left, the data bus is released and floats high. On the right, the data bus begins to float high then is almost immediately driven low (you see a brief spike after the ACK when the clock goes low). We're not sure whether it's the host or slave BQ78350 driving the data signal low, but suspect it's the BQ78350. Is this potentially the BQ78350 reacting to multiple transmission errors trying to reset the SMBus? Note that this condition is reproducible using the BQ76940EVM with EV2300 as well as our design, which closely follows the BQ76940EVM reference design layout and components.

Any advice and troubleshooting ideas are welcome.

Thanks,
Mark

  • Yes, I think its important to discover which pin (or pins) of the device is being subjected to excessive electrical stress. The unwanted current from the event is likely causing the gauge to reset repeatedly. An electrical overstress event on almost any pin could cause an internal ESD structure to conduct and the resulting current through the power pins may cause a reset.

    I would suggest that you first scope each pin, with the shortest possible scope ground lead. Its best if the ground lead can be no longer than 1 cm. For any pin that looks suspicious, try to decouple the noise with small capacitors and see if things improve. One method is to back off on the noisy load until communication is just marginal, then any improvements from the capacitors will be more obvious.