Other Parts Discussed in Thread: BQSTUDIO, EV2400, BQ78350
Hi , I'm having issues properly communicating with the BQ78350-R1.
I have a fault case when trying to read data from the gauge.I'm not using PEC.
The problem is the following : after sending the address of the chip following the repeated start, the gauge does a NACK condition 1/3 of the time. The other 2/3 of the time, the transaction happens correctly. What I've noticed from the NACK case is that the slave stretches the clock after the 7th bit. On the 8th bit my READ bit is sent, and after that I receive NACK. I will put a screeshot of my logic analyzer to show the faulty situation. I've used an oscilloscope too just to make sure my signal integrity is good and everything seems fine. I will join the oscilloscope screenshot of the faulty situation too.
The things that appears weird to me is that the slave stretches the clock before the 8th bit. When looking at other situation when it stretches the clock, it always does it after the 8th clock transition right before the ACK/NACK bit.
I've purchased a battery that uses this gauge. When I look at the battery's datasheet , it says the following regarding a read transaction :
1. Battery address(0x16) + W =>2. Command code => 3.Battery address(0x16) + R => 4. Slave outputs the data.
However I seem to find contradictory information when I look at the following document found on TI's page of the product : https://www.ti.com/lit/an/slua475/slua475.pdf?ts=1612753931768&ref_url=https%253A%252F%252Fwww.google.com%252F . Looking at figure 2 of page 2 what they actually seem to do is the following :
2. Battery address(0x16) + W. => 2. Command code => 3.Battery address(0x16+1) + W => 4. Slave outputs the data
Does anyone has ever had a similar issue ? Could someone shed some light on this issue? I've also added screenshots of the datasheet & the figure 2 of the web page.
Thanks a lot guys!
