Always getting NCK.

RE: Always getting NCK.

Kim — Tue, 09 May 2017 01:16:28 GMT

I solved the problem. It was just cold solder joint.. I didn't consider about soldering because I did SMT. I'm sorry and thank you very much.

RE: Always getting NCK.

Susheel Nuguru — Fri, 28 Apr 2017 09:51:52 GMT

Your oscilloscope output looks correct. I can see clean start bit and the device address 0x6a (110 10101) and the write bit but a Nack from slave. If the slave is able to give NACK on the 9th byte then the proper stop bit. You said that the sensor is Acking with other masters. Only thing we can do is to put the oscilloscope output side by side and where the timing issue is.

RE: Always getting NCK.

Kim — Thu, 27 Apr 2017 08:43:08 GMT

I did it already. But if I try as like you told me, slv_addr will be sent 0xA8 and return NCK. (Slave address << 1) I think 0xD4>>1 or 0x6A is correct. Anyway, thanks for your help.

RE: Always getting NCK.

Daniel Sullivan — Thu, 27 Apr 2017 07:56:56 GMT

I had a similar issue. Try just uint8_t slv_addr = 0xD4; (I seem to recall the twi_tx automatically does the offset)

RE: Always getting NCK.

Kim — Thu, 27 Apr 2017 00:30:42 GMT

I just attached oscilloscope capture files for SDA and SCL.

RE: Always getting NCK.

Susheel Nuguru — Wed, 26 Apr 2017 08:32:43 GMT

can you show some logic analyzer pin output for SDA and SCL so that we can see why the slaves are NCKING

RE: Always getting NCK.

Kim — Wed, 26 Apr 2017 08:20:16 GMT

Thanks for your answer. I got NCK for address byte. I think so LSM6DS3 chip is not understanding. So I tried to the other sensors. (Of course, I changed slave address for each sensors.) But they are also return NCK.

RE: Always getting NCK.

Susheel Nuguru — Wed, 26 Apr 2017 07:20:43 GMT

is it an address byte or data byte that you get the first nack? If it is an address byte then it is obviously the address the TWI slave is not understanding.