LM70 without CS

Hi John,

This is a very good question, and I believe that other people might want to know about this as well. To be honest with you, I have never done this before. I wouldn’t suggest tying the CS to ground because the state machine of the chip depend these two edges to determine each transactions. How would you re-synchronize the master and slave when the device detects the glitch (out of sync) on the SC? For example, let’s say that I send 16 clocks to the device; however, the device detects that there are 17 clocks because it is infinite. The state machine goes into unknown stage where it can’t reset the SPI communication.

I hope this will help. Please let me know if you have any further question.

Aaron

Hi Aaron,

You are correct it does appear that by holding the CS line low the LM70 never updates the temperature (state machine halted), I was able to get data back via the SPI by cycling the SC line up to 32 times looking for a 111000 bit pattern (which indicates the end of the 14bits of temp data and start of receive stage where SI/O goes tri-state), but the value returned was always the same (around 1.75 degrees). Have added control over the CS line and the system works fine now.

Thanks

John