Reading faulty chip temperature

I have now tested to add a 1 millisecond delay between writing to register 0x135 and 0x134. It seems to solve the problem.

Can you explain why this helps? This would make me more confident in my solution. And 1 millisecond a good choice? Why not 1 microsecond or 10 milliseconds?

Regards

Hi smurfen ,

Thank you for your interest in AD9162. We'll look into your query and get back to you.

Regards,
Zaeefa

Hi.
Do you have an update on my question? Or do you have an estimate on when I can expect an answer?

Regards

Hi smurfen ,

Apologies for the late response.

You are correct that there should be a delay between enabling the temperature sensor through register 0x135 and updating the temperature reading through register 0x134 to get a correct measurement. This is due to the limitation in the processing speed of the ADC used for the sensor. As for exactly how much this delay should be, we are still trying to determine the minimum number of ADC clock cycles required by the AD9162 between writing to the two registers to obtain a proper reading of the die temperature and will let you know once we find out.

Hope this information helps, but let us know if you have additional questions.

Regards,
Zaeefa

Hi.

Any update on the timing?

Hi smurfen ,

Unfortunately, timing data on the temperature sensor ADC of the AD9162 is currently unavailable and may take some time to be measured. May I ask if you are you setting register 0x135 to 0xA1 as the datasheet suggests when enabling the sensor? Doing this also toggles the register's bit 7 which speeds up the temperature sensor settling time and, in turn, may affect the delay required between writing to registers 0x135 and 0x134. 

Regards,
Zaeefa

Hi.

Yes, we are writing 0xA1 to register 0x135.

Regards