What is the best way for measuring the temperature for a weatherstation. The temperature would be between about -30°C to 30°C and I would like to have a accuracy of about 0.1°C to 0.5°C. I was thinking about using some Temperature Resistors in parallel but I think I won't get the high accuracy for this. With another temperature sensor like the lm75 there is a too high inaccuracy.

  • \$\begingroup\$ Surely there are 0.5 degC temperature sensors out there. Look around. This also can't be answered because you haven't defined what criteria go into "best" other than 1/2 degC accuracy. \$\endgroup\$ – Olin Lathrop May 1 '18 at 14:41
  • \$\begingroup\$ The ADT7420 might be in the rough ballpark. But measuring temperature in weather stations is an art entirely its own. Stevenson screens and wet-bulb vs dry-bulb are commonly found. So an IC may not be the best choice. \$\endgroup\$ – jonk May 1 '18 at 17:13

Resistance temperature detector (RTD)

Historically, this has been the preferred solution for temperature sensing with the kinds of accuracy you desire.

If you want to build an RTD signal conditioning circuit out of OpAmps, this application note might help. There are also ADCs with integrated front ends for RTD.

Precision thermistors

Over the last two decades, the thermistor materials have improved, and we have thermistors which can achieve ±0.1°C accuracy.
This thermistor, for example.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.