I'm looking at performing accurate temperature measurements with a three wire RTD, and I'd like to be able to measure the resistance and thus the temperature using an arduino. The accuracy I'm looking for is better than 0.1Ω @ 100Ω.

I was told that to get good accuracy, I would first need a stable accurate low current source and very accurate low temperature coefficient resistors in a wheatstone bridge.

I was wondering if there was to better utilize the arduino's capabilities to perform this measurement, or is the external current source + wheatstone bridge the way to go?

  • \$\begingroup\$ With the many analog inputs you can easily make it somewhat autoranging. You only have a 10 bit DAC (1024 steps) so you need to figure out something smart to reach .1% resolution. \$\endgroup\$ – jippie Sep 6 '13 at 20:57
  • \$\begingroup\$ @jippie -The range is quite small, between 100 and 150 ohms. \$\endgroup\$ – nbubis Sep 6 '13 at 20:58
  • 1
    \$\begingroup\$ @nubis: Have a look at LMP90080 series from TI, for an integrated solution for 3-Wire or 4-Wire RTD. \$\endgroup\$ – AKR Sep 7 '13 at 3:57
  • 1
    \$\begingroup\$ @AKR +1. AD7793 (and family) are in the same class as LMP90080. \$\endgroup\$ – Nick Alexeev Sep 7 '13 at 17:51

enter image description here

Source: application note AN687 by microchip. It shows one of the methods of signal conditioning for 3-wire RTD. Notice that this method doesn't require a Wheatstone bridge.

  • \$\begingroup\$ Nice :) does the final accuracy strongly depend on accurate resistor values? also, I guess one would have to get a stable voltage from the arduino \$\endgroup\$ – nbubis Sep 6 '13 at 22:17

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.