0
\$\begingroup\$

I am thinking of replacing standard ratiometric RTD measurement (of temperature) with two sequential measurements.

The reason is I have to use an ADC already on the PCB and cannot replace its reference voltage with voltage across a reference resistor, as standard ratiometric RTD measurement dictates.

First, the ADC would sample across the RTD. Second, the ADC would sample across the reference resistor, then the ratio would be calculated.

Are there any pitfalls? Perhaps noise (from switched power supply) could be an issue for these sequential readings but repeated measurements could take care of this.

Does anyone have experience with this?

\$\endgroup\$
2
  • 1
    \$\begingroup\$ It's commonplace. \$\endgroup\$
    – Andy aka
    Dec 14, 2021 at 8:35
  • \$\begingroup\$ What are your expectations in terms of noise, dynamic range and sampling frequency? \$\endgroup\$ Dec 14, 2021 at 11:56

1 Answer 1

1
\$\begingroup\$

If you do sequential measurements, you'll lose some common mode noise immunity, which may or may not be a problem depending on your application.

But if you're going to use a fixed reference, consider using a voltage divider configuration and taking a single measurement of the midpoint. If the voltage divider and the ADC can share the same reference voltage (either the ADC has an external reference input, or it has an internal reference pin capable of sourcing enough current), you'll recover immunity to noise in the power supply.

The transfer function won't be linear, so you will need to make sure the ADC resolution is high enough to give you enough precision over the desired range.

It all depends what your expectations are in terms of accuracy, speed, noise and precision, and whether you want a 2-, 3- or 4- wire RTD configuration.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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