I want to use an AD7124-8 or AD7124-4 to measure an RTD and a Thermistor. In the reference design for the 3 wire RTD, CN-0383, the inaccuracy for the current sources is removed by generating the reference voltage by applying the current source to a precision resistor. I would like to also be able to read other temperature devices, like a 22K thermistor. To that end I would like to use the internal reference and the two current sources as described in CN-0383, but instead of having the reference resistor in line with the RTD, have a set of reference resistors all hooked up to their own A/D channels such that I can measure the current sources with a full scale reading. I.e. a precision 2.2k resistor to calibrate the 1ma source. 5.11k for the 500uA, etc. each hooked up to their own A/D mux input and current source output.
A couple of issues come to mind.
1) The short term stability of the current sources. Will they change values in a short period of time? I.e. the time it takes to switch them from one mux output to another? From the temperature device to the reference resistor?
2) The effect of the sink impedance on the current sources. Will the current sources change value if I switch from a 10K thermistor in a 100 degree bath, approximately 800 ohms, and then switch it to my reference 10,000 ohm +- 0.1% low drift resistor?
3) Any effect that the internal MUX has on the accuracy of the current sources. Is it safe to assume that this is zero? Or is there a practical way to calibrate this out?
4) Must I assume that all of the six current sources are independent? I.e. If I calibrate the 1mA source will its cal scale to the 50uA source?