I wanted to confirm that the CR1000 FieldCal functions cannot be used to calibrate thermocouples because thermocouples are nonlinear. Is that correct? If so, do you have any recommendations for thermocouple calibration procedures using the CR1000?
Most thermocouples are reasonably linear over a limited temperature range anyway. However, if you use a thermocouple instruction in the datalogger with the correct type entered the typical non-linearity of the thermocouple will be corrected for.
Whilst variability from one thermocouple to another may have some small component of non-standard non-lineariy, you should find that for all practical purposes, use of a multiplier and offset, i.e. a linear calibration, of the output of thermocouple instruction should be OK, so you can use the FieldCal function or a similar manual process.
The reason to do otherwise to this might be is you are taking measurements that span hundreds of degrees C.
In addition, you can use either the offset or zero function of the FieldCal instruction. This method would still use the the polynomial to calculate the temperature, but would peg the curve to a specific point for greater accuracy.