Mail Archives: geda-user/2015/03/06/16:18:59
On Mar 6, 2015, at 12:51 PM, John Griessen <john AT ecosensory DOT com> wrote:
> On 03/06/2015 11:24 AM, John Doty wrote:
>> Some delta-sigma ADCs have differential inputs for both reference and measurement. This allows you to make a direct ratiometric
>> comparison between a reference resistor and an RTD. Wire them in series, make the voltage across the reference resistor the
>> reference voltage, and the voltage across the RTD the measurement. This reduces the analog accuracy problem to a very small
>> number of components.
>
> That sounds great!
>
> If you also used high purity wire able to take the heat, and kept
> the ref R close to the RTD on a conductive block right in the oven, there would be no thermocouple effects to speak of.
It doesn’t really help to put the reference resistor at the hot end. You don’t get thermocouple effects if:
1. You’re isothermal at both the sense and measurement end of the conductor pair.
2. The conductors in the pair have identical, uniform composition.
If this balance is violated, you have to arrange for the imbalance to be identical between the reference and sense devices, which I think is generally harder than getting the balance right in the first place. Remember that physically, the potential gradient is a material-dependent function of the temperature gradient. When you do the integrals to relate the potential difference to the temperature difference the length of the conductor drops out, which superficially makes it look like the thermoelectric potential difference arises at the joints. But it’s really a phenomenon of the bulk conductors. As long as every temperature difference around a circuit is balanced by an opposite temperature difference in an identical material, the net thermoelectric potential will be zero.
John Doty Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd AT noqsi DOT com
- Raw text -