[SOLVED] How read RTD temperature without lookup table

[prmurthy]

How to read RTD temperature without lookup table

Hi Everybody,

Can we read RTD temperatures without lookup tables. If yes can somebody explain me how.

Thankyou all

 

[dfullmer]

Re: How to read RTD temperature without lookup table

If I understand your question I think the answer is yes. I have used temperature readings from these devices and simple equation to multiply so many degree C per so many bits read gives temperature result. Sometimes I do avaergaing to smooth it out but it is fairly straight forward.

Hope the answers your question or maybe I am offbase.

dave

 

[FvM]

Polynomials are a suitable way to represent a RTD characteristic with an accuracy of your choice.
A standard polynomial for platinum RTD can be found here: Resistance thermometer - Wikipedia, the free encyclopedia

 

[prmurthy]

Re: How to read RTD temperature without lookup table

Yes you understood correctly what I mean was can we convert RTD values from say ADC and convert to temperatures without using lookup tables.

I have used temperature readings from these devices and simple equation to multiply so many degree C per so many bits read gives temperature result. dave

Will your equation will be same for 10bit, 12bit, 16bit etc. Suppose if I frame the equation for 10bit and later if I want use 16bit without changing the software is it possible??

Thankyou.

 

[dfullmer]

No the software would need to know the degree per bit and that would change with a higher resolution device. The only way around that would be a percentage of full scale times a constant but even then the software would need to know how many bits are in the scale. I suppose you can put a hardware toggle in the circuit to tell the software if its 10 or 16bit and to make the adjustment that way. This would mean writting the code to do the calculation either for 10 or 16 bit based on another input port status.

Well thats one way to do it there must be others.

dave

good luck

 

[prmurthy]

Thankyou Dave. I think you are correct everbody has some solution or ways to to do.

Thankyou FvM. I think you have posted about the same quite some time back in other forum.

In my opinion, the most adequate solution is to fit a polynominal for the reversed function to a Pt100 resistance table. You can choose the temperature range of the fit to get optimal accuracy for your application. The acceptable temperature error determines the required polynominal order.

A spreadsheet program as Excel can be used to calculate the fit.

t = b0+r*(b1+r*(b2+r*b3));

I think the following method also can do the job if I want to convert from a given resistance to temperature for PT100.

Temperature = ((Resistance Measured/RTD Resistance at 0 DegC)-1)/(CoeffecientB +CoeffecientA)

The resistance can be found by

Resistance = ((ADCoutput/ADCresolution) x ADC Vref)/ ADC Gain

I would like more comments on this as I am getting errors at very high temperature though the above equations are supposed to give me 0.01 accuracy.

Thankyou

 

[kapilp]

This is what i did for my project:
1: Simulate a resistance sweep corresponding to the temp range
2: R2V converter with output range matched to that of ADC input
3: Calculate the ADC values for various resistance stages
4: Table in ms-excel --> polynomial (not more thank 2nd order as high accuracy was not required)
5: Execute poynomial in uC code.