Go Down

Topic: Thermistor (Read 189 times) previous topic - next topic

renoks

Hello!

In my project i am using thersmistor, and to get real temperature value i am using Steinhart-Hart equation.


But i concluded that with this equation i can't mesure 0 0C, even to get 1 0C my thermistor resistance has to be over 100 000 Ohms, that also is very high value and thermistor does not offer so high resistance values.

T0 value = 298.15 K.
B value = 3435
R0 value = 10 000 Ohm
R value depends on thermistor, my thermistor at 25 0C is 10 000 ohms.

Does every one has used this temperature sensor on arduino and Steinhart-Hart equation to mesure temperature close to 0 0C and know how to do it?

jremington

#1
Apr 26, 2019, 12:37 am Last Edit: Apr 26, 2019, 12:38 am by jremington
Quote
But i concluded that with this equation i can't mesure 0 0C
The equation is for Kelvin, the international absolute temperature scale.

0 0C = 273.15 K

renoks

In equation i put Kelvins, after that i convert in 0C. That doesn't change anything.

wvmarle

But i concluded that with this equation i can't mesure 0 0C, even to get 1 0C my thermistor resistance has to be over 100 000 Ohms, that also is very high value and thermistor does not offer so high resistance values.
Then what actual resistance does your thermistor give you when you place it in a bath of ice water?
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

renoks

Then what actual resistance does your thermistor give you when you place it in a bath of ice water?
18 000 ohms

wvmarle

Putting that number in the function above (which is not the Steinhart-Hart equation, by the way) gives a temperature of 10°C.

If I solve the function above for R, and calculate it for T=273K, I find R = 28,736Ω. Not exactly accurate, which is expected from the B-coefficient formula.

That 18k is expected for a thermistor with your beta coefficient (see e.g. the TTF3A103 curve on p.3). For accurate results, use a calibration table - most likely you can find a table of resistance vs temperature for your thermistor.

Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

renoks

#6
Apr 26, 2019, 12:50 pm Last Edit: Apr 26, 2019, 12:52 pm by renoks
Putting that number in the function above (which is not the Steinhart-Hart equation, by the way) gives a temperature of 10°C.

If I solve the function above for R, and calculate it for T=273K, I find R = 28,736Ω. Not exactly accurate, which is expected from the B-coefficient formula.

That 18k is expected for a thermistor with your beta coefficient (see e.g. the TTF3A103 curve on p.3). For accurate results, use a calibration table - most likely you can find a table of resistance vs temperature for your thermistor.


Putting that number in the function above (which is not the Steinhart-Hart equation, by the way) gives a temperature of 10°C.

If I solve the function above for R, and calculate it for T=273K, I find R = 28,736Ω. Not exactly accurate, which is expected from the B-coefficient formula.

That 18k is expected for a thermistor with your beta coefficient (see e.g. the TTF3A103 curve on p.3). For accurate results, use a calibration table - most likely you can find a table of resistance vs temperature for your thermistor.


In my calculations 18K ohms gives 18,5 0C.

I don't know how you find R, but when i put 10K in equation i get exacly 25°C.

And at the datasheet at 0 °C resistance should be close to 30K.

wvmarle

I just entered the formula in a spreadsheet using the parameters you gave before, and that's the value I got.

The curve as given in the datasheet I linked to is for a thermistor with the same B-coefficient as yours, and it shows about 18k at 0°C.

If your data sheet says 30k at 0°C, it must have a different B-coefficient. Or they calculated the value based on the B-coefficient, but didn't actually measure it: if I enter 30000 in the formula above for R, it gives me a temperature of -1°C.

Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

jremington

Take the time to calibrate your thermistor at three different temperatures, using this web calculator, and it will work well.

renoks

Thank you for answers, i got it!

wvmarle

Care to share the core of your problem and your solution? I'm curious.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

Go Up