7 Replies Latest reply on Jan 5, 2015 10:04 AM by D_Hersey

# Thermistor problem...

I have tried to decipher the data sheet provided but no avail... any assistance would be great, thanks!

• ###### Re: Thermistor problem...

Hi Nicholas, You are right that wasn't a very good data sheet. It looked more like a sales flyer. Since it is a thermistor you will have to build it into some sort of a voltage divider and then read the voltage across one of the resistors in the divider circuit. The voltage that you get for a specific temperature will depend on the resistances that you choose to go into the divider. If I were doing it I would use an empirical approach and once I had the divider set up and I was reading the voltage I would put the thermistor in contact with melting ice and read my voltage. This voltage corresponds to 0 degrees C. Next I would calculate the temperature of boiling water for the atmospheric pressure where I live. Then I would put the thermistor into the boiling water and once again measure the voltage from my divider resistor. This voltage corresponds to the temperature of boiling water, roughly 98 degrees C at my altitude. Next I would do the math to see how much the voltage has changed in the range of 0 degrees to 98 degrees C and from that I could get a good approximation of the temperature for all the voltages in between. you might check around as there might be a better data sheet available but you will still determine the voltage of a given temperature by the design of your voltage divider.

John

• ###### Re: Thermistor problem...

I figured this would be the case... I am using it with a Raspberry Pi and an ADC so I can read/write the voltages for desired temperatures (0C and 100C) and then solve for the line... Was hoping there might be a better solution than this lol.

Thanks!

• ###### Re: Thermistor problem...

Not really, unless you adopt the RC Timer system mconners  posted a little while ago, that eliminated the ADC, use the thermistor instead of the Photoresistor

Regards

peter

• ###### Re: Thermistor problem...

John's two-point calibration method is going to work if the thermistor is linear in the 273 - 373K range.  Thermistor transfer functions can be up to fourth-order polys, IIRC.  To do it rightly, you need to use (for the low point, stir the slurry rapidly) distilled water (use a plastic ice tray to avoid ions) and correct for air pressure in the case of the (use glass) boiling water.  An LM393 is linear over this range, pretty much, as an alternative.  Also, John, to be canonical, if we wanted to compare our thermistor to a resistor, we could also use a Norton amp.  Then our resistor and thermistor wouldn't be in a ladder, but each into a different input of the amplifier.  Another possibility is a CCS rather than a resistor for a load.

Any bipolar transistor (and most diodes) can be used as a TS if it has been cal'd up.  Your sensing range would be dependant upon what elements it was fabricated from.

• ###### Re: Thermistor problem...

"Rt/R25 = exp{A + B/T + C/T 2 + D/T 3 }

where T = temperature in K"

Yeesh, thermistors.