simulate this circuit – Schematic created using CircuitLab

I'm using the voltage divider equation to solve for thermistor resistance. I have:

V_source = 4.9v
V_therm = 2.95v
R_known = 9kohm

I rearrange the equation so that it looks like:

R_therm = ((R_known * V_therm) / (V_source - V_therm));

And this results in a resistance value of approximately 13.6kohm, however, when I take the thermistor out of the circuit and measure its resistance directly (being sure not to touch it), the resistance measures ~11kohm. This 11kohm value makes more sense because it translates into an accurate temperature reading in Kelvins using the Steinhart-Hart equation. So, what am I doing wrong here?

edit: I know I'm converting the raw 10-bit ADC value to the V_therm value properly, because it's exactly the same value that my multimeter shows.

Basically it seems like I need to let the Arduino run for a bit, heat the thermistor, cool it, then after a while it starts giving proper resistance readings. Is there a reason it doesn't work properly from the get-go and/or anything I can do about it? I'm using a TTC-103 10k thermistor.

  • \$\begingroup\$ Schematic would help \$\endgroup\$ Sep 23 '16 at 15:22
  • 2
    \$\begingroup\$ What is the input impedance of your voltage measuring circuit? \$\endgroup\$
    – Arsenal
    Sep 23 '16 at 15:25
  • 1
    \$\begingroup\$ Arsenal beat me to it. Your maths is correct IF you assume there is no current flow into the device measuring Vtherm \$\endgroup\$
    – Andrew
    Sep 23 '16 at 15:29
  • \$\begingroup\$ I'm not sure what the input impedance is, I'm using an Arduino Uno connected to my laptop that is getting 4.9v. This: electronics.stackexchange.com/questions/67171/… looks like it says its < 10kohm? Also the resistance just suddenly changed to the correct (measured) value after running it for a bit and re-uploading the exact same code to the board a couple times. Any idea why this change could have occurred (so I can prevent it in the future)? \$\endgroup\$
    – Austin
    Sep 23 '16 at 15:40
  • 1
    \$\begingroup\$ Have you verified the thermistor voltage with a voltmeter, and not just your A/D? That would be my step 1. \$\endgroup\$ Sep 23 '16 at 20:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.