# Resistance in a thermistor

Last week, I did an experiment in class. It consist in determine the parameters $R_0$ and $B$ of the thermistor, from the equation

$$R(T)=R_0e^{B/T},$$

where $R$ is the resistance and $T$ the temperature. To do this, I immerse the thermistor in glicerine, and the glicerine in water that I heated gradually. Then, taking the resistance of the thermistor in intervals of one centigrade degree, I could adjust the points to get the parameters.

After that, I observe that, when the thermistor inmersed in the hot water is cooling (without providing heat), the resistance at some temperature doesn't much with the information I took before. For example, the resistance while the thermistor at $77^{\circ}$C was been heated was 12.404 $\Omega$, but the resistance at the same temperature while it was cooling (naturally) was 12.857 $\Omega$. It happens for all the values of the temperature I took.

I want to know why does it happens, a physical explanation.

• How did you measure the temperature of the thermistor? – sammy gerbil Nov 25 '16 at 22:24
• I would guess that your thermistor and your thermometer are not at the exact same temperature during the heating and cooling cycles, and that is what is giving the appearance of hysteresis between heating and cooling. If you want further feedback, it would be best if you supplied a diagram and/or picture of your experimental setup showing precisely where the various components (i.e., heating unit, thermistor, thermometer) are located with respect to each other. – user93237 Nov 25 '16 at 23:33

1. You let the setup cool down naturally. This means that the temperature distribution of your setup was certainly inhomogeneous and therefore, the temperature that your thermometer displays does not need to reflect the temperature of your resistance (especially if this resistance has a considerable extension, there is a good chance that the resistor itself was not uniformly at one temperature). If you were heating up slowly and waiting for the setup to thermalise before measuring during your measurements before, this would be a clear source for discrepancies. Given that you report a higher value on cooling, this to me seems plausible (I assume that $B>0$.).
2. A deviation of $0.4 \Omega$ is simply within your uncertainty of the whole measurement setup. As you have not given any indication as to how reliable your measurement data is, that cannot be ruled out.