Understanding, Measuring, And Using Thermistor Thermal Time ConstantsFebruary 24, 2017 By: Mehdi Samii, Ametherm
Thermistors are very accurate thermal transducers, indicating temperature through their resistance. Using them as a sensor, one measures the temperature by simply applying a voltage, measuring the current, and translating the resistance into a temperature. However, thermistors can also act as variable resistors in a circuit, affecting behavior through increases or decreases in resistance, depending on whether the temperature coefficient is positive or negative.
The response to temperature changes take time, and the principal parameter measuring this response is the thermal time constant (TTC). The materials and assembly of a thermistor have a critical impact on the TTC, so a team of engineers at Ametherm ran a number of experiments to show just how much the TTC can vary. Used in an application, we will then see just what kind of impact the TTC can have.
Construction Impacts the TTC
A thermistor consists of a resistive element that absorbs heat as it changes temperature. The response time is due to both the specific heat of the mass and the thermal conductivity of the mass and anything surrounding the mass. Commonly made of sintered ceramics, the masses or the element can also be made of silicon.
The TTC is an intrinsic device property that is independent of the rate of ambient change. When measuring the TTC, you need to apply a temperature change, but if that change is too slow then you're measuring the rate of change of ambient temperature, not the transducer's response. So it's important to use a temperature change that's as close to instantaneous as possible.
The response rate changes throughout the response, slowing down quasi-asymptotically as the device approaches steady state at a new temperature. Waiting until true steady state has been achieved would make for a difficult measurement to standardize, so the TTC is instead defined as the time it takes for the temperature to reach 1/e, or just over 63% of the full transition (see figure 1).
Fig. 1: The TTC measures response at 63.2% of the transition. The blue curve shows a cold-to-hot transition, and the green curve shows a hot-to-cold transition.
There are several variables that affect the TTC:
- The mass of the thermistor
- The shape of the thermistor (surface area vs. volume)
- The potting material used for encapsulation
- The external housing that encloses the thermistor
- The nature of the "ambient" – the gas or liquid in which the thermistor operates
- The methodology used to measure the TTC
If we were comparing different thermistor materials, then the specific heat of the material as well as the temperature coefficient – positive or negative – would also have an impact. These were not considered since all measured devices were of sintered transition metal oxide (NTC material). Sintering affects the resistivity and slope of the resistance temperature curves as well as the stability by closing the pores between different oxide particles.
Because the measurement method matters, it's extremely important, when comparing TTCs of different thermistors, to be sure that they're using the same measurement technique. TTC is expressed as an absolute time. So, for instance, if one device was measured with a 0°C to 100°C temperature change, while the other experiences only half that change, then the first device – even if identical to the second – will have a shorter measured TTC, since the TTC is driven by the temperature difference.
The two major variables Ametherm investigated were chip size, which affects both mass and shape, and encapsulation type. The first varies with the transducer itself; the second varies with the materials around the transducer.
Most Read Articles