Basics of temperature measurement—thermistors

Dale Cigoy -- Control Engineering, 5/1/2006

Whether in industrial applications, process industries, or in laboratory settings, accurately measuring temperature is a critical part of success. Temperature measurements are needed in medical applications, materials research in labs, electrical/electronic component studies, biology research, geological studies, electrical product device characterization, and other areas.

Many sensors are available to measure temperature. One common device is a thermistor. It typically offers higher sensitivities than either RTDs or thermocouples. It has a negative temperature coefficient in general, meaning that the resistance decreases with increasing temperatures.

Thermistors are also less linear than RTDs. This nonlinearity requires a correction to linearize it. The Steinhart-Hart equation helps approximate individual thermistor curves for linearity. It describes the resistance change of a thermistor as related to temperature. The equation can be written as:

1/T = A + B x (lnR) + C x (lnR)³ where:

• T is degrees Kelvin;
• R is thermistor resistance;
• A, B, and C are curve fitting constants determined through a calibration process; and
• ln is the natural log function (log to the base e).

Although thermistors generally are non-linear, a great deal of work has gone into the development of linear thermistors. Traditional methods for linearization involve the use of external matching resistors to linearize the characteristic curve. However, the issue of linearization is becoming less important, because contemporary data acquisition systems have built-in linearization correction features making hardware linearization unnecessary.

There are several techniques for measuring temperature with a thermistor, the most common being the two- and four-wire techniques.

The two-wire technique forces a current through the thermistor, measuring the resulting voltage. The benefit is that it's a simple method, using only two wires, making it easy to connect and implement. Its main drawback is that its lead resistance is part of the measurement, which can cause some error.

In the four-wire technique, a current is forced through the thermistor and a voltage is measured. However, the current is sourced on one set of leads, while the voltage is sensed on a different set of leads. The voltage is sensed at a different spot from the source current. This means the test lead resistance is completely out of the measurement path. In other words, the test lead resistance is not a part of the measurement.

Thermistors have several distinct advantages over other temperature sensors. For starters, they are simple to set up and operate, using a two-wire measurement scheme. They are also fast, because they can be made small and can respond to temperature changes quickly.

On the down side, the non-linear properties of thermistors require their linearization. They also have a limited temperature range and are fragile. Because they are semiconductors, they are more likely to have de-calibration issues at high temperatures. Thermistors also require a current source and have self-heating characteristics, which must be recognized.

Not allowing for self heating, and selecting a device with an inadequate temperature range are common mistakes made when using thermistors. There are a few ways to reduce self-heating effects, including using as small a test current as possible and using a pulse current method instead of a continuous current.