Industrial Motor Control: Temperature Sensing Devices

Home | Articles | Forum | Glossary | Books


  • Describe different methods for sensing temperature.
  • Discuss different devices intended to be operated by a change of temperature.
  • List several applications for temperature sensing devices.
  • Read and draw the NEMA symbols for temperature switches.

There are many times when the ability to sense temperature is of great importance. The industrial electrician will encounter some devices designed to change a set of contacts with a change of temperature and other devices used to sense the amount of temperature. The method used depends a great deal on the applications of the circuit and the amount of temperature that must be sensed.

Expansion of Metal

Fgr. 1 Metal expands when heated.

A very common and reliable method for sensing temperature is by the expansion of metal. It has long been known that metal expands when heated. The amount of expansion is proportional to two factors:

1. The type of metal used.

2. The amount of temperature.

Consider the metal bar shown in Fgr. 1. When the bar is heated, its length expands. When the metal is permitted to cool, it will contract. Although the amount of movement due to contractions and expansion is small, a simple mechanical principle can be used to increase the amount of movement (Fgr. 2).

The metal bar is mechanically held at one end. This permits the amount of expansion to be in one direction only. When the metal is heated and the bar expands, it pushes against the mechanical arm. A small movement of the bar causes a great amount of movement in the mechanical arm. This increased movement in the arm can be used to indicate the temperature of the bar by attaching a pointer and scale, or to operate a switch as shown. It should be understood that illustrations are used to convey a principle. In actual practice, the switch shown in Fgr. 2 would be spring loaded to provide a "snap" action for the contacts. Electrical contacts must never be permitted to open or close slowly. This produces poor contact pres sure and will cause the contacts to burn or will cause erratic operation of the equipment they are intended to control.

Hot-wire Starting Relay

Fgr. 2 Expanding metal operates a set of contacts.

A very common device that uses the principle of expanding metal to operate a set of contacts is the hot wire starting relay found in the refrigeration industry. The hot-wire relay is so named because it uses a length of resistive wire connected in series with the motor to sense motor current. A diagram of this type of relay is shown in Fgr. 3.

Fgr. 3 Hot-wire relay connection.

When the thermostat contact closes, current can flow from line L1 to terminal L of the relay. Current then flows through the resistive wire, the movable arm, and the normally closed contacts to the run and start windings. When current flows through the resistive wire, its temperature increases. This increase of temperature causes the wire to expand in length. When the length increases, the movable arm is forced down ward. This downward pressure produces tension on the springs of both contacts. The relay is so designed that the start contact will snap open first, disconnecting the motor start winding from the circuit. If the motor current isn't excessive, the wire will never become hot enough to cause the overload contact to open. If the motor current should become too great, however, the temperature of the resistive wire will become high enough to cause the wire to expand to the point that it will cause the overload contact to snap open and disconnect the motor run winding from the circuit.

The Mercury Thermometer

Another very useful device that works on the principle of contraction and expansion of metal is the mercury thermometer. Mercury is a metal that remains in a liquid state at room temperature. If the mercury is confined in a glass tube as shown in Fgr. 4, it will rise up the tube as it expands due to an increase in temperature. If the tube is calibrated correctly, it provides an accurate measurement for temperature.

The Bimetal Strip

Fgr. 4 A mercury thermometer operates by the expansion of metal.

Fgr. 5 A bimetal strip.

Fgr. 6 A bimetal strip warps with a change of temperature.

The bimetal strip is another device that operates by the expansion of metal. It is probably the most common heat sensing device used in the production of room thermostats and thermometers. The bimetal strip is made by bonding two dissimilar types of metal together (Fgr. 5). Since these two metals are not alike, they have different expansion rates. This causes the strip to bend or warp when heated (Fgr. 6). A bimetal strip is often formed into a spiral shape, as shown in Fgr. 7. The spiral permits a longer bimetal strip to be used in a small space. A long bimetal strip is desirable because it exhibits a greater amount of movement with a change of temperature.

If one end of the strip is mechanically held and a pointer is attached to the center of the spiral, a change in temperature will cause the pointer to rotate. If a calibrated scale is placed behind the pointer, it becomes a thermometer. If the center of the spiral is held in position and a contact is attached to the end of the bimetal strip, it becomes a thermostat. A small permanent magnet is used to provide a snap action for the contacts (Fgr. 8). When the moving contact reaches a point that's close to the stationary contact, the magnet attracts the metal strip and causes a sudden closing of the contacts. When the bimetal strip cools, it pulls away from the magnet. When the force of the bimetal strip becomes strong enough, it overcomes the force of the magnet and the contacts snap open.

Fgr. 7 A bimetal strip used as a thermometer.

Fgr. 8 A bimetal strip used to operate a set of contacts.

Fgr. 9 Thermocouple.


In 1822, a German scientist named Seebeck discovered that when two dissimilar metals are joined at one end, and that junction is heated, a voltage is produced (Fgr. 9). This is known as the Seebeck effect. The device produced by the joining of two dissimilar metals for the purpose of producing electricity with heat is called a thermocouple. The amount of voltage produced by a thermocouple is determined by:

1. The type of materials used to produce the thermo couple.

2. The temperature difference of the two junctions.

The chart in Fgr. 10 shows common types of thermocouples. The different metals used in the construction of thermocouples is shown as well as their normal temperature ranges.

Fgr. 10 Thermocouple chart.

The amount of voltage produced by a thermocouple is small, generally in the order of millivolts (1 millivolt = 0.001 volt). The polarity of the voltage of some thermocouples is determined by the temperature. E.g., a type "J" thermocouple produces zero volts at about 32F. At temperatures above 32F, the iron wire is positive and the constantan wire is negative. At temperatures below 32F, the iron wire becomes negative and the constantan wire becomes positive. At a temperature of +300F, a type "J" thermocouple will produce a voltage of about +7.9 milli volts. At a temperature of -300F, it will produce a voltage of about -7.9 millivolts.

Since thermocouples produce such low voltages, they are often connected in series, as shown in Fgr. 11. This connection is referred to as a thermopile. Thermocouples and thermopiles are generally used for making temperature measurements and are some times used to detect the presence of a pilot light in appliances that operate with natural gas. The thermocouple is heated by the pilot light. The current produced by the thermocouple is used to produce a magnetic field that holds a gas valve open and permits gas to flow to the main burner. If the pilot light should go out, the thermocouple ceases to produce current and the valve closes (Fgr. 12).

Resistance Temperature Detectors

The resistance temperature detector (RTD) is made of platinum wire. The resistance of platinum changes greatly with temperature. When platinum is heated, its resistance increases at a very predictable rate; this makes the RTD an ideal device for measuring temperature very accurately. RTDs are used to measure temperatures that range from -328 to +1166 degrees Fahrenheit (-200 to +630 C). RTDs are made in different styles to perform different functions. Fgr. 13 illustrates a typical RTD used as a probe. A very small coil of platinum wire is encased inside a copper tip.

Copper is used to provide good thermal contact. This permits the probe to be very fast-acting. The chart in Fgr. 14 shows resistance versus temperature for a typical RTD probe. The temperature is given in degrees Celsius and the resistance is given in ohms. RTDs in two different case styles are shown in Fgr. 15.

Fgr. 11 Thermopile

Fgr. 12 A thermocouple provides power to the safety cut-off valve.


The term thermistor is derived from the words "thermal resistor." Thermistors are actually thermally sensitive semiconductor devices. There are two basic types of thermistors: one type has a negative temperature coefficient (NTC) and the other has a positive temperature coefficient (PTC). A thermistor that has a negative temperature coefficient will decrease its resistance as the temperature increases. A thermistor that has a positive temperature coefficient will increase its resistance as the temperature increases. The NTC thermistor is the most widely used.

Thermistors are highly nonlinear devices. For this reason they are difficult to use for measuring temperature. Devices that measure temperature with a thermistor must be calibrated for the particular type of thermistor being used. If the thermistor is ever replaced, it has to be an exact replacement or the circuit will no longer operate correctly. Because of their nonlinear characteristics, thermistors are often used as set point detectors as opposed to actual temperature measurement. A set point detector is a device that activates some process or circuit when the temperature reaches a certain level. E.g., assume a thermistor has been placed inside the stator winding of a motor. If the motor should become overheated, the windings could become severely damaged or destroyed. The thermistor can be used to detect the temperature of the windings. When the temperature reaches a certain point, the resistance value of the thermistor changes enough to cause the starter coil to drop out and disconnect the motor from the line.

Thermistors can be operated in temperatures that range from about +100 F to +300 F.

Fgr. 13 Resistance temperature detector.


Degrees C, Resistance (Ohms)

0, 100 50, 119.39, 100 138.5, 150 157.32, 200 175.84, 250 194.08, 300 212.03, 350 229.69, 400 247.06, 450 264.16, 500 280.93, 550 297.44, 600 313.65


Fgr. 14 Temperature and resistance for a typical RTD.

Fgr. 15 RTDs in different case styles.

Fgr. 16 Solid-state starting relay.

One common use for thermistors is in the solid state starting relays used with small refrigeration compressors (Fgr. 16). Starting relays are used with hermetically sealed motors to disconnect the start windings from the circuit when the motor reaches about 75% of its full speed. Thermistors can be used for this application because they exhibit an extremely rapid change of resistance with a change of temperature. A schematic diagram showing the connection for a solid state relay is shown in Fgr. 17.

When power is first applied to the circuit, the thermistor is cool and has a relatively low resistance.

This permits current to flow through both the start and run windings of the motor. The temperature of the thermistor increases because of the current flowing through it. The increase of temperature causes the resistance to change from a very low value of 3 or 4 ohms to several thousand ohms. This increase of resistance is very sudden and has the effect of opening a set of contacts connected in series with the start winding. Although the start winding is never completely disconnected from the power line, the amount of current flow through it's very small, typically 0.03 to 0.05 amperes, and does not affect the operation of the motor. This small amount of leakage current maintains the temperature of the thermistor and prevents it from returning to a low resistance. After power has been disconnected from the motor, a cool-down period of about 2 minutes should be allowed before restarting the motor. This cool-down period is needed for the thermistor to return to a low value of resistance.

The PN Junction

Another device that has the ability to measure temperature is the PN junction, or diode. The diode is becoming a very popular device for measuring temperature because it's accurate and linear.

When a silicon diode is used as a temperature sensor, a constant current is passed through the diode.

Fgr. 18 illustrates this type of circuit. In this circuit, resistor R1 limits the current flow through the transistor and sensor diode. The value of R1 also deter mines the amount of current that flows through the diode. Diode D1 is a 5.1 volt zener used to produce a constant voltage drop between the base and emitter of the PNP transistor. Resistor R2 limits the amount of current flow through the zener diode and the base of the transistor. D1 is a common silicon diode. It is being used as the temperature sensor for the circuit. If a digital voltmeter is connected across the diode, a voltage drop between 0.8 and 0 volts can be seen. The amount of voltage drop is determined by the temperature of the diode.

Fgr. 17 Connection of solid-state starting relay.

Fgr. 18 Constant current generator.

Another circuit that can be used as a constant current generator is shown in Fgr. 19. In this circuit, a field effect transistor (FET) is used to produce a current generator. Resistor R1 determines the amount of current that will flow through the diode. Diode D1 is the temperature sensor.

If the diode is subjected to a lower temperature, say by touching it with a piece of ice, the voltage drop across the diode will increase. If the diode temperature is increased, the voltage drop will decrease because the diode has a negative temperature coefficient. As its temperature increases, its voltage drop becomes less.

In Fgr. 20, two diodes connected in a series are used to construct an electronic thermostat. Two diodes are used to increase the amount of voltage drop as the temperature changes. A field effect transistor and resistor are used to provide a constant current to the two diodes used as the heat sensor. An operational amplifier is used to turn a solid-state relay on or off as the temperature changes. In the example shown, the circuit will operate as a heating thermostat. The output of the amplifier will turn on when the temperature decreases sufficiently. The circuit can be converted to a cooling thermostat by reversing the connections of the inverting and non-inverting inputs of the amplifier.

Fgr. 19 Field effect transistor used to produce a constant current generator.

Fgr. 20 Solid-state thermostat using diodes as heat sensors.

Expansion Due to Pressure

Another common method of sensing a change of temperature is by the increase of pressure of some chemicals. Refrigerants confined in a sealed container, for example, will increase the pressure in the container with an increase of temperature. If a simple bellows is connected to a line containing refrigerant (Fgr. 21) the bellows will expand as the pressure inside the sealed system increases. When the surrounding air temperature decreases, the pressure inside the system decreases and the bellows contracts. When the air temperature increases, the pressure increases and the bellows expands. If the bellows controls a set of contacts, it becomes a bellows type thermostat. A bellows thermostat and the standard NEMA symbols used to represent a temperature operated switch are shown in Fgr. 22.

Fgr. 21 Bellows contracts and expands with a change of refrigerant pressure.

Fgr. 22 Industrial temperature switch.

Fgr. 23 Cut-away view of a smart temperature transmitter.

Smart Temperature Transmitters

Standard temperature transmitters generally send a 4 to 20mA signal to indicate the temperature. They are calibrated for a specific range of temperature such as 0 to 100 degrees. Standard transmitters are de signed to operate with one type of sensor such as RTD, thermocouple, and so on. Any changes to the setting require a recalibration of the unit.

Smart transmitters contain an internal microprocessor and can be calibrated from the control room by sending a signal to the transmitter. It is also possible to check the transmitter for problems from a remote location. A cutaway view of a smart temperature transmitter is shown in Fgr. 23. The transmitter illustrated in Fgr. 23 uses HART (Highway Addressable Remote Transducer) protocol. This transmitter can accept RTD, differential RTD, thermocouple, ohm, and millivolt inputs. A smart temperature transmitter with meter is shown in Fgr. 24.

Fgr. 24 Smart temperature transmitter with meter.


1. Should a metal bar be heated or cooled to make it expand?

2. What type of metal remains in a liquid state at room temperature?

3. How is a bimetal strip made?

4. Why are bimetal strips often formed into a spiral shape?

5. Why should electrical contacts never be permitted to open or close slowly?

6. What two factors determine the amount of voltage produced by a thermocouple?

7. What is a thermopile?

8. What do the letters RTD stand for?

9. What type of wire is used to make an RTD?

10. What material is a thermistor made of?

11. Why is it difficult to measure temperature with a thermistor?

12. If the temperature of an NTC thermistor increases, will its resistance increase or decrease?

13. How can a silicon diode be made to measure temperature?

14. Assume that a silicon diode is being used as a temperature detector. If its temperature increases, will its voltage drop increase or decrease?

15. What type of chemical is used to cause a pressure change in a bellows type thermostat?

Top of Page

PREV: Solenoid and Motor Operated Valves NEXT: Hall Effect Sensors Guide Index HOME