BASIC RTD THEORY

A Resistive Temperature Detector (RTD) is a metal resistor that increases in resistance with increasing temperature.

RTD's are commonly made from platinum, copper, and nickel. By far the most common industrial RTD is made of platinum in two forms; as wire wound on a form, or as a platinum film vacuum deposited or sputtered on an insulating surface.

Thin film RTD's cost less than wire wound units, but have a lower temperature limit.

The most common platinum RTD has a resistance of 100 ohms at 0 Deg C and increases .385 ohms/Deg C. The temperature coefficient of the RTD is commonly referred to as it alpha. The alpha would be stated as .00385 ohms/ohm/Deg C. Temperature limits for the wire wound unit are -200 to 800 Deg C. The thin film version temperature limits are -200 to 650 Deg C, but varies somewhat with manufacturer.

The American standard has an alpha of .00392 and the most common unit has a resistance of 100 ohms at 32 Deg F. Temperature limits for the wire wound unit and the thin film versions are the same as the .00385 alpha RTD.

Copper RTD's are usually built at 10 ohms because of the low resistance value of copper. Because of the low resistance value, measured temperature spans must be large to get an adequate signal to process. Copper RTD's have very specific and limited applications because of these two factors.

LEAD WIRE COMPENSATION

Copper wire has almost the same alpha (.004) as platinum and becomes a factor in signal conditioner cost when long lead wires are required from the RTD to the signal conditioner.

2 WIRE INPUT

The lowest cost method is to keep the signal conditioner close enough to the sensor so the copper wire resistance change does not introduce an excessive error in the measurement.

The signal conditioner will be in its simplest form.

3 WIRE INPUT

Lead compensation with 3 wires requires 2 conductors attached to one end of the RTD and a single conductor on the other end.

This method requires the 3 lead wires to track each other in resistance change versus ambient temperature. It is effective for a total copper lead resistance equal to about 10% of the RTD value. The error introduced increases proportionally as the lead wire length increases.

The signal conditioner requires additional circuitry.

4 WIRE INPUT

A 4 wire input provides the most accurate measurement. Excitation current flows through 2 wires and the other 2 wires are used to measure the voltage across the RTD. The wire leads have no influence on the accuracy of the measurement.

This method requires 4 leads from the RTD and a more complicated signal conditioner.

LINEARITY OF RESPONSE VS TEMPERATURE

All RTD's have a nonlinear resistance change versus temperature, but the non-linearities are small and in the case of platinum RTD's the response can be made extremely linear at no cost.

PLATINUM RTD PHYSICAL CHARACTERISTICS

Thin film RTD's can be made very small and can be mounted in any fashion desired to accomplish the desired measurement. General industrial practice is to protect the RTD element in a metal tube and bring out lead wires to connect to the signal conditioner. A single element can be housed in a tube only 1/8 in diameter. Two elements can be placed in a 3/16 diameter tube.

Wire wound RTD's are larger and much more expensive than thin film units. They are generally used where a higher temperature limit is required and the absolute temperature measurement must be very accurate.

SIGNAL CONDITIONING FOR RTD SENSORS

An RTD is usually excited by a constant current and the voltage across the RTD is measured to determine the change in resistance. The larger the current the larger the signal from the RTD. The larger the signal the less costly the amplifier required to meet ambient temperature stability requirements.

The current limit through the RTD is limited by self-heating of the RTD due to power dissipation. Currents generally range from 1 to 10 mA, and this represents a 10 to 1 signal level increase from the RTD.

A platinum RTD can be linearized over its temperature range by simplifying increasing the excitation current as a function of temperature. This is usually implemented by a single resistor. A 100 ohm platinum RTD can be linearized from –200 to 800 Deg C to about +/- 0.7 degrees by this technique. The linearity improves with lower spans.

COST FACTORS OF RTD INPUT SIGNAL CONDITIONERS

1. Absolute accuracy of temperature measurement.
2. Stability versus ambient temperature.
3. Environment (hazardous, humid, corrosive, etc.)
4. Span of measurement. (Span is the difference between the zero scale temperature and full scale temperature.) The narrower the span the more costly it is to meet stability requirements.
5. Lead wire compensation requirements. None, 3 wire or 4 wire.