ence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray

Overview

A thermocouple is a commonly used type of sensor that is used to measure temperature. Thermocouples are popular in industrial control applications because of their relatively low cost and wide measurement ranges. In particular, thermocouples excel at measuring high temperatures where other common sensor types cannot function. Try operating an integrated circuit (LM35, AD 590, etc.) at 800C.

Thermocouples are fabricated from two electrical conductors made of two different metal alloys. The conductors are typically built into a cable having a heat-resistant sheath, often with an integral shield conductor. At one end of the cable, the two conductors are electrically shorted together by crimping, welding, etc. This end of the thermocouple--the hot junction--is thermally attached to the object to be measured. The other end--the cold junction, sometimes called reference junction--is connected to a measurement system. The objective, of course, is to determine the temperature near the hot junction.

It should be noted that the "hot" junction, which is somewhat of a misnomer, may in fact be at a temperature lower than that of the reference junction if low temperatures are being measured.

Reference Junction Compensation Thermocouples generate an open-circuit voltage, called the Seebeck voltage, that is proportional to the temperature difference between the hot and reference junctions :

Vs = V(Thot-Tref)

Since thermocouple voltage is a function of the temperature difference between junctions, it is necessary to know both voltage and reference junction temperature in order to determine the temperature at the hot junction. Consequently, a thermocouple measurement system must either measure the reference junction temperature or control it to maintain it at a fixed, known temperature.

There is a misconception of how thermocouples operate. The misconception is that the hot junction is the source of the output voltage. This is wrong. The voltage is generated across the length of the wire. Hence, if the entire wire length is at the same temperature no voltage would be generated. If this were not true we connect a resistive load to a uniformly heated thermocouple inside an oven and use additional heat from the resistor to make a perpetual motion machine of the first kind.

The erroneous model also claims that junction voltages are generated at the cold end between the special thermocouple wire and the copper circuit, hence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray

ield conductor. At one end of the cable, the two conductors are electrically shorted together by crimping, welding, etc. This end of the thermocouple--the hot junction--is thermally attached to the object to be measured. The other end--the cold junction, sometimes called reference junction--is connected to a measurement system. The objective, of course, is to determine the temperature near the hot junction.

It should be noted that the "hot" junction, which is somewhat of a misnomer, may in fact be at a temperature lower than that of the reference junction if low temperatures are being measured.

Reference Junction Compensation Thermocouples generate an open-circuit voltage, called the Seebeck voltage, that is proportional to the temperature difference between the hot and reference junctions :

Vs = V(Thot-Tref)

Since thermocouple voltage is a function of the temperature difference between junctions, it is necessary to know both voltage and reference junction temperature in order to determine the temperature at the hot junction. Consequently, a thermocouple measurement system must either measure the reference junction temperature or control it to maintain it at a fixed, known temperature.

There is a misconception of how thermocouples operate. The misconception is that the hot junction is the source of the output voltage. This is wrong. The voltage is generated across the length of the wire. Hence, if the entire wire length is at the same temperature no voltage would be generated. If this were not true we connect a resistive load to a uniformly heated thermocouple inside an oven and use additional heat from the resistor to make a perpetual motion machine of the first kind.

The erroneous model also claims that junction voltages are generated at the cold end between the special thermocouple wire and the copper circuit, hence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray

unction Compensation Thermocouples generate an open-circuit voltage, called the Seebeck voltage, that is proportional to the temperature difference between the hot and reference junctions :

Vs = V(Thot-Tref)

Since thermocouple voltage is a function of the temperature difference between junctions, it is necessary to know both voltage and reference junction temperature in order to determine the temperature at the hot junction. Consequently, a thermocouple measurement system must either measure the reference junction temperature or control it to maintain it at a fixed, known temperature.

There is a misconception of how thermocouples operate. The misconception is that the hot junction is the source of the output voltage. This is wrong. The voltage is generated across the length of the wire. Hence, if the entire wire length is at the same temperature no voltage would be generated. If this were not true we connect a resistive load to a uniformly heated thermocouple inside an oven and use additional heat from the resistor to make a perpetual motion machine of the first kind.

The erroneous model also claims that junction voltages are generated at the cold end between the special thermocouple wire and the copper circuit, hence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray

conception of how thermocouples operate. The misconception is that the hot junction is the source of the output voltage. This is wrong. The voltage is generated across the length of the wire. Hence, if the entire wire length is at the same temperature no voltage would be generated. If this were not true we connect a resistive load to a uniformly heated thermocouple inside an oven and use additional heat from the resistor to make a perpetual motion machine of the first kind.

The erroneous model also claims that junction voltages are generated at the cold end between the special thermocouple wire and the copper circuit, hence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray

ence, a cold junction temperature measurement is required. This concept is wrong. The cold -end temperature is the reference point for measuring the temperature difference across the length of the thermocouple circuit.

Most industrial thermocouple measurement systems opt to measure, rather than control, the reference junction temperature. This is due to the fact that it is almost always less expensive to simply add a reference junction sensor to an existing measurement system than to add on a full-blown temperature controller.

Sensoray Smart A/D's measure the thermocouple reference junction temperature by means of a dedicated analog input channel. Dedicating a special channel to this function serves two purposes: no application channels are consumed by the reference junction sensor, and the dedicated channel is automatically pre-configured for this function without requiring host processor support. This special channel is designed for direct connection to the reference junction sensor that is standard on many Sensoray termination boards.

Linearization Within the "useable" temperature range of any thermocouple, there is a proportional relationship between thermocouple voltage and temperature. This relationship, however, is by no means a linear relationship. In fact, most thermocouples are extremely non-linear over their operating ranges. In order to obtain temperature data from a thermocouple, it is necessary to convert the non-linear thermocouple voltage to temperature units. This process is called "linearization."

Several methods are commonly used to linearize thermocouples. At the low-cost end of the solution spectrum, one can restrict thermocouple operating range such that the thermocouple is nearly linear to within the measurement resolution. At the opposite end of the spectrum, special thermocouple interface components (integrated circuits or modules) are available to perform both linearization and reference junction compensation in the analog domain. In general, neither of these methods is well-suited for cost-effective, multipoint data acquisition systems.

In addition to linearizing thermocouples in the analog domain, it is possible to perform such linearizations in the digital domain. This is accomplished by means of either piecewise linear approximations (using look-up tables) or arithmetic approximations, or in some cases a hybrid of these two methods.

The Linearization Process Sensoray’s Smart A/D’s employ a thermocouple measurement and linearization process that is designed to hold costs to a practical level without sacrificing performance.

First, both the thermocouple and reference junction sensor signals are digitized to obtain thermocouple voltage Vt and reference junction temperature Tref. The thermocouple signal is digitized at a higher rate than the reference junction because it is assumed that the reference junction is relatively stable compared to the hot junction. Reference junction measurements are transparently interleaved betwe