Advanced Search
RSS LinkedIn Twitter

Journal Archive

Platinum Metals Rev., 1984, 28, (3), 107

Monitoring Thermocouple Usage

New Device Indicates when Re-Calibration Required

  • R. A. B.
SHARE THIS PAGE:

The precise control of temperature during manufacturing processes has assumed an increased importance in recent years. Particularly within the semiconductor industry, this has arisen as more complex devices and higher product yields have been sought. For example, during the processing of silicon wafers the stages of diffusion and epitaxial growth are carried out under very strictly controlled conditions. Diffusion furnaces used during these stages require not only accurate control to within narrow limits, but also the maintenance of temperature profiles. In the past, the measurement of profiles was carried out using single junction thermocouples, although the use of multi-junction assemblies is increasing and more recently such assemblies are being used for direct furnace control.

To achieve the precise control of temperatures, noble metal thermocouples are generally employed because of the accuracy to which they can be calibrated and their stability in service. While both these factors are important, stability presents users with the greater imponderable. Calibrations can be carried out to high degrees of accuracy in laboratories approved by the British Calibration Service, such as the Calibration Laboratory of Johnson Matthey Metals Limited, where standards are traceable to those of the National Physical Laboratory. The stability of thermocouples, on the other hand, is relative, as all thermocouple types are subject to drift during use as shown in Figure. 1 and therefore require re-calibration from time to time.

Fig. 1

The drift characteristics of platinum versus 10 per cent rhodium-platinum thermocouples in air at temperatures of 750 and 1000°C. The points represent individual calibrations and the blue lines are a least squares fit

The drift characteristics of platinum versus 10 per cent rhodium-platinum thermocouples in air at temperatures of 750 and 1000°C. The points represent individual calibrations and the blue lines are a least squares fit

In the past obtaining a reliable guide to the need for re-calibration has proved difficult, not only because the rate of drift is dependent upon the environment, although this variable can be much reduced by the use of a suitable sheath, but also because of the problems associated with monitoring temperatures and lengths of time of thermocouple usage. Such problems have led the operators of processes depending upon precise temperature control to adopt one of two practices. Either thermocouple usage is monitored continuously by sampling the temperature at frequent intervals under computer control and processing this information to give an estimate of accumulated drift, or thermocouples are re-calibrated or replaced at pre-set intervals of time. The first approach is expensive and requires an extensive knowledge of drift characteristics under conditions of time and temperature, while the second is somewhat arbitrary.

To help overcome the problems of deciding when to re-calibrate thermocouples, Johnson Matthey Metals have developed a small electronic device named DeltaLog which can form an integral part of their range of quartz sheathed multi-junction thermocouples. Housed within the end connector, the device monitors the time and temperature of thermocouple usage and indicates when re-calibration is necessary.

The indicator consists of a small dot within a glass tube which moves along an adjacent scale at a rate determined by the temperature of usage. When the dot has reached full scale deflection, thermocouple re-calibration is recommended.

A DeltaLog device fitted to a modified Lemo connector of a 3-junction profiling thermocouple is shown in Figure. 2. It is completely self contained, being driven by a battery inserted into the device immediately prior to despatch, giving an active life of approximately one year. Thermocouples returned to the Calibration Labratory of Johnson Matthey Metals Limited are checked and re-calibrated while the DeltaLog indicator is re-set to zero and a new battery inserted in preparation for further use.

Fig. 2

The DeltaLog device, shown fitted to a modified Lemo connector of a 3-junction profiling thermocouple, is battery driven and completely self contained

The DeltaLog device, shown fitted to a modified Lemo connector of a 3-junction profiling thermocouple, is battery driven and completely self contained

BACK TO TOP

Acknowledgements

DeltaLog is a Trademark of Johnson Matthey, registration and patents have been applied for.

Read more from this issue »

BACK TO TOP

SHARE THIS PAGE: