13.4.1 ERROR ANALYSIS

The error in the measurement of a thermocouple temperature is the sum of the errors in the reference junction temperature, the thermocouple output (deviation from standards published in NBS Monograph 125), the thermocouple voltage measurement, and the polynomial error (difference between NBS standard and CR10 polynomial approximations). The discussion of errors which follows is limited to these errors in calibration and measurement and does not include errors in installation or matching the sensor to the environment being measured.

REFERENCE JUNCTION TEMPERATURE

The junction which is created when a thermocouple is wired to the wiring panel is referred to as the reference junction. The temperature of the reference junction must be known in order to calculate the absolute temperature of the measuring junction. The CR10TCR Thermocouple Reference Temperature (CR10TCR) is used to measure the temperature of the reference junction (terminal strips).

The CR10TCR uses a thermistor to measure the relative temperature of the terminal strips. The accuracy of this measurement is a combination of the thermistor's interchangeability specification, the precision of the bridge resistors, and the polynomial error. In a "worst case" example, all errors add to

SECTION 13. CR10 MEASUREMENTS

yield a ±0.4°C error in the range of -24°C to +48°C and ±0.9°C over the range of -38°C to +53°C. It is emphasized that this is the worst case. Campbell Scientific's experience shows that the overall accuracy is typically better than ±0.2°C. The major error component in the -0°C to +50°C range is the

±0.2°C thermistor interchangeability specification. When a CR10 is outside of this temperature range, the polynomial error becomes much worse (Figure 13.4-1), and may necessitate the use of an external reference junction to improve the accuracy.

If the terminal that the thermocouple is wired into is at a different temperature than the CR10TCR thermistor, this difference in temperature becomes an error in the thermocouple temperature measurement. With the CR10 in one of the enclosure options (Section 14) this error will generally be less than 0.3°C.

THERMOCOUPLE LIMITS OF ERROR

The standard reference which lists thermocouple output voltage as a function of temperature (reference junction at 0°C) is the National Bureau of Standards Monograph 125 (1974). The American National Standards Institute has established limits of error on thermocouple wire which is accepted as an industry standard (ANSI MC 96.1, 1975). Table 13.4-1 gives the ANSI limits of error for standard and special grade thermocouple wire of the types accommodated by the CR10.

TABLE 13.4-1. Limits of Error for Thermocouple Wire (Reference Junction at 0°C)

 

 

Limits of Error
ThermocoupleTemperature(Whichever is greater)

Type

Range °CStandard

Special

T

-200 to 0

± 1.0°C or 1.5%

 

 

0 to 350

± 1.0°C or 0.75%

± 0.5°C or 0.4%

J

0 to 750

± 2.2°C or 0.75%

± 1.1°C or 0.4%

E

-200 to 0

± 1.7°C or 1.0%

 

 

0 to 900

± 1.7°C or 0.5%

± 1.0°C or 0.4%

K

-200 to 0

± 2.2°C or 2.0%

 

 

0 to 1250

± 2.2°C or 0.75%

± 1.1°C or 0.4%

13-13