Chapter 3
Signal Connections
©
National Instruments Corporation
3-13
For best results, use a well-calibrated MIO board so that you can ignore
offsets.
You can eliminate offset error, however, by grounding one channel on the
AMUX-64T and measuring it. This value is the offset of the MIO board,
and it can then be subtracted in software from all other readings. For the
best results, you should use an average of many readings (about 100 or so).
When you take these measures, typical accuracies are about ±2 °C.
Finally, thermocouple wire error is caused by inhomogeneities in the
thermocouple manufacturing process. These errors vary widely depending
on the thermocouple type and even the gauge of wire used, but a value of
±2 °C is typical. For more information on thermocouple wire errors and
more specific data, see Application Note 043, Measuring Temperature with
Thermocouples.
Thermocouple Measurement Accuracies
Table 3-4 lists the expected thermocouple accuracies in degrees Celsius,
subject to the following conditions:
•
The MIO board must be correctly calibrated.
•
The temperature of the screw terminals equals the temperature of the
board (no gradients on the board).
The uncertainties listed apply at either 0 °C (Type J, K, E, T) or 600 °C
(Type S, R). The linearization errors of the NIST polynomials in Table 3-2,
measurement error of a calibrated MIO board, and the ±1 °C cold-junction
sensor error are included. Thermocouple wire error is neglected because of
dependence on several factors as listed above. Finally, these uncertainties
are for the gains listed with a ±10 V input range.
Table 3-4.
Thermocouple Measurement Accuracies
Gain
Type
J
K
E
T
S
R
100
±2.7 °C
±3.7 °C
±2.9 °C
±3.6 °C
±9.8 °C
±8.5 °C
500
±1.4 °C
±2.1 °C
±1.8 °C
±1.9 °C
±3.6 °C
±2.9 °C