8.0 CHECKING CALIBRATION
8.1
The BAT-12 thermometer can have a ‘worst case’ error of 0.2°C. The ‘worst
case’ is rarely met as it assumes that all sources of error are at their maximum, and
that all are additive.
A further possiblity of error is in the sensor. This is a maximum of 0.1°C at 50°C,
0.2°C at 100°C, rising to 3/8% of reading, i.e. a possible 1.5°C at 400°C. This is the
maximum error for Special Limits Copper-Constantan thermocouples permitted by
ANSI (American National Standards Institute.)
In practice, these errors are usually much lower unless poor measurement techniques
are being applied.
8.2
If errors are suspected, check F in Section 7 should first be made. Then the
instrument should be checked against a glass thermometer with 1/10th degree cali-
brations traceable to NIST. Note that the calibration of a glass thermometer can be
invalidated by overheating -- ensure that this never happens.
8.3
Bind the thermocouple probe to the tip of the glass thermometer so that they
are close together. Do not compress the glass thermometer bulb.
8.4
Insert probe and thermometer in a Dewar flask or Thermos bottle containing
liquid at a suitable temperature. Do not allow the probe or thermometer to touch the
sides of the vessel. Ensure that the thermometer is immersed to its proper depth.
STIR WELL.
8.5
Allowing for some reading error in the glass thermometer (there can be none
in the digital BAT-12), readings should agree with the limits noted in 8.1. If readings
are outside these limits, the instrument should be returned for further checks. (See
section 10.0.) Recalibration of BAT-12 should not be attempted unless you have an
accurate temperature standard and a voltage standard such as a potentiometer with
one microvolt resolution.
A service manual is available. This includes circuit diagram, servicing information
and calibration instructions. Price is $25.
10
Summary of Contents for BAT-12
Page 2: ......
