6 calibrating the concentration measurement – K-Patents PR-21-S User Manual
Page 60

54
PR-21-S instruction manual
6.6 Calibrating the concentration measurement
mA OUTPUT
FIELD CALIBRATION
CHEMICAL CURVE
n
D
CCD
TEMP
Pt-1000
Indicating
transmitter
DTR
Sensor
CONC
CALC
1.
2.
3.
4.
5.
6.
DAMPING
Figure 6.12
The six layers of
concentration calibration
The concentration calibra-
tion of the K-Patents inline
refractometer PR-21-S is or-
ganized in six layers.
1. The information from the
CCD element and the Pt-1000
temperature element. The
position of the shadow
edge (Figure 2.2, “Image
detector system”) is de-
scribed by a number called
CCD and scaled from 0–100 %.
2. The sensor calibration: The
actual refractive index n
D
is calculated from the CCD
value. The process tem-
perature is calculated from
the Pt-1000 resistance.
The sensor output is n
D
and temperature TEMP
in Centigrade. Hence, the calibrations of all PR-21-S sensors are identical,
which makes sensors interchangeable. Furthermore, the calibration of each
sensor can be verified using standard refractive index liquids, see Chapter 12.
3. The chemical curve: The Indicating transmitter DTR receives n
D
and TEMP and
calculates the concentration value according to chemical curves derived from
available chemical literature and K-Patents expertise. The result is a tempera-
ture compensated calculated concentration value CALC.
4. Field calibration: Adjustment of the calculated concentration value CALC may
be required to compensate for some process conditions or to fit the measure-
ment to the laboratory results. The Field calibration procedure, Section 6.6.3,
determines the appropriate adjustment to CALC. The adjusted concentration
is called CONC. If there is no adjustment, CALC and CONC are equal. Thus the
chemical curve is kept intact as a firm base for the calculation, the adjustment
is merely additional terms.
5. Damping: See Section 6.1.
6. Output signal: The range of the 4 to 20 mA signal is defined by its two endpoints
on the CONC scale, see Section 6.5.1.