LI-COR LI-6400, Руководство пользователя

Страница: 459 / 846

Calibration Issues
CO2 and H2O Analyzers
Using the LI-6400 18-3
CO
2
and H
2
O Analyzers
Factory Calibration
The factory calibration of the infrared gas analyzers (IRGAs) consists of de-
termining the coefficients of the polynomial
f(x) for water (Equations (14-5)
and (14-6) on page 14-5), and g(x) for CO
2
(Equations (14-8) and (14-9) on
page 14-6). The source for CO
2
concentrations is a series (usually 13) of stan-
dard tanks, ranging in concentration from 0 to about 3000 µmol mol
-1
. Water
concentrations are generated using a LI-COR LI-610 Dew Point Generator.
These measurements are made at a series of temperatures (typically 30, 40,
and 50C), with the entire instrument in a temperature controlled chamber.
The data are then standardized (concentrations are scaled by temperature,
voltages are scaled by pressure), and the calibration curves generated: a 5th
order polynomial for CO
2
, and a 3rd order polynomial for H
2
O. The coeffi-
cients for these curves are provided on the calibration sheet, and are entered
in the instrument as the values following the
CalCO2= and CalH2O= config-
urations commands (see Figure 16-12 on page 16-13). The calibration also
provides data on the IRGAsÕ drift with temperature, and this information is
provided via the
CalZero= calibration command.
User Calibration (Zero and Span)
User calibration actions consist of checking and/or adjusting the zero and
span. There are two parts: ÒzeroÓ, which is checking the reading with dry,
CO
2
-free air in the cell, and ÒspanÓ, which is checking the reading with a
known concentration in the cell. A zero adjustment shifts the entire response
curve a fixed amount. For example, if an IRGA reads 2 µmol mol
-1
with
CO
2
-free air in the cell, and you do a zero adjustment to make it read 0, then
you have changed the response by -2 µmol mol
-1
at every concentration. By
contrast, a span adjustment affects the sensitivity of the response. For exam-
ple, suppose an IRGA reads 990 µmol mol
-1
when there is 1000 µmol mol
-1
in the cell. If you adjust the span to correct this, you will have increased the
response by 10 µmol mol
-1
at 1000, but at 500 µmol mol
-1
, the response will
be increased by only 5 µmol mol
-1
, and at 2000 µmol mol
-1
, it will be about
20. (The numbers are not quite right, since the span adjustment is linear with
IRGA signal, not concentration.) At 0 concentration (or rather, at zero volt-
age), a span adjustment has no effect at all.