Reference Manual
748384-C
September 2003
1-20 Introduction
Rosemount Analytical Inc. A Division of Emerson Process Management
NGA2000 Reference
Finally, you may have a problem with your
dilution system, or your reference bottles.
No matter how badly messed up the ana-
lyzer may be by this time, it should still read
the same on the same gas. Make your
blender produce the same concentration
gas as a reference bottle, disable any lin-
earizer, and read the two samples with the
analyzer. If they don't read the same,
something is wrong with either the dilution
system or with the reference, and you will
have to resolve this before you can resolve
the linearity questions.
A common source of problems is the sam-
ple system in a CEMS installation. If you
are losing concentration on the way down
from the stack, due to leaks or water ab-
sorption or something, you may be unable
to pass a linearity check when you intro-
duce the reference gases at the probe. You
will have to debug this. You may have lost
concentration in a water trap, or you may
have leaks. If the source adds ammonia to
control NOx output, you may be condensing
this in the sample cooler, and this will strip
all NOx out of the sample.
Linearization: Well I tried all that, but I
have to make it work anyway…
If all else fails, and you have to make the
analyzer work without being able to resolve
the real problem, you can use the tweak
function described below. Remember that
you are probably masking a real problem by
doing so. Verify that the analyzer is really
working correctly before resorting to this.
Some esoteric analyzer problems that can
make it difficult to linearize include: FID -
use of N
2
as the fuel diluent, and stratifica-
tion at low hydrocarbon levels.
Saturation of the signal due to incorrect
carbon content setting, sample pressure
setting, or calibration gas value.
CLD -
Inadequate ozone production, either
caused by excessive sample flow, inade-
quate ozonator air flow, failure to use oxy-
gen for higher than 1000 ppm concentration
readings on a low range, or 2500 ppm on a
high range unit, or a bad or tired ozonator
lamp. Saturation of the signal caused by
incorrect capillary setting or installation, and
incorrect calibration gas values.
NDIR -
Use of the incorrect cell spacer for
the range you are working on, or incorrect
settings of the temperature compensation
factors.
All analyzers - Saturation of the analog sig-
nal path due to incorrect settings of gain re-
sistors, or sample conditions, or specific
parameters.