T400 Ozone Analyzer Operator’s Manual EPA Protocol Calibration
215
10.3.5. O
3
LOSS CORRECTION FACTOR
In spite of scrupulous cleaning and preconditioning, some O
3
may be lost on contact
with the photometer cell walls and the gas-handling components. Any significant loss of
O
3
must be quantitatively determined and used to correct the output concentration
assay. In any case, the O
3
loss must not exceed 5%.
To determine O
3
loss:
1. Calibrate a stable ozone analyzer with the UV calibration system, assuming no
losses.
2. Generate an O
3
concentration, and measure it with the analyzer as close as
possible to the actual inlet of the photometer cell.
3. Measure the concentration as close as possible to the outlet of the cell.
4. Repeat each measurement several times to get a reliable average.
5. Measure the concentration at the output manifold. The tests should be repeated at
several different O
3
concentrations.
The percentage of O
3
loss is calculated as,
Equation 10-3
100
C
2
)CC(
C
lossO%
m
oi
m
3
×
+
=
Where
C
i
= O
3
concentration measured at cell inlet, ppm
C
o
= O
3
concentration measured at cell outlet, ppm, and
C
m
= O
3
concentration measured at output manifold, ppm.
For other configurations, the % O
3
loss may have to be calculated differently. The
ozone loss correction factor is calculated as:
L = 1 - 0.01 × % O
3
loss.
10.3.6. SPAN DRIFT CHECK
The first level of data validation should accept or reject monitoring data based upon
routine periodic analyzer checks. It is recommended that results from the Level 1 span
checks be used as the first level of data validation. This means up to two weeks of
monitoring data may be invalidated if the span drift for a Level 1 span check is 25%.
For this reason, it may be desirable to perform Level 1 checks more often than the
minimum recommended frequency of every 2 weeks.
10.4. AUDITING PROCEDURES
An audit is an independent assessment of the accuracy of data. Independence is
achieved by having the audit made by an operator other than the one conducting the
routine field measurements and by using audit standards and equipment different from
those routinely used in monitoring. The audit should be a true assessment of the
measurement process under normal operations without any special preparation or
