
with
¼
n
ð1 RÞ
1 R þ
i
ð2:63Þ
The theoretical exponential can be fitted to the experimental points, as shown
in Figure 2.11 depicting an actual experiment. The concentrati on increase in
the supply duct, C
0
3
C
3
, quickly reaches its steady-state value, while a good
fit of the exponential can be obtained within two time constants, allowi ng the
determination of the steady-state concentration and time constant without
waiting for equilibrium that is reached after at least five time constants.
Appropriate method for assessing the
recirculation ratio
In Equations 2.33, 2.34 and 2.41 the concentration difference C
6k
C
3k
is at the
denominator, and these two concentrations are close to each other at steady
state when the recirculation ratio is high. This leads to a large confidence
interval of the calculated airflow rates. In units with large recirculation
ratios, it is better to inject the tracer gas at location 3 instead of location 2.
The supply airflow rate can then be determined with better accuracy, using:
Q
24
¼
I
3
C
3
0
3
C
33
ð2:64Þ
Assuming that the confidence interval is the same for both concentrations, the
confidence interval is:
Q
Q
ffi TðP ; 1Þ
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I
I
2
þ 2
C
C
0
C
2
s
ð2:65Þ
0
5
10
15
20
25
30
35
40
Tracer gas concentration (mg/m
3
)
0
0 2 4 6 8 10 12 14 16
2
4
6
8
10
12
14
16
Concentration difference [mg/m
3
]
C3
C3¢
C3¢–C3
Tracer injection
Time (h)
Figure 2.11 Tracer gas concentrations in the supply duct, upstream (3) and
downstream (3
0
) of the tracer gas injection port
Note: Points are measured concentrations, while lines are exponential fits.
Source: Roulet and Zuraimi, 2003.
36 Ventilation and Airflow in Buildings
Commentaires sur ces manuels