variations. The results of the variations C2_S1,2 and
C1,2_S1 will be described for both displacement ventilation
and mixing ventilation. However, as the mixing ventilation
does not create a rather distinctive air movement in
the space, only the airflow pattern with displacement venti- 8 lation will be shown schematically Figs. 7 and 10 . Ž .
The following Figs. 9 and 11 present the profiles of the
local air temperature difference Dtair, x for the tests with
the cooled ceiling and displacement ventilation, as only
this combination causes significant temperature profiles.
The air temperature distribution with mixing flow was
always almost uniform. Figs. 8 and 12 present the relative
contamination efficiency directly comparing the tracer gas
concentrations with displacement flow and mixing flow.
All these figures show the data vs. the height for three
different axes, where each point in the graphs represents
the average of two values measured in one respective
height compare Fig. 6 . Ž .
5.1. Variation C2_S1,2
The contaminant removal efficiencies measured with
displacement flow are lower than with mixing ventilation
in the entire room Fig. 8 . The data of the lower region Ž .
Ž . 0.1 to 1.0 m indicate that the fresh air was supplied to all
axes, although the tracer gas concentration with displacement
flow increases towards the measuring axis 3 due to
recirculating room air contaminated from the ceiling see Ž. Ž
Fig. 7 ..
The uneven mixing of supply air and room air when
mixing ventilation was used is one of the reasons for the
relatively better air quality at all heights with displacement
ventilation. While the displacement flow resulted in contaminant
removal efficiencies of 0.05 to 1.1 0.1 to 1.7 m Ž
height , the corresponding values with mixing flow were .
between 1.1 and 1.4, with the higher values below the