The k-factor were calculated for all the leak geometries. It was
found that they are a function of geometry in contrast to the regular
formula which expresses it in terms of static pressure loss
and mean velocity through the leak. The flow coefficient (C) and
pressure exponent (n) were found for all leak geometries and
at each Reynolds number using the power law model also. Also,
the leakages calculated by the CFD code and with the formula
used to estimate duct leakage in mine ventilation were compared.
The use of the form of Eqs. (1) and (3) is simply to allow the
current research to borrow from the current literature some of
the appropriate forms of how to quantify the characteristics of
these chosen leakage hole geometries. To be more exact the literature
did not yield any theoretical results yet of these types
of leakage holes rather there are results of typical classifications
of semi empirical values of leakage in ducts in units of L/s/m2
of duct area. These holes are somewhat more idealized in this
study obviously in their geometries but do offer a variety of situations
as to what type of leakages might be confronted in the
field. In fact an informal survey of some field installations and
some of the current observations in the laboratories at UNLV that
were involved in part of this study did indicate indeed that a variety
of shapes exist for these leaks and that no one shape would
depict all types of leaks. Hence it is thought that this study can
provide a reasonable ground on which somebody attempting to
measure air duct leakage in the field can base his calculations on.
Previous studies [11] did present some suggested values for the
exponent “n” in Eq. (1) of around 0.6 (variation given between
0.5 and 1.0 depending on the leakage shape) although there was
no definition of the geometry of the leakage shapes at all. Comparing
that with the n values from Table 1 one can observe that
the average of these indices is a little on the low side compared
to that value by about 20%. This is not surprising as the leakage
geometry here is more defined and would probably be much
“smoother” than that of a typical jagged real duct leak. The original
study in which Eq. (3) was used [4] in the calculation of k
was for a motorized damper placed in a duct and not for leakage
of air flow where the hydrodynamics of the flow in these
two cases are very different and cannot be compared together.