Page 57 - Valve Selection Handbook

44 Valve Selection Handbook

the air than the lower frequencies. A second beneficial effect of the diffuser
is to distribute the flow more evenly, over the cross section of the pipe.
Ingard has shown that the normalized acoustic resistance of a perforat-
ed flat plate mounted across the pipe is directly proportional to both the
Mach number of the flow through the perforations and the factor

28
where a = open area ratio of the perforated plate. Although this can-
not be directly related to noise attenuation, it would appear that the Mach
number should be as large as possible, and a as small as possible. For
practical purposes, a maximum Mach number of 0.9 is suggested. If the
available pressure drop across the diffuser is limited, a Mach number
with a lower value may have to be chosen. Practical values for the open
area ratio may be taken as between 0.1 and 0.3. Practical values lower
than 0.1 may result in an excessively large diffuser, while values higher
than 0.3 may result in too low an attenuation.
The peak frequency of the jet noise is also inversely proportional to
the diameter of the jet. Therefore, from the point of noise attenuation, the
diameter of the perforations should be as small as possible. To avoid the
nozzles from becoming blocked, nozzles with a minimum diameter of
5mm are frequently used.
If the flow velocity in the pipe downstream of the silencer is high, the
boundary layer turbulence along the pipe may generate a noise compara-
ble with the attenuated valve noise. Experience suggests that this will not
be a problem if the Mach number of the flow in the pipe is kept below
about 0.3.
Predicting valve noise and silencer performance is a complex matter.
Discussions on these subjects, including the design of silencers, may be
found in the References 29, 30, 31, 32 and 33. Further discussions on the
generation and radiation of piping noise may be found in References 34
and 35.