Reference Manual

6−4
levels can be produced in a two-inch line with as
little as a 200 psi pressure drop. Major sources of
aerodynamic noise are the stresses or shear
forces present in turbulent flow.
Some of the sources of turbulence in gas
transmission lines are obstructions in the flow
path, rapid expansion or deceleration of
high-velocity gas, and directional changes in the
fluid stream. Specific areas that are inherently
noisy include headers, pressure regulators, line
size expansions, and pipe elbows.
Aerodynamic noise is generally considered the
primary source of control valve noise. There are
several reasons for this:
D This type of noise has its highest energy
components at the same frequencies where the
human ear is most sensitive - between 1000 and
8000 hertz.
D Large amounts of energy can be converted
to aerodynamic noise without damaging the valve.
In the past, the noise was considered a nuisance,
but as long as the valve did its job, it was not of
major concern. Today, with increasing focus on
environmental issues, including noise, there are
guidelines on the amount of noise a valve can emit
in a given workplace. Research has also shown
that sustained noise levels above 110 decibels
can produce mechanical damage to control
valves.
High noise levels are an issue primarily because
of OSHA’s standards for permissible noise limits
and the potential for control valve damage above
110 dBA. Additionally, loud hydrodynamic noise is
symptomatic of the more dangerous and
destructive phenomenon known as cavitation.
Noise Prediction
Industry leaders use the International
Electrotechnical Commission standard IEC
534-8-3. This method consists of a mix of
thermodynamic and aerodynamic theory and
empirical information. This method allows noise
prediction for a valve to be based only upon the
measurable geometry of the valve and the service
conditions applied to the valve. There is no need
for specific empirical data for each valve design
and size. Because of this pure analytical approach
to valve noise prediction, the IEC method allows
an objective evaluation of alternatives.
The method defines five basic steps to noise
prediction:
1. Calculate the total stream power in the process
at the vena contracta.
The noise of interest is generated by the valve in
and downstream of the vena contracta. If the total
power dissipated by throttling at the vena
contracta can be calculated, then the fraction that
is noise power can be determined. Because power
is the time rate of energy, a form of the familiar
equation for calculating kinetic energy can be
used. The kinetic energy equation is:
E
k
+ 1ń2mv
2
where,
m = mass
v = velocity
If the mass flow rate is substituted for the mass
term, then the equation calculates the power. The
velocity is the vena contracta velocity and is
calculated with the energy equation of the first law
of thermodynamics.
2. Determine the fraction of total power that is
acoustic power.
This method considers the process conditions
applied across the valve to determine the
particular noise generating mechanism in the
valve. There are five defined regimes dependent
upon the relationship of the vena contracta
pressure and the downstream pressure. For each
of these regimes an acoustic efficiency is defined
and calculated. This acoustic efficiency
establishes the fraction of the total stream power,
as calculated in step one, which is noise power. In
designing a quiet valve, lower acoustic efficiency
is one of the goals.
3. Convert acoustic power to sound pressure.
The final goal of the IEC prediction method is to
determine the sound pressure level at a reference
point outside the valve where human hearing is a
concern. Step two delivers acoustic power, which
is not directly measurable. Acoustic or sound
pressure is measurable and, therefore, has
become the default expression for noise in most
situations. Converting from acoustic power to the
sound pressure uses basic acoustic theory.
4. Account for the transmission loss of the pipe
wall and restate the sound pressure at the outside
surface of the pipe.
Steps one and three are involved with the noise
generation process inside the pipe. There are