Industrial-process control valves – Part 8-4: Noise considerations – Prediction of noise generated by hydrodynamic flow
|Publication Date:||1 August 2005|
|ICS Code (Pressure regulators):||23.060.40|
|ICS Code (Noise emitted by machines and equipment):||17.140.20|
|ICS Code (Industrial process measurement and control):||25.040.40|
This part of IEC 60534 establishes a method to predict the noise generated in a control valve by liquid flow and the resulting noise level measured downstream of the valve and outside of the pipe. The noise may be generated both by normal turbulence and by liquid cavitation in the valve. Parts of the method are based on fundamental principles of acoustics, fluid mechanics, and mechanics. The method is validated by test data. Noise generated by flashing flow is not considered in this standard.
The transmission loss (TL) equations are based on analysis of the interaction between the sound waves inside the pipe and the coincidence frequencies in the wall of the pipe taking into account that commercial pipe tolerances allow a relatively wide variation in the thickness of the pipe wall. Ideal straight piping is assumed.
The method can be used with all conventional control valve styles including globe, butterfly, cage-type, eccentric rotary, and modified ball valves. Tests so far have only been conducted with water. The applicability of this method for fluids other than water is not known at this time.
This standard considers only noise generated by hydraulic turbulence and fluid cavitation. It does not consider any noise that might be generated by mechanical vibrations, unstable flow patterns, and unpredictable behaviour. In the typical installation, very little noise travels through the wall of the control valve body. The noise is measured at the standard measuring point of 1 m downstream of the valve and 1 m away from the outer surface of the pipe.
This prediction method has been validated with test results based on water covering more than 90 % of all known valve types at inlet pressures of up to 15 bar. This method is considered accurate within ± 5dB(A) except in the range of xF = xFz ± 0,1, when xFz is calculated using equations (3a) or (3b).