Just starting to look at a project involving a requirement for a number of control valves for propylene gas,inlet pressure 14 barg,outlet pressure 2 barg, done a sizing calculation and the software indicates supersonic flow with a decibel level of 90@ 1 metre,valve size is 1O0mm with the outlet size opened up to 200mm using a reducer,the software indicates the use of a 3 stage cage trim,dont have the flow to hand as I am doing this from home,but available tomorrowDoes anyone have any experience of this sort of application,things to look out for etc,Would a differnt trim help?,is it benificial to keep away from supersonic flow and what are the possible negatives,the valve characteristic is equal percentage and the actuator fails open.

You do not mention the purpose of this study and your position in the project.

Before going into details for any technical valve solution, I have always found it sensible to look into 'what is available at already existing (standard) commercial solutions' and what do the suppliers of the commercial solutions suggest.

To make such a survey you will need not only to focus on the detailed components itself (in this case valves) but also purpose and overall target for the total application.

1. The flow will be sonic, not supersonic.
2. The noise of 90dBA can be mitigated by heavier wall piping downstream and/or insulation if lo-db trim is deemed undesirable.
3. Depending on flowrate, a v-ball with diffuser could reduce the noise below 85 dbA.  If flow and pressure drop are constant then a lo-db diffuser would be adequate.  There are many low-noise options.  Ask Fisher, Masoneilan, Valtek, Neles etc and they can guide you.
4. The 3 stage trim sounds overkill for the relatively low upstream pressure although they likely look at the pressure ratio.  I think a single stage drilled hole cage or grille should be able to lower the noise by 5 dBA.

I agree with scotsinst.  Supersonic flow is a rarity pretty much confined to converging/diverging nozzles as seen on the tail end of rockets.  Inside valves and pipes we'll experience standing shock waves and lots of noise.  I have not run the numbers, but 90dB sounds light for that pressure ratio and valve size.  

Insulation of the pipe, and heavier wall pipe both limit radiated noise, but above about 110dB, mechanical damage becomes a concern. You can shake off the accessories and damage the valve internals, even if you can't hear the valve screaming for all the insulation.  

Globe valves with noise attenuation trim are a valid choice as you have evaluated, but it is much less expensive to stick a noise-reducing plate downstream, after a pipe reducer to increase the diameter of the pipe.  Gases expand when throttled, and the primary noise reducing strategy is to keep the velocity low.  Anything over about 0.7 Mach is likely to scream, and 0.3 mach generally gives a comfortable noise level.  

Do your ideal gas calculations, and size the downstream piping to limit the velocity.  Stick a noise-attenuating plate of that diameter in the pipe, then work back upstream to the valve.  Calculate the Cv required for the valve and the Cv for the plate should be about 1.5 times as great.  ANd remeber that a low-noise plate does not have a single orifice, but jillions of small holes.  Characteristic frequency generated is inversely related to hole diameter. A single big hole may sound like a Saturn V at blastoff, but Small holes radiate in a frequency that is lightly weighted on the dB(A) scale.  If you don't hear it, it's not noise.  

MORE NEWS