PDA

View Full Version : Scew Compressor Pulsations



AOkasha
06-04-2012, 04:11 PM
Hello all,

I am trying to measure the dynamic pressure out of a screw compressor (Atlas Copco GA7) with WCT-312 pressure transducer, but I cannot capture any pulsations

I have investigated GA7, and I think that it doesnot contain a muffler or silencer.

Does anyone can help me to know why the pulsations are suppressed.

Thank you

aramis
06-04-2012, 05:14 PM
Sorry no experience with air compressors!

My stupid ideas are:

Even if the compressor had a muffler you should be able to measure pressure waves.

If the sensor has the correct response time for dynamic measurements then the problem could be your sampling rate. Are you sure you are sampling at the correct frequencies?

Where you measure is also important, large discharge lines could act as mufflers too.

Try changing flow conditions.

Hope you solve it soon!

AOkasha
06-04-2012, 06:54 PM
Dear Aramis,
Thanks for your reply.
These ideas are not stupid ideas at all : )

The pressure transducer that I use has a usable bandwidth up to ~20kHz if the transducer tip is mounted flush to the flow.

The first harmonic is expected to appear from 480 Hz to 1275 Hz. I have used 6400 Hz sampling rate.

Measurement Test Points is near the source, and I already tried to change the flow conditions many times.

aramis
06-04-2012, 08:58 PM
I supposed you already checked and calibrated your equipment but rechecking might help.

I’d try to use the same equipment in front of a loudspeaker with a wave generator.

Verify you are not using any filters or the wrong ones.

Check that nothing with dampening effect is affecting your test point, like an obstruction upflow that may cause eddy currents.

At this point I’m out of even stupid ideas!:eek:

Good luck!

aramis
06-04-2012, 09:38 PM
... well more stupid ideas fell from nowhere, please remember I don’t know these compressors.

Does the compressor have coalescent oil filters installed? They surely would have a large dampening effect.

And if they don’t, oil could cover your sensor and also have a dampening effect. If the flow is high enough placing the sensor on top of the discharge would not help you can get oil in anyway. So remove the sensor and check for oil.

Do you need to cool the sensor? If cooling media is too cold you could get condensation (unless the air is dry enough). Just in case also check for water.

I don't think these problems would work on their own, you would also need stagnation points or eddy currents working with them together and this depends on the actual geometry of the discharge.

Just brainstorming!

AOkasha
07-04-2012, 09:54 PM
I think that I have to try using the same equipment in front of a loudspeaker with a wave generator or something like that.

Are you sure that oil filters can make a large dampening effect ?

aramis
08-04-2012, 03:56 AM
I think that I have to try using the same equipment in front of a loudspeaker with a wave generator or something like that.

Just to make sure. Maybe I’m too inclined to some superstitious behavior before my experiments! :)


Are you sure that oil filters can make a large dampening effect?

Yes, I don't know by what factor but can tell you this:

- The more efficient the filter and larger the area would have a greater effect.

- The larger the pressure drop in the filter means that oil has penetrated inside and this would cause a larger effect.

- Even more if your compressor hast two filters.

Screw compressors with coalescent filters don’t need mufflers (…in refrigeration!).

Read this:
http://www.parker.com/literature/Finite%20Filter/Literature%20PDFs/Finite_Exhaust_Coalescing_Silencers_1300_330_USA.pdf

If the loudspeaker test comes OK maybe you should rethink the position of the probe but you will have an oil film on the sensor. I think a thin layer of oil would only have a small effect.

Hope dies last!

glenn1340
08-04-2012, 05:26 PM
Hi,
there aren`t any silencers or mufflers on any screw air compressors as far as I`m aware. You`ll have to fit the transducer directly after the air end discharge, I think there`s a temperature sensor you can take out and replace with the transducer, just make sure the compressor isn`t running hot before hand. It`s no good fitting it elsewhere as any pulsations will be eliminated by the seperator element.

Hope this helps ,
Glenn