Restriction Orifice in Liquid Service Making Noise 3

[StoneCold]

Greetings all
I have a restriction orifice in the spill back of an ether transfer line. The orifice makes a horrible howling noise when it is running, and I want to fix the problem. I don't think it is hurting the pipe but the noise is iritating.

The details of the orifice are as follows.
Pipe size 1" sch 40 carbon steel.
Orifice size is 0.28" flange mounted.
Thickness is 0.125"
MOC is 316ss
Orifice type is sharp edged.

Process conditions are
Upsteam Pressure 65 psig
Downstream pressure is 20 psig
Flow through the orifice is about 12 gpm
Operating temperature is around 80F
Material is ethyl ether.
Specific gravity is about 0.7
Viscosity is about 0.2 cP

What I am looking for is an equation or book that describes how to predict the noise generated by the pressure drop through the orifice. I would like to know that changing the current orifice from a single hole to a three hole would fix the noise issue.

Thanks
StoneCold

 

[MiketheEngineer]

Possibly the orifice is vibrating?? or you are causing such cavitation that the noise is produced??

 

[katmar]

A quick estimate indicates you have cavitation problems. The way to overcome it is to use multiple orifices in series. If this is what you mean by a three hole orifice then yes, but putting three holes in the same plate is not going to help.

12 gpm through a 0.28" hole is going to give a velocity of over 60 ft/s and a velocity head of around 18 psi. The velocity head in the vena contracta will be even higher. This is the head that is recovered to give you the 20 psig you have downstream. Bernoulli indicates that the static pressure in the throat will be very low, although in reality the losses from the sudden expansion means that the real static pressure in the throat will be a bit higher than indicated by Bernoulli.

A good reference and recipe for designing multiple orifices is Tung and Mikasinovic, Chem Eng, Dec 12, 1983, pgs 69-71. If you search here (or Google) on the author names you will find lots of discussion.

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[Latexman]

A previous company had a rigorus spreadsheet for multiple series orifice design in liquid service. It's objective was avoiding cavitation. It did not have a noise correlation. I'd be interested in a noise correlation if one is found.

Good luck,
Latexman

 

[hacksaw]

the plate will only be sharp edged for a few hours, there are noise calcs offered by some manufacturers


having said all that, most spill back orifices are thick plate or "gun barrel" designs, rather that thin plates, give a ring to a local pump manufacturer he'll set you straight.

 

[casflo]

The cause of the noise provoked by the restriction orifice is probably cavitation. According to the provided data, your restriction orifice has a cavitation index of almost 2 and the cavitation parameter for a fluid vapor pressure of approximately 10 psia, the upstream pressure of 70 psia and the pressure drop of 45 psi, is 1.33 that is well below 2, so a severe cavitation exists.
To avoid the cavitation substitute the actual restriction orifice plate by a multi-hole plate (I recommend you 5 holes instead 3) and so the cavitation index of this new plate will be less than 1.33.
Take into account that the multi-hole plates have different flow coefficients than the single hole plates, so to mantain the same operational conditions (12 gpm flow and 45 psi pressure drop) the pass section of the multi-hole plate must be lower than the pass section of the actual restriction orifice (0.0616 in2). The decreasing could be approximately 10%.
Regards
casflo

 

[ione]

http://peutz.fr/lacoustique/articles/equipements/Pipe_noise.pdf

Regards

 

[StoneCold]

Thanks everyone for the input.
I changed out the orfice plate yesterday. The 1/8" think plate looked fine, no errosion, but it only ran a few hours. The new orfice plate is 3/4" thick. There is no noise coming from the new orfice plate. Which indicates to me that the noise being generated was vibration of the orifice plate and not a lot of cavitation. I will keep an eye on it but this seems to have fixed the howling problem.

Thanks
StoneCold

 

[stanier]

Spirax Sarco has taken over a company called Mitech who manufacture multiport orifice plates for the expressed purpose of noise elimination and anti cavitation. Mitech used to distribute their software via the local dealer for modelling such items.

Copes Vulcan and CCI manufacture tortuous path valves for servere service applications such as turbine bypass on HP steam. They also make orifice plates using the same technology.

On solution to avoid the noise is to use heavier wall pipe downstream of the orifice plate. The cavitation remains and may ultimately erode the pipe wall

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[casflo]

If the material of the restriction orifice plate and the pipe is stainless steel, there is no damage due to cavitation. But if the downstream pipe is carbon steel and the restriction orifice cavitates the damage is focused in the pipe.

casflo

 

[zdarlight]

Hi Stone Cold,

I don't think cavitation has anything to do with the howling noise coming from your orifice plate.

Noise production in orifice flow is at its loudest when operating at the peak frequency. The peak frequency of the stream is a function of jet diameter at the Vena Contracta, fluid velocity and the corresponding Strouhals number of the jet.

Although an orifice plate is a rigid component there will always be a small dynamic response under process conditions.

I believe there is a chance the howling noise could have been created by resonance between the plate and the jets peak frequency.

The natural frequency of the plate and the peak frequency may have came close to coinsiding with each other, hence unusual levels of noise.

By altering the plate thickness you have drastically altered the natural frequency of the plate. Hence, no more noise!