Cavitation in mutistage pump 1

[terrorist]

Hi everyone,
We have a pumping system to circulate amine in closed loop consisting of columns, filters etc and back to the storage tank.
Now this pump is a axially split casing type 4 stage pump making the fluid reach 60 kg/cm2 at pump discharge. Tank is atmospheric pressurized tank.and level in tank never goes below 6 meters. Pump suction line is in same horizontal plane always I.e. whatever the head is available in tank, same reached to pump suction. There is one suction strainer installed in the line and suction line is around 25 meter long.
I feel cavitaion like sound coming from the pump. I saw pump test curve which shows that NPSH required at this point is 5 Meters. Operating temperature of fluid is 40 deg C.
I have few doubts,
* I assume the cavitation may be taking place in first stage only. Is it possible for cavitaion not occuring in 2nd and third stage as pressure is high in these stages and liquid may not be vapourising?
* pump suction line is 16" in size and there are 4 elbows in total in suction line.Strainer is clean.
* other than vaporisation of liquid, is there any other scenario that may be happening and making that cavitation type noise.
* NPSH required is 5 meter absolute pressure and available is 5 meter in tank + 10 meter corresponding to atmospheric pressure - max 5 meter for frictional losses I.e. not less than 10 Meters. From this calculation I don't think that pump should cavitate. But in real it sounds like cavifation.
Please give your valuable inputs.
Thanks a lot.

 

[LittleInch]

You've given us quite a lot of information, but missed out flow rate / velocity and density and vapour pressure of your liquid at 40C. This is the critical issue - a high vapour pressure would wipe out your spare in on time at all.

You don't add enough for pressure drop even through a "clean" filter can be the equivalent of 2-5m.

Be aware that NPSH and cavitation limits are NOT the same thing.

In general cavitation can start to occur anywhere from 1 to 4m HIGHER than NPSH. NPSH is measured by throttling the input until the differential head drops by 3%. The pump can be cavitating well before that, but as it is not directly measurable then no one really quotes it. The graph below, if you assume NPSH R=0 is the "cavitation" line, this illustrates the point.

PS - I'm not really sure I like your handle....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[JJPellin]

In my experience, amine systems exist for the purpose of absorbing gas from a process stream. In our processes, we would be using amine to absorb H2S, primarily. Whether the amine is considered rich or lean, it would often be saturated with dissolved gases. The way you describe circulation through filters suggests that you are dealing with lean amine. But, even so, we would consider the vapor pressure of the amine to be equal to atmospheric pressure and would not take credit for the atmospheric pressure when calculating NPSHa. Cavitation can occur from vaporization of the product, or from dissolved gasses coming out of solution. Only one of our amine systems had the sort of atmospheric tank that you describe. We eliminated the tank, in part, because of chronic cavitation in the amine circulation pumps. We now use elevated pressure vessels rather than atmospheric tanks sitting on the ground.

Johnny Pellin

 

[pkg80]

Hello terrorist

First , you should 1st confirm that this is cavitation ,you are judging now only by sound, at least you can perform the vibration test and be sure that there is cavitation in pump.

 

[Artisi]

Yep, vibration and sound analysis to start with, unless of course you have inspected the 1st stage impeller and have confirmed cavitation. Noise could be many things, gas in solution, poor entry into the impeller eye, pre-rotation, operating off BEP etc.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[terrorist]

Thanks O lot everyone,
I will collect required details like flow/velocity, density and vapor pressure of amine soon and let you know. Vibrations readings on De and nde side do not exceed 3 mm/S which I think is quite in limit.
And it's me only who is suspecting cavitation to occur, other people in plant has literally no interest in all these things. So pump can't be opened to see any symptoms of cavitation. Dissolved gases may be causing this problem. Please guide how to approach further. Practically I won't be allowed to do much. I am quite sure that cavitation is occuring either due to vaporisation or dissolved gases. How can I prove that.

 

[Artisi]

If nobody else is showing any interest, why are you?
How long have theses pumps been installed, has cavitation been considered a problem since installation?
If it is cavitation, what do you intend doing about it?
If you can't resolve the problem, what do you see as the outcome?
Many pumps run under cavitation conditions for years without any serious problems / damage, whereas some can be destroyed in a very short period of time, especially if there is any corrosion / erosion component involved or if cavitation is severe.

Think you need to sit back and ask yourself- why am I getting involved / what am I getting involved in?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

You should be able to show that the pump is operating very close to or below the NPSH limit once you get all the required information and do the simple calculation, including a more accurate friction loss calculation.

If it has been operating for some time then you might have damaged the initial stage already or, as artisi says, it might be alright. Only by looking at the differential head being produced can you see if the pump is failing or not.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.