Pump Cavitation

[ragequit]

Hi,

We have a pump connected to a tank at our work place, the problem is, the pump's base is orinted such that an elbow is required to connect the pump to the tank. the pump is cavitating quite heavily. i believe there should be sufficient suction pressure available. but the pump is still experience cavitation. can the elbow alone be responsible for causing cavitation, due to uneven flow? Is the reason why elbows are not recommended at the suction end of a pump purely due to head loss, or can the shape of an elbow cause uneven flow or bubbles.

Thanks is advance

 

[1gibson]

What type of pump? Horizontal double suction (centrifugal) would probably be the worst for this configuration.

You can purchase an elbow (or add to existing) turning vanes, but you are probably stuck with the configuration so I won't suggest correcting it. Google images "pipe elbow turning vanes."

 

[ragequit]

Thanks for the reply!

Its a PD pump

 

[zdas04]

Cavitation in PD pumps is exceedingly rare. Cavitation is the formation and subsequent collapse of steam bubbles in the flow. A PD pump does not depress suction pressure appreciably (the way a dynamic pump would), so phase change on the suction is quite rare.

A PD pump cavitating "quite heavily" would probably pulverize the pistons (in a plunger pump) or destroy the stator (in a PCP) within a few hours. Are you sure you don't mean "vibrating" instead of "cavitating"?

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[Latexman]

Elbows cause irrotational flow which can lead to vibration and /or cavitation. Thats why pump manufacturers want so many diameters of straight pipe length right before the pump suction nozzle. I recall 5-10 diameters as a rule of thumb.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[JJPellin]

I would offer that both reasons are valid. An elbow imposes some pressure drop that reduces NPSHa. But, we require a straight run of pipe at the suction even if there is more than adequate NPSHa. The uneven velocity and turbulence created by the elbow can cause cavitation even of the NPSH margin is high. We want 10 diameters of straight pipe with 5 diameters as the bare minimum.

Johnny Pellin

 

[LittleInch]

Ragequit, You will find people on this forum very helpful, but we all need data to help you or point you in directions you might think of changing. If you look up NPSH on this site you will find multiple threads as it is a common issue.

You say you "believe" there is sufficient suction pressure. If you give us the fluid, its vapaour pressure (sometimes called Reid Vapour Pressure or RVP) at the temperature you are pumping at, your elevation above sea level, minimum liquid level above the pump, sketch or description of the inlet piping (size, length, number of fittings) and flowrate we might be able to check this and see where the issues lie. Temperature can be very importaant - has this only just started or is it a new pump?

Otherwise I agree with the posts above, an elbow close to the pump is not normally recomended, but as many of these issues are velocity related, an option may be to increase pipe size as much as possible up to the pump and then reduce down (concentric reducer). Another option is a flow straightener or flow conditioner plate which I have seen used before where there is not enough room for 5 to 10 D of pipe going into the pump (google image - ISO flow conditioner ).

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Latexman]

Does your PD pump have pulsating flow? If so, did you account for acceleration loses in the NPSH calculation? I have found a good % of engineers are do not know to apply this. I've seen cases where acceleration loses was the major factor in NPSH.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[ragequit]

Hello guys

Thanks for all the replies!

Im new to this forum and new to engineering aswell. so please forgive my ignorance and lack of knowledge.

I looked at other pumps in the area and they are all running fine, they have the same configuration, and roughly the same amount of piping from the tank to the suction, but they do not have elbows. so i assumed that 1 elbow would not impede the NSPH too much.

So i tried to do the calculations, and i am getting numbers that do not make sense. this is my first time calculating pressure heads and head losses so please bear with me.
on the first page of the attachment you can see a drawing of a tank. ignore the "assume 10, 5, 1m" . i looked at the tank data, and it is usually at 10m level above the level of the pump

So our available head pressure available should be 10 m

so now i needed to calculate major and minor losses. to do that they require velocity.

So using the equation
V=2gh
V=196m/s (too high?)

so using that velocity and calculating the head losses meant the losses were way too high, so high in fact that they were greater than the available pressure head. so i think my mistake is at this step. for calculating velocity.

I would really appreciate your help

some more data that i have

liquid temp-175 C
SPEC GRAV- .91
VISCOSITY; 60 cS

 

[ragequit]

Not sure why last link didnt work

 

[LittleInch]

Please look at the responses carefully and give us answers and data. Why do you think its cavitating? What type of pd pump is it?, what is line size and flow rate?, at 175 c, what is your vapour pressure?, what is this stuff?, are all the other tanks identical fluid?

Velocity is volumetric flowrate per second divided by cross sectional area all in metres.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Latexman]

V=2gh? What does that have to do with the velocity created by the pump volumetric flow rate? Check out the units in that equation. m/sec = m[sup]2[/sup]/sec[sup]2[/sup]? The units do not equatee! You can put a big X through that approack. It's a dead end. Try another approach.

Good luck,
Latexman

Need help writing a question or understanding a reply? forum1529

 

[zdas04]

I was going to ask the same thing. It looks like yet another bastardization of Bernoulli's equation. If v^2/2+P/ρ=constant and you assume P=ρgh and you assume that the constant=0 then v^2=2gh. Perfectly logical right? He just forgot to take the square root. Life gets so much simpler when you can assume the constant equals zero and that the only pressure source is hydrostatic head. The fact that both assumptions are pure garbage shouldn't stand in the way of partial Engineering.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[mec5216]

Man, what kind of pump is it? A pistion pump, vane pump, gear pump, screw pump...? And one other crucial information is how much is the flow.
I work as well with pumps in a refinery, and an elbow just before the suction flange can lead to turbulence and noise very easily. If the flow is high, and therefore velocity in the pipes is high, it leads to turbulence and consequently noise. If flow is not eough high, turbulence is more diffuclt to occur. But these are not due to cavitation, they are due to an fast change in the direction of the fluid in the suction of your pump, then leading to turbulence.
I would recommend you to try to decrease flow and chek if it stops in an lower flow.
If the flow is high, a pipe elbow turning vanes may be a good idea.

Good luck!

 

[lukman143]

I have never got such problem with PD cavitation, even though there is a more complicated suction line than you said here.

The only problem I have is blockage suction strainer with dirt or sludge. So, please look at your liquid. Is there any possibility to form sludge after you loading the liquid into the tank?

Once again please look at your pumping liquid too (liquid temperature, etc). It is extremely important.

http://chemicalengineeringmagazine.com,

Free Chemical Engineering Magazine Subscription

 

[Artisi]

Quote. " the pump is cavitating quite heavily" how do you know this, is it a fact, it guess work, or just the term normally mis-applied every time there is a noisy pump?

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.)