Pump Cavitation

[airbus]

If a pumps is dead headed, how would you determine the time it would take to cavitate? How this time compares to reality? Could it cavitate inmediately?

 

[Artisi]

When a pump runs against a "dead" head it will not cavitate.
cavitation is a function of flow into the eye of the impeller.
Running against a "dead" head can cause numerous other problems: high temperature within the pump case, high levels of vibration, broken pump stafts, bearing failure, seal failure, noise - to name a few of the problems.

Suggest you look at the following site for more info.

http://www.mcnallyinstitute.com/index.html

Naresuan University
Phitsanulok
Thailand

 

[Bjegovic]

Artisi is right.
note also that 'immediate' effect of cavitation is loss of power/head accompanied with noise, vibrations, etc.

but for pump itself (regarding the damage inflicted), cavitation is a long term process with cummulative effects, which may or may not be connected to loss of power/head characteristic for fully developed cavitation!

 

[Zapster]

The energy input into the fluid will raise the temperature of the fluid. Depending on the fluid and pipe configuration, couldn’t the fluid be raised to the boiling point? Then any minor changes of pressure as the fluid is worked by the pump’s impeller will cause cavitation????

 

[checman]

I would agree with Zapster. But many things would influence the outcome. Suction pressure, the fluids specific heat and other characteristics, the pumps efficiency, the pumps clearances and so on. If the pump has a high specific suction speed you would most likely develop incipient clouds of Cavitation the minute you drop below a certain flow rate. Suction recirculation could cause some Cavitation of the fluid in the pumps suction. If you have a check or the valve shut on the suction side you might burst a pipe or something from the fluid expanding. Or you might lock the pump up from the pump parts expanding and contacting each other. I do not recommend trying it to find out what happens.

 

[sprintcar]

Running a pump against shut head will generate heat inside the pump casing. This will build until the fluid turns to steam and the pump can explode

"If A equals success, then the formula is: A = X + Y + Z, X is work. Y is play. Z is keep your mouth shut."
-- by Albert Einstein

 

[JJPellin]

I respectfully disagree with some of the other replies. Many pumps that I work with have an NPSH(r) curve that increases at very low flow. It is not uncommon for some of these pumps to cavitate immediately if run dead-headed. This would depend on how much NPSH available there is. But, I agree with my predecessors in the fact that the primary risk to the pump is not the immediate cavitation, but the build-up of heat which will cause the fluid to boil and the pump to run dry. How long it will take depends on the pump. We have some very high energy barrel pumps that would suffer major damage within a couple of minutes. We also have some very low energy circulating pumps that could probably run dead-headed for half an hour with little or no affect. At all depends on the particular pump. Pump manufacturer's have a program that can estimate temperature rise and give you an idea how long the pump may survive.

 

[Artisi]

Quote. "Many pumps that I work with have an NPSH(r) curve that increases at very low flow"

My understanding of this is that it is not an increase of NPSHr in terms of a requirement to avoid cavitation as the NPSHr requirement drops with decreasing flow but a head (pressure) requirement to get the water onto the impeller blade/s due to a mis-match of the flow into the impeller eye at the reduced flow. However, you would read the NPSHr curve as the head requirement for the duty /flow.

I think you will also find this is true for all centifugal pumps - it is not usually shown due to the fact it is not normally tested for standard production pumps and, in most cases is really unnecessary as well as misleading due to the fact that operating in the region of reduced flow introduces other problems not associated with NPSHr/a.



Naresuan University
Phitsanulok
Thailand

 

[Tmoose]

I've been around a few pumps that were mighty noisy, with gravel-y banging like severe cavitation when running with a severely restricted discharge (for required system testing). Mostly pumps with a few blades.

 

[Artisi]

Tmoose (Mechanical) 22 Feb 06 22:46
I've been around a few pumps that were mighty noisy, with gravel-y banging like severe cavitation when running with a severely restricted discharge (for required system testing). Mostly pumps with a few blades.


What you are hearing is the noise generated by the mis-match of the inlet flow onto the impeller blades -not cavitation. Coupled to this could be internal recirculation which introduces other problems including noise and localised cavitation in some cases.


Naresuan University
Phitsanulok
Thailand

 

[ekschwab]

There are established formulas for determining how long it will take for the fluid to boil in a dead headed centrifugal pump. It is a function of fluid specific heat, power input, mass of the fluid in the pump, etc. I don't know how real-world the formulas are but they are out there.

By the way, cavitation is the formation AND subsequent collapse or implosion of gas bubbles in a liquid due to dynamic action.