LPG PUMP CAVITATION PROBLEM 3

[Energeer]

I am told by field personnel that a vertical multi-stage pump (can) cavitates when pumping butane from a bullet tank, but does not cavitate when pumping propane from the same bullet tank (they are basing this on sound by the pump). This problem is worse in the winter. I understand the vapor pressure relationships between the 2 LPGs as well as the temperature releationships, but I'm stuck --- In the NPSHa calcs., the vessel pressure and vapor pressure basically cancel each other out. The only factors that influence NPSHa are then static head (same for butane and propane)less the frictional losses (very close for butane and propane). Why is this occuring? What am I missing? Thank you in advance.

 

[checman]

Are the flow rates the exactly the same? The NPSHA maybe very close to each other but the NPSHR maybe greater for the propane service depending on where or how fast you are discharging it. Cavitation sounds may get worse in the winter because the delta T may be greater if so then the energy involved would be greater.

Regards checman

 

[checman]

I meant the NPSHR of the butane pump may be more

 

[Energeer]

Thanks checman. The same pump is being used at the same flow rates for both butane and propane.

 

[WebWizz]

Believe you are on the right track with your calculations. A good analysis of the suction and discharge system and matching the head with the pump curve will give you the flowrate.
From your description I would look expect the temperature of the fluid to be different, after all the main difference in the physical properties is the vapour pressure. Is there a temperature gauge on the tank?

From your writing I understand you already compared the vapour pressure graphs. (I did the same using

http://www.engineeringpage.com/calculators/properties/pure_fluids.html

for fluid names propane and butane).

If all other data really turn out to be the same it could perhaps also be the difference in the heat of vaporisation (data on same webpage as above). The heat of vaporisation is higher for propane (426 kJ/kg) than for butane (366 kJ/kg). If it takes more heat to evaporate one would expect it to evaporate more difficult resulting in less cavitation.

To solve it you could place the pump lower or reduce the pressure drop by a larger pipe diameter.

 

[25362]

The effect noted by Energeer is apparently a function of the volume of vapor formed at the impeller entry at a given NPSH. You may compare the saturated specific volumes of propane and butane vapors to see what I mean.

There is a chart published by the Hydraulic Institute in 1983, that allows for a reduction in NPSHR from the normally required NPSH for cold water.

This chart shows that at, say, 70 deg F, the reduction for propane would be 10 ft and for n-butane, only 3 ft.

These reductions are to be taken with a series of limitations and are considered just a safety margin. However, they clearly show that liquid propane may need a lower NPSH than butane.

 

[Energeer]

25362 - This may be the reason. I searched the

http://www and

was not able to find the chart yoy reference. Any tips on how I could find it? Thanks.

 

[quark]

http://www.mcnallyinstitute.com/Charts/NPSH_reduction_chart.html

 

[Energeer]

Thank you quark. I believe that these results, at least on a relative basis, are very helpful and may very well be the reason for the problem I outlined above.

 

[quark]

Energeer,

You should also have a look at this link

http://www.mcnallyinstitute.com/12-html/12-01.html

to check how to calculate the reduction in NPSHr

25362,

If I am correct, it is the compound effect of both latent heat and vapor specific volume. Can you have a look at our old thread798-129328

 

[25362]

To quark, thanks for the reminder. That resuscitated thread was really instructive. The last link you raised is indeed useful as a guideline in applying the "NPSHR reduction".

The latent heats at saturation at, say, 300 K are not very far apart: C₃, 79.3 kcal/kg; C₄, 85.8 kcal/kg; while their respective vapor specific volumes, m³/kg, are in fact quite separate: 0.046 and 0.153.

Thus, I reckon that at equal NPSHA, and at saturation (of course, at different absolute total pressures) the effect would be mainly conditioned by the vapors' volume. Agree ?

 

[quark]

I know you have something up your sleeve, but it is fun (besides knowledge)for me to engage you in discussion. Generally, it ends up with a coup-de-grace (of a brilliant derivation or some basic data which I tend to overlook), yet maintaining your calm and patient nature.

PS: The question mark is redundant

 

[25362]

It is relevant to notice that the pressures at suction differ, being near 10 atm for propane and less than 3 atm for butane. That alone would serve to give an idea why the generated volume of butane vapors is greater than that of propane vapors at equal temperatures and comparable evaporation enthalpies.

As usual quark's explanations are "comme il faut" and unerring, and there isn't much to be added.

Vapor-filled cavities appear due to energy transfer to the liquid, and as quark rightly says, by supplying enthalpy of vaporization, and as a direct result of higher temperatures or lower pressures, or both.

In my example above, based on thermodynamic properties at equal temperatures, the same heat transferred at boiling conditions would produce 3 times more vapor volume for n-butane than for propane.

 

[Montemayor]

Quark:

I still remember that old thread you reference and what I stated then: "This thread should go down in the archives as one to refer to when the subject of pumping saturated liquids comes up again – as I’m sure it will."

I knew then that a lot of brain power went into developing that thread and that it represented a lot of know-how and sharp engineering input. Why doesn't someone convert it into an FAQ? Is RMW reading this?