6 replies to this topic

[anthonydsa]

I have a problem and need technical opinions. I have a screw pump requiring NPSHr as -0.4 kg/cm2. The pump is adjacent to the oil tank and is around 1.5 meters lower than the normal oil level. The suction pipe rises to the tank top and has length of 2 meters then goes horizontal for around 800 mm before coming down to the pump located 1.5 meters below. Now during running of the pump, will the negative head and positive head cancel out each other?

[breizh]

Hi ,
For better understanding can you draw a sketch and upload it ?
What is the unit for NPSHr ? should be meter of liquid .

Breizh

[pavanayi]

Anthony,

Firstly, as mentioned by breizh, NPSH (both required and available) is normally stated in meters of liquid.

Secondly, the static liquid head contribution (positive or negative) for calculating the NPSH is taken as the difference in height between liquid level in the tank and the suction eye of the pump. It does not affect if the pipe is looping up and down.

Thirdly, what will affect due to pipe layout is the contribution of the frictional loss within the pipe.

Edited by pavanayi, 14 December 2011 - 03:00 AM.

[kkala]

Screw pumps are self priming http://www.electronaval.gr/en/?p=1011, but to what extent? A communication with pump supplier, along with a configuration sketch and fluid properties / vapor pressure, could clarify the question.
As pointed out by breizh, NPSHR value creates doubts. Probably mentioned -0.4 kg/cm2 g is minimum required suction pressure for the pump to be self primed?
Note: Below concerns non self priming pumps, e.g. centrifugal pumps.
Actually NPSH is not practically affected by loops at suction line (apart from some increase in frictional ΔP due to loops).
But if a suction loop is high enough to exceed tank liquid level, this high point of suction line will not be full of liquid when the pump is not operating. So the pump will not be a self priming case. You will have to insert liquid into suction line to start the pump up. Once primed, the pump can have smooth operation (for heights as stated), but after a break (intentional or unintentional) it has to be primed again.
I spend more than half an hour (~1977) trying to prime a cooling tower pump, even if the suction line ended to a simple non return valve to theoretically help filling the line. Priming may be easier for lines of smaller diameter though. But try to locate all suction line below LLLL of the tank, if you can. It is worth while.

Edited by kkala, 14 December 2011 - 12:27 PM.

[ankur2061]

anthonydsa,

To know more details about piping installation for screw pumps refer the link provided below which exclusively deals with screw pumps:

http://85.185.231.19...REW%20PUMPS.pdf

The above link provides the chapter on screw pumps from the famous "Pump Handbook" by J Karassik.

Additionally, another link provides very good general guidelines on pump suction piping:

http://www.pdhengine...es/m/M-3035.pdf

Regards,
Ankur.

Edited by ankur2061, 14 December 2011 - 02:36 PM.

[anthonydsa]

Hi everyone,

Thanx for the replies. Breizh, am attaching the sketch for better appreciation. One more thing, will NPSHa have different values in static condition & dynamic condition i.e. with pump stopped & pump running?????

Anthony D

Attached Files

•  Sketch to upload.JPG   51.88KB   39 downloads

[GS81Process]

The NPSHa will be different under static and dynamic conditions, because under dynamic conditions (i.e. flowing oil) there will be frictional pressure drop. However, I have to ask why you care at all about "static" NPSHa. The pump will not have an NPSHr if it is not running. Static NPSHa is not a concern.

I do not think that your suction piping configuration is very good. Why is there an intermediate high point between the tank the screw pump? Is there a vent on this high point? Otherwise how would you get rid of air?

Why do you suspect cavitation? I suspect the oil has a very low vapour pressure. Is there a chance that you are experiencing air lock in the pump suction?

To clarify one point brought up earlier: NPSHr for positive displacement (PD) pumps is typically specified in pressure and not height of liquid because the operation of PD pumps is not affected by liquid density. It is a constant pumping volume operation.

One other interesting note: the NPSHr of positive displacement pumps is greatly affected by the acceleration of the fluid.

Edited by GS81Process, 21 December 2011 - 04:54 AM.