NPSHa - One large and one small pump on same suction manifold

[Esporp]

Hi

I have come across a problem that and I don't quite know where to start.

The problem is as follows:
We have a 13 meter high tank, water level can be anywhere between 0 and 13 meters.
A small pump (3 m3/h, NPSHr = 3 meter) is placed 6.5 meters above the ground. The suction pipe is located on the suction pipe of a large pump (850 m3/h, it is all I know about this pump)1 meter in front of the pump inlet flange (yes, I know it is far from ideal). The large pump is being used to fill and empty the tank and located 2 meters above ground. The small pump is used to sample the water as the tank is being emptied, the large and the small pump is running at the same time.

Who do i solve the NPSHa for the small pump?
1. Normal NPSHa without taking into account the low pressure the large pump will create at its suction
2. Consider the suction piping for the large pump as a separate tank (for the small pump) with the pressure conditions we will see there and do NPSHa based on those values.

Or is there another way?

All help would be very much appreciated.
--Esporp

 

[JJPellin]

Determine the static pressure at the tie point where the smaller pump suction comes off of the main line. You know the tank level, fluid conditions, total flow rate, elevation change, line size, etc. Velocity at that point in the line may be significant depending on the line size. At this point, you know the static pressure at the start of the branch line leading to the smaller pump. Then, you can calculate the pressure drop from that point to the suction of the smaller pump. This does not sound like a particularly tricky system. Be careful about your units. If you start with absolute head in meters and stick with that through to the end, you will end up with NPSH available. If you work through in units of gauge pressure, then you have to convert to absolute head at the end.

Johnny Pellin

 

[Artisi]

Why not sample direct from the discharge of the large pump or do you need to sample without the big pump in operation?
Without putting too much thought into it at this stage, your suggestion 2 sounds like the correct approach, establish the pressure at the pump inlet pipe at the interface with the main inlet line (worst case) and calculate from there.

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.)

 

[Esporp]

Thanks for your replies!

Sampling from the suction of the large pump was not optimal, but necessary due to adding of other chemicals downstream. The major problem was the original placement of the sample pump. On other installation this design has worked, however then with the sample pump installed lower than the large pump.

Thanks again!

--Esporp