Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Calctd NPAH(A) is 286 feet , but data sheet value is only "20"feet 2

Status
Not open for further replies.

patrickraj

Chemical
Apr 5, 2003
87
We have a horizontal centrifugal pump for ammonia service. Pump is taking suction from a horizontal cylindrical tank located at 20.0 feet elevation(from tank bottom).

Pump suction pressure : 279.7 psiA
Temperature : 100°F
Sg : 0.58 4
Vapour pressure at 100°F : 212 psiA
Suction pipe : 6 " dia
Length : 30 Feet with a gate valve & 5 nos 90° bend.

I have calculated the NPSH(A) as 286 feet. But the data sheet reads only "20 + " feet. NPSH(R) indicated is 11.0 feet. NPSH(A) calculation details :

System pressure : 279.7x2.31/0.584 = 1106.4 feet
Vapour pressure : 212 x 2.31 / 0.584 = 838.6 feet.
Friction loss : 2.2 feet
Velocity head : 0.46 feet
NPSH(A) :Sys Press + Ele. +Vel Head – VP - Friction loss
NPSH(A) =1106.4 + 20+0.46 - 838.6 – 2.2
=286.0 feet

What could be the reason for the NPSH(A) indicated value of only "20 + " against the calculated value of 286 feet.
What is actual NPAH(A)

Thanks in advance.

 
Replies continue below

Recommended for you

Can't say for sure, best guess, which I've seen before, is that whoever filled out the data sheet just said "I've got 20 feet of actual static head, assuming the ammonia in the tank is saturated, the surface pressure equals the vapour pressure, so I can neglect that term." If they knew ahead of time that the surface pressure would actually exceed vapour pressure, then they can still safely say at least 20' NPSHA, provided they knew friction loss would be minimal.

One thing regarding your calculation method, however, is where you're taking your system pressure from. You stated 279.7 psia at pump suction, and then in your equation added the static head for the liquid level. If you're using a gauged pressure at the pump suction, your static head is already included in that pressure. The form of equation you're using, you should be starting with the surface pressure on the ammonia, not the suction pressure at the pump.
 
When the NPSHr of a pump is less than 20 ft (determined by supplier), it is OK to state the NPSHa (in the data sheet)as 20+ feet. This is enough for pump selection. Should the NPSHr needs more, supplier will tell you.
 
I'm curious about the difference between the tank pressure and the vapor pressure. Is there an inert gas in the tank because otherwise, assuming the tank is partially full with a vapor/liquid interface, the vapor pressure should be the tank pressure.

As Scipio has commented, I suspect the original designer assumed the ammonia in the tank would/could be saturated as a reasonable worse case. If you didn't, and got a pump that required say 50' of head, you'd have problems if conditions in the tank ever changed such that system pressure = vapor pressure. This way, the pump will still work.
 
The vapor pressure - temperature curve of Ammonia is very steep. Recycling ammonia for even short periods of time will considerably heat up and the suction liquid and this is a condition that will occurr on many occasions. The heated up liquid has a much higher vapor pressure and NPSHA will reduce considerably.

By placing 20 + feet on the design sheet the designer mentions that he needs a pump that has the lowest NPSHR possible so they can recycle and idle the plant as long as possible.

 
Dear friends,

It is true that presure in the tank is maintained higher at 279.7 PSIA,compared to the VP of 212 PSIA at 100 DEG C. Now i too think that the vendor would have supplied the pump for the worst condition of assuming saturated ammonia in the tank.

Any comment on the SCIPIO'S calculation method.

Thanks to all for your views.
 
patrickaj:

The vendor should have supplied the pump to what the engineers specification was and had really little to nothing to do with the system. It sounds like the vendor was not given a specification????? And someone relied on them to engineer the system specifications????

Just curious...

BobPE
 
Scipio's view is correct. Considering that ammonia pumps usually have pump feed vessels immediately before the pump. Static head would be the only thing that can be relied upon.

The data sheet value of 20 should be taken by the pump supplier for the pump selection. If The supplier wants to provide a pump with higher NPSHR he should ask for a deviation from the customer so all are aware that there is a potential NPSH problem.
 
to patrickraj, scipio is right, when the pressure reading is taken at the pump suction, neither the friction drop in the line nor the static head difference should be taken into consideration, since they are already included in the (absolute) pressure reading.
You have a typing error in your last message the temp. is in deg F not deg C.
I agree with TD2C concerning vap. pressures and system pressures. Could it be possible that the liquid ammonia has a uncondensable gas above the liquid interface, so as to consider the liquid ammonia to be "subcooled" ? Otherwise, either the temperature or the pressure readings would have to be wrong, and the VP would equal the system pressure.
Comments by tad123 and scalleke on NPSHr are right, scalleke's comments on VP vs T are important. [pipe]
 
Dear friends,

Indicated Pump suction pressure of 279.7 psiA is the tank pressure,not the pressure at the pump immediate suction.
Is my calculation still wrong and need correction.

Once again thanks to all for your views
 
So long as that's the pressure in the tank vapour space, then it looks like you've got the calculation right.
 
Dear scipio,

I too checked in few textbooks.My calculation appears to be correct.
 
Patricraj,

Please check the velocity head. If 279.7 PSIA is the tank pressure, then the velocity head is created at the expense of tank pressure + static head. Therefore it appears that this should no be included.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor