best pump for suction lift 2

[EVEN01]

I was wondering if any body could lead me to a better understanding of suction lift. I've been able to learn a few concepts from the Cameron Hydraulic Data book, but from what I've seen, the use of a set of centrifugal pumps in parallel we have at the plant is not ideal for this kind of application. Any leads?

 

[CH5OH]

don't quit understand the question...
suction lift or suction head is the amount of height the pump has to overcome to get the liquid in the pump.
it isn't an application as such
running two centrifugal pumps parallel can cause some problem, one pump choking the other one

 

[Artisi]

I also don't quiet understand your question, are you asking about multiple pumps on the same inlet pipework?

There is nothing wrong with multiple pumps on suction lift on a common inlet system.

However, saying that there are important considerations to take into account.
You would treat each pump as an individual unit for NPSHr/a requirements but the inlet pipe must be considered for the sum of the installed units as far as head loss and the effect on the NPSHa of each unit.

Isolation of each pump from the other is necessary in the event that one or more pump/s is/are not operating so that the inlet systen is not open to any non operating pump/s, this means a NRV of some type on either on the inlet or the discharge of each pump.

You need to carefully review the effect of the inlet pipework on each of the units so as not to introduce inlet problems, ie, pre-rotation, flow favouring any particular pump on the system under all operating conditions, air trapped in any dead ends of the pipework etc.

 

[impeller1]

If your pumps are drawing fluid from a tank at atmospheric or +ve pressure and the level of the liquid in the tank is below the impeller centreline of the pump or if it is a sealed tank in which there is a vacuum, then the pumps will be operating with a suction lift. This would affect the NPSH requirements of the pump. You can use these pumps in parallel once their suction pipes are not very close to each other or close to any wall or partition.

 

[EVEN01]

Sorry if I misled you all with my explanation. I have a set of three parallel pumps that draw water from an open pit. Ideally we'd like to have two pumps running and one on stand by but with current conditions all three pumps are running all the time. Now, we are undergoing an expansion and I've been trying to determine if I'll be better off with a set of vertical pumps and replace the centrifugal ones or add a fourth pump.

 

[cvg]

won't your "vertical" pumps be centrifugal? Your vertical pump bowls should be set below the low water elevation in your pit.

 

[impeller1]

Are your current pumps horizontal pumps? If they can match your pressure and flow requirements for expansion as is, i see no reason why you should replace them..and if you do need more flow, add a fourth pump once you have sufficient room to do so.

 

[Artisi]

Too many unknowns to make any coherent comments.

Are the pumps surface mounted horizontal, with seperate or a common inlet line and seperate or common discharge line?

How many duty pumps do you want running after the upgrade 3 or 4?

Think it might be a good idea if you think thru' what you have now, what you are trying to achieve and some reasoning to your thoughts, this way people can make some informed suggestions.

 

[lonovane]

Horizontal pumps are perfectly capable of working on suction lift aslong as NPSH required is less than NPSH available. When two pumps are running in parallel and one pump fails, the operating pump will intersect the system curve at a higher flow called the run-out flow. Generally, you need to check your suction lift condition and NPSH available calculations to make sure that sufficient NPSH margin is available at the run out flow. Horizontal pumps are much less sensitive to sump disturbances compared to vertical pumps.So, I suggest that this analysis be done before replacement. If one knows the duty, speed and suction lift - a more useful suggestion can be attempted.

Lonovane

 

[EVEN01]

thank you all for the suggestions, I checked the specs for this system and the NPSHA/NPSHR margin is only 1, which from literature reviews indicates is not enough (literature indicates at least 5). The pumps are installed in parallel, with a common header. We tried doing a velocity study but we could not find a good section of pipe to attach the instrument to. The section we ended up using turned out to be inconvenient because it was a 90 and apparently, with high levels of turbulence. This is the reason why I am studying the possibility of adding vertical pumps to replace the horizontal ones we have in service.

Regards