How to size a pump suction from top to discharge bottom?

[Nickwong]

Hello experienced engineers out there,

I am currently deciding on which pump to use or how i can size it up.

This is my setup.

1. An enclosed tank that is filled entirely with solution. Pump will have to suck from the top cover of the tank to the bottom of the tank to recirculate the system. The suction pipe cannot be submerged into the solution because of the nature of this experiment and it can only be attached to the top cover.

Assuming that the pump is on ground level and pipelines are filled with solution as well, how do i size this pump base on a negative NPSH?

2. Similar concept just that my tank is not entirely filled with solution(assuming there is 3-5 cm gap of air in the enclosed tank. Vacuum Pump is required?

Thanks!

 

[LittleInch]

There's quite a few odd things here:

1) recirculation systems normally require the system to be pressurised so that these issues are not important. As the inlet and discharge of the pump are connected via the tank the pump needs to be part of the loop under pressure so that as the fluid leaves the tank it is replaced by the same amount of fluid form the discharge. Any slight discrepancy and the pump will cavitate as it tries to flow more liquid than is coming back in and hence create lower pressure / vacuum somewhere.

2) Even a vacuum pump can't make the liquid expand to reach the impellor of the pump. From your description ( a sketch or drawing always helps) this simply will not work.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[bimr]

LittleInch is correct. There is no pump made that will work for this application.

Have you considered using a tank mixer instead of a pump?

 

[LittleInch]

Nickwong,

Thinking a bit more, in reality for option 2, your connected pipe (assuming it is full of liquid and there is a valve in place on the top cover) would simply drain back into the top of the vessel leaving the air gap at the high point of the pipe. A self priming pump mught be able to induce some sort of flow, but that air has to go somewhere and if it's straight back to the tank then flow will be poor with a lot of air bubbles in it. If the pipe is empty then there is zero chance of getting flow unless your liquid is quite expandable under vacuum like conditions.

This sounds like some sort of lab type experiment?

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[chicopee]

If an air gap is necessary in the system for whatever reason, I would connect a "T" fitting on the top of the tank cover with one end of the upper leg of the "T" connected to the tank cover and the other half of that top leg with a nipple capped so as to fill the system and leave an air gap; the bottom leg (90 degree branch) would be connected to the pump and would be filled with the solution with no air in it. I would extend the upper leg with the cap several inches so that when the pump is in operation, no air gets trapped in the inlet of the pump. Also the upper portion of the "T" with the cap will trap any air leak thru the pump seal.

 

[racookpe1978]

Can you put the pump inside the tank: Pull a suction from near the bottom, discharge a little bit down, but aimed into a corner of the tank to create a mixing effect?

Alt: take a suction from the side of the tank halfway down, discharge into the bottom.

 

[ARHeit]

I'm going to guess you are a student and this is a homework problem, the systems sound to theoretical and loosely defined with just enough information and focus.

http://www.pumpfundamentals.com/tutorial2.htm

Checkout the website above, it might have some better visuals to describe suction and head that help you sort out the problem.

"Whether you think you can or think you can't - you're right." - Henry Ford

http://www.eng-tips.com/faqs.cfm?fid=376