Suction Pump Submergence

[dicksewerrat]

I am working on a problem with 'vortexes' forming in a MH . there are 2 18 inch HDPE suction lines to 2 different pumps. They are moving 5300 gpm down 2 18 inch bypass lines. The suction lift is 18 feet. Submergence is 5 feet. I have a graph to pull minimum submergence from and a calculator for the same. the graph is based on velocity and calculator is based on inside Dia. and Q. Would separating the suction lines help? Or do we need to increase submergence ? And do you guys have any other suggestions. Thanks

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[Artisi]

Vortex breakers ahead of the inlets may help, moving the inlets apart may help as might a bit more submergance. All maybe's - a careful study of the installation is really needed, careful observation, running each pump separately etc. Where / how are the pumps installed - photo's or a sketch could throw a bit of light on the subject.
The last vortex problem I was involved with was large pumps (3000kW)consisting of 2 units installed in each forebay, one behind the other all in accordance with good practice and theory, after much hand wringing and hair pulling the problem was finally given to a hydraulic lab to analyse - to cut a long story short, the solution was to straighten the flow entering the forebay - this overcome the problem. This highlights the point that the inlet pipe / suction bell / submergence / location isn't always the cause of vortex formation.


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.)

 

[dicksewerrat]

Here is a sketch of the MH and Pumps.We are at 18 feet to water level now and are hoping to go to 14 feet to reduce suction lift and increase submergence.

Richard A. Cornelius, P.E.
WWW.amlinereast.com

 

[bimr]

With the limited space in the manhole, about all you can do is to deepen the water level.

The Hydraulic Institute has an equation for the vortex depth:

http://www.pumpfundamentals.com/help11.html

 

[dicksewerrat]

Thanks bimr. I found that one in pump website. That one gave us a submergence about what we had. I think we were right at the limit. We are going to increase the submergence and add a pump to reduce the velocity in the suction pipes. We thought about some kind of baffle but with 3 18 inch pipes in the manhole there is no room for one. Maybe the third pipe will act as one. too. Thanks for the info guys.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[Artisi]

Are the inlet pipes fitted with anti- rotation devices?

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.)

 

[dicksewerrat]

They are HDPE SDR17, butt-welded to the 90 above ground. Why? And who offers a decent pump online class? or a 1-2 day class south of Washington, DC. I am getting more and more bypass stuff to do. i should go to school.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[Artisi]

Why??? Do you wish to stop pre-rotation / vortexing??

https://www.bing.com/search?q=pump+...id=7b5d806fff2c4afcba7cd1f3f73cfa3b&FORM=QBRE

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.)

 

[Artisi]

Foregoing what has already been posted, can you physically see any vortex on the sump water surface and what if any is the effect on the pump/s and is it the result of any vortex?

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.)

 

[dicksewerrat]

You can see the vortex and it is causing cavitation in the pumps. At this time we are going add a third pump throttle them all back. We will be lowering the intake velocity to 6.8 fps from 8.7 fps. Also increase the submergence to 9 feet.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[Artisi]

Seems like an expensive exercise to eliminate vortexing when other remedies are available and worth considering/ trying. As for cavitation, not a function of inlet vortex- pumps may be noisy resuting from a mismatch of flow into the impeller eye.

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.)

 

[LittleInch]

For me you could try any of the options above or, make the bottom 3ft say 24" or bigger or drill lots of holes in the bottom 2ft to reduce velocity into the open end. Is the end actually cut at 45 degrees?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[dicksewerrat]

Artisi, The pumps are already there. There are 2 back up pumps and 2 main pumps. Just cost more for fuel. The Bypass manifold has 2 extra capped connections.
Little Inch, The pipes are cut at an angle. I will pass that on to the pump people for future reference . On this project it may be too difficult to get 3 24's into the pipe.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[LittleInch]

The other thing I've seen is to cut slots out of the end say 2feet long by maybe 2"wide spaced equally around the circumference.

There's many things to try before you stick another pump in which might not make a big difference.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

For flow of 5300gpm through an 18inch line, suction losses appear to be tolerable. However, the suction lift of 18ft leaves you with only less than 10ft of NPSHa ( pump strainer losses not included), so what is the NPSHr for this pump at max flow? If the strainer burns up say 3ft of head, that would leave you will only 7ft or less for NPSHa, and the pump NPSHr would need to be less than this.

Agreed, a vortex breaker of some sort should be fitted on each suction line also. The submergence will help with reducing vortexing somewhat. As suggested, the Hydraulic Institute has sketches on intake arrangements.

Changing suction lift may not be possible in this case ? If so, suggest adding a low NPSH booster pump ahead of the main pump for each pump run.

 

[dicksewerrat]

Little inch,
Cutting the slots would reduce the velocity at the bottom of the pipe and reduce vortex forming. I will pass that along also.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[Artisi]

One other point not discussed is the actual sump design / shape etc., this can be as important - I some cases more important than anything else in the installation.

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.)

 

[LittleInch]

Artisi, as you would expect, has a very good point. I've not really understood where the 56" pipe came in and a plan view on this arrangement would be good. If there is any swirl induced by the incoming fluid, then this can act as a starter for vortexing.

As I think you know, the noise you're hearing is probably air going through the pump and not cavitating. Normally adding a bit of air reduces cavitation noise.

If you're getting cavitation without the vortex actually reaching the inlet then you're exceeding the lift capacity of the pumps.

In any event two pumps in parallel, assuming they go into the same header, are never truly equal, so one could easily be at a higher flow than the other.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

2 x 18" pipes extracting 10,000 gpm from a 2m diameter "sump" appears to be a disaster waiting to happen.

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.)

 

[dicksewerrat]

This sump is formed by blocking the pipe. We don't have the luxury of constructing a nice, big, reinforced concrete sump pit. We work for the pipe owner. They don't want to pay for that. The sump would have to be a foot lower that the bottom of the pipe, 19 feet deep by 8 foot wide and a minimum of 10 feet long. This is a 50-60 year old pipe.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com