Suction lift

[jack880]

Hi. I'm designing a system using a Grundfos JP5 suction lift pump. It can be used to lift up to about 5. The suction pipe has to rise out of the reservoir and fall back down to the pump as shown in the sketch. I know that the lift the pumps needs to deal with is 2m + pipe friction losses, but how do I take the 2m height of the hump into account? I appreciate the issues I may have with keeping the pipe primed etc, but assuming the pipe is full of water when the pump is running, I want to know the suction lift that the pump has to be able to deal with.

Thanks very much

 

[JJPellin]

Using the assumptions as you stated, the pump is not affected by the 2m "hump". The head loss up to the top of the high point in the suction line is recovered fully on the line down from that high point to the pump. The pump will see a suction pressure of negative two meters, gauge.

In order to keep the high point of the suction line primed, I would recommend that you consider an automatic vacuum pump that would kick in to evacuate any vapor. I have seen these as standard equipment on portable rental pumps that we used for temporary cooling water applications when the pump had to lift from the cooling tower sump down below. Unfortunately, I don't recall the manufacturer of that little vacuum pump or the supplier of the big rental pump.

Johnny Pellin

 

[jack880]

Hi Johnny. Thanks very much for the reply. Presumably though there is a limit? For example, what would happen if the hump was 100m high? What determines whether or not the water can make it up the vertical bit at all, regardless of the head being recovered on the way down?

 

[micalbrch]

jack880: The max. geodetic suction lift a pump can do is (in theory) 9.81 m.

Johnny, was it perhaps a Cornell pump you saw? The have big pumps with a small diaphragm pump directly fitted on the suction side.

 

[FACS]

How does the 4m of pipe get filled in the first place? if the pump has to do it, then it needs to be able to create the vacuum required. If you are manually going to prime the pump and the pipe, then the operating lift only needs to be the 2m (plus friction loss).

I woild say 4m of lift plus friction loss.



Charlie

http://www.facsco.com

 

[JohnGP]

I'm with Johnny Pellin, once the pipe is primed, the static suction lift for the pump is only 2m.
Cheers,
John

 

[BigInch]

Sorry guys. The pump is indirectly concerned with the hump. Strictly for NPSHR what you are saying is true, once there is water in the downcomer, the pump doesn't care, however you must consider the highest point in ANY part of the suction piping as the initial suction lift, otherwise you will never get water into the downcomer. If the lift is less than about 31 feet, its theoretically possible that a proper pump for the application could prime itself, however at dimensions over much less than that, problems with some slow self-primers could be encountered. And that applies to other than first-time operation, since it is entirely possible and probable that the suction line segment containing any trapped liquid is pumped out at some time during operation, or that vapor of some kind will get trapped there when the pump is eventually turned off. If you exceed fluid vapor pressure at any point in the suction piping, there is a very good chance that after that, any liquid trapped there will be pumped out and all you'll get is nothing but pure vapors.

"The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward X-CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[FACS]

I agree with BigInch.


Charlie

http://www.facsco.com

 

[JohnGP]

Yes, all good, but I was going with the OP's assumptions (rightly or wrongly).

 

[BigInch]

Yes, but since when is the OP's question important. It won't likely remain full of water. And I would imagine that somebody would be really PO'd if he designed it like that, especially considering that it only needs a relatively small 2 meters to easily eliminate that potential problem entirely.

"The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward X-CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[Artisi]

The answer to the OP's question is yes,you have to account for the 4 metre lift but only if concerned with initial self-priming capabilities which in this case isn't a problem as the pump is capable of 5 metres lift (according to the OP). Once primed and operating the 2 metre "hump" is cancelled from the lift consideration and the suction lift becomes 2 metres plus any friction losses.

Now whether the pump can operate within the 'system' in terms of initial prime, re-prime etc is another issue to which we do not have enough information to pass judgement.

 

[rmw]

Be careful with your pump intakes. With this arrangement, if there is any vortexing at all, you will get air in the suction "hump" and either cause you to lose prime or build up a air pocket that the pump could "swallow" on a sudden flow change. Be sure to get the right.

rmw

 

[JohnGP]

We're (well not me exactly) about to commission some pumps in a similar configuration, so I'll see first hand how they go......

 

[jack880]

Guys, thanks loads for your replies, this is an incredibly useful forum! I appreciate the issues with priming the pump and air etc - which brings me on to my next question.

If the foot valve leaks a bit when the pump isn't running and air gets into the pipe and collects at the top, firstly is there a way i can assess how likely it is to either accumulate there or get sucked into the pump? The pipe internal diameter, flow rate and speed will be either 32 mm - 0.65 l/s - 0.8 m/s or 22 mm - 0.35 l/s - 0.92 m/s.

Secondly, what sort of valve/device would I used to get rid of trapped air?

Thanks again - maybe one day I'll know enough to give people advice!

 

[micalbrch]

You will not get air in through the foot valve as it is below liquid surface.

It is very difficult to find suitable valves to get the air out. There are some de-aeration valves available in the market (I know that company Grundfos has them) but they are not really performing well, especially not when the pump is operating. You need a pressure difference between pipe and atmosphere to get the air out. That means the pressure in the pipe must be higher than the atmospheric pressure which is unlikely the case in a pump's suction line.

 

[BigInch]

With a negative pressure in the riser, it's possible that air can leak back in through a poor piping connection and even the pump, from seals, or if the discharge line is not full of water.

"The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward X-CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com