Rising main

[SMIAH]

Im am having an argumentation with a colleague about 2 options for a discharge line.

Option 1 = actual design (discharge line ending up at the hill point, then gravitary).

Option 2 = design modified (discharge line ending up to the downstream point).

He's pretending that the pumping flow rate would be higher with option 2 than it was with option 1 (using the designed pump for option 1).

I think that he's taking in account a vacuum effect that would form after the hill point. How could the flow rate be higher in option 2 than 1 using the same pump?!

 

[LittleInch]

A bit difficult to tell without knowing the discharge head of the pump (10m, 100m,100m?)and the actual lengths, but what I assume here is that option 1 pumps to the top of the hill then flows out of the pipe at atmospheric pressure then down the hill?

The hydraulic gradient from pumping station to outlet at 145.75 m is the key. If this is less than the hydraulic gradient from discharge point to outlet option 2 ( red dashed line) then it may be slightly more flow because the pressure at the high point would be below atmospheric and hence get you a little bit more flow.

However if the opposite is true (blue dashed line)then you will get less flow because there is a higher total resistance.

Of course I could have misinterpreted the arrangement....

My motto: Learn something new every day

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

 

[cvg]

most pumps will pump higher discharge with lower head. option 1 may have a higher TDH, so theoretically he may be correct. you would have to run the numbers to know for sure.

 

[LittleInch]

Forgot about that - if it's a centrifugal pump then maybe, but a PD pump - No

My motto: Learn something new every day

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

 

[SMIAH]

Thanks! It was supposed to be 16,9 l/s for TDH = 8,2 m

Pump to option 1 = 468 m of HDPE 150 mm
Pump to option 2 = 626 m of HDPE 150 mm (with an Air/Vacuum Valve).

Pump curve is attached.

 

[LittleInch]

At that pump head then your colleague is correct, but only if you delete the vacuum valve as this is what is giving you a couple of metres extra, i.e. the fluid d/s of the high point can flow faster than the flow feeding it and hence reduces in pressure to below atmospheric as it flows away. Normally a couple of metres head makes no real difference, but here it does.

You are then operating in slack flow conditions which can create problems, but for a 150mm PE pipe with an open end these are not important.

how much more flow is difficult to judge, because as flow increases your pump head decreases and any minor thing can have an important effect, but you could get maybe another 1-15% more flow, but as said, only if you delete the vacuum valve.

My motto: Learn something new every day

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

 

[SMIAH]

Could we use another pump with less head for the same flow? What we want is 16,9 l/s and not... 24 l/s or so.

 

[LittleInch]

Yes you could, or just install the vacuum valve or fit a PD pump of some sort and run at a fixed speed or fit a control valve on the outlet of the pump and waste a few cents per litre.

many options, but your system is very sensitive to even small variance from predicted flow to real life. One metre extra or less head can make a lot of difference to the flow for your system - most systems are not that sensitive. If you want 16.9l/sec +/- 2% then you need to add some sort of variable control (pump speed or control valve)

My motto: Learn something new every day

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

 

[bimr]

Your colleague is correct, but you probably will need a larger pump for more velocity.

Without knowing the details of your project, you typically have to pump at a higher velocity going downhill because of the velocity of the air bubbles that are moving backwards. Otherwise you will end up with an air bound pipe or two phase flow.

You don't have to worry about this if the flow is gravity as in Option 1 and you can pump at a lesser rate going uphill.

There is a discussion of this in Sanks' Pump Station Design.

http://www.amazon.com/Pumping-Station-Design-3rd-Edition/dp/1856175138

 

[LittleInch]

At 16.9 l/s in a 150mm PE pipe you have a fluid velocity of around 1.2m/sec. At that sort of slope (5 m in 150m on the downhill side) I really don't think you have an issue with velocity. though it it true to say that high points in this sort of line are generally to be avoided if you can.

So long as the line is full and assuming an NRV on the pump discharge, the pressure just d/s of the high point at rest should be about 5m absolute. This isn't enough to pull a vacuum, but some dissolved gas might come out. However on re-start this will just move along the line providing you don't have a vacuum relief valve at the top at which point some of the water will just flow away...

My motto: Learn something new every day

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

 

[bimr]

The siphon principle is valid when liquid flow is free from air and vapors so that the densities of the liquid in the rising and falling pipes are alike. It is this principle that can limit the height of the siphon effect because the flowing liquid can vaporize if its temperature gets too high or the pressure in the pipe gets too low. A water siphon is limited to somewhere between 26 and 28 feet.

With only 5 meters of elevation, the siphon effect should not make much difference on the pump unless the pipeline is very short.

You typically don't want a vacuum release unless the vacuum release is necessary to protect the pipe from collapsing. At 5 meters elevation, you do not need a vacuum release.

 

[LittleInch]

I think we agree!

My motto: Learn something new every day

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

 

[SMIAH]

Last time I did one pumping station like this, I did so the pressure is always near or above 14,7 psi. Was I right? I didn't want to rely on a partial vacuum effect somewhere.
The pump is already in order... I just didn't want to be too far from what we need (16,9 l/s) or we'll have to work with the water levels in the pumping station (start/stop).

 

[LittleInch]

14.7 psia, yes.

As said, with such low heads, if you get within 10% of your calculated flow you're going to be doing quite well so if you need to control it more accurately once installed, include some sort of flow control at the pump end.

My motto: Learn something new every day

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

 

[bimr]

The 626 m of HDPE 150 mm has a headloss of 3.4 meters at the desired flow so it should work.

The psi should be gage pressure. 14.7 psig.