Water Hammer Analysis for a forcemain that rises and then runs flat

[m_elkordy]

Dear All,

I am performing water hammer analysis for an 18km long forcmain that transmits raw water from a river to a water treatment plant. The forcemain is 2 x DN1200mm pipes in parallel and flow is about 4.6 m3/sec.

Starting from the pump station at the lowest point (elevation 8m), the line steadily rises for the first 3 km to an elevation of 140m, and then it continuous perfectly flat for the remaining 15 km till the destination with elevation of 140m also. The line discharges at the destination to a tank that is only 10m higher than the pipe (water level 150m).

I think the protection devices would be:
1. A hydropneumatic tank at the pump station
2. One way surge tank at the high point at station 3km.
3. Precautions should be taken to make sure the vertical pipe entering the downstream tank will not get empty during transient after power failure.

The proposed solutions above are not enough to protect the line from negative pressures following pump power failure especially for the horizontal reach. More protection devices are needed along the 15km horizontal reach. However I don't want to propose rediculous solution with multiple stand pipes and/or one way surge tanks along the line in the horizontal reach as these will not be accepted aesthetically by the client. We have also investigated options to break the line by an elevated tank at station 3km such that the first 3km section operates by pumps and the remaining 15km operate by gravity, however the option is rejected by the client too due to high cost of the elevated tank.

Any clues and tips on how to deal with such problematic profile? I will really appreciate any help on this matter..

Thanks in advance
Mohamed Elkordy, PhD

 

[LittleInch]

I'm a bit confused here as to whether you've actually done the analysis yet and need to make some protection or just thinking ahead.

It is usually best to try various solutions to avoid the surge in the first place - slower closing valves, flywheel on the pump, that sort of thing.

If 4.6m3/sec is equally divided then I get that to be about 2m/sec - not a large velocity. For the system you describe where there is a sizeable static element to the overall pressure drop (some data would be useful here on frictional losses) then the issue of loss of pump power shouldn't be seen on the flat section.

Some more data and graphs would be good to see. also advise what assumptions you're making - some might be seen as overly conservative.

What program are you using to simulate this?

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

 

[m_elkordy]

Hi LittleInch,

Thank you for your reply. I am using Bentley Hammer for the Analysis. At the time I wrote the original post I was just thinking ahead, but now I have performed some preliminary analysis and came up with 2 options for the protection devices and a third option that involves a change in the design. The attached PDF files explains a draft for the proposed solutions along with hand sketshes of how various tanks and deceeration chambers should look like to achieve their hydraulic purpose. The flow if 4.6m3/sec and the velocity is around 2m/s as you have mentioned.

I would appreciate if you take a look and tell me what do you think. I haven't presented the solutions to the client yet, however I think some options might look too much or exaggerated. I would really appreciate if someone can propose something different. May be there is an idea that is missing and could provide huge cost savings.

Best Regards,
Mohamed Elkordy, PhD