Water hammer analysis for high pressure - low flow water network

[melkordy]

Dear All,

I am designing a hydropneumatic tank (air vessel) to protect a water distribution network from pump trip (or power failure). The design operating point of the pumps is 150m head and only 25 lit/sec flow. This is mainly because of the hilly topography of the area and the low demands.
Transient model runs states that using a hydropneumatic tank at the pump station with initial air volume of 4 m3 is enough to protect the system from negative pressures. What concerns me is that the final air volume through the simulation is only 4.2 m3!! this means that the total tank volume will be 5 m3 with 4m3 of it is air. This sounds weird to the contractor and to me of course.

Do you have any special concerns on such system with high head and low flow.. any help would be highly appreciated.

Best regards
Mohamed Aly ElKordy, M.Sc.
Resources and Environment Dept.
Dar Al-handasah (Shair and partners)

 

[Lan123]

I usually make the tank alteast half full with air before the surge event

 

[micalbrch]

4 m³ of air in the 5 m³ tank at atmospheric conditions or while the pump is running against 150 m head?

 

[BigInch]

Could be very reasonable, given a small volume low capacity and low velocity system. An equally relatively small released volume would thereby give an overly proportional larger pressure reduction than might be normally expected.

"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

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

http://virtualpipeline.spaces.live.com

 

[stanier]

Prof ARD Thorley gives an exhaustive explanation of sizing such vessels in his book Fluid Transients in Pipeline Systems. Suggest you invest in the book and do some manual calculations to check your software output. Alternatively you could have another analyst run the same model using different software and then compare the results.

 

[melkordy]

That you all for your precious and immediate replies.
The 4 m³ of air in the 5 m³ tank is when the pump is running against 150 m head. I have requested another run using AFT impulse to compare its results with my software (Bentley Hammer) and am waiting for the results. I believe that they will lead to the same results. However, I would be more comfortable to increase the tank volume to at least 6 m3 so that it will be only 1/3 full at steady state.. This could be returned to the ease of operation and control of the tank under other flow conditions. What do you think of that? I don't want to have the client yelling of an unnecisatedly high tank volume.
Thank you very much for listening

 

[Lan123]

Client can think of anything but you as the designer should be comfortable with your design. 1/3 full looks reasonable to me. How complex (# of pumps, length, # of nodes? etc) is your pipe line. How many high points do you have in the system? Is it a single line or has several branches.

 

[micalbrch]

What will happen with the 4 m³ air pocket when the pump is stopped? It will expand to 15 times the volume (assuming atmosperic pressure in the system when the pump is not running). Will this create a problem for your system?