Modifying System Curve Utilizing Throttling Valve

[colowater]

Hi all,

I have been searching for a reasonable solution to a hurdle I have encountered on a recent project. I am designing a pump station that will be delivering a large range of flows to two seperate locations. A system of two pumps will be used to cover the entire range of flows (approximately 200 gpm to 9000 gpm). It is feasible based on intake water levels that the static head between the two systems could vary between 31 feet and 3 feet. We will need to cover the entire range of flows under both conditions. The friction losses further complicate things making our range of total dynamic head varying from nearly 80' (at 9000 gpm and 31' static head) to 3' (200 gpm and 3' of static head).

I am interested in designing a system that will provide adequate, dynamic backpressure that will essentially align the second system curve (to the low elevation outlet) with the first system curve (to the high elevation outlet). I believe that I can accomplish this using an electronically actuated throttling valve (either butterfly, globe or pinch). My confusion lies in how to adequately calculate the minor losses for varying cross sectional areas through the valve. Do the equations at the following link apply, or do they not because the pipe diameter would again increase downstream on the valve (in other words is my A2 the value through the valve, or the value just downstream of the valve)? Any other thoughts?

http://nptel.iitm.ac.in/courses/Web...CS/lecture-14/14-7_losses_sudden_contract.htm

Thanks in advance for any responses. I have been a lurker for years and have found this forum to be an excellent resource.

 

[77JQX]

That's a big range of flow and TDH for only two pumps. You have your work cut out for you. I think I'd try to design this as if it were two pump stations in one building - one with a low flow-low head pump, and the other as a high flow-high head pump, and see if I could get an acceptible hand-off between the two as flow and head changes. That way, you're not trying to make the small pump operate under high flow conditions, and vice-versa. How fast would the flow and TDH conditions change?

But to address your question: If I understand, you're proposing installing a valve to increase friction in the system to raise the low TDH system curve to the higher TDH system curve. Is this correct? And the valve closes to increase friction in response to lower flows or static head? If I understand right, you'd only need to know the CV for the fully open valve to be sure the valve doesn't add too much friction to the high flow-high TDH scenario. That full open Valve CV should come from the manufacturer, rather than a calculation. In lower flow or static head scenarios, you'd control the valve position to vary the CV all the way down to zero as your conditions dictate. Ypu wouldn't need to pre-calculate the partially open CV - you let your controls set it to whatever it needed to be. I doubt any single valve will have good control characteristics for the full range of flows. Good luck.

 

[colowater]

Thanks for the response. Unfortunately both pumps will have to be capable of delivering water to each location, under various levels of head. I'm starting to think that you may be correct that I would not be able to assign a single valve to adequatly throttle the flow. Due to the high flow requirement we are utilizing an 18" discharge pipe. This required opening to create adequate head during the low flow requirements would be miniscule. I may need to look at a discharge pipe gallery of various sizes for varying conditions. Back to the drawing board I suppose...

 

[77JQX]

For 9,000 gpm, I'd use 24" to keep head loss and energy costs down.

 

[QualityTime]

Wow, the proposed solution sounds way too complicated. Remember the KISS principle. You will not be anyone's hero when the whole thing does not work. Your rationale that you were trying to save money with this method may sound good now but I guarantee it won't sound too good to the client after it has been built and it does not work. You will be in deep shit. If it takes 1.5 years to build you will be worrying for 1.5 years HOPING that it will work. Keep the pumping systems separate. Generally speaking, if you need to PLC to do all this type of sophisticated control you will be in a lot of trouble trying to run the station in the manual mode when the PLC goes down. No one will understand how the system is supposed to run. Total waste of energy also. Remember the KISS principle. Ditch the idea

 

[stanier]

Separate systems seems the most likely way to go if not for the steady state but then to avoid transients. If you want to play big boys games you need big boys toys. It will cost to do it right.

If you go to

http://www.pumpsystemsmatter.org/content_detail.aspx?id=110

you can down load a cut down version of AFT's Fathom. Its called PSIM? This will enable you to model the various scenarios, adding valves with open% versus Cv etc. Then you can see the range of opening you will encounter. Beware of cavitation if you throttle the valve too much. Valve damage and noise issues are the concern.

You could use Epanet or any other similar software if you are more familiar.

If fluid dynamics are not your area of expertise then get someone else to do it for you.

Once you sort out the steady state you will need to do a waterhammer analysis. Again if this is not something you do regularly get someone who does. If you have to go down that path they could do the steady state evaluation as part of the exercise.

Generally I would not recommend this link to anyone who is not well up on fluid dynamics and you are an unknown quantity. It will same you time and give you a chance to do a thorough review of options.

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[QualityTime]

Just goes to show how sometimes solutions can become so convoluted and why bouncing ideas off of experienced people is so important

 

[bimr]

It would seem that you can design the pump for the larger flow, and use a clavalve type of valve to supply the lower flow and pressure for the second application.

 

[Artisi]

Just bear in mind that anything is possible however impractical even if you have unlimited funds to throw at the project.

You need to look carefully at the likely operating conditions, how often at the high flow/ low flow / intermediate flows. It could well be worth considering 1 x 50% pump and 2 x 25% pumps or 3 x 33% or some other combination, don't get hung up on a 2 pump scenario.

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.)