Effect of Valve Throttling on Upstream Pump Suction Flow 2

[vpl]

I need some help with probably a very basic fluid flow problem. I'm reviewing a flow model evaluating the effect of adjusting a control (butterfly) valve. The valve changes from about 34% open to 40%. It is located fairly far down the line (there's a half dozen valves and a lot of piping between it and the pump.)

The flow model shows the flow going into the suction of the pump increasing by about 1% for every % change in the valve (i.e., the valve goes from 34% to 36%, the pump suction flow increases by by 2%.) This is in addition to the increased flow through the valve. I thought that in order for the suction flow to go up, the pump would have to change its operating point - which would imply a fairly big delta-p change. Since the valve's the same, would opening it a relatively small amount have that big a perturbation?

I've read through my Crane (#410), and don't see this addressed, but I could be missing it. Any help would be appreciated. Patricia Lougheed

 

[vanstoja]

It's not clear from your description whether or not the butterfly valve is principally controlling pump flowrate or is acting as a vernier flow adjustment with some other valve(s) controlling the pump operating resistance characteristic. You mention 6 other valves upstream of the butterfly valve. What are they doing and how wide are they open? Where is the pump operating with respect to its best efficiency flowrate when the correlation is made between pump suction flow and butterfly valve flow? The relationship between pump flow and head change depends on the slope of the head-flow curve which varies from low to high flowrates. More details are needed to avoid a "garbage in - garbage out" solution to your problem. vanstoja

 

[TD2K]

Patricia, I suspect you have a fairly flat pump curve. If the curve is fairly flat, a relatively small changes in head will produce (or result) in large changes in flow rate. You need to take a look at the entire piping system to see if increasing the valve opening by this amount would change the system resistance curve enough to result in the flow change you are seeing. I'd be happy to take a look at this in more detail if you want, let me know.

 

[vpl]

The valve I'm looking at is controlling flow into a heat exchanger. The other valves mentioned are also controlling flow into or out of heat exchangers. They won't be changing position when this valve's position is changed. My understanding of the system is that the pumps are running basically full out - the system is underdesigned and is having trouble meeting minimum flow requirements. However, the questions that you posed are very helpful. I'm going to talk to the guy who did the calcs tomorrow, and will ask him to answer them, or at least explain why the flow is changing to the degree it is. Appreciate the shove in the right direction. Patricia Lougheed

 

[TD2K]

Feel free to drop me a sketch of the system at testdog2000@yahoo.com or fax me at 508-448-8536 if you like Patricia.