How an automatic flow control valve such as a Griswold affects the piping system curve

[BronYrAur]

I just want to better my understanding of pump/system curves.

Am I right in my thinking about this? If I put an automatic flow control valve (like a Griswold) into a system as a main balancing valve, it would appear as a vertical line on the system curve, correct? I threw together a chicken-scratch sketch (see attached). With no Griswold, my system curve follows a basic quadratic shape and intersects the pump curve. I pulled numbers out of the air, so let's say that point is 1,200 GPM @ 50' of head.

Now let's say I insert a Griswold that is sized for 1,000 GPM over a delta-P range of 12'-75', So, won't my system resistance curve follow the quadratic until I get to 1,000 GPM, then it will turn vertical until it intersects the pump curve? So my operating point to 1,000 GPM and let's say that corresponds to 60' of head.

So finally, if I just used a regular circuit setter valve instead of a Griswold, I would just get a new steeper quadratic curve that passes through 1,000 GPM @ 60', correct?

 

[LittleInch]

In essence yes you are correct IMHO. The point on either option is that the pump will very rapidly get to that point (1000GPM) as there is nothing stopping it.

try to think of it in terms of the pump, i.e. the pump will go until it reaches the flow the valve is set a then drops head vertically until it reaches the system curve. The pump is what makes it all go round so is the important thing.

Which one to use - the flow control valve will be much more accurate as changes in either the pump curve or the system curve can be accommodated up to a certain point whereas depending on the slope of the curves the fixed valve option could vary by probably 250 GPM in either direction unless you keep going and changing the valve

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

 

[Drazen]

you cannot tell without analyzing complete circuit. it would be great if one valve would solve all troubles.

in general, flow control valve is not designed to interact with pump directly, but to control flow of some partial circuit. shape of system curve does not mean much if there is no intersection with pump curve.

 

[EnergyProfessional]

You have to consider you have multiple of such valves. Each at various states of between open and close. Overall what you want to do is to limit the pump pressure to the maximum you need. So basically to the point where your graph goes up vertically.

I love the PICV as they eliminate all balancing and prevent excessive flow. i use the Honeywell PICV as I found they have less overall pressure drop than the Belimo and Griswold valves. i do like the Griswold pressure independent balancing valves (the PICV without the control valve part).

A side advantage of PICV is they are ball valve... much less hassle than globe valves (I don't even know how many globe valve actuators we had to replace as they all destroy themselves over time)

The system curve is a theoretical object. As soon as you introduce control valves you are constantly changing pressure. for example, according to the system curve more flow means more pressure. but as a valve throttles, more pressure will mean less flow. system curve concept assumes all valves open, and all elements to follow a quadratic relationship, which also is not correct. Valves behave differently than fittings than straigth pipe than equipment.