control valve sizing for extremely low flows 1

[moltenmetal]

Does anyone have good, proven sizing equations for extremely small control valves (Cv less than, say, 0.01) for liquid services, where the flow through the trim (much less the body!) is clearly laminar?

Unless I've missed something, the typical low flow control valve manufacturers (Kammer, Baumann (now Fisher), Badger Research etc.) don't give you anything useful on this subject. It's not mentioned in Fisher's otherwise very useful little black book. Viscosity corrections exist and are given but don't apply to this case. And I can find equations for laminar flow control valve sizing on the 'net, but without references as to where they're from and what data they're based on etc. I'm about as suspicious of using that sort of stuff as I am about eating food I find lying on the ground...

 

[BigInch]

If its laminar, can't you just relate to velocity head?

"Make everything as simple as possible, but not simpler." - Albert Einstein (1879-1955)
***************

http://virtualpipeline.spaces.live.com/

 

[moltenmetal]

Imagine a needle valve with an annulus between plug and seat that is so small that flow of liquids, even relatively thin ones, remains laminar while passing through the annulus- even at pressure drops of hundreds of psi. The vendor doesn't give you the annulus dimensions- they just say that the valve has a "Cv = 0.001", which of course they base on the (turbulent) flow of a gas through the same annulus. And unlike a conventional (large) globe control valve, these units are all to some extent "seat-guided"- meaning the plug extends through the seat, forming an annulus even when the valve is "fully open". So it doesn't ever simply reduce to flow through an orifice. Wouldn't matter anyway since I don't know the orifice dimensions!

Aside from experimentation (which we're doing!), is there any way to a priori know the pressure loss across this valve, such that you can select the right one for the job- based only on the turbulent flow Cv value the vendor gives you?

I can't imagine I'm the first person to have asked this question! People have been using needle-type control valves for a very, very long time...but so far, a web search turns up nothing I feel comfortable using...

 

[rcooper]

An alternative approach might be to use a small positive displacement pump fitted to a variable speed motor, such as a peristaltic pump or diaphragm pump to allow you to control the low flows.

 

[BigInch]

Moltenmetal, Maybe I don't understand. Do you have a Cv curve, or just one Cv value? What do you know abuot this valve?

"Make everything as simple as possible, but not simpler." - Albert Einstein (1879-1955)
***************

http://virtualpipeline.spaces.live.com/

 

[moltenmetal]

rcooper: we do this when we can. But you can't let pressure DOWN that way...

BigInch: single Cv value- that's all you get to choose from when you buy 'em.

 

[BigInch]

Must be Cv at full open I suppose.
Is there some reason you can't use that Cv?
Is it linear, equal percentage, or what type?
If you know the type, you can relate Cv to %open.

"Make everything as simple as possible, but not simpler." - Albert Einstein (1879-1955)
***************

http://virtualpipeline.spaces.live.com/

 

[JLSeagull]

Without knowing the flow rate and pressure drop, I wouldnt assume that the flow is laminar within the valve. I have selected Research valves with tripple zeros after the decimal. Select small valve and trim trim that accommodates trim replacement with larger or smaller trim and without affecting the valve body.

 

[moltenmetal]

OK, I'll take it that nobody has an equation to help me out right at their fingertips, which makes me feel like less of an idiot for not finding it myself...

They do exhibit reasonably good Cvs for gases, even down to the triple zero Cv size. But they're in error for viscous liquids, and the standard viscosity correction factors don't work.

 

[JLSeagull]

Calculating viscous crude oil on some large valves I found that the viscosity must be very high, near 1000 cP before the valve capacity coefficient begins to change due to viscosity.

 

[Loopee]

I have a very good book called the Control Valve Primer by Dr Hans Baumann that has a section on valve sizing with high viscosity flow and/or very low flows - even with a tapered plug as you describe- that may be helpful. It's available through ISA.

 

[moltenmetal]

loopee: thanks, I have heard of Baumann's book but haven't seen it. Will ask for it through the vendor (Fisher/Emerson own Baumann).