Low viscosity impact to CV of control valve 2

[Nickchen007]

I am using the CV curve from control valve vendor to calculate the liquid flowrate, I found the actual flowrate is higher than predicted.
I know the CV value of control valve was generated with water, for viscosity fluid, a factor should be used for correction, and actual CV will less than calculated CV.
The service I have calculated is lower viscosity (0.2cP), does it means I will gain a higher CV than calculated. How to correct the CV value of control valve for a very low viscosity fluid?

 

[Latexman]

No, I've never seen this documented in the literature.

Good luck,
Latexman

 

[DRWeig]

Here are some treatises that deal with viscosity correction.

There's a nomograph in the Emerson literature.

Emerson
Siemens
Womack

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[Latexman]

Like all the literature I recall, those do not address liquid viscosities < 1 cP or 1 cSt. They do address "viscous liquids" (viscosity > 1 cP) very nicely though.

Liquid viscosities < 1 cP seem to be a "non-issue". The flow is becoming a little more turbulant. No change in flow regime. Whereas liquid viscosities > 1 cP, and definately >> 1 cP, are heading into a different flow regime, laminar flow. So, large changes are to be expected.

Good luck,
Latexman

 

[DRWeig]

Thanks for that word, Latexman -- I didn't read the OP thoroughly I think. Interesting...

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[Nickchen007]

Thanks Latexman!
I have calculate the friction loss of the fluid (viscosity=0.2cst) for pipeline, it was 6% lower than fluid with 1cst. I suspect a similar thing may happen to a control valve, which mean lower pressure drop for low viscosity (higher CV).
What do you think?

 

[moltenmetal]

When flow through the valve trim is turbulent, the Cv of a valve is a constant independent of viscosity. It's only when the flow through the valve trim slips through transitonal to laminar (i.e. as viscosity increases) that viscosity starts to matter, and it's under those conditions that the viscosity corrections apply.

For small pressure drops, you can use the Cv value to reasonably estimate the performance of the valve with a gas, much less with a slightly less viscous liquid.

By the way, if you end up with a reduced port globe valve or worse still a needle-type control valve with a viscous liquid, don't put too much stock in the viscosity correction. Even the ones who publish values for the geometry factors (Cd etc.) necessary to calculate the ISA control valve viscosity correction Fr, often haven't based those numbers on actual testing using viscous liquids. In our experience, the results of the viscosity correction are frequently wrong, and sometimes by many trim sizes.

 

[Latexman]

Seems like a reasonable approach.

Good luck,
Latexman

 

[StoneCold]

If you are really worried about it you can go with a Samson Control Valve, They have interchangeable seats and trims. Then if you miss your size, you can change the seat and trim and be on your way with very little trouble.

Standard Disclaimer, I don't work for Samson, I have just bought some of their valves.

Regards
StoneCold

 

[moltenmetal]

StoneCold: Samson aren't the only ones- pretty much all the small control valve manufacturers sell valves with interchangeable trimsets, as do many larger control valves. We have a box full of expensive interchangeable trims for small control valves from various manufacturers arising from the "experimental" sizing approach to laminar and transitional flow, but that got tiresome. We've now got our own sizing method based on our own measurements which works seamlessly from turbulent through transitional into fully laminar flow in the trim, for geometrically similar trims. We like that better, and may share it with others some day when we get around to it. The control valve guys don't really care to do these measurements- they are happy to sell trims.

 

[Latexman]

In a chemical complex/plant, CV purchases are usually limited to certain make(s) and models. The instrument technicians that work in the plant usually carry the interchangeable trimsets that goes with these certain make(s) and models. Isn't "standardization" wonderful!

Good luck,
Latexman