Ball Valve Non Linearity 1

[stanier]

I am modelling a system with a ball valve for emergency shutdown. In one reference document the term "non linearity" is used to describe the type of ball valve. I take it that this is the relationship of the open area or Cv to % rotation of the stem. Factors betweeen 3 and 7 have been quoted.

I am looking for a text that explains how the linearity factor can be applied to the Cv vs % open characteristics.

 

[JLSeagull]

You can find the capacity coefficient for every 10 degrees of travel for ball-style control valves. However you should consider a conventional ball valve as a quick opening valve. Lots of capacity exists at a much smaller degree of opening. I am sure that you can find details but conventional ball valves are not intended for throttling.

 

[stanier]

The question remains unanswered!

 

[DLite30]

Take a look at this thread:

http://www.eng-tips.com/viewthread.cfm?qid=280761

 

[stanier]

Dlite30,

Thanks for the link. I was aware of this thread as I had participated in it.

What I need to understand is this term "non-linearity" in relation to a ball valve.

My take is that the valve is controlled in opening and closing by a PID loop and thus the valve Cv vs % open can be characterized when Cv vs time data is used in a transient analysis. However it may perhaps be used to express the physical characteristics of the valve.

What I don't understand is the "factor" associated with this non-linearity. I am after an expression that can be used to adjust the valve Cv vs time in a transient model.

I have asked my client but the answers are slow in coming back. I have searched the web before posting the question. have contacted valve manufacturers.

 

[DLite30]

I hope this is what you're asking about:

"Non-linearity" in terms of a ball valve's flow coefficient versus position, simply means that there's not a constant relation between the two.

That is, for example, that when opening a ball valve, the first 10% of position change will result in the Cv value increaseing to 40% of the full open Cv value...while moving the valve the last 90 to 100% open only increases the Cv value say from 99% to 100%.

This is as opposed to a linear globe valve type control valve, where the percentage open and percent Cv value track almost one for one.

So in a Cv versus time model, you can curve fit the valve chart into a into a polynomial or exponential equation, and assume a valve closing speed, and there you derive a Cv versus time equation to use however you want.

 

[JLSeagull]

Perhaps by posting details on the context of the reference document that uses the term "non linearity" one could better understand your question.

 

[vpl]

stanier

somebody posted this document a while back -- I thought it was helpful and kept a copy.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[DRWeig]

Yah,

Good on ya Patricia, I looked all over and couldn't find my copy anywhere.

Stanier, your ball valve may vary. However, it's a matter of making your own curve-fit to the diagrams on page two of vpl's post. Then adjust based on your own testing if needed for accuracy.

Folks have used this method to measure flow in situations that do not require great precision.

Good on ya,

Goober Dave

 

[gerhardl]

... Note and question? 'Non -linearity closing' by ESD could that not also shoow to an 'outside' influence throtteling or slowing the actuator to obtain a fast first-step closing (say down to about 80% closed) and the rest (second step) slowed down to avoid water/fluid hammer?

Is it clear in the factual description that this common closing down sequence above is not the 'non linear' movement referred to, as all ballvalves, more or less, will have the non-linear flow-through as described in the attached curves, and it seems a bit special to describe or specify details in a common ball-valve's flow-through curve as 'a requirement' if this is the case ?

 

[stanier]

Thank you for your responses.

it would appear that this non linearity factor to describe a ball valve performance is not in common usage. I shall return to the source of the report and try and gather an explanation. When I do I will post same here.

 

[rmw]

Can you use ball valve and performance in the same sentence?


rmw

 

[ione]

Each type of valve exhibits an inherent valve characteristic, which shows how valve capacity (through Cv value) changes vs position (open %). Trim design is what determines the changes in valve capacity vs open %.

Despite the fact ball valves are not intended for control purposes and are should be managed fully open or closed, in some instances (when performing transient flow analysis) it could happen that variation of Cv vs open % becomes important.

In this link you can get further info

http://www.spiraxsarco.com/resource...n-actuation/control-valve-characteristics.asp

Nonlinearity could also be interpreted as the relation existing between minor losses through the valve and velocity (and consequently flow).

h = K*v^2/(2g)