Minor Head Loss Coefficients for Valves

[Gigawatts]

Hello,

I am working on a problem where I'm trying to approximate the minor head loss in a scenario where one branch includes a mostly-closed globe valve. I say "mostly-closed" because there is an obstruction keeping the valve-seat from being fully engaged with the valve plate. I can't find any information that would give a good estimate of the minor head loss coefficient for that valve. I saw that a ball valve 2/3 closed is about 200, so I would think a leaking globe valve would be closer to 400. Does anyone know where I could find a more exact approximation for a mostly-closed, 6 inch globe valve? Thanks for your help.

 

[EnergyMix]

The problem is how do you estimate how much the valve is closed? About the only thing you can do is measure the flow and back calculate a value as there really isn't a way to estimate it. You can put in your "guestimate" and see how it fits with reality, but in real life it matters what the obstruction is and how it's going to impact the long-term operation of the valve. Is it going to be stagnant, is it going to grow, or is going to erode the valve seat? I've seen all three scenarios.


Want to know the do's and don'ts of Eng-Tips? Read FAQ731-376.
English not your native language? Looking for some help in getting your question across to others or understanding their answers? Go to forum1529.

 

[gerhardl]

To EnergyMix : Agree. Good comment!

To Gigawatts : Usually there is a more interesting general problem behind a question like this. Suitable for this forum?

 

[Gigawatts]

Thanks for your response. This analysis is for a one time event where I need to approximate flow rates to each branch to the best of my knowledge. Unfortunately, I am not in possession of the faulty valve in question but must somehow estimate a reasonable minor head loss. This doesn't have to be exact, but I'd at least like to be using the correct order of magnitude here. I can't understand why all of the head loss coefficient tables all give the same list of components and those don't include a globe valve that is anything other than 100% open. I would assume that as it closes, the coefficient should rise exponentially, but I'm not sure if that would lead to 20, 200, or 2000! It seems as though someone at some point in time must have done research resulting in further data for minor head loss coefficients, but if such a paper exists, I certainly haven't been able to find it! Thanks again for the help.

 

[77JQX]

Can you ask the manufacturer? Below are some links for approximate CV corrections for % open for one manufacturer. This manufacturer sells an "accurate" CV curve for each diameter valve they sell with the intent that one can then use pressure drop across the valve and % open to measure flow.

http://www.cla-val.com/documents/pdf/E-X133.pdf

page 5:

http://www.dfc.co.za/UploadedFiles/ClavalCatalouge.pdf

 

[rconner]

I agree with original replies - there does not appear to be any way to dependably/defensibly answer this question, as one really doesn't know exactly how big is the flow path in the mostly closed valve, and then how much of that flow path is occupied by the obstruction. That being said, if we are basically talking instead about a basically closed valve that just has trash resulting in a leak, I would think head loss could be HIGH (not really "minor"!)