Flow through partially open valve

[MTPipeliner]

Is there an acurate way to predict the amount of flow through a partially opened valve? Particularly a plug valve flowing natural gas.

 

[zdas04]

Patrick,
No there isn't. Some valves (e.g. needle valves, variable chokes) have quite linear valve position vs. mass flow characteristics and are very easy to predict the position vs. flow. On/Off valves like gate valves, ball valves, butterfly valves and plug valves have very non-linear performance. They are all very poor throttle valves because they will pass approximately 100% of their capability at some very small proportion of their travel (usually 10-15% open will pass 100% of the mass flow rate).

These valve tend to make a lot of noise while they are 1-20% open, but this noise really doesn't reflect the magnitude of the mass their passing.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[MortenA]

If you know the valve chrateric as a function of position the i would say yes. Contact the valve evndor and they most likely can supply with a Cv vs. degree of opening chart.

Best regards

Morten

 

[zdas04]

MortenA,
Have you ever looked at one of those Cv vs. degree of opening charts for a plug valve? The one I saw was obviously prepared by an inventive marketer and it showed Cv on a 2-cycle log scale to try to make the un-informed see linear. On cartesian scales, plug valves approach 100% flow within the first 10% of travel and further opening doesn't change mass flow much.

This "Cv vs. degree of opening" chart for block valves is not something I would ever try to program into a PLC or to estimate a flow-rate from. It has been said many times in this forum, but I must say it one more time - it is rarely a good idea to operate an inherently on/off device in a throttled state.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[saxon]

zdas04, We've developed and use specially characterised plug valves in polymer service as control valves all the time. And have developed the "inherent flow characteristic curves" (%ratedtravel vs. %maxflow)as well as %open vs. cv rate with the manufacturers. This also applies to ball valves and butterflys. Gates are the only valves good for on-off service only.

saxon

 

[khan101]

Still it is hard to predict the flow vs opening because inherent characterstic curves are very different than intalled characterstic curves. You can have a good guess let's say + - 20%.

 

[zdas04]

Saxon,
Wow. Is there any way you could share that data? What I've seen from valve manufacturers causes me to feel strongly that in throttling service I want a globe, needle, or v-ball, but I can be convinced by compelling data. No valve manufacturer has come across with compelling data yet. One manufacturer has told me flatly that their line of ball valves should never be subjected to extended periods of partial open.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[rmw]

Ball valves not designed for throttling service let process fluids in behing the seats, which is not much of a problem if the ball valve is full open or full closed. Those designed for throttling have provisions to prevent or minimize the detrimental effects of exposure to the process.

rmw

 

[MortenA]

Ahh sorry David. I should read a little more carefully before writing I did not get to the "plug flow" thing.

I agree with your opinion re. the value of trying to use Cv vs opening for a on/off valve.

For ball valves i do know the Q ball trim (Metso formerly Neles) wich is quite suited for control but still retains much of the good things from ball valves (tight shut off, 1/4 turn closure, low resistance when full open etc.) I

Best regards

Morten

 

[katmar]

There are plug valves available (from Tuflin I think) that do have specially profiled openings to improve the flow control characteristic. It would make it something like a V-ball valve in characteristic, but with tight shut-off.

A trade-off of getting tight shut off is that you need a large actuator to get accurate and stable positioning of the valve.

The way I read the original post, the valve was a standard isolating plug valve and I would agree totally that for this valve it would be very difficult to correlate flow against opening position.

 

[saxon]

zdaso4, Check the following sites:
For Baumann Valves,

http://www.emersonprocess.com

For Durco Valves, fcd.flowserve.com/valves

These should provide you with all the info you require or at least a source for the full tech. info.

As for the curves we've developed; sorry, those are proprietary.

saxon

 

[zdas04]

Saxon,
I must be slow, but I just went all through the Emerson web page and all of the Baumann valves I found were real control valves (globe and proportional disk for the most part). There were some butterfly valves, but they have modified plates to improve linerity. I didn't find any graphs or tables of flow vs. percent open for any of the valves.

On the Flowserve page, even the "Characterized Ball" (similar to the Fisher V-Ball) doesn't have any data on it's linearity.

I still don't see any data that would lead me to believe that a conventional ball, or a flat plate in a butterfly, or a conventional plug are suitable for the "accurate" throttling that PatrickR started this thread with.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[MTPipeliner]

Thanks for all the reply's. I was able to get some info from the vendor that should be pretty helpful. Approximately 200 pages so there must be something good in there.

FYI,
We use plug valves in our system as side valves for purging, filling, blowing down, etc. The other day we were pigging a line and someone asked how long it would take for the pig to go through to the other end. I couldn't thinkg of a good way to predict that without knowing how much gas had been used to push the pig. Anyway, that's where the question originated.

 

[sailoday28]

Wouldn't the time for pushing the pig be dependent upon mass of the pig, length and resistance of the piping, back pressure expected on pig in addition to the inflow of natural gas?

 

[Montemayor]

Patrick:

Am I correct in interpreting that you have plug valves in your lines as main block valves through which you plan to pig? Unless I miss my guess and you are using a "special" type of pig in your process lines, you may may not be able to carry out your pigging operation with conventional plug valves in your lines. Another way I could be wrong is if you are using over-sized plug valves in your main lines as block valves (which is an expensive way to go). I hope that I am wrong and that you simply are referring only to the plug valves used to introduce the pressure medium behind the pig (air?).

What I'm referring to is that every plug valve that I've bought and operated had a conventional reduced port. If you can visualize the way and manner that a plug valve is fabricated and the way it works, you will agree that unless you install an over-sized model in the line (complete with concentric expanders) the pig will stop dead in its tracks the instant it hits the under-sized, conventional "bore" of the plug valve. In fact, depending on the type of pig and the way you run the pigging operation, you may have problems with the pig jamming itself at the concentric expander junctions due to the pig getting mis-aligned with the concentric "bore" of the valve. This used to be a problem when gates (with inherent reduced bore) were the standard in pipelines being pigged. Today, with "full-bore" ball valves being the valve of choice in most pipelines, the problem has gone away. One way to resolve the mis-alignment problem is to elongate the pig -i.e., make the pig long enough so that it remains fairly stiff and straight as it negotiates its way through the over-sized valve and both expander/reducers. However, when you do this, you make it harder for the pig to go through ells and off-sets within the piping system. I hope I'm way off target with this comment.

My point here is not to make or raise new problems for you but to alert you should you not have taken the above into consideration - if, in fact you have plug valves in your main lines. I'm sure other engineers experienced in pigging operations will want to comment. My ears pricked up the moment you mentioned PIGGING - which you failed to mention before in the thread.

I hope this helps to straighten things out rather than confuse them.

Art Montemayor
Spring, TX

 

[1969grad]

FYI. Plug valves were often used in natural gas metering stations in the 1960s and before whenever larger sized regulators were required. Since that time the v-ball, etc. has been developed and you never see them.