Use of Butterfly Valves

[amiraz]

I am installing automatic valves on one of our propylene refrigeration compressor to isolate the unit in case of a compressor shaft seal leak, in order to isolate the casing and vent the casing gas to the flare. One of our engineers here recommended using full port ball valves instead of butterfly valves for better valve cv. The cost of ball valve (8", 10" and 12", 300# flange) is much higher than a butterfly valve. These valves would not be directly installed on the compressor nozzles, but would be located about 10 feet away on the piping feeding the compressor (suction, discharge and 2nd stage). The engineer's concern was about disturbing the flow of gas to the compressor in case if a butterfly valve was used. This is a 3000 HP horizontally split case centrifugal compressor.

I personally would like to use the butterfly valves. Does anyone see any issues with this approach?

 

[zdas04]

There is only one reason that ball valves are more expensive than butterfly valves--BECAUSE THEY ARE WORTH IT. I use butterfly valves in trash service. If I don't care about flow profile, positive isolation, or dP, then I'll use them. I never use a butterfly valve in mission critical service, they just are not up to the task.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[amiraz]

I heard you loud and clear. I personally share a lot of the same concerns. However, what do you think about high performance butterfly valves with stainless steel construction, triple offset design with metal seat? Limited space also make the use of butterfly valves more attractive.

 

[zdas04]

The triple offset valves still leave a bluff body in the flow. I've seen too many of the traditional valves bust off part or all of the flapper in a high velocity flow stream to ever be comfortable putting one in front of an expensive compressor.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[rotw]

Zdas04,

Your point is also clear to me why the butterfly valve would not be smart choice. I have some questions that would extend the subject of the topic.

What is your experience with big butterfly valve used as suction control valve (say with a CV from 10000-15000 usgpm and higher)?
I am interested to know about reliability of such big sizes and about the dynamic characterstics when they are actuated (e.g. quick opening)?
I have also no idea if on this range of diameters the most encoutered/reliable actuators are motor type or pneumatic.

I also wonder what would would be the main difference in terms of reliability and suitability when using the valve for insulation or for process control.

Also when you say that these valves are not up to task, could you please elaborate a bit on this ?

Beside this, I have personal experience with the selection of a suction throttle valve about 8'' size for a cent. compressor where minimizing dP losses with valve fully open was a critical project requirement. Both angle valve and globe valve turned out not to be a good choice with respect to dP at full passage. It appeared that only the butterfly type satisfied the low dP losses requirements also at low costs. In that project, the use of flangeless valve driven the cost down even further.

 

[zdas04]

People used to publish percent open vs. percent flow curves for their valves. I haven't seen one in many years, but they were very telling. As I recall, the butterfly valve was similar to a gate or ball valve--by about 35% open they would pass 100% of flow. Not a good characteristic for a control valve.

You keep talking about price. It sounds to me like you are interested in saving a few hundred dollars on a half million dollar machine. I've inherited a few of those machines over the years and they are truly difficult to operate. I generally ended up throwing the cheap crap away and installing a proper valve.

If I have to do big inch suction control and want a zero dP when fully open, I'll use one of the "profile ball" valves like a Fisher V-Ball. When fully open it has about a zero dP, but has great linearity over most of the range.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[amiraz]

Thanks to everyone for showing your interest in the subject. Please note, these valves are NOT going to be control valves and they will not modulate at all. They will be strictly automatic actuated on/off valves. Talking to folks at FS Elliott and IR and looking at their skids, butterfly valves is what they use for the most part. The butterfly valve technology today is much more advanced than ever before. Double offset high performance, triple offset metal seated and even quadruple offset butterflies are very reliable I think in an on or off position. My only concern is whether there would be any significant propylene gas flow disturbances, through the 14" BF valve located about 5' upstream of the machine suction nozzle?

Thanks again!

 

[rotw]

Thanks Zdas04 for the info.