Butterfly versus Gate Valves 1

[rzjim]

I apologize if this topic has been discussed before. I could not find it using the search tool. I work for a water supplier and I am responcible for writing specs. We have repeatedly argued over what valve is superior in a distribution system and a transmission system. Our system is over 100 years old. We have pressures ranging from 40-PSI to 150-PSI. We originally installed double disc gate valves with integral bypass valves. We are seeing failures of some of these valves and are writing specs to replace them. Should we go with butterfly valves? What are the pros and cons to each? We are also considering resilient wedge gate valves, but some are questioning the life expectancy of the rubber. Is that an issue?
Thanks in advance for any help!!!

 

[pennpiper]

I see the biggest problem you may have in replacing a Gate Valve with a Butterfly Valve is the vast difference in the face to face dimension of the two valves.

Modifying one hundred year old piping to accept the different face to face dimension might introduce more trouble than you can stand from a cost standpoint.

Think it over and talk it out before you make the final decision.

 

[zdas04]

That is a good point penpiper, but it is deeper than that. Valves made in the 19th century were not built to a standard flange-to-flange dimension, and virtually any modern valve that you buy will require piping modification--adding a 1-inch pup is pretty similar to adding a 6-inch pup.

My experience with gate valves (both as an operator and a designer) is that they are quite tedious to operate. A quarter turn valve is a lot less tedius (even with a gear operator). You've limited your universe to gate valves and butterfly valves. Some reason you've excluded ball valves? If long-term maintenance is a driving characteristic I would bet that ball valves would give you superior performance for about the same money as gate valves.

I'm not a fan of butterfly valves in most applications. My experience with them has been that the resilliant seat doesn't last decades, and that for high-velocity flows, the von Karmen streets downstream of the open flapper can represent a lot of force on the rod holding the flapper in place. A lot of people use them "successfully" (whatever their individual definition of "success" is), I just never have.

David

 

[reverman]

Biggest con I see with butterfly valves vs gate valves is that you have the disc sitting in the stream with butterfly valves which means less flow and wear on the disc. I would imagine that the maintenance on them is much greater.

the pros is that they are lighter weight and only quarter turn so they operate faster and since they are usually soft seated they have a better shutoff vs metal seated gate valves.

I think it depends on the size when talking ball valves. On smaller sizes I think the extra cost of the ball might be worth the switch But large diameter ball valves are quite expensive and I'm not sure the switch from gate valves is worth the extra cost.

You didn't really say but it appears that the gate valves in your system lasted quite a long time so another fact to consider is it worth changing the system parameters for something you know worked for decades?

 

[rzjim]

Thanks so far!!!

I think a little clarification is needed. Valves were mostly direct bury with either bell by bell or spigot ends with poured lead joints. Replacing those we would cut the cast iron pipe on each side and replace with mechanical joints. Small spool pieces of ductile iron would be inserted into the mechanical joints and couplings would secure the new valve to the old cast iron. In addition, some type of clamp, such as the steel restraint clamps supplied by Ford, and 1-inch threaded rods all wrapped with wax paper for restraint.

Valves in vaults may have been flange by flange, so the above about spacing and cost could be correct.

We still have the ongoing disagreement over which valve is superior or more correctly which valve should be used when and for what reasons.

Here are some of the arguments I have to address:

butterfly valves are easier to operate, they cost less, they do not seal tightly, they can be throttled, the seats could be damaged during throttling

gate valves are hard to operate, they cost more, they are designed to seal tightly, they can not be throttled,

the rubber used for resilient wedge gate valve and/or the resilient seat on butterfly valves may not have a design life of 100 years and may be as little as 30years

Should I be considering going back to a metal seated double disc design???

 

[stanier]

If butterfly jams open how do you get it out?

Butterfly valves are not considered a satsfactory form of isolation in a double isolation scenario.

Wafer type butterfly valves are not an effective form of isolation when the downstream piping is removed for maintenance.

A full bodied flanged double or triple eccentric butterfly valve (ie a real one) will cost you more than a gate valve.

Resilient seats can be ripped out in a waterhammer event.

If the spindle of a butterfly valve is in the vertical the bottom bearing gets contaminated and will wear prematurely.

Any debris in the stream will jam the disc of a butterfly valve and damage the seat.

You cannot pass a pig or sponge ball through a butterfly valve if scouring out the crypto or giardia in the sludge.

 

[gerhardl]

I see you fail to mention the common diameters of the valve, and also your country. By watermains in Europ with longlife requirement, it is most common to use double eccentric butterfly valves, but not!from high-quality producers.

Mains in this case would be pipelines for pressure class PN10, PN16 and PN25 (bars) and dimensions from about DN150/DN200 or somewhat higher, for instance from DN250

See for instance

http://www.erhard.de

(several others)

Note: with lifetime IP68 gear.

With gear placed correctly (when ordered), valve stem will remain horizontal with gear-operaator spindle up.

For theese types of double eccentric butterfly valves the quality of valve, gear, sealing construction and general flow charecteristic, way of operation and lifetime expectancy will be far above centric butterfly valves, and much cheaper than ballvalves.

The seat sealing construction (seals with integrated sealrings secured adjustable and mounted on the disc) will not be ripped out by waterhammer for this type of valve.

Valves might be throtteled for smaller delta Ps with slightly better charecteristics than a common centric valve, and somewhat increased range, but will of course cavitate by throtteling at higher delta Ps' as centric valves.

The geometrical and seal construction will give thight lifetime sealing.

For 100 years lifetime for theese dimensions the double eccentric ballvalve (sample: same factory) could also be considered.

For smaller valves in the street distribution net: different type of gatevalves or ballvalves.

You need to do a cost over lifetime study before you choose types, but overall valve quality is highly important, and when asking for (preliminary?) bids you should ask for alternative types and quotes from all suppliers and a technical argumentation following the bid, and also references.

Very often today quotes tend or have to be ordered from cheapest offer, not highest quality of else (only seemingly) socalled 'equal types' from different manufacturors.

Advice: do not stress your inquiries with 'overspecification', stress instead lifetime and standtime importance and let the factory suggest or offer additions, materials and even types of solutions to give the best result.

An easy understood example: a 'gear' on a valve can be many different things, but in reality vastly different in operational torque, construction, materials and protection.
If you look at a commen valvesheet this will very often be described as 'with gear and (x) length extension.'

See the point?

If you do not ask for alternatives of highest available quality, or allow for this in competition comparison, you will end up with the cheapest gear available - not the best cost/lifetime solution.

- and for the valves it is even more complicated: small differences in construction can give large differences over 100 years, for instance exposed or not exposed inner bolts, and details on bolts and material quality.

Good luck!

 

[Ashereng]

In most water service, I see gate valves. One reason maybe that there is nothing in the flow path for debris to hang up on. With a butterfly, if there are debris, it may prevent the butterfly from closing properly/completely.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[HitokiriAoshi]

Resilient seated valves will provide 0 leakage for the beginning of it's life and leak more and more as it ages. I would not use any resilient seats for valves difficult to maintain. Nor would I recommend such valves for 100 year service...

Some things to consider:
- Do you need bubble tight shutoff? (resilience vs metal seat)
- How often can you afford to change seats? (butterfly)
- Would the valves be used as isolation valves? (butterfly type, number of valves)
- How often would the valves need to be actuated? (quarter turn vs multi turn)
- Is throttling necessary? (butterfly/ball vs gate) please note many valves are not designed to be throttled for extended periods
- Emission requirements? (quarter turn is better)
- Pigability? (normal gate/butterfly out of question)
- Solids in the medium (resilient vs hardfacing vs material selection)
- Dead end service

Personally, it seems the double wedge valves were used for DBB isolation service, the best replacement for it would be.. the same thing... or as reverman say, ball valve. I don't see resilient butterfly valve a good candidate and you would need two butterfly valves for safe DBB.