Leakage Question 1

• Mar 21, 2001

• #2

svanels

Petroleum

Oct 24, 1999

387

SR

Well my common sense tell me that there are no acceptable leakage rates for gas for ball valves.

Would you work on a line which is supposed to be isolated, and is leaking through a defective valve?

Steven van Els
SAvanEls@cq-link.sr

 

Upvote 0 Downvote

• Mar 23, 2001

• #3

pkelly54

Mechanical

Jun 1, 2000

4

US

All valves leak - although the leakage of many properly operating valves is insignificant for real world applications. There are industry standards addressing valve seat leakage. Organizations such as the Manufacturers Standardization Society, Fluid Controls Institute, and Instrument Society of America have published standards covering seat leakage Some standards have multiple classes of leak rates. The valve may comply with the specified leakage class but that leakage class may be improperly selected.

I agree with Steve van Els. Valves for gas service should be "bubble tight". More likely the valve seat is scored or the valve is assembled or installed incorrectly. An inspection of the valve is wise.

 

Upvote 0 Downvote

• Mar 27, 2001

• #4

struggling

Mechanical

Mar 23, 2001

6

US

If all your valves leak, you should consider either a different valve design or double isolation. I agree with svanels. From a maintenance safety standpoint, unless you can maintain some kind of continuous purge, there is no acceptable leakage for gas.

 

Upvote 0 Downvote

• Mar 27, 2001

1
• #5

kstaylor

Mechanical

Jan 8, 2001

86

US

The Americam Petroleum Institute (

http://www.api.org)

has a standard for valve leakage. Look for API-598. It covers a wide variety of valve types as well as sizes and pressure classes. It has acceptable leakage rates for liquid as well as gas testing. All valves built to the various API standards are required to meet API-598 leakage criteria prior to shipment from the manufacturer or supplier.

I agree with pkelly54 in that all valves leak to some extent. Bubble tight is a matter of definition (how many bubbles per minute and at what pressure). You need to determine the acceptable leak rate for your application and then determine why your valves are leaking at an unacceptable rate. Disassembly and inspection are most definitely required. Also mentioned was that the valves in question are equipped with soft seats. There may be some trace compounds in the gas stream that have degraded the elastomer seals. Analysis of the gas may be in order to understand all components of the stream.

 

Upvote 0 Downvote

• Apr 9, 2001

• #6

rusman

Chemical

Apr 8, 2001

35

MY

Agreed with all comments by the previous commentator. Though the leakage is unavoidable and it's up to certain allowable leakage rate (bubble rate) at a certain valve class tightness, eventhough it's soft seated valve. To mitigate this, try install double block and bleed valve, with pressure gauge at the bleed valve.

Hope this help

Process

 

Upvote 0 Downvote

• Apr 10, 2001

• #7

moodley

Petroleum

Apr 3, 2001

8

ZA

Both API 598 and API 6D (Spec for Pipeline Valves) specify zero leakage for soft seated valves. However, most pipeline operators determine their own acceptable leakage rates for valves in service (assuming no threat to life). Operational conditions and requirements will help determine the acceptable level of leakage before valve replacement/ refurbishment.
Most Ball valves are supplied with emergency sealant injection systems that can be used to assist seal integrity. Depending on the extent of seat damage - valve sealants can restore seal integrity.
To my knowledge, most of the larger valves are cycled infrequently, so sealant use would be minimal.
The main reasons for seat failure would be incorrect elastomer selection (kstaylor)either on the part of the buyer or seller. The belief that "Viton is Viton is Viton" is incorrect. Correct elastomer selection plays a vital role in valve reliability. The least admitted reason is poor installation practice (shipping, storage, commisioning etc..) Any place where dirt and grit can get between the ball and seat.
Although I can't give you an "operational" leakage figure, I hope this helps.

 

Upvote 0 Downvote

• Apr 11, 2001

• Thread starter
• #8

Noob

Chemical

Mar 21, 2001

2

SG

Thanks for the feedback.

 

Upvote 0 Downvote

• Apr 17, 2001

• #9

MikeONeill

Mechanical

Apr 3, 2001

23

US

Your valves are most likely trunion style ball valves with RTFE seat and seals. The original leakage rate for these valves is 'bubble tight' (zero leakage at full rated Dp). If the valves are trunion style, they are in-line repairable. The seats are designed using a molded polymer with 'memory' charicteristics designed in. After a period of time, the seats tend to lose that memory. A repair kit consisting of soft goods (seats & seals) will solve the problem.

Mike

 

Upvote 0 Downvote

• Aug 1, 2001

• #10

lefty424

Specifier/Regulator

Aug 1, 2001

2

US

I disagree that all valves leak. Valvtechnolgies, Inc. metal seated ball valves are guaranteed zero leakage. All valves are tested against high and low pressure and certified before shipment. They meet and exceed API 598 and are beyond Class VI ANSI FCI 70-2.

 

Upvote 0 Downvote