Cryogenic Seat Leakage Test For SPE-DPE Trunnion Mounted Ball Valves (BS6364) 1

[mbelly84]

We have a number of cryogenic TMBV's, 3x2 class 150, lip seals, PCTFE seats.
Seat configuration is SPE-DPE.
Valves have been tested 100% at ambient temperature according to API 6D (leaks detected through body cavity, DPE seat tested in both directions) with no leaks.

The problem is that BS6364 (cryo valve spec) does not specify the leakage detection point.
My proposal is keeping the API6D procedure and checking the leakages through the body cavity.

Manufacturer does not agree and during the cryogenic leak test at -196 deg is detecting the leakages downstream.
Which is a nonsense especially for a SPE-DPE design (any SPE leakage would be hidden by the DPE seat).

What is your opinion on this, would you accept a cryogenic seat test for such a valve with leak detected downstream?

Is there any cryo test specification specifically written for ball valves that can help us proving our point?
Unfortunately BS6364 is a general spec, so it's not really detailed when it comes to special design ball valves like SPE-DPE.

Thanks

 

[Danlap]

Hallo,

Inspector usually do not (necessarily) go in depth with valve design. Therefore whether it is SPE, DPE, etc. the seat leakage rate should be the same as per valve (general) type (ball,gate,globe,check) and as per agreed standard or consensus agreed prior.

Try ISO 21011.
[ul]
[li]If this test is considered as prototype ⇒ @ cryogenic temperature, cycle 2000 times for category A valves or 100 times for category B. Seat leakage rate afterwards shall be equal or less than 200 mm³/s x DN. Applicable for both metal and soft seat[/li]
[li]Standard cryogenic test at -196°C ⇒ seat leakage shall be equal or less than 100 mm³/s x DN[/li]
[li]Standard production test at ambient temperature ⇒ seat leakage as per Class A – ISO 5208 aka no visually detectable leakage.[/li]
[/ul]
Assuming you’re already aware with the P/T rating, extension bonnet requirement, extra NDE test, etc.
PS: Some end users or certain operational requirement slightly alter above requirements and make it more stringent. I know some even requires Class A – ISO 5208 at cryogenic temperature. Or in general 100mm³/s x DN for metal seated and 30 mm³/DN for soft.

Regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376

 

[mbelly84]

Danlao, this is a production test, 10% sampling and we are using 50 mm³/s x DN as allowable leakage, but this is not my point.

the concern i am raising to the valve manufacturer is the following:

if they are detecting the seat leakage through the body cavity during the ambient temperature test (as normal practice for any trunnion ball valve, and as stated by API6D), why would they detect the seat leakage downstream during the cryogenic test? why changing the procedure?

this is, of course, a retorical question: MFR is taking advantage of a poorly written spec (BS6364). in fact detecting downstream for a SPE-DPE valve means that any SPE seat leakage would go totally unnoticed.

anyway IMHO this is not reasonable for an SPE-DPE trunnion valve.
leakages should ALWAYS be monitored through the cavity, regardless of the test temperature.


best regards

 

[Danlap]

Hi,

Thanks for the explanation. I am partially agree with you. the wording "leakages should ALWAYS be monitored through the cavity, regardless of the test temperature." is not correct for part of monitored through cavity.
Some trunnion mounted ball valves do not have cavity vent / drain connection. My company (not to be taken as sole reference) particularly do not approve this feature.
API 6D quote "Leakage from the upstream seat shall be monitored via the valve body cavity vent or drain connection, where provided. For valves without body cavity or drain connection, or downstream seated valves, seat leakage shall be monitored at the respective downstream end of the valve (the valve end downstream of the pressurized test fluid)." unquote

There are no stand alone standards for this application, especially when combining API, ASME and BS in this regards.
Medium will always be in the cavity...whether it is going to be 'entrapped' and subject for thermal expansion due to temperature/sun/etc that is something else
After installation, concept of valve is that no medium passes (zero or within tolerable leak rate)towards donwstream side...unless both of end user and manufacturer agrees that at certain pressure this valve act as DBB. Maybe for such as safe 'live' activities might be required on the downstream side of the piping, thus end user might drain the cavity first to check whether it is pressure free or not.

Without any prejudice, I am with the manufacturer on this case.
- Cavity relief testing @ambient is to find out the relief pressure of upstream seat. Yes, this might be considered as acceptance criteria.
- cavity relief testing @cryogenic is to find indication of relief pressure at cryogenic. Good for reference but not necessarily subject for rejection shall relief pressure is less than above (@ambient)/ leak rate is bigger measurement. Unless both you and manufacturer agreed in writing on parameter such as relief pressure at cryogenic and position of valve installed will be the same as tested (stem vertical/horizontal/angle/side)
- Seat leakage acceptance criteria is always measured on the downstream side as per API 6D/API598/ISO5208/50 mm³/s x DN. Yes, normatively this is regardless the design temperature the value shall remain within tolerance.

I will not go in depth with SPE-DPE valve, as you have better knowledge of it.. and also since it is not covered by all the standards mentioned in this topic

Regards,
MR

https://nosuchvalve.com

All valves will last for years, except the ones that were poorly manufactured; are still wrongly operated and or were wrongly selected

http://www.eng-tips.com/faqs.cfm?&rat1=2&fid=376

 

[mbelly84]

Danlap, thanks for your point of view.
Allow me to disagree on your statement "Seat leakage acceptance criteria is always measured on the downstream side as per API 6D/API598/ISO5208/50 mm³/s x DN".
This is not true, in fact API 6D clearly states:

11.4.4 Seat test procedures for block valves
11.4.4.1 Uni-directional
With the valve half-open, the valve and its cavity shall be completely filled with test fluid. The valve shall then
be closed and the test pressure applied to the appropriate end of the valve.
Leakage from the upstream seat shall be monitored via the valve body cavity vent or drain connection, where
provided. For valves without body cavity or drain connection, or downstream seated valves, seat leakage shall
be monitored at the respective downstream end of the valve (the valve end downstream of the pressurized
test fluid).
11.4.4.2 Bi-directional
With the valve half-open, the valve and its cavity shall be completely filled with test fluid. The valve shall then
be closed and the test pressure applied successively to both ends of the valve.
Seat leakage shall be monitored from each seat via the valve body cavity vent or drain connection, where
provided. For valves without a body-cavity vent or drain connection, seat leakage shall be monitored from the
respective downstream end of the valve.

Checking leakage downstream is only allowed in case valve is not provided with a cavity drain connection.

The above, in my opinion, should be applied regardless of the test temperature.
A leakage test is a leakage test.. at ambient, high, or cryogenic temperatures.

Regards