Fire Testing requirement for Soft seated Valves (Elastomers) 4

[ET1983]

Hi all,

In one of our projects, we are using soft seated ball valves, with the requirement of fire testing in accordance with API 607.

When i had gone through the API 607, i observed that the valves would be under flame at 760 Deg celcius for 30 minutes times and shall be checked for the leakage rate. If this is the case, all the soft seated materials, (Elastomers and polymers) can withstand to a maximum temperature of only 240 Deg Celcius, based on vendor details.

Hence, what is the purpose of carrying out API 607 on soft seated valves.

 

[Danlap]

Hi,

You can also cross reference with ISO 10497 - Fire Safe testing requirement.
Why 30 minutes? General accepted time for fire brigade to distinguish pool of fire (before it becomes bigger or prior exothermic reaction events)

In summary most of us will only concern with 2 things:
a. External leakage (through Body to Bonnet and Stuffing Box). During and after 30 minutes on fire it may have leakage, however the amount must be lower so the leakage rate will not enlarge the pool of fire (e.g. few drops of kerosene per minute). Some say that if this sealing are mainly using PTFE packing/gasket, it would never be fire test (officially) certified.
b. Internal leakage. See introduction of ISO 10497, lower pressure class Ball valve (below PN 40 and below Class 300#) may have Soft seats. This testing was intended to simulate whether during and after 30 minutes (on fire) the internal seat leakage rate will contribute to bigger (exothermic reaction) on its downstream side.

Will the soft seat melt/damage after 30 minutes on fire? Most likely yes. However since this is done while in close position, the 'melted' soft seat will form itself and filled up the crevice/gap between Ball and Seat.

Hope I inform you sufficiently,
Regards,
MR

Greenfield and Brownfield have one thing in common; Valve(s) is deemed to "run to fail" earlier shall compared to other equipments

 

[bcd]

The purpose of the test is to prove that the amount of leakage through the valve, or to the outside of the valve is not so much that it causes a fire to grow larger. The group that developed the test standard considered refinery fires and times of response and wrote the test procedure to duplicate the situation.

Unless it is a large valve, where the test equipment does not provide enough heat input to fully melt the parts, any PTFE seals and elastomers completely melt and burn away. They are usually completely gone at the end of the test.

In the case of floating ball valves, the ball will move until it contacts the metal portion of the valve body. The contact surfaces are typically machined so that they seal well enough to pass the test. Any stem seals or body seals usually have some graphite somewhere which provides sealing before, during, and after the fire test.

 

[moltenmetal]

bcd has explained it for you. What we need is a valve which does not leak to the exterior during a fire, so no soft goods are permitted there- it must all be expanded graphite on body seals and stem seals. The seats, which may give class VI tight shutoff during normal temperature operation, go away entirely during the test- but the floating ball moves down and seats against a secondary metal seat machined into the body/end flange. This gives poor sealing, but enough to reduce the amount of fuel this valve would feed to something downstream which is on fire.

Remember also that during a fire, the aluminum actuator will eventually melt- I've seen pictures of piping after fires with nice bright splotches of aluminum on top of the pipe from the melting actuators. You want, indeed need, to have any actuated valve fitted with a piece of plastic tubing on the air line, such that a fire will melt that tubing off and allow the valve to go back to its failure position before the actuator gets too hot to work. Running the entire air line with stainless steel tubing is not safe! And of course any valve which requires the actuator to remain in place to stay shut off, will not be fire safe.

 

[ET1983]

Thanks MR, BCD and Molten metal. Your responses were really helpful.

Hence, i understood that, the fire testing is mainly for the Body, Bonnet and stuffing box of a valve, irrespective of the seating material. And even if the seats are burned away, the closure member shall be forced to mate with the seat to form the tightness, which ultimately would avoid triggering the fire bigger. is my understanding right??

And whether, it applies for the Rubber lined Butterfly Valves, with the same philosophy.??

 

[gerhardl]

bcd and moltenmetal: excellent answers!