Fire-safe test failure? 2

[adrenalin]

Hi everyone,

I am having trouble with the fire test of big dimension floating ball valves. And I will appriciate your comments if you have an idea about the possible reasons.

The testing of small diameters(DN50) were succesful for both floating and trunnion ball valves.
But when we tried the DN150 floating; although the burning period was OK with very little amount of leakeage; after the burning period, During forced cooling, the valve starts leaking with a high flow rate of water. We experienced the same failure twice and I need to find a solution to be certified by a third party.

Any similar experiences or comments on this issue??

 

[belowzero]

adrenalin (Mechanical)
Is this a soft seat or metal seated ball valve?
Solid or hollow ball?
The method of testing is critical for either, I've had many a failure because of the wrong methods used by a test house, there is a real 'art' to getting through these believe it or not. Will respond when I know more.
B

 

[belowzero]

Ohh, what standard are you testing too API or BS?
B

 

[gerhardl]

Geometrically a linear increase of the diameter will give a second degree curve increase of opening area and a third degree curve increase of volume. This will to a certain degree be true for all main parts of the ball valve.

This will again lead to different heat amount transferred and contained in the different parts, different amounts of material exposed to different temperatures at different time-curves, different expansion and contraction and different mechanical forces from size to size.

What exactly is causing the failure is of course impossible to guess without knowing the exact construction and damage after the test, but something constructional has obviously to be altered/strengthened in the larger dimension ball valve, to compensate for the different forces working on the larger valve.

I would imagine the total seat sealing construction, material and geometry is the obvious part to start, but other parts might also contribute to your problem, for instance different expansion and contraction of body and ball compared to the smaller size.

 

[belowzero]

I have Fire tested 8" Class 150 and 6" class 300 floating ball valves to both API and BS6755, see faq408-589 (which I wrote as Burdy when I was doing this work).
The class 300 is a heavier ball and was harder to pass. Despite 'seat sealing construction, material and geometry ' you must know and have built up a knowledge of how your valve performs and operates in various conditions. This 'Product knowledge' is what enables you to not just manufacture but 'Design' a valve which works. You just cannot beat experience, we appear to be losing most of it as time served people retire, without passing thier knowledge on.
B

 

[gerhardl]

belowzero, ..you stated clearly what I was groping to express...!

 

[adrenalin]

Thanks for the reply,
The sealing of the valve is a soft one with PTFE and the ball is hollow. It is not that heavy. And the failure occurs duing the high pressure testing. I guess it is not related with the weight of the ball.
Actually it is almost obvious that different contraction and expansion rates and heat transfer problems cause the failure but until now, I could not come up with a solution yet.
Any suggestion will be appriciated!
Regards,

 

[NGiLuzzu]

adrenalin,
is there a thermocouple within the valve during your trials?
I guess it would be useful to know what is the maximum temperature reached in the area where PTFE is, how long that lasts and which is the shape of the temperature graph when cooling starts...


Hope this helps, 'NGL

 

[belowzero]

This is the art of successful firetesting a softseated valve, if the body is hotter in one region than another then the seat will deform under load quicker in that area, this causes the ball to become misaligned with the rest of the seat and a leak occurs. Does the stem hole go all the way through the ball or part way, is it a slot?
Both API and BS Firetest standards dictate the placement of thermocouples so this is fixed. Hollow balls are actually harder to test especially under high loads because they deform.
B