We are looking to purchase some valves which need to be fire safe. We are running into the difference between the safe designed valves and the tested. We are aware of the test criteria for API 607 and what it will mean to be tested. But, what is the criterion to call yourself a fire safe designed valve. Is there a small amount of time that the valve needs to not leak in a fire, like 2 minutes? Is it just having a stem packing rated to a certain temperature, I have seen the most common being 1000f even though the fire test will see 1800F. Any thoughts or factual difference?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
fire safe "design" verse "tested"
- Thread starter breaker34
- Start date
Several API specifications apply to fire tested valves such as API SPEC 6FA, 6FC; API STD 589 and STD 607 etc. Other standards include FM Class 7400, approval guides, British Standards Institution BS EN ISO 10497 etc. Some client specifications also apply. My guess might include ExxonMobil, Dow, etc.
I can see why you are confused, valve manufacturers may have a design of valve which has been Fire Tested and is therefore 'Certified' Fire safe, whether by design or not, it holds a unique certificate for its BOM.
A simple change in build i.e seat material, may then not have been tested(and therefore doesn't hold a Fire Test certificate) but the designers have used the same design as that which was tested. So the valve no longer has a unique certificate for that build but is marketed as a 'Fire Safe by Design' valve in order to cover it. If you really want to be sure then the complete valve you are buying must have its own fire test certificate matching your BOM. Well thats how we used to do it and I don't think its changed in all the years.
A simple change in build i.e seat material, may then not have been tested(and therefore doesn't hold a Fire Test certificate) but the designers have used the same design as that which was tested. So the valve no longer has a unique certificate for that build but is marketed as a 'Fire Safe by Design' valve in order to cover it. If you really want to be sure then the complete valve you are buying must have its own fire test certificate matching your BOM. Well thats how we used to do it and I don't think its changed in all the years.
Hello Breaker34,
Example:
When performing a fire test, you logically ruin the valve for future use. So when a manufacturer is making expensive valves e.g. 48" control valves, he will not be too happy to perform a test (even if he is allowed to test a 24" valve to certify the 48" size). It will jeopardise his competitiveness.
He might argue that the 20" valve size that is certified has the same design and would be much more susceptible to fire damage, since less material is less heat absorption. So the 48" valve will easily withstand the fire test.
Concluding:
- the 20" valve can be called fire safe tested/qualified
- the 48" valve can be called fire safe designed
I am not aware of any standard mentioning 'fire safe design'.
It is up to you to decide what is acceptable to you.
Gr.
Terje
Example:
When performing a fire test, you logically ruin the valve for future use. So when a manufacturer is making expensive valves e.g. 48" control valves, he will not be too happy to perform a test (even if he is allowed to test a 24" valve to certify the 48" size). It will jeopardise his competitiveness.
He might argue that the 20" valve size that is certified has the same design and would be much more susceptible to fire damage, since less material is less heat absorption. So the 48" valve will easily withstand the fire test.
Concluding:
- the 20" valve can be called fire safe tested/qualified
- the 48" valve can be called fire safe designed
I am not aware of any standard mentioning 'fire safe design'.
It is up to you to decide what is acceptable to you.
Gr.
Terje
Some Manufacturers claim that their valves are "fire safe by design" or "intrinsically". I still haven't found any international standard recognizing and defining such a concept... but for sure a polymeric lined valve - for example - is very different from a metal seated one, with regards to fire resistance; or the scope of fire testing, at least, should be different. In the first case it should be considered as a mandatory, not banal proof; in the second one it may be considered just as a possibly superfluous, further confirmation of what expected based upon materials selection and design features (and Customers may elect to give it for granted).
Coming to terje61's post, all the standards I ever read provide for fire testing of valves not larger than ND 8” because an actual fire on larger pipe sizes “would cause substantial damage to adjacent pipe and equipment before it would cause valve failure” (see, for reference, API 607 standard, 4th edition 1993, Note to TABLE 3 on page 7).
Hope this helps,
'NGL
________________________
Coming to terje61's post, all the standards I ever read provide for fire testing of valves not larger than ND 8” because an actual fire on larger pipe sizes “would cause substantial damage to adjacent pipe and equipment before it would cause valve failure” (see, for reference, API 607 standard, 4th edition 1993, Note to TABLE 3 on page 7).
Hope this helps,
'NGL
________________________
All postings above confirm what I suspected. There is a 'an open gate' for the use of the 'firesafe design' as a concept.
'Firesafe design' is widely used, but to my knowledge there is no common or firm description or rules limiting or qualifying the use of this concept.
If one roughly tries to describe the scale from top and down you will in my opinion have at least three levels:
Top level:
Fire safe tested valves: Valves built and tested (with fire at given intensity and time and with following operational and leakage tests) according to a given firesafe standard. The standards will accept within limitatations equally constructed valves in near pressure classes and sizes as 'firesafe tested' if the firesafe test for one size and pressure class is passed.
The customer will either accept an earlier passed test or require a new test for a certain (bulk) order of 'firesafe tested' valves, within the range described, as a part of the contract.
Middel level.
'Firesafe design' from a factory that have at an earlier time fire tested with passing results equally constructed valves (according to a given firesafe standard), but where the actual valvesize or pressure class (or near sizes and pressure classes as allowed under the given standard) have not been tested earlier. (Usually for economical or time-consuming reasons, or because standards does not include the sizes or pressure classes.)
Tests may or may not be required from the customer, with or without economical compensation (according to market competition, risk evaluation and cost).
Bottom level
Firesafe design from a factory not actually at any time having firesafe tested a valve, or with 'firesafe' tests with weak actual quality or relevance, and/or large deviations in construction.
Conclusion:
1. The end user has to qualify what he means by 'firesafe valves'
2. The end user has to qualify what he can accept on test background.
3. Advice: It would anyway probably be best to use standard test values, and not construct your own deviations, for instance lower temperature for stem sealing exposure.
Who shall bear the cost of an actual deviating test? For large bulks you could probably press the producer for this, but should bear in mind the cost of witnessed tests.
A good factory will already have a test background for higher temperatures against the argument from a cheaper supplier: 'We have actually never done a test, but we are sure our construction will pass a test at a temperature and time lower than the ones given in the standard, it is after all a 'firesafe design'!(?)'
- Thread starter
- #8
- Status
Similar threads
- Locked
- Question