13 Apr 09 11:05
Comparing shell to shell spacing table 18.104.22.168 of the 2008 ed. to table 2.7 of the 1999(96) ed. there is no change. I don´t have the 2000 ed.
Apart from this, some personal thoughts.
Yep, the 3´ looks to me quite small.
Thinking of the oil industry, API documents (recommended practices) related to fire protection (doc API RC 2001) refer the distances to NFPA 30 and of course free room for other criteria due to especial risk considerations.
API fire reports don´t give much information about the importance of shell to shell distances. If I remember well they mention that in actual fires, cooling exposed tank walls with spray systems have not been a very important issue related to losses on adjacent tanks. (Although API has a RP 2030 doc. where they recommend exposure protection with fixed water spray).
So it gave me the impression that it is not very well defined how important is the shell to shell distance based on actual fires; when I see photos of actual fires in the oil industry I think that it is difficult to generalize a criteria about when shell to shell spacing is important. There are people doing fire exposure modeling with advanced calculation tools, but I haven´t heard if they are giving us valuable data to revise the prescriptive NFPA rules. With the actual oil industry situation and the terrorism risk, I was expecting that some aspects were going to change.
Other industrial standards I´ve seen require bigger distances but they look to me that they are not well updated.
If your are working with methanol, a volatile one, maybe it would be better to think of complementing the NFPA with other criteria, and consider the expected fire fighting practices.