Good afternoon. I have a question for the CII collective.

We are doing a project to install four 60' diameter API 650 atmospheric welded steel storage tanks. The design condition is 260° F and 0.5 psig. The problem consists of the design of the 24" vapor recovery header on the tank roofs. The header is 24" NPS sch10S 316L. The tanks are connected in series, in a straight line in plan view, to the vapor header, i.e. the header spans the roof of the four tanks before dropping down. The header is connected to each tank with a 4' riser and 24" butterfly valve. Pressure loss is critical in this system so no bends are allowed to increase flexibility of the piping.

We are trying to estimate the magnitude of the thermal displacement reactions imposed on the tank roof nozzles by the piping to ensure we have not overloaded the nozzles. As you know, API 650 provides a method to calculate nozzle flexibility BUT the published method only applies to tanks much larger than these, thus I am reluctant to extrapolate it to this particular situation. The situation can be fixed with flexible elements (bellows) but our client is adverse to the use of bellows. I can probably talk him into using a few bellows but not enough to relieve the large reactions.

However, if I can accurately model the roof flexibility, I can more accurately predict the magnitude of the reactions on the roof and hopefully leave out a couple bellows. Our initial thought is to use the RISA structural software to calculate all six stiffnesses of the roof nozzle and then place those into a CII flexible anchor.

Any thoughts/opinions/criticisms/cheap shots on this? Has anyone else used any other methods to calculate flexibilities of small thin-wall storage tanks?

Thanks ! ! !
Thought for the day:
Good judgment comes from experience;
Experience comes from bad judgment.