Calculate allowable shell and roof nozzle loads on a tank?

[KernOily]

Here's yet another nozzle load question for you fellers. I am Mr. Piping Reactions this week...

I am involved with LOTS of smaller storage tanks, mostly for water, crude oil, and oil-water mixtures. Most are API 650 welded but some are API 12J bolted and ALL of them are 100,000 bbl or less, the vast majority being 5,000 to 20,000 bbl cone-roof. I frequently have occasion to hook up hot lines to existing tanks, so I am constantly in need of nozzle allowables. As you know, API 650 only gives allowable nozzle load calculation methods for tanks MUCH larger than these. So how to calculate the allowable reactions for smaller tanks?

What I've done in the past is: (1)use FEA (expensive) (2) use some 'judgment allowables' (3) calculate the allowable using API 650 and then reduce it based on judgment (hard to defend that one...) (4) calculate the allowables using API 650 and then increase(!) them because a smaller Ø tank is supposedly stiffer, all other parameters being equal.

The tankies are usually of no help, believe it or not. And that doesn't help anyway if you're re-using an existing tank or changing its service.

So, short of applying an expansion joint at the nozzle every time, which is hardly aaceptable, what do you guys do/what have you done? Any thoughts/ideas/miscellaneous ramblings are welcome... Thanks!
Pete
P. J. (Pete) Chandler, PE
Principal Engineer
Mechanical, Piping, Thermal, Hydraulics
Processes Unlimited International, Inc.
Bakersfield, California USA
pjchandl@prou.com

 

[stanier]

Have you tried the

http://www.algor.com

site. They have a product called PV Designer that is based on FEA but is custom designed at the front end for engineers to design tanks and vessel nozzles.

 

[MJCronin]

Pete,

This is an excellent question...... and it has been asked many times before. For flat-bottomed, cone-topped tanks - my experience has been that each engineering firm addresses this issue in a different way.

The problem, of course is a varied tank geometry and the lack of an expensive empirical study.......

While there are calculation methods such as WRC 107/297 that incorporate empirical data (in the form of curves) no such expensive study has been performed on API-650 style tanks. No book or text or anything other than an "in-company" informal guideline has ever addressed this issue. (I assume that you are aware of the recent API-650 appendix that addresses nozzles low in large diameter tanks)

Some design firms have guidelines that put an expansion/flexible joint at every piping nozzle on these tanks. The stated purpose is for "tank settlement" which may or may not occur...... and oh, by the way, your nozzle reaction problem has now disappeared.

I suggest that you post this on the ASME bulletin board and also e-mail the people at API. They can add this question to the FAQs on thier website.

Try:

http://www.asme.org

and good luck !!

MJC