Weak Shell to Roof Joint NFPA 30 Tank Subdivisions 2

[engrjulz10]

Our team is designing a tank farm and we need guidance in interpreting a particular statement of NFPA 30 on tank farm subdivisions . Note that our tanks are NOT designed to be frangible tanks, as defined by API 650. However, they are atmospheric tanks with vertical cone roofs and in consideration of its overall construction, the seam between the roof and the shell is the weakest connection in the tank having a single weld overlay vs. the rest of the body and bottom of the tank having a double weld overlay. I quote the following entry of NFPA 30, sec 22.11.2.6.3.1 , which is the subject of my query:

"Where stable liquids of vertical roof cone tanks of weak roof-to-shell seam design or in floating roof tanks, one subdivision ...."

I seek guidance if the said statement is to be interpreted as describing a frangible tank as a general understanding in the Tank design linggo. I am of the opinion that the statement describes a frangible tank, and as such, our tanks are NOT falling into this categorization of NFPA, as they were not designed as frangible tanks according to API requirements. However, other members of my team are of the opinion that it is not necessary for the said statement to equate to a frangible tank as long as the construction of the tank is of vertical roof cone and with a weak roof-to-shell seam and thus qualifies for this category.

 

[fegenbush]

See 22.7.1.2 in NFPA 30 for the approved designs for emergency venting purposes. Then, contact the Fire Marshal or inspector to determine if they are in agreement with you regarding the definition.

If the inspector and/or Fire Marshal will not permit emergency pressure relief devices per API 2000 in lieu of a frangible roof joint per API 650, consider using a lifter roof.

 

[JStephen]

I interpret the "week roof to shell seam design" to mean a frangible roof. I don't think you could automatically deduce that the roof seam was weaker than the corner weld based on number of weld layers; it's a different geometry, sees different pressures, etc.

 

[Duwe6]

If your proposed tanks are not anchored to their pads, the square Shell-to-Floor 'chime' weld will be the weakest joint. Only way, other than anchorage, to avoid this is to follow the Frangible Roof details in 650. And I agree with you, NFPA's "weak roof-to-shell seam design" equates to API's Frangible Roof.

Tanks are tanks. Either design the tank to allow the roof to reliably blow off, or provide an Emergency Vent big enough to handle the infamous Pool Fire. Up to 50-ft diameter, a weighted room manway vent like ProtectoSeal's 53300 at 20-inch diameter will work. Above 50-ft, Frangible Roofs can be built.

and there is no Easter Bunny, nor a tank smaller that 30-ft with a frangible roof

 

[JStephen]

Cool video if you haven't seen it, by the way...

https://www.youtube.com/watch?v=6qcrwNM74sg

 

[MJCronin]

Kudos and a star to you, Duwe6 ! for insightful and practical knowledge !

The frangible roof question has been around for decades.

In 2005, API published some guidance:

API 937-A Study to Establish Relations for the Relative Strength of API 650 Cone Roof Roof-to-Shell and Shell-to-Bottom Joints

For historical insights, this is a good read and it supports Duwe6's recommendations :

http://www.thunderheadeng.com/wp-content/uploads/downloads/2010/09/Zhi_Lu_Thesis.pdf

Other information here:

https://www.nistm.org/houstonSept2009/docs/2009_HoustonConf_Pres/relief_venting.pdf

MJCronin
Sr. Process Engineer
Venture Engineering & Construction

http://www.VentureEngr.com