Concentrated load on tank roof plate

[abehong]

Hello:
I am running into a situation that we have to install some of operating platform support on the roof plate of one existing API 650 tank. So it is becoming a design case to apply a concentrate load at various location on flat plate. I could not find any design guideline in API 650 code or some other tank design manual. It would be appreciated if someone any share experience or knowledge with me.

 

[fegenbush]

A good reference for stress in a flat plate is Roark's Formulas for Stress and Strain.

An API 650 roof is likely not flat, but 3/4:12 is common. In that case, there are rafters to support the roof plate also. I would assume this as a flat plate simply supported at the rafter maximum spacing under consideration. If there are no rafters, the roof should be pitched no less than 2:12 and I would use a different analytical method to determine the stress in the conical member.

 

[abehong]

Hello, fegebush:

Thank for your response. It is fairly large diameter tank with rafters . The spacing between rafter is about 1899mm.It is also a cross brace installed in the middle of rafter. The flat plate design criteria basically is a concentrate load at various location on 1899 mm W x 5085mm L rectangular flat plate. The problem for me is that I could not find one case in Roark's book really suitable for my design. I tried to use design case for concentrate load at the center of rectangular plate as per case 8b under table 11.4, but the results ( stress and deflection) seems to be non-realistic because the supports are not located on the center of plate. Most of supports are actually located closed to the rafters ( Max offset 15 in ) . Is there any other analytical method to determine stress and deflection more accurately ?
Thanks
Abe

 

[JStephen]

The flat plate formulas in Roark assume small deflections, and will not be applicable to roof plate.
You could treat a strip of plate as a catenary, neglect bending, and derive an allowable load.
Bednar's pressure vessel handbook has a simple method for deriving allowable line loads on a shell, but it will usually give very low allowable loads in a case like this.
You could put stiffeners over the rafters.
If you try to analyze the problem more exactly, one problem is that you don't now the exact geometry in the first place. Theoretically, it's a perfect cone, in reality, it has buckles and waves several times the plate thickness.

 

[LittleInch]

Why not build a separate platform / supports directly over the rafters? This might increase the overall weight, but would allow you to analyse the rafter(s) as a beam with the known weight of the existing roof plus your new loads. You might need to add a secondary external rafter to re-inforce it, but seems less complex than what you're trying to do at the moment.

Also for a circlar tank, surely the rafter spacing reduces as you go towards the centre of the roof? As ever a sketch or drawing of what you're trying to do will help everyone a lot.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way