Horizontal tank mounting supports

[dmh2819]

I am working on a 16" dia by 34" OAL horizontal tank designed to 312 psi. Shell is .250 SA 53B pipe. Heads are 2:1 elip SA414G. The desired mounting is to use 2x2x1/4 angle running parallel to the shell axis such that the tank will be flush to the mounting surface. This is not a saddle design, skirt, lug or legs for that matter. I am wondering what types of calculations I need to perform and where I might find them? Any suggestions would be a big help.

 

[SnTMan]

dmh2819, it sounds like you need to consider an attachment to a cylinder, perhaps a unit length/loading (not sure I correctly understand your geometry). A WRC-107 analysis, often included in commercial software, should be capable.

Regards,

Mike

 

[dmh2819]

Thanks SnTMan,

The geometry is simple. Think about laying 2 angles on a table, parallel to each other about 12 inches apart. Then sit the horizontal tank down on the angles and weld them up. I am looking at the attachment calcs. I have no software and doing all the number crunching by hand.

The other question is that I have to confirm the mounting will withstand static loading in any direction equal to four times the weight of the container and it's contents.

 

[SnTMan]

dmh2819, that is more or less what I had pictured.

These calculations can be done by hand, if you had a copy of WRC-107 or a similar method, but it won't be easy.

Maybe find some freeware / cheapware?

Regards,

Mike

 

[lganga]

YOU CAN CONSIDER IT AS A PIPE THEN YOU CAN USE AN STANDARD PIPING SUPPORT

 

[JStephen]

I don't think WRC 107 will cover a case where the support runs from one end to the other.

There are some load cases in Roark's Formulas for Stress and Strain for a pipe supported by two load points like that, but they don't account for the effects of the ends.

 

[Duwe6]

The ends sort of 'disappear', if the saddles are fairly close to the ends. If the bending stress on the middle section is OK, the short, cantilevered ends will calc OK also. Just have to apply the full design load to the saddles.

 

[JStephen]

If you read the description above, these are not saddles. The angles run lengthwise, not across like a saddle.

 

[innovation2]

Use software like NozzlePro. Everyone should have a copy.

 

[racookpe1978]

At 16 inch dia x only 34 inches long with 1/4 inch thick walls, he is really looking at very light loads.

16 x 3.14 x 1/4 x 34 x .283 = 120 lbs of steel for the walls.
About 30 lbs for the heads.

He's got a complex "beam" (pipe plus two angle iron in cross section) with a continuously supported reaction down the length of the angle iron. Looking at an axial cross-section, he's got a cylinder supported by two angle irons: again, the 16 inch dia cylinder resisting "bending" by the upper and lower arches of the cylinder, supported by the two angle iron.

But at so light a dead weight load, does he need to show the calc's for that near-zero bending load?

 

[SnTMan]

I wouldn't think so, unless there is more...

 

[chicopee]

For design purposes,I would treat each leg supporting the vessel and its content as fixed beams with end loading acting radially from the tank to the end tips of the upright legs. The end loads would have horizontal and vertical components and the analysis from both components would be per unit length. Stiffeners may be required in each angle iron to prevent leg separation. Anchoring the two angle irons to a base seems to be more critical than welding the vessel to the angle irons since thermal and vibratory stresses may be factors. I probably would rest the tank on the angle iron upright legs instead of welding. Crushing the shell is probably unlikely but you'll need to check for that too.
Instead of 1/4 thick leg, I would be more apt to use 1/2" thick legs thereby eliminating the need for stiffeners.