Adding Doors to a Shipping Container 3

[dreber]

Adding a Door to a Shipping Container

There was recently a request for more technical content on here, so here's one...

I have an unusual request - to add a door to the side of a 40ft shipping container "40HC".

Technical drawings for a 40HC Shipping Container:

https://flex-box.com/images/pdf/specs/40HCDD.pdf

The walls of the container behave as trusses or built up beams with the top rail of the container serving as the top flange/chord and the bottom rail being the bottom flange/chord. The wall acts as the web.

This beam spans between support points at the two far sides.

Wall Elevation of 40HC Shipping Container with Door Midspan

Since the door is directly in the middle of the span, shear forces at this location will be minimal compared to those at the corners, so the door should be more feasible in the center of the container.

There is another fly in the ointment that the walls are corrugated:

I created a simplified FEA model assuming non-corrugated steel shells and omitting the opening with these preliminary results:

This was done to establish a baseline of stresses in the corrugations - my model indicates a maximum of 8.4ksi (my cargo container is not evenly loaded, so there are higher stresses on one side than the other). This gives me a baseline to compare the next model against.

I then added the opening, framing it with HSS posts and a header, and I get 18ksi in the corners next to the openings. This, being double my previous maximum, seems like more than I should be allowing.

I added PL3/16x3 straps at 45 degrees to try and stiffen that corner, but since the RISA model considers the straps to be just about as stiff as the wall panels (modeled as flat where they are really corrugated), it doesn't change the stress concentrations much.

So what would you do? Add the diagonal straps? Omit them? Find some fancy-dancy way to accurately model corrugations?

It seems logical to me that if there are stress concentrations in the corners that the corrugations will undergo limited plastic deformation until the stress concentrations are more uniform and at acceptable levels, so by that logic, straps wouldn't be necessary. Am I insane?

 

[Tomfh]

What is being done with it? Is it a house?


I don’t believe your stress concentrations at the bottom corners of the doors are real. It’s just the model spreading the rail load quickly up into the web. In reality it’ll spread out over longer length of web if need be.

I’d include the posts, and the header, and extend the header into the web if necessary.

 

[Kevin James]

We do this very often in rail cars, cutting holes in the side to add doorways for passengers and equipment. You're lucky in that the hole is in the middle. We typically have to add them at points of highest shear. The corrugation immediately to the left and right will double in shear load. I would make some free body diagrams of your side wall and do some hand calcs to find the bending moments in each of the corners. When we do the MOD we typically use door framing members and connections designed to take these moments at the corners of the door. We also extend the bottom and top into the side wall on each side to eliminate the hot spots you see in the door corners. Typically about 1 door width on each side. Not sure what your doing with the container, but if it is dynamic, watch for diagonal jacking. That puts a large amount of shear into your side wall.
James

 

[EdStainless]

Will this need to pass as a real container afterward?
There are ones made like this. You should be able to find details on some of these.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[dreber]

Tomfh:
Some relatively heavy industrial equipment will be permanently mounted inside and the container will be shipped by truck. I agree agree that the stress concentrations in the corner will spread more widely, but there has to be a limit to what would be considered allowable.

James:
Yes, placing the door in the middle is helpful.

When you say "The corrugation immediately to the left and right will double in shear load." Are you referring to the door posts?

I'm not sure what you mean by "diagonal jacking" but other than extending the top header of the door frame, it seems as though you aren't worried about the wall panel itself, since the shear at the door should be lower than the shear at the ends.

EdStainless:
It will be transported by truck. I don't think I'll have access to any typical details for these sorts of modifications, unless someone has a lead on that info.

http://www.kadent-structures.com

 

[Kevin James]

Dreber,

When you make a cut-out in a shear panel, the shear loads are transferred to adjacent panels above, below, and to the left and right. Sometimes the skin can carry this extra load but often it creates hotspots as you have seen. To mitigate this, what is done is to add framing around the door to carry the shear load that was carried by the now removed panel. This shear load on the framing translates to axial and bending forces. The bending forces end up governing the design of the corners.
To create a good connection between vertical and horizontal parts of the door frame, extend the horizontal parts into the side wall on either side. This also will cary load away from the corner and alleviate your hot spots.

When the container is lifted and set down, sometimes what it is set on will not be flat resulting on all the weight being on only two diagonal corners. This creates a twist in the "tube" causing the walls, roof and floor to carry large shear loads. In trains, this load often governs the design of large cargo doors. Note the this shear load is constant throughout the length of the wall. The door frame should be able to maintain the original rigidity of the structure.
Maybe for cargo containers it doesn't matter that they flex a bit, or that there is a bit of yielding. I am not familiar with their design or standards that govern their construction.
It also depends on the use of the container.

Hope this helps,

 

[r13]

I suggest to double panels at the corner, or beef up the chord members.

 

[KootK]

This would be my recommendation for reinforcing. Conceptually, it's very similar to:

1) how we reinforce roof and floor diaphragm openings.

2) shear wall coupling beams.

I would assume a shear of at least 1/3 of your maximum end shear at the opening. Few things truly emulate uniform load.

 

[dreber]

Thanks for all of your help everyone, it has been invaluable. I believe that I have my results:

=========================

http://www.kadent-structures.com