Flexural Bending and Shear in Built-up Round HSS with holes in wall 1

[Harriss1972]

I have been asked to strengthen a round pipe that will have three symmetrical ‘windows’ cut into it on 120 degree centers, with wide-flange shapes welded to the outside of the pipe section as pictured in the attached photo (I'll try to make the screenshot attach). The goal is to provide at least the same strength (axial, bending, and shear capacity) as the uncut pipe with the same support conditions. The entire assembly of the pipe and strengthening steel will be supported as a ‘flagpole’ cantilever structure, fixed at the base and free at the other end.

Based on determining the moment of inertia of the built-up section using the parallel axis theorem and the properties of rotated geometric sections (Table 71-27 of the 13th Ed. SCM), I determined the wide flange strengthening shapes that are needed to meet or exceed the section properties of the continuous round pipe and to exceed the axial strength of the continuous round pipe section.

However, the question was raised by the client regarding the maximum size (distance along the length of the pipe) of the windows and the minimum size of the 'strips' of pipe left in place between the windows, with the built-up shape still having at least equal bending, axial, and shear properties of the original pipe. The limit state that comes to mind is buckling of the strips and how shear flows around these ‘windows’ when the section is exposed to shear and bending. This seems somewhat related to how stresses work in a castellated beam, but a round shape instead of a wide-flange shape.

Does anyone have any recommendations, suggested reference publications, or other pointers for hand calculations on how one should approach this problem, given that the client does not want to delve into a finite element analysis or buckling analysis?

Thank you for any guidance or suggestions.

 

[ToadJones]

Can we ask what this is for?

 

[ishvaaag]

"the client does not want to delve into a finite element analysis"

but this is what most likely will give you some kind of correct answer for your problem. Model it either wholly 3D or with plates, and some initial imperfections. Need not even to be thoroidal or sweep-loft for the initial imperfections, just imitate a non straight cylinder and circular sections, then place holes of the investigated size and additaments, include P-Delta. See if remains stable at the wanted factored loads, and compliant with limit strength. This way you can decide in what extent you want to stretch your holes.

 

[Harriss1972]

This is for an oilfield project where the client (an oil company) has a corroding 10" diameter production casing inside the 20" diameter conductor casing. The wellhead sits on top of the casings, and the client wishes to excavate the 20" casing, cut windows into it, and access the 10" interior casing to make repairs and strengthen the corroding 10" casing. They feel that by excavating the top 15 feet of soil and strengthening the 20" casing and cutting windows into it for access will be cheaper than paying $2 million to bring in a 'workover rig' for each corroding well casing. This solution will be ideally used in the places where the corrosion is in the top 15'.

AISC Solutions Center didn't want to touch this problem--they said this is normally done with FEA. Maybe I just need to tell the client they need to cough up some more money for further calculations.

thanks for the posts so far.

 

[prex]

So you have essentially no bending and no shear in the pipe, as I assume the wellhead could be laterally supported somehow during the operation.
Then it's basically a problem of axial strength and axial buckling. Am I taking it correctly?
As far as axial buckling of the reinforcing beams is concerned, you should treat them as isolated beams with an elastic support at each solid pipe section, the support constant being given by the pipe wall behaving as an arch.
But you should also be able to connect the three beams together with a triangle of beams or gussets at each solid pipe section: in this way the three beams could be treated as a single truss, disregarding any collaboration from pipe wall (except for the strip welded to the beams acting as a reinforcement to beam flange).
In all this of course you need to be sure that the beam ends are correctly supported, especially the upper one. If you really need it to act as a cantilever, then the most critical part is perhaps how to transfer the loads from beam lower ends to the unreinforced pipe below.

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads

 

[ToadJones]

Will the access holes need to be patched ?

 

[Harriss1972]

The client had us do away with a tie-off system we proposed for the top, and asked us to design it for without. So, there would be flagpole, cantilever action, although the expected lateral loads at the top would only be wind load on the wellhead 'christmas tree', about 1 kip. But, we have no guarantee that they aren't going to apply this standard solution/detail to operating wellheads that may have pipes connected with thermal stresses pushing the wellhead laterally. I suppose that's the basis for the client requiring us to design, not for any expected loads, but for the full strength of the uncut exterior casing.

Yes, the holes will need to be patched back, but we are accomplishing that by wrapping the entire assembly with some pieces of 36" diameter corrugated drain pipe before backfilling.

Prex, good point about how the loads are transferred to the unreinforced pipe below. I'll take another look at that part in particular.

 

[KootK]

As a quick and dirty method, how about the following:

1) Figure out the load that the original tube could take. Use that as the design load for what follows.

2) Work out the worst case compression in one of your reinforcing columns. Check to see that the columns don't buckle between the solid annular strips of tube.

3) Take the lateral load on the tube and lump it into point loads to be applied to the solid annular strips. Check those strips for buckling & strength using ring formulas from Roarks.

4) Model yourself a 2-D vierendeel truss to look at the shear transfer between reinforcing columns. The truss will be as deep as the straight line distance between your reinforcing columns (the chords of the truss). Use 100% Ix for the top chord and 50% Ix for the bottom chord (it is shared between two trusses). Use 0.577 times the lateral load on the truss (0.5 / sin(60 deg)). Make sure that the portions of the annular rings that form the webs of the truss don't shear buckle, or yield in combined shear + bending.

5) Weld the reinforcing columns to the tube ensuring that the welds are strong enough to take VQ/I forces from the composite section.

6) Make sure the lap between the bottom of the columns and the reinforcing beams is stuffiest to deliver the requisite axial loads.

I suspect that reconstituting the capacity of the original tube will be difficult. You might be better off trying to work out realistic loads and designing to those.

 

[KootK]

Also, make sure that you've got a generous radius on the corners of your "windows".

 

[KootK]

I was also thinking that, in many respects, your situation is similar to a built up lattice column with the solid segments of the original tube acting as the lattice bars.

As such, it might behoove you to try to adhere to the proportioning guidelines for built up columns. The AISC manual and Timoshenko's stability book are both good resources in that regard.