Weld Repair of HSS Column

[BMart006]

I have a circular HSS column subjected to cyclical compression stresses that has significant corrosion pitting to the extent that the section needs repair for strength reasons. The pitting is limited to a 4" area along the length of the column but is approximately half the depth of the 7/8" wall thickness. Normally I would consider doing some clad welding and welding a section over to bypass/reinforce the localized area, but this being a rounded surface subjected to cyclical loading, I'm not sure this is an appropriate choice. Does anyone have any suggestions for resources or solutions that address this type of repair (where fatigue conditions are a concern)?

 

[TehMightyEngineer]

Encase in concrete?

Ian Riley, PE, SE
Professional Engineer (ME, NH, VT, CT, MA, FL, CO) Structural Engineer (IL, HI)

 

[BMart006]

This is part of an articulating structure so I don't think we'd be able to do that on this project. That's a good idea for a common column though.

 

[TehMightyEngineer]

Composite fiber wrap? Cut out and groove weld in a new section and carefully grid and profile welds to limit fatigue stress risers?

Ian Riley, PE, SE
Professional Engineer (ME, NH, VT, CT, MA, FL, CO) Structural Engineer (IL, HI)

 

[BMart006]

We had considered replacing the deteriorated section with new HSS, but thought that the risk to weld quality was greater than just "filling in" the pitted area and reinforcing somehow. I hadn't considered using a column wrap though, so that may just do the trick.

 

[Agent666]

If fatigue checks out, what's your particular concern with the patch approach? I'd imagine getting a plate curved to the correct radius wouldn't be that much of an issue.

Another way might be to weld thinner plates on their edge to the section as stiffeners at appropriate centres around the full perimeter of the section, making them long enough to transfer load around the damaged area. Combine with cleaning up and addressing the original damage.

If there's a cause for the pitting, ensure you mitigate this so it doesn't occur again, or continue to occur if the damaged area is still exposed.

 

[KootK]

1) I don't understand what a concrete or fiber wrap would accomplish here? Confinements not the problem is it?

2) I feel that, with the right detailing and accounting, you could do reinforcement. Hopefully less that I'm showing below.

3) I could get behind a full penetration welded replacement piece so long as:

a) Good QC and maybe testing per your own concern and;

b) Not too onerous for the guys doing the work.

HELP! I'd like your help with a thread that I was forced to move to the business issues section where it will surely be seen by next to nobody that matters to me:

http://www.eng-tips.com/viewthread.cfm?qid=456235

 

[Agent666]

Yeah kootk's sketch is what I was describing, "sun stiffeners". Could also use equal angles and weld both legs to section face (kind of easier to hold in place to weld, and easier access to weld).

Agree that FRP or carbon fibre wrap won't do too much here. Need to replace steel area.

 

[TehMightyEngineer]

My thoughts with a wrap of concrete or FRP would be to develop the compressive force in the wrap and transfer a portion of the compressive load to the wrap around the damaged area while simultaneously providing protection and avoiding as much of a fatigue issue as possible (though the concrete to steel bond is dubious at best under fatigue).

Ian Riley, PE, SE
Professional Engineer (ME, NH, VT, CT, MA, FL, CO) Structural Engineer (IL, HI)

 

[Agent666]

the fibres in FRP matrix are generally only good in tension though.

 

[KootK]

Another thing that might be worthy of consideration here is that fatigue in compression is much less of an issue than fatigue in tension. So much less that some folks don't even believe in it. A detailed critique of this is beyond my MatSci pay grade however. That said, if this repair were taking place well away from the mid-span of the member, where some potential for flexural stresses might exist, I'd not be losing too much sleep over the fatigue business.

HELP! I'd like your help with a thread that I was forced to move to the business issues section where it will surely be seen by next to nobody that matters to me:

http://www.eng-tips.com/viewthread.cfm?qid=456235

 

[SteelPE]

What are you talking about in terms of loading and cycles per day? I would suspect if you are worried about it your loading is quite high and your cycles/day are quite high as well.

As a point of comparison, I have done a small amount of work with industrial buildings. Nothing crazy, mostly cranes 20-tons and under. I can seem to remember anywhere in the design guides where compression fatigue was even mentioned. However, I do know they are a little particular in regards to the design of the weld for the composite W+C support girders.

In regards to Aggent666's sketch. I have attempted this before on a round column.... and depending on the size of the column getting the angles in might be difficult. In the end I settled for 2 large angles vs 4 small angles as shown.

 

[Agent666]

Regarding fatigue, some of the welds that I'm envisioning might be involved, do result in some of the worst detailing categories for fatigue though, so I wouldn't ignore it if the stress range is large without a brief look to make sure you are comfortable in ignoring it. No doubt there are already similar welds at end connections though, so maybe you're heading down the rabbit hole of exposing a previously unthought of issue.

Seems in terms of codes some consider compressive loads for fatigue while others completely neglect it, a recent discussion was had here in this thread outlining some of the different approaches which may be of interest.

 

[TehMightyEngineer]

the fibres in FRP matrix are generally only good in tension though.

Generally but they have definite compressive strength as well. For what sounds like a small repair area the cost of extra FRP layups and resin seem negligible. That said; I'm still struggling to understand exactly what the OPs equipment is undergoing and what the design criteria is. I fully agree that a series of sun stiffeners is likely easier and better but wanted to propose non-obvious routes since OP likely has considered those.

Another thing that might be worthy of consideration here is that fatigue in compression is much less of an issue than fatigue in tension.

Well, the welds would likely be in shear and shear does have a tension component to it subject to fatigue. That's one reason I proposed cutting out the damage and groove welding the repair in place to put the fatigue critical welds in direct compression. But, yes, put me in the camp of compression fatigue as being a non-issue for 99% of all structures.

Ian Riley, PE, SE
Professional Engineer (ME, NH, VT, CT, MA, FL, CO) Structural Engineer (IL, HI)

 

[Earth314159]

With cyclical compression stresses, is fatigue a significant issue? It is usually just differential tensile stresses that need to be considered for fatigue.

 

[KootK]

Well, the welds would likely be in shear and shear does have a tension component to it subject to fatigue.

No argument there. I really brought that up as it might pertain to any number of options for a repair in general, including doing nothing at all.

This is the compression fatigue thread that live in my archives: Link

HELP! I'd like your help with a thread that I was forced to move to the business issues section where it will surely be seen by next to nobody that matters to me:

http://www.eng-tips.com/viewthread.cfm?qid=456235

 

[Agent666]

The benefits gotten from FRP and carbon fibre are sensitive to initial loading (i.e. dependant on strains), for applying it to an already loaded member the FRP isn't really seeing any load so is basically doing nothing, all the load is in the original member. Might as well say because you applied some paint to a section it can take more load!

Due to the variance in load here, it might see some load but they don't work by redistributing the loads from the steel if it is overloaded to the fibre. I'm not on board with FRP or carbon fibre working in compression either way in a practical design sense.

I think anyone from the US won't like/consider fatigue in compression because as per the thread I linked to your codes don't see it as an issue and those guys probably think anyone who needs to consider it has gone bonkers. However some of the rest of the world is thinking you've gone bonkers because you don't consider it. It's all about perspective. If there are residual stresses then it is considered a real effect that can cause fatigue in many codes and shown in research based on my brief delve into it based on that other thread.

 

[BMart006]

Thanks for all of the responses. The design is a one-off, but can be compared with a hydraulic derrick with some wave action at the base. The area in question is not near the middle so flexural behavior is of no concern. Stresses range from 2ksi T to 22ksi C. After reading the thread KootK posted and some other articles, I believe I’ve eliminated my worries on compressive action fatigue.
Thanks again everyone, this community is always helpful.

 

[SKJ25POL]

BMart006 (Structural)(OP),
Could please provide couple pictures so I can get a better understanding of situation?

Thank you