Strengthening of Steel Column 1

[strucguy]

I am looking for some guidance on strengthening existing steel columns for a major renovation project. All columns are A992 Gr. 50 steel and will be reinforced with A36 grade plates welded to the flanges. But, I am having hard time estimating the compressive strength of the stiffened member. Is there any guidance available on estimating Fcr for a member that uses two different grades of steel? Your thoughts are greatly appreciated. Thanks in advance.

 

[ToadJones]

There is a paper out there from the 4th Quarter, 1988 Engineering Journal called:
"Reinforcing Loaded Steel Compression Members" by J.H. Brown.

I can't find my copy of it.
I know it is available through AISC for free for members and for $10 otherwise.

Not sure if it covers different materials.

 

[Lion06]

Toad,

Here is the paper you're talking about.

 

[Lion06]

Here is another one

 

[strucguy]

Thank you for your responses.
I did go through the papers...but, I couldn't find a case where two different material strengths are addressed.

 

[Gumpmaster]

Do you have to use A36 plate? When I have this issue, I normally call for A572 Grade 50, and then I don't have to worry about it. I don't think there's much of a price premium.

 

[strucguy]

Gumpmaster...we did specify Grade50 plates. But the fabricator is asking if they could use A36 plate steel. I can't just say no without giving him a valid reason.

 

[paddingtongreen]

If the column is in the elastic buckling range, could you not consider both to be A36.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[strucguy]

paddingtongreen -

The KL/r for the column isn't big enough for stresses to be in the elastic range.

 

[ToadJones]

Wouldn't a column with a large value for KL/r be more likely to in the elastic range?

 

[strucguy]

Toadjones -

You are right. But, the KL/r isn't large enough for the column to fail because of inelastic buckling.

 

[strucguy]

type in my earlier post

"....fail because of elastic buckling"

 

[Lion06]

Can you assume 36ksi for the whole section and get it to work? You're not only increasing the area, but you're also increasing the radius of gyration, so the allowable stress goes up, too.

 

[ToadJones]

EIT-
I have read the second paper your posted in the past and I'm not sure I'd ever go that route. The whole process seems way too academic to me.
Thoughts?

 

[ToadJones]

I have had the same problem with 50 ksi wide flanges with 36ksi cap channels....I am trying to remember how I handled it.

 

[strucguy]

StructuralEIT-

I could definitely do what you said. But, wouldn't that be conservative and force me into using thicker plates? I just wanted to know if there is any other rational approach to the problem, and get away with a little additional steel as possible.

 

[paddingtongreen]

On consideration, I think you have to go with SEIT's suggestion, the A36 will be connected to the A992 but if you must not strain the material at the interface beyond the A36 limit, it becomes a limit for the A992 as well.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[Lion06]

What about this?

Use the new kl/r to get the allowable stresses in the plates only (call fa1). Use the new kl/r to get the allowable for the WF only (call fa2), then subtract the existing stress from the WF (call fe). Use the difference (call fa3 = fa2-fe).

Hopefully fa3 is close to fa1 so that you don't need to have different strains in different parts of the column to reach full capacity. I would probably use the lower of fa3 and fa1 on the entire section. I don't think I would want to get any more refined than that.

My biggest concern is that if the plates yield while the WF is still elastic, they are going to soften and effectively reduce the E of those plates. That would have detrimental effects on the overall column. I would limit fa3 above to no more than fa1.

 

[ToadJones]

Sounds like a good article for next quarters Engineering Journal.

 

[csd72]

I agree with Structural EIT, it is important to remember that higher grade steel still has basically the same E which means it has to yield further to achieve the higher yield strength.