Compressive Strength of Steel

[BikeDaily]

Greetings,

I am looking at a cantilevered rack system. So a pretty straight forward set of calculations. Or so I thought. For the rear vertical member (backbone as I call it) I have come up with a stress due to bending of 12,001 psi and a stress due to compression (from the weight of load and structure) of 2,338 psi. This would result in -14,339 psi compressive stress and 9,663 psi tensile stress at the edges of the "backbone". I typically design for a safety factor of at least 3. But, with this being made from standard structural steel components with a typical yield of 36ksi, I only get a safety factor of about 2.5 in compression.

I have two questions.

1.) Assuming my bending and compressive stresses are correct, you can add them together to get the resultant stresses as I did, correct?

2.) As long as question 1 is true, what is a typical value to use for the compressive strength of A36 steel? I have always used the yield stress of 36ksi. But I recently saw some numbers as low as 22,000 psi.

Thoughts?

 

[dhengr]

BikeD:
I think you should have more than two questions. What about buckling and instability of the various members in this rack system? What about secondary loadings and stresses? What about the various connection details on this system, and the concentrated loads and stresses at these points? What supports the backbone, and what shape and size is it? What is the load on each shelf or level, is it a uniform load, or can it be applied as a point load out at the tip of the canti.? Should I ask a senior engineer or my boss for some help and guidance on this? Should I consult the AISC Steel Construction Manual on this? Can I even buy A36 material any longer?

 

[dhengr]

And, what about a sketch or two with loads, dimension, some details, so we could understand what you are trying to do?

 

[rb1957]

answering the questions asked ...
1) yes, you can superimpose bending stress and axial stress.

2) depends, have you heard of crippling (thin sections can fail with a compression stress less than fcy).

personally 12001psi looks suspiciously like 12ksi (ie 5 significant digits ??)

Quando Omni Flunkus Moritati

 

[BikeDaily]

I have heard of crippling. When I use thin sections I use Bruhn's "Analysis and Design of Flight Vehicle Structures" to check for the crippling stress. But the member in question is a W21x62 beam. Pretty small for a 21" I-Beam. But I don't believe crippling is a concern here. The length at this high stress is pretty short, ~40". The bending moment is 1,492,812 in-lb, the moment of inertia is 1,309 in^4 and the farthest element in stress is 10.495" away from the central axis.

Bending Stress = My/I = 1,492,812*10.495/1,309 = 11,969 psi.

I will go ahead and look at crippling. Besides crippling and buckling, are there additional concerns for this area of concern I need to look at? What is the 22ksi compressive yield I saw on a few sights used for?

 

[fegenbush]

BikeDaily

From the original question, I thought this was a cold formed member. However, it is clear now that you should be using the design principles in the AISC Steel Construction Manual. Chapter H1 specifically deals with the design of doubly and singly symmetric members subject to flexure and axial force.

If you do not have a copy of this reference, AISC 360 is available as a free download. If in doubt, please consult with a structural engineer in your area. There can be a considerable amount of liability in the design of such structures, and good engineering practice is prudent.

 

[rb1957]

"What is the 22ksi compressive yield I saw on a few sights used for? " ... no idea (but then 36ksi is really low for me !)

instead of crippling, you might consider the flanges as flat plates in compression, SS on 3 sides, free on the other.

getting back to your original problem ... if you want a SF of 3 and are only getting 2.5, then don't you want to up-size some of the structure ?



Quando Omni Flunkus Moritati

 

[chicopee]

As dhengr stated provide a sketch so that everyone is on the same page, otherwise, we could be and in turn you would be misled.

 

[JStephen]

Depending on the circumstances, the beam can potentially fail at a considerably lower load than yielding, in which case the safety factor based on yielding is meaningless. The AISC specification mentioned above includes requirements for beams failing in various buckling modes, and also includes interaction equations for combined bending and axial load (which is not necessarily the same as just superimposing the stresses).

If you've got any substantial load supported, consider seismic design as well, and, if outdoors, wind loading.

 

[wannabeSE]

I agree with what the others have said. I just wanted to answer question 2. A wide flange beam with A36 steel have a minimum yield stress of 36 ksi. 22 ksi is probably what some people use for allowable stress. If the W21x62 is new, it will probably be A992 steel with a 50 ksi minimum yield stress.

 

[tr1ntx]

What wannabeSE said . . . structural shapes nowadays are Grade A992, which is, as I understand, the same as A572 Gr.B HSLA steel, thus the allowable tensile stress is more than 36k . . . 50 or 60 (off the top of my head, late at night). I think the rule of thumb for most steels, not high strength brittle stuff and exotic alloys, is to assume about 60% of tensile strength as compressive strength. If I recall correctly, I've seen that and used it for A500 Gr.B tubing, getting experimental results which matched the theoretical.

Sorry for not being definitive. I don't want to go on a hunt for the information. It is available in engineering literature, I've found it myself before.