Multiple Section Moduli with Non Composite Bending 5

[UtahAggiePE]

Here is the background. With design of glass facades for buildings you often need to strengthen the vertical members (mullions) that transfer the wind loads to the floor slabs by inserting steel cross sections that fit snugly so the steel member and the aluminum member act together, but they are non-composite. This is done for controlling deflection or reducing the stress on the aluminum section.

For deflection behavior it is a simple matter to multiply the steel moment of inertia by the modular ratio add it to the aluminum moment of inertia.

The question I have is, for this situation, how does one combine the section modulus of the steel and aluminum when computing stresses? It seems overly conservative to simply add the two, but it doesn't make obvious sense to multiply the steel by the modular ratio either.

Thanks, in advance.

 

[Celt83]

If they are non-composite then the procedure you describe using the modular ratio is not appropriate and the steel and aluminum will only share load based on relative stiffness assuming an adequate load path from the aluminum to the underlying steel.

Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

 

[Denial]

For each of the two components M = CEI where C is the curvature.[ ] Hence
[ ][ ][ ][ ]M[sub]total[/sub] = M[sub]s[/sub] + M[sub]a[/sub] = C[E[sub]s[/sub]I[sub]s[/sub] + E[sub]a[/sub]I[sub]a[/sub]]
because (we assume) the curvature is the same for each component.

 

[BAretired]

Agree with Celt83. Stiffness is measured by EI, so the difference in material properties is already taken into account.

Using two dissimilar metals in contact with each other presents a potential galvanic corrosion problem. Are you planning to wrap the interior shape with plastic strips or similar material to prevent contact?

BA

 

[JAE]

You compare the relative stiffness (EI) values of the two members and apportion the load to the two according to that comparison.
Each will deflect exactly the same since deflection is directly related to the two EI's.

You can then just take the portion of load attributed to each and calculate the stresses for each.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[structSU10]

What you need is the separate forces in each member. The load is shared based on stiffness. I like to do it as a ratio, and then apply that percentage of the total load to each component. for example if the steel is represented by EsIs, and aluminum by EaIa, the moment in the steel for a simply supported beam with uniform load would be M = (w*L/8)*(EsIs/(EsIs+EaIa). That ratio of stiffness can be used for each force, and then stresses checked for each individual component.

 

[UtahAggiePE]

Thank you, that is very helpful.

Regarding BAretired's question about galvanic corrosion. So far as I know, it is standard practice in the industry to not use any wrap. The steel is protected from heavy corrosion by being contained inside the aluminum. Also the relative amount of aluminum (which would act as the anode) is such that galvanic corrosion is negligible over the whole member over the expected life span.

 

[JAE]

The stars come out tonight!!

Check out Eng-Tips Forum's Policies here:
faq731-376