Aluminum versus Steel Design

[LearningAlways]

We have a subcontractor working on a project with us who is in charge of designing a metal frame made of aluminum. This frame system takes a high amount of wind pressure as we are in Florida. After reviewing his calc package I realized he isn't designing using the Aluminum Design Guide. Our building code requires the use of Aluminum Design Guide (ADG) when designing aluminum systems.

I know squat about how to design aluminum systems. The subcontractor's approach thus far has been to use AISC but incorporating very large safety factors in design, their Safety Factors are usually twice what AISC requires.

How much difference is there between steel and aluminum design? "Enough to write another specification." is probably what you're thinking. Would you feel comfortable with their approach of using larger safety factors? My gut says there is enough difference between aluminum and steel design that AISC doesn't just say "use larger S.F.'s in your design of aluminum", because if it were that simple why not just add it in.

I've had them revise and resubmit a couple of time noting that they need to use ADG, they still haven't done it and we're about to come to a head between schedule and them needed to use the ADG.

 

[KootK]

1) I feel as though it's a pretty big ask for the sub to dodge the use of proper Aluminum design guides and codes. I'd not be willing to accept that unless:

a) The client really wanted this to preserve schedule and;

b) I could somehow divorce myself from any associated liability, which seems unlikely.

2) To play devil's advocate, were there no such thing as aluminum codes, I'd probably do something like what the sub has done so long as some heed was paid to the "aluminisms":

a) Welding issues

b) Fatigue issues.

c) Proper Fy & E

d) Conventional shapes. No thermally broken voodoo.

e) Proportions similar to AISC shapes.

As you surely know, Aluminum is considerably less stiff than steel. If the members are sized for drift and deflection using the correct elastic modulus, and well braced, odds are pretty good that everything else will wind up being auto-solving to a large degree.



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

 

[hokie66]

Agree with KootK. Steel and aluminum are very different materials, and there are a lot of aluminum alloys, with a host of different properties. In his list 2)a), welding changes the aluminum in the area of the weld, usually lowering the strength.

Steel and aluminum are both great materials, but there are more differences than similarities.

 

[Ron]

AISC is definitely not appropriate for aluminum design. Aluminum is significantly different in design than steel...the most notable is the reduction in allowable stresses near (within 1 inch) of a weld. I am in Florida and have designed over 600 aluminum structures....some simple, some very complex including trusses with over a 100 ft clear span.

Since you are working under the Florida Building Code and your subcontractor is providing a delegated design, I assume you have review (SEOR) responsibility. Reject it!

The ADG is referenced in the Florida Building Code (See Chapter 35). As such, it is included in the code as if written there in its entirety (See Chapter 1 of the code). You have every right to reject the submittal and require compliance with the code. If you don't do so, you are a party to a code violation and you are deviating from your standard of care....which in Florida, can be considered negligence under Chapter 471 and the rules of the Board of Professional Engineers, Florida Administrative Code Section 61G15.


A Great Place For Engineers to Help Engineers

Follow me there.....

 

[NorthCivil]

What are we looking at? if it is a small accessory structure, just a 2 sheet design, I might accept it.

If its more than that, unacceptable. Personally, if I had to submit calculations for aluminium structures that I attempted to complete with the steel design code, I would be embarrassed. If I did so and they were accepted, I would be a bit surprised (and worried).

Aluminium is quite similar to steel in its behavior when it comes to your primary stresses (bending, web crippling, stability of slender flanges etc). but very different when looking at welds. and there is a huge variety of material types. If the guy running the numbers isn't sharp on aluminium, then the guy checking the numbers probably should be.

 

[EdStainless]

KootK hit the biggest one, that is stiffness. Al is only half as stiff as steel.
You can't fix this by making things thicker, they need to be made deeper in the load direction.
Find someone that designs in Al for a living.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[racookpe1978]

Aluminum corrosion factors in a salt and water environment are nasty problems.
Yes, not all Al parts will corrode quickly (exposure, fatigue, wear matter), and not all Al alloys corrode the same way.
But they all corrode differently than the practices and allowances and assumptions that would be correct for steel members.

 

[Agent666]

In our local Aluminium design code (NZS1664), depending on the fundamental type of load (compression/tension/shear/bearing) being considered in an element there is often a difference in the basic design yield stress of 5-25MPa between tension and compression yield strengths for example that might be used directly in design equations for a given 'grade' of material. For some aluminium grades this value is less than half the typical strength of steel, so simply aiming for a conservative 0.5 capacity factor can be flawed depending on the grade and temper of the aluminium.

However, many of the fundamentals with respect to design share many similarities with steel design, but they are subtly different in some aspects and the formulas for most limit states take quite a different form to their steel counterparts.

Aluminium stress strain curve follows a Ramberg Osgood relationship, there is no yield plateau like structural steel. So design values are usually limited to the 0.2% proof strains to ensure there is limited permanent non-recoverable deformations.

Al is only half as stiff as steel

Aluminium is generally around E = 70GPa, so its considerably less than half the stiffness of steel at E = 200GPa, i.e. a ratio of ~ 35% of steel stiffness.

 

[GC_Hopi]

In agreement that the design should be done to code, FBC, which means Aluminum Design Manual as a referenced code. I have witnessed aluminum structures that have failed at the welds due to hurricane force winds. Its not pretty.

 

[glass99]

Yes you should get calculations ADG calculations, but assuming they are not complete idiots and know that aluminum is weaker/more flexible than steel, it would probably be ok with both codes. Compressive stresses in aluminum are often limited by local buckling formualae in the ADG, but they only matter when you have slender sections. In general, the ADG has a factor of safety of 1.65 vs 1.67 in steel.

 

[Earth314159]

I am not American so I don't know your codes. However, I have done both steel and aluminum design and it is a big mistake to equate the two. There is a plethora of aluminum grades and you have to be careful which grade you use and how you use it. Many grades depend on work hardening for strength and as soon as you weld the aluminum you loose the strength. If you weld near the maximum moment, you can loose over half your strength (depending on the grade). 6061-T6 goes from 240 MPa to 110 MPa. There are also many other issues. You get into higher classes with lower aspect ratios since the E is so much less. All the instability/buckling equations (local and gross buckling) are different. There are also bi-metallic, concrete, and PT wood separation issues. The welding electrode also has to be properly matched with the grade.