Need a help with AL design, are my interpretations correct?

[dlclarkii]

Hi. I have the Aluminum Design Manual and Al Structures book as references. I'm trying to size AL 6061-T6 AL standard beams.

--------------------------------------------------------------------------------------------------------
S12x17.3
L=16'
Lb=16' (unbraced)

AL 6061-T6
Ftu=38 ksi
Fty=35ksi

Section 3.4.2/3.4.4
Fbt=19.5ksi (tension flange)
Fbt=28 ksi (tensino web)

Section 3.4.11 (Gross Compression)
Fbc=21 ksi (fully braced in weak axis)
Lb=16'
ry=1.03
S=Lb/ry=79 (S1=21, S2=79)
Fbc=2.50ksi

Section 3.4.15 (Flange Compression)
S=b/t=3.63 (S1=6.5, S2=10)
Fbcf=21ksi

Section 3.4.18 (Web Compression)
S=h/t=15.55 (S1=48, S2=75)
Fbcw=28ksi

Section 3.4.20 (Shear)
S=h/t=15.55 (S1=36, S2=64)
Fs=12ksi
--------------------------------------------------------------------------------------------------------

Assumption: Fallow=2.50ksi (minimum)

So I design based on an allowable stress (2.50 ksi)???? Or can I do a weighted average of the flange and web compression instead with the expection of getting a higher allowable design stress?

I don't design AL and I would appreciate a second opinion whether I'm interpreting the AL DM correctly. Thank you!

 

[dlclarkii]

Correction:

Section 3.4.11 (Gross Compression)
Fbc=21 ksi (fully braced in weak axis)
Lb=16'
ry=1.03
S=Lb/ry=186 (S1=21, S2=79)
Fbc=2.50ksi

 

[NHstructural]

dlclarkii,
I don't have the code handy to look over your list, but you can design/analyze aluminum beams in RISA. Doing so with an S12x17.3, spanning 16 ft, unbraced weak axis, I do not get such a low value as a limiter.

I do not think it is appropriate to do a weighted average, but perhaps you wouldn't need to if you can determine why RISA gives a better value than you are finding.

 

[rb1957]

averaging the different modes of failure doesn't make sense.

remember that the very low alloawable is for compression load on the beam. sanity check (i don't use your codes) ...
euler ... P = pi^2EI/L^2 ... s = pi^2*E/S^2 = 10*1E7/186^2 = 3 ksi ... so 2.5 ksi is "reasonable".
but remember, this is for axial compression. how much compression is in the beam ? wouldn't your allowable for bending be higher ?

different modes of failure apply against different loads.

 

[dlclarkii]

Thanks for the replies. I just don't know whether to use Fc from Section 3.4.11 (Gross Compression) OR the lesser Fc from Section 3.4.15 (Flange Compression)and Section 3.4.18 (Web Compression)? I don't have RISA or any other software to determine the allowable strength, just forumulas from the AL DM.

 

[dlclarkii]

edit: Based on 3.4.11, the single web shape is subject to lateral buckling (bent about the strong axis w/o continuous support) therefore I use the lesser of all three, so gross compression controls, thus Fc=2.5 ksi. Is this correct???

 

[rb1957]

is the member axially loaded in compression or is it being loaed in bending (so only a flange is in compression) ?

 

[dlclarkii]

Simple beam with a point load in the center. Bending only, no axial. Top flange is in compression.

 

[rb1957]

so then the gross compression allowable doesn't apply, yes?

and your allowable flange comression includes the effect of the unbraced length ?

 

[JoshPlumSE]

There is an alternative calculation allowed by section 3.4.11 that you may calculate Fc by replaying ry with the rye value given in section 4.9.

The use of ry instead of rye is conservative. When looking through my ADM, I circled the comment about allowing rye instead of ry with a hand written note that RISA uses rye. Therefore, that would seem to explain why the RISA calculation comes up with a higher allowable stress.

 

[rb1957]

what's the difference, ry <=> rye ?

 

[dlclarkii]

Rye=(Lb/1.2PI)*SQRT(Me/ESc). Section 4.9. For a loading that does not cause torsion or lateral bending a more accurate value of rye is determined according to this section. Otherwise revert back to 4.3.

Thanks

I download a trial copy of Digital Canal AL Design and it's computing the same S beams that I am with the AL DM. Risa must be using Rye.

 

[rb1957]

that looks odd to me (but then i don't use your code, don't know your terms). i thought ry was radius of gyration = sqrt(I/A); that's the way i used it in Euler above.

i'm thinking Sc is section modulus = I/y, so playing a little i get rye^2*E*Sc*1.44*pi^2 = Lb^2*Me, Me = 1.44*(pi^2*E*I/Lb^2)*rye^2/y, ... not sure where this is going (and don't Really care)