Unbraced Length of Continuous Steel Beam

[lutein]

I was wondering what have you all used for unbraced length for continuous steel beams over steel columns with joist bearing on t/beam at 5'-0" o.c.?
Presuming only gravity case, i.e. no uplift. Could we utilize the inflection point? in other words, for the beam with positive moment - Use the joist spacing as the unbraced length?

Your valuable input would be very much appreciated.

Thanks in advance.

 

[271828]

If you're using the AISC 3rd Ed. LRFD Manual or 13th Ed. LRFD/ASD Manual, the inflection point cannot be used as a braced point. Check out Appendix 6 of the 13th Ed. Manual. You're stuck with using the entire girder length unless you have some bottom flange braces somewhere along the span.

Note that, in reality, you should (but currently can't) be able to use the entire length but then calculate a whopper Cb value because the top flange is more or less continually braced. There are Cb equations out there that take this into account. The Spec. does not currently say anything that allows one to use such an equation. Hopefully, it will be there next time.

 

[JAE]

Could we utilize the inflection point?

Not according to AISC. Use the full span length and calculate the correct Cb ratio for that span.

Here's a couple of previous threads on the topic:

thread507-189712

thread507-148451

 

[lutein]

271828 - Thanks for your reply. what you described is actually what I have been doing. however, I am currently working on an analysis of existing structure, and by using this current code will make most roof beams fail. I am currently researching to see if there is anyway I could use inflection point.

Thanks for your quick response.

 

[lutein]

Thanks to both 271828 and JAE. I studied previous 2 forums as reminded by JAE.

271828 - You mentioned about Yura and Helwig's method in Cd calculation, do you know where could I find such equation? Could you please provide me more information on that?

Thanks.

 

[DaveAtkins]

I agree with everything that has been posted. However, Salmon and Johnson indicate in their steel textbook that if the Cb factor is NOT used, the inflection point can be used as a brace point. You would not assume the beam is braced at 5' oc--you would assume it is unbraced from the tip of the cantilever to the point of inflection.

DaveAtkins

 

[jike]

Many years ago, it was fairly common practice to assume a braced point as the inflection point, but it is no longer that way since the code has been more specific. It is not unusual to find existing buildings with this type of construction that no longer check out.

 

[271828]

Dave, I think that worked out only because Cb was huge, perhaps up to about 4-5 for such beams. Then, using the wrong unbraced length (like everybody did pre-2001) and no Cb kinda cancelled out. Pretty haphazard way to do it, though.

lutein, the Cb equation is in the Yura/Helwig seminar notes. There are various versions of these, but I suspect it's in most or all of them. It *might* be in the latest SSRC Guide also.

 

[JAE]

I sat in a seminar some years ago in Texas with Yura and raised my hand and asked him if using column-to-inflection point as the unbraced length with Cb = 1.0 was OK.

He thought for a moment and then said "yes".

However, two years later, Yura had developed more research and a new Cb formula and an engineer I knew had asked the same question and gotten a "no" for a response from Yura. So the thinking then (about 1985) was that inflection points should not be used as a brace point.

In the past we did use inflection point distance, but added about 20% to that length with Cb = 1.0 so we rationalized that doing it that way was OK...but as 271828 says, "pretty haphazard" with no reasearch or logic to back it up.

 

[DaveAtkins]

I agree, JAE--not the right way to design. But in this case (existing condition), good old engineering judgment is needed.

Related to this issue, lutein, are there stiffener plates each side of the beam where it passes over the column? I read an article almost 20 years ago about a roof collapse, due to buckling of a beam flange where it passed over a column. Ever since I read that article, I ALWAYS show stiffener plates above the column in this type of construction (even if the AISC formulas say you don't need stiffener plates). So, if you are reinforcing the roof structure anyway, I would add plates if they are not already there.

DaveAtkins

 

[csd72]

The fact that it has no moment there has nothing to do with being braced. It definately is not braced.

Besides, the critical flange for a cantilever can sometimes be the tension flange i.e. the same flange as the critical flange for the span!

 

[lutein]

Thank you all for your valuable input. I deeply appreciate it and I am glad we are helping each other in this industry.

After my research on some textbooks, I will specify on drawings to extend the joist bottom chord to brace the beam at several points - defenitely on top of column, and intermediate.

 

[GalileoG]

Okay, can someone please explain to me or provide me with a reference explaining what the relationship is between inflection points and bracing? Is it preferable to brace where there is an inflection point, or? I was not able to find anything about this in the few texts that I have. Thank you!

 

[271828]

Clansman, check out Appendix 6 of the 13th Ed. Manual. It answers your question.

In particular, read App. 6.3 and 6.3.1.

 

[hippo11]

There is a good article about this called

"Design Guideline for Continuous Beams Supporting Steel Joist Roof Structures" by James Rongoe...available at

http://www.aisc.org

to Members...in the ePubs sections