Monorail beam Continuous or Simply supported

[McGill10]

Hello,
I am designing a monorail beam that is supported by roof beams. As usual, the monorail is attached to the bottom flange of the roof beams and trolley runs on the bottom flange of monorail.

Here top flange is braced at supported points but the bottom flange is not braced. But due to continuity, negative moment develops at support regions and bottom flange will be in compression and no lateral torsional bracing.

I have question about how to you handle such situation. I see people design monorails as simply supported beam and don't worry about bottom flange bracing.But I want to understand the mechanism and find the best solution.

thanks
Mcgill

 

[sponton]

Well you can always argue that usage of the distance between inflection points as the bracing length is correct, since compression would only be continuous through that span, plus it wouldn't be that big of a deal considering the fact that it's not constant [which is how the formulas were derived so cb>1], i think other thing to take into consideration is the fact that the beam is loaded through the bottom flange which affects the lateral buckling stability.

designing it as a simple beam means that you have a larger moment to design for [rather than the reduction in moment you get from having a continuous beam], so it's a bit conservative and genuinely, you have to have supports spaced at an incredible distance to have it fail by lateral torsional buckling of the bottom flange. since you would use either the distance between the inflection points or the real bracing distance [whatever is larger]

 

[McGill10]

Typical spans being 20 ft and total of 5 spans = 100ft long run.As bottom flange can be braced only at the ends which puts Lb =100 ft.
I understand using inflection point as braced points. But as far as I understand, per AISC, inflection points are not considered as braced point.

 

[TehMightyEngineer]

I can confirm, AISC does not let you consider inflection points to be bracing points (with good reason).

One thing to remember about monorails with underhung trolleys is they are self-limiting to a degree in regards to lateral-torsional buckling. When the top flange is braced, if the bottom flange starts to roll out of plane then, by virtue of the load being below the centroid of the beam, the bottom flange can only deflect so far before the eccentricity on the load holds it back. Thus, I've typically designed monorails where braced top flanges are considered braced points for the beam. If you have an exceptionally slender beam or something like that then half depth stiffeners may allow your trolley to pass while still providing enough bracing to the section.

All things equal I would go simple span so that you don't have to deal with splicing members without interfering with the trolley run. It will also prevent thermal shrinkage/expansion from causing issues on such a long beam.

Another thing to consider is deflection. Monorails can easily be deflection controlled and having those continuous spans may help a lot with deflection.

Lastly, make sure you check flange bending. This can often control as the wheel loads on the edges of the flange can cause some extreme loads.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

http://www.americanconcrete.com

 

[sponton]

Well it doesn't make sense using lb=100 ft since the beam isn't subjected to compression through all the span. In any case is it possible to weld a stiffener connecting the flanges such that individual lateral deflection of the flanges is constrained? That given you could use the same unbraced length can be used for both conditions. Salmon's book mentions stiffeners as a means of preventing lateral buckling and it also mentions using inflection points as unbraced lengths which I am almost certain isn't prohibited by aisc. Let me check some stuff I had regarding the topic.

 

[sponton]

From an old thread on the topic

JAE (Structural)15 Feb 13 12:34
BA - In the "old" days we used to use inflection points as "brace points" and for negative moment checks we used Lb = column-to-inflection distance and then kicked that up by a factor of 1.2 just for feel-good.

I sat in a seminar given by Yura some years ago in Texas and asked the question - "can we use that IP to column distance for negative moment unbraced length". He paused a minute and then said yes - as long as you use Cb = 1.0. Within a couple of weeks after that - I learned that he back-tracked on that and he and others spent efforts revising the Cb formula to its current form we see in the AISC spec today.

Based on that - the current "correct" method is to use the full span for Lb for negative moment checks and calculate a Cb ratio for that particular check....with the Cb using moments at quarter points between points of bracing. The Cb factor does get fairly significant and appears to work fairly well but it is sort of counter-intuitive when you see that the compression flange typically jumps up to the top where you have joists resisting lateral translation.


So I guess the cb value should indeed take into consideration continuous beams and offset the taking of the whole thing as unbraced length.

 

[TehMightyEngineer]

Yeah, sponton brings up a good point. Ignoring Cb in this case can be a grossly conservative assumption. Cb for a monorail can have a big impact on the lateral-torsional bucking failure mode.

To expand on the inflection point bracing:

AISC 360-10, App 6.3: "In members subject to double curvature bending, the inflection point shall not be considered a braced point unless bracing is provided at that location."

There is also addition bracing strength requirements for braces near inflection points per appendix 6.3.1a and bracing at inflection points must brace both the bottom and top flange.

The commentary lists the research by Yura (2001) and Ziemian (2010) that sponton references above.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

http://www.americanconcrete.com

 

[McGill10]

Thanks for pointing out the exact section of AISC, which rules out the use of inflection point as braced point.

As I read the code more deeply,I find two types of beam bracing as lateral bracing and torsional bracing.In our situation, even though, the bottom flange is laterally unbraced, it is torsionally braced at every support point.Per AISC App 6.3 2a- "it is permitted to attach torsional bracing at any cross-sectional location, and it need not be attached near the compression flange".

Now I feel comfortable using the beam span as unbraced length for both flanges.Also, take advantage of Cb as much as we can.