Page 9 - Modern Steel Construction (April 2019)

If you’ve ever asked yourself “Why?” about something

related to structural steel design or construction, Modern steel

Steel’s monthly Steel Interchange is for you! Send your interchange

questions or comments to solutions@aisc.org.

All AISC Design Guides mentioned can be found at www.aisc.org/dg, Local buckling is not a consideration for connection elements
and all Engineering Journal papers can be found at www.aisc.org/ej. with a rectangular cross section. This is because the buck-
All other AISC publications, unless noted otherwise, refer to the current led shape is characterized by lateral translation and twisting,
version and are available at www.aisc.org/specifications. which defines the lateral-torsional buckling limit state. This was
addressed in the fourth quarter 2016 AISC Engineering Journal
Field Verification and Flange paper “Stability of Rectangular Connection Elements,” which
Thickness Tolerance shows that rectangular connection elements subjected to flexure
Structural engineers are commonly asked to confirm that can be designed using AISC Specification Section F11. Only the
an existing structure can support a new rooftop unit, but we lateral-torsional buckling limit state is required for these ele-
often don't have the original drawings and must make field ments. An example of this is in Part 9 of the 15th Edition Manual,
measurements to determine the member sizes. Sometimes where Specification Section F11 (with a modified C b factor) is
the flange thickness does not precisely correspond to the used to evaluate the flexural strength of double-coped beams.
flange thicknesses provided in the AISC Steel Construction Extended single-plate connections in Manual Part 10 use the
Manual. Is there a mill tolerance for flange thickness? same design procedure. You can find further information on this
in the first quarter 2014 AISC Engineering Journal paper “Local
Stability of Double – Coped Beams.” As discussed in the Manual,
No, there is no mill tolerance for beam flange thickness. These
dimensions are indirectly controlled through weight and area tol- the C b factors in Part 9 are only for cases where the plate is
erances. In ASTM A6 Annex A2, footnote “A” under flange thick- braced at both ends. You will need to use your judgment to deter-
ness states: “Actual flange and web thicknesses vary due to mill mine C b for other situations.
rolling practices; however, permitted variations for such dimen- Bo Dowswell, PE, PhD
sions are not addressed.”
It is acceptable to calculate cross-sectional properties based Vibration and Low Natural Frequency
on field measurements. This is stated in Section 5.3.1 in the I am currently designing a long plate girder supporting an
AISC Specification for Structural Steel Buildings (ANSI/AISC 360), office space. I computed and checked the floor vibrations
which states: “All dimensions used in the evaluation, such as under walking excitation as per AISC Design Guide 11: Vibra-
spans, column heights, member spacing, bracing locations, cross- tions of Steel-Framed Structural Systems Due to Human Activity.
section dimensions, thicknesses and connection details, shall be I am getting a very low acceleration of less than 0.1% of g, but
determined from a field survey. Alternatively, when available, it is the natural frequency is approximately 1.95 Hz, which is lower
permitted to determine such dimensions from applicable project than the minimum 3-Hz requirement. Can you advise on the
design or shop drawings with field verification of critical values.” applicability the natural frequency limit?
Also note that AISC Design Guide 15: Rehabilitation and Retro-
fit also provides tables of dimensional and material properties for In my own designs, I try to avoid natural frequencies below
historic steel shapes that would be helpful. The design guide also 3 Hz. Some of the most dramatic problems I have seen have
contains example problems as well as guidance for handling other been when the natural frequency was below that limit. In those
design or analysis nuances that arise when dealing with historic cases, rhythmic group loading caused extremely high accel-
steel structures. erations. In your case, the 1.95-Hz natural frequency is at the
Jonathan Tavarez, PE average step frequency of humans, so I would expect resonance
on a regular basis.
Local Buckling of Rectangular Bars If your system must have a natural frequency below 3 Hz, then
I am looking for guidance on designing a connection plate I recommend the following approach using Design Guide 11:
subject to flexure. AISC Specification Section F11 addresses Evaluate the floor for walking using a modified form of Equa-
the design of rectangular bars. Yielding and lateral-torsional tion 4-1 that accounts for the first harmonic of the walking force
buckling are the only limit states considered. Isn’t local and a higher R factor.
buckling also a potential failure mode? Are there any guide- Evaluate the floor for vandal jumping using the Normal Jump-
lines or current research on how to evaluate local buckling ing category in Table 1-2, along with the equations in Chapter 5.
of a plate subjected to non-uniform flexural stresses? Brad Davis, SE, PhD

Modern Steel Construction | 9