Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: steel - vibration analysis

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Well, I read non-orthogonal as more skewed than you did. I did forget that the 7.2 atrium girders got columns, but that's a girder problem - the beams could have been just about anything and the girder still would have been an issue. None of the triangular areas, nor the beams around the atrium appeared to require work.

Again, I read the previous comments as applying to areas with varying beam lengths - a special, limited case of non-orthogonal framing. The exaggeration of the problem was just that - non-prismatic would have been sufficient to describe the difference. Having spent most of my pre-building career in aerospace, I find vibrations mostly intuitive now (I say that knowing the analytical knowledge is locked somewhere in my mind, I just can't remember where). Most of my cases come in two flavors - simple enough to apply the DG11, which for me means regular orthogonal bays, or a job for a dynamic modeling package. Since I own neither RISA nor RAMSS, I can't really comment on either's implementation. I've found most building software analysis (limited though it may be) to be somewhat simplified for the user. That's a good thing for what we do, for the most part (I've incurred the wrath of the staff at RAM before for having a black box approach...I swear - that's a good thing in this case).

We do agree that there are cases where some non-orthogonal framing can result in a framing issue, but there are enough common conditions where it precludes the beam participation in the problem (most triangular bays in an otherwise normal grid, for example) I felt it necessary to chime in on Allen's behalf. For what it's worth, I think the discussion is good - if for no other reason than to get more engineers (who read the list) to remember the vibration issue during design.


Josh Plummer wrote:
Jordan -
1) Allen's message essentially said that it was impossible to have vibration
problems in non-orthognal areas of a building.  The main point of my
response was that this stance of his was patently absurd. I take it that you
at least partially agree with me on that.
2) Please take a closer look at that example in 7.2 of Design Guide-11. Your
characterization of that problem is COMPLETELY inaccurate. I'm not trying to
flame you here... I just think you didn't read the example very closely. The
text of that example actually says the following:
"Unacceptable walking vibrations occurred throughout most of the floor, more
so adjacent to the atrium".
When you look at the figure, the area adjacent to the atrium is the one with
the non-orthogonal framing.  Now, look closer at the skewed girders.  They
added in new COLUMNS at the mid-points of the girders.  That's a waaay more
significant stiffening effect than what they did to the other girders.
Personally, I'm confident that our RISAFloor program would produce results
which show the skewed area being more sensitive to vibration than the rest
of the floor.

3)The fact that you characterized non-orthogonal plate vibration as akin to
a "non-linear non-prismatic curved member" is a fairly gargantuan
exageration. Structural vibrations just aren't that complicated.  If it
were, then you'd need a PhD to understand modal analysis, even more so if
you were doing a Response Specta Analysis. I understand that some folks
(perhaps even you) can find vibrations to be a truly intimidating subject.
But, we're talking about common framing configurations with girder to joist
angles of less than 90 degrees. This just isn't that big of a difference.
We're not talking about extending the rectangular panel mode concepts to one
of those freaky Frank Gehry buildings that look like they're melting.

Like you pointed out, a true modal analysis (which RISAFloor can do through
it's interaction with RISA-3D) can be appropriate for many of the funkier
non-orthogonal cases (like the Gehry buildings).  But, doing that for a
nearly orthogonal bay would be a shocking amount of extra work for something
that is almost certainly going to produce the same kind of results as the
DG-11 calculations.

Finally, I said before that I thought that I had read every article that
AISC has published on Floor vibrations. And, I still haven't seen anything
in any of their articles that implies that non-orthogonal areas are immune
from floor vibration problems.  Nor have I seen any warnings or cautions
about extending these concepts into systems that are not perfectly
orthogonal.   But, if I'm missing something important here, then I'd like to
know about it.

Sincerely, Josh Plummer, SE
RISA Technologies

-----Original Message-----
From: Jordan Truesdell, PE [mailto:seaint1(--nospam--at)] Sent: Tuesday, June 12, 2007 4:42 AM
To: seaint(--nospam--at)
Subject: Re: steel - vibration analysis

Josh - when I read  Allen's post, I generally agreed with it, though I may
have misread what I wanted to read.  Where framing changes length within a
bay, it is correct that the variation in frequency tends to reduce or
eliminate footfall vibrations because of the large number of different
vibration modes at play - there is no large area in which the vibrations may
be sustained.

The example in 7.2 is also a very regular structure, with the diagonal areas
both on the perimeter and around the atrium apparently not affected by the
vibration problems - the primary areas of stiffening were in back-to-back
regular orthogonal bays.  (I only skimmed the article to see what you were
referring to, let me know if I missed some of it) You can still have issues
in non-orthogonal bays if the primary frequency is related to the girder,
but the joists/beams should tend not to promote resonance unless they are
all of similar stiffness, or all fall in the lousy range.
Vibrations of plates and shells is greatly complicated by non-regular
geometry, and can effectively only be performed numerically.  If you have an
issue with an unusual structure, the first thing to do (barring obvious
deduction) is to determine your mode shapes, frequencies, and modal mass
participation - then you can start to isolate where the issues are. Your
comparison of extending the vibration theory from wide flange to tube is not
analogous- extending from a wide flange to a non-linear non-prismatic curved
member would be more appropriate. And, of course, the theory doesn't extend
to such members.

(now where did I put that set of nomex khakis...)


******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
* * This email was sent to you via Structural Engineers * Association of Southern California (SEAOSC) server. To * subscribe (no fee) or UnSubscribe, please go to:
* Questions to seaint-ad(--nospam--at) Remember, any email you * send to the list is public domain and may be re-posted * without your permission. Make sure you visit our web * site at: ******* ****** ****** ****** ******* ****** ****** ********

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
* * This email was sent to you via Structural Engineers * Association of Southern California (SEAOSC) server. To * subscribe (no fee) or UnSubscribe, please go to:
* Questions to seaint-ad(--nospam--at) Remember, any email you * send to the list is public domain and may be re-posted * without your permission. Make sure you visit our web * site at: ******* ****** ****** ****** ******* ****** ****** ********