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

Re: More problems with the 1997 UBC

[Subject Prev][Subject Next][Thread Prev][Thread Next]
This message responds to a few points from Rick Drake's email:  First of
all, Rick, thank you very much for taking the time to respond to my
concerns.

Quotes from Rick's email are shown with a preceding ">".

> From: Rick.Drake(--nospam--at)fluor.com
> To: seaint(--nospam--at)seaint.org
> Subject: Re: More problems with the 1997 UBC
> 
> Gary:
> 
> Your understanding of the 1997 UBC provisions for your 
> equipment hung from a concrete slab and supported by vibration isolation
springs is 
> essentially correct.  This is a dynamic problem involving 
> flexibly-supported equipment at the roof of a flexible building.
> A few thoughts that might help your situation.
> 
> 1) If you want one design that will work on all floors, you 
> have to use the roof value and must expect the design force to be higher.
If you 
> provide two or three designs for different elevations within your
building, 
> you can use Formula 32-2 to provide lower design forces lower in the
building.

If you check my email, I did exactly that.  Even if I am installing
equipment at midheight in the building, the supposedly rational equation
32-2 gives me a higher value than the simplified Formulas 32-1 or 32-3.
Even if I break the design into three separate designs, the top two-thirds
of my building are governed by Formulas 32-1 and 32-3.

> 2) Section 1632 allows you to rationally evaluate the value 
> of a-sub-p.  If you know the fundamental frequency of the building , and
the 
> fundamental frequency and damping ratio for your vibration-isolated
equipment, you 
> may be able to reduce a-sub-p from 2.5.

Unfortunately, the fundamental period of the building is unknown since I do
not have a computer model of the building (unless using Method A to compute
the period is acceptable, which it probably should not be); I do not know
the fundamental period of the equipment; and I also don't know the damping
ratio.

> 3) Expansion anchors have a history of "failure" during 
> earthquakes and deserve the design force penalty.  You don't seem to be
considering 
> undercut anchors in your evaluation.  They are very good for tension
applications.

A number of people have suggested to me that the penalty for expansion
anchors may be undeserved.  A number of people have also suggested undercut
anchors.  However, I looked up Hilti undercut anchors and it appears as if
they still expand and put some tension into the concrete, though
significantly less than a typical expansion anchor.  In either case, this is
a shallow concrete deck and any anchors I put into the concrete are going to
be shallow, so the mandatory increase in force is still required.

> 4) Even in your worst case evaluation, your seismic design force is only
3.02
> Kips.  This doesn't seem like a very high number to design to.  If you are
using
> structural steel supports, your minimum AISC ASD connection load is 6
kips.  As
> for the anchorage to the concrete, a couple of undercut anchors will
easily take
> the load.  And you may need more than two anchors anyway because of the
geometry
> of your details.

I see your point for small items.  However, I actually intended my example
to be more generic than just this single example.  If the equipment weighed
8 kips, the design forces would be ten times larger and then we would be
talking about a significant lateral design load of 32 kips.

My point is more along these lines:

If the 1997 UBC design forces are four to six times larger than the 1994 UBC
forces, then that implies that we (as structural engineers) have been
producing designs which are significantly and seriously unconservative for
the past thirty years.  Even moderate earthquakes should be causing
tremendous damage to equipment which has been designed and installed
according to past building codes.  I do not believe that this is what has
been observed in recent earthquakes.  I know we sometimes are shown images
of damaged equipment, but most of the time, the equipment was not even
braced in the first place.  Where are the analyses and field observations
which look at equipment which have been through earthquakes, check if the
equipment was designed and installed correctly according the original
governing code, and then show whether or not the equipment actually failed
in earthquakes?

We know that roof accelerations tend to be much higher than the
accelerations experienced by the rest of the building due to higher mode
effects.  I would support increasing the 1994 UBC non-structural code
provisions by a modest amount for the roof only;  say 50% or 100%.  But to
increase the code design forces by 400% to 600% for over half of my example
building seems ridiculous.  With some larger pieces of equipment, the
calculated design overturning forces may become so large that I may not be
able to get my slab or floor beams to work.  Have we really seen equipment
causing floor slabs and floor beams to fail in earthquakes?  Have we really
seen massive amounts of equipment failures in moderate earthquakes?  Do we
really need a table 23-O which has almost as many footnotes as entries in
the table?  Do the design procedures have to be so complicated that it is
difficult for intelligent and educated engineers to arrive at similar design
loads?

These concerns form the basis of my criticism with the non-structural
provisions of the 1997 UBC code.

Gary R. Searer, P.E., S.E.  -  gsearer(--nospam--at)wje.com
Wiss, Janney, Elstner Associates, Inc.