Need a book? Engineering books recommendations...

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

Re: Structure Magazine Questions

[Subject Prev][Subject Next][Thread Prev][Thread Next]
I understood your remarks in your earlier post, and really do not take
any exception to them.  I was kind of trying to open a new thread on
this issue.  Since Bill Allen no longer posts on this list, we don't
always get many controversial issues to discuss anymore.

Why is it that we do not use the actual accelerations of a building when
we do overturning design?

Why is it that a steel frame building using ordinary moment frames will
have a different overturning moment than a special moment frame
building?  It does not make any sense to me.  

>From a quick first look, you would think that for overturning forces, Rw
should always be 1.  

I think we do not really understand how this works, or we have not
really gone into it in depth and  modified our design based on the
actual performance of the building.  I think our current design approach
is conservative.  I just think people should recognize the conservative
Code design requirements for what they are.  

I was also hoping for some people to call me to task on this issue and
point out where I might be wrong.  I really thought more people would
jump in and challenge me on this issue. I thought it would generate some
interesting discussions.


Mark K Gilligan wrote:

> Lynn:
> I was trying to make the point that the earthquake forces on retaining
> walls from the soil are calculated differently than those for buildings.
> In the former case the forces are an approximation of the actual forces the
> structure will see, while in the later case the calculated seismic forces,
> as specified by the code, are often less than the forces the structure has
> to resist. My point was also that in the first case a factor of safety less
> that 1.5 for seismic loads may be appropriate while in the second case
> designing the retaining wall for the code forces would overestimate the
> factor of safety of the wall.
> I was not suggesting that we design the shear wall or the retaining wall
> for an Rw=1.0, but rather that we design the retaining wall for the forces
> that can be transmitted to it from the rest of the structure.
> The use of an Rw>1.0 for the super structure is justified because of the
> ductility that we have designed into specific elements.  Unless the
> retaining wall is able to exhibit some ductility and still remain stable,
> then it needs to be designed for the forces that can be transmitted to it.
> This statement is consistent with the rest of the code but because it is
> not explicitly stated in the code, many people do not follow it, and
> instead just design for the code forces.
> We have to be careful with statements such as; "Retaining walls which are
> properly design for static loading conditions do not fail in overturning
> during earthquakes".  This sort of thinking has lead us to be complacent
> about some aspects of seismic design only to be rudely reminded of the laws
> of physics when an earthquake occurs.   Some of the problems with making
> decisions solely based on previous earthquake experience arise from changes
> in design and construction practices, and because the past earthquakes may
> hot have been all that large.
> The reason why most retaining walls do not fail in earthquakes was
> addressed in the second part of my posting.  If the only earthquake forces
> on the retaining wall are from the soil mass behind the wall, there is
> rarely a failure.   The problem is that the original question did not make
> it clear how the earthquake forces were computed.
> On the other hand, I have seen designs that rely on a cantilever retaining
> wall to resist building seismic forces acting normal to the face of the
> wall.  Admittedly, this is not a normal situation so it is not surprising
> that few such failures have been reported, but that is not to say that this
> failure mode cannot occur.
> Mark Gilligan