Need a book? Engineering books recommendations...

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

Pendulum response

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Bill,
I thought you disappeared as I had written you three emails regarding the '98 
UBC Wood Seminar Examples which you created and received no responses. I'll 
get to those in a minute:

The main problem that I see with your comments is that the suggested 
methodology is at best an approximatation of the expected performance (worst 
case) of a plywood diaphagm. As your design example suggests, a comparision 
between flexible and rigid values (and I assume wind loads) should be made at 
each line of shear and the worst condition applied. This would mean that in 
one line of shear, the walls will be designed to resist loads resulting from 
wind, torson+seismic reactions or proportional distribution of forces through 
a flexible diaphragm. The stiffness of the resisting walls will not be 
balanced from grid line to grid line on the true performance of the diaprhagm 
(which is only an approximation) but upon the stiffness of the worst 
condition - the unknown. In otherwords, nobody really knows how the diaprhagm 
will perform so let's assume the worst. This is a very complicated 
methodology for such a vauge understanding of a true plywood diaphragm.

My point is that there is no way to "prove" deformation compatibility since 
we can only approximate the performance of the diaphragm and in no consistent 
way at that.

If I was hog-tied to accept the concept of a truely rigid diaphragm, I would 
accept it only when the system is as close to homogeneous as possible - 
thereby, flat (less than 2:12 pitch with a glued diaphragm and tight 
blocking. In otherwords, I am more apt to accept the analogy of a rigid 
diaphragm in a floor system than I am in a roof. One major reason for my lack 
of confidence is the lack of emperical data available to the professional 
community on various residential models consisting of gabled roofs, scissor 
trusses, vaulted ceilings, hips, valley's etc. If it has been done, it has 
never been brought to the attention of this List during our months of 
discussions and debates with those who were involved.  Even a low sloped roof 
can be designed with skipped sheathing or with OSB that requires a 1/8" 
expansion gap between panels creating an questionable rigidity.

Getting specific, I believe the exception for embedded columns should relate 
to all residential structures with the exception of those of 
multi-residential design with sub-terranian parking or a lack of redundant 
partitions at the first or lower level. 

My next question has to do with the design examples. In all due respect, the 
examples assume engineers have an understanding of the analysis for Center of 
Mass and Center of Rotation. Therefore, the example lacks an explanation as 
to where the dimensional values originate. The plans and elevations are not 
keyed to the numbers used in the design and often, units are ignored. This 
requires the engineer to learn a new method (such as the majority of us that 
restrict our practices to wood framing and have historically done little with 
rigid diaprhgams) and to waste time trying to justify numbers.
Take, for example, the conditions where the individual lateral loads are 
calculated at the roof and floor. The confusion comes from trying to 
determine the dimensions to the center of each uniform load - which is not 
indicated on the sketch (only the grid dimentions are noted). 
Another example assumes the value for hn as 20' yet it is not explained that 
this is the height from top of foundation or finished slab to top plate at 
roof. The elevations provided do not indicate this as the second floor 
framing is not dimensioned to assume a plate height of 10' at each level. 
In the analysis for Base Shear, your example assumes a Soil Profile type as 
Sc yet does not explain how this value was determined. The code is clear that 
if the value does not originate from a soils report, the Profile type shall 
be assumed as Sd. 

The Diaphragm Shear Stress provide formula's that I am sure we all should 
know. However, the one thing drumed into my head as a student was to show all 
of my units. In this example units were omitted which forces the engineer to 
backtrack to determine whether the number represents feet or pounds.

There are many other examples that do not clearly explain how the results 
were determined.

My point of this email is to suggest a means to help engineers who are 
struggling through this new methodology as our clients are losing patience 
(most of them think that it is clearly defined and easy to follow in the code 
- if only they knew). 
Residential construction is rarely as simple as the design examples in the 
'98 seminar or the examples (which I believe originated from yours) in the 
ICBO Design Guide Vol II (which I was one of the engineers who volunteered to 
checked problem number 2). The discussions that we hold on this list demand 
more informed answers for the various conditions that we face. 
I understand that Bill Warren has suggested a Website and/or list for those 
who attended the SEAOC seminar to be able to discuss and receive advise on 
interpretaion of the methods. However, I think it is important if members of 
seismology committee or Wood committee who were involved in the creation of 
the methods would avail themselves to actively field the questions arising on 
the SEAINT Listservice.
At this time, we are not only debating issues between ourselves, but stand 
the prospects of having to argue our cases with the plan checker who may or 
may not be as informed as we are. 
I was very active in the practice of the RGA for URM buildings in the late 
80's and understand how time consuming and often frustrating it was to debate 
the methodology at each plan check session and then to start from scratch 
each time a new plan checker came to work at the EQ division. 
I don't think this is practical for the amount of work hinging on acceptance 
of this new code. Therefore it becomes imparative that better communication 
is maintained between those who created the code and those who much design by 
it. Attending the seminars is simply not sufficient to understand and design 
practically.

Sincerely,
Dennis Wish PE