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

RE: How about a private listservice for

[Subject Prev][Subject Next][Thread Prev][Thread Next]

Several painful occurrences in my professional career have forced me to
produce design calculations that can be followed and understood.  I have had
my calculations reviewed internally, reviewed by clients such as the Corps
of Engineers, NRC, peer reviews, and reviewed by municipalities all over the
U.S.  It became essential to create a logic path for the reviewers to
follow.  I also like to use MathCad, and use the automatic page number
generation (I am lazy).  Another factor is the size of the projects that I
work on require calculations by several structural engineers that need to be
knit together.

Using MathCad templates has required me to develop 2 sets of templates, a
general set and a project specific set.  I always start with the general set
that includes every variant that I have ever run across with all the
annotations that allow me to use it over and over again (I am long on
forget, and short on remember).  The annotations are where I list the logic
and input comments from past reviewers.  When I perform a calculation for a
specific project, I delete what does not apply to that particular project.
If I run through additional checks (like crack width calculations) I save
that section in the general set as well as on my project specific set of
templates.  It takes a bit more time on the first few calculations, but I
save time on the many calculations that follow.

What I found was that a good table of contents is easy and essential.  It
forces me to organize my design methodology.  I organize it similar to the
	1.	General (project identification, scope, applicable
references and codes, design material properties, Basic structural system
assumptions (shear walls, moment frames etc.).  	Genxx.xx
	2.	Load assumptions	
		2.1	Seismic 	LdEQxx.xx
		2.2	Wind		LdWndxx.xx
		2.3	Live Loads	LdLLxx.xx	
	3.	Framing members (hor.)
		3.1	Floor Beams  	Bm1stFlxx.xx, Bm2ndxx.xx, etc.
		3.2	Roof Beams	BmRfxx.xx
		3.3	Girts		
	4.	Columns
	5.	Braced frames
	6.	Slabs
	7.	Foundations	
	8.	Masonry Partitions
	9.	Misc.
	Appendix (Computer runs, photos, field surveys, etc.)  

I use alphanumerics to create a logical prefix on the page numbers.  If I
shuffle the order of the major topics in the final assembly, it is not a big
deal; I just change the table of contents page.  The page number system also
allows me to insert calculations without changing all of the page numbers in
the set.  Many reviewers have a lot of heart burn when I use spreadsheet
templates unless I have a lengthy verification.   I generally use
spreadsheet templates to get the most economical solution, and then use a
MathCad template to develop the final design calculations.  This approach is
generally acceptable to reviewers.  I also ask the reviewing authorities in
advance for a list of  software that is acceptable.  

This is just my humble attempt to create order.  Are there any other more
well developed systems out there?

Harold Sprague

> -----Original Message-----
> From:	SEConsultant [SMTP:seconsultant(--nospam--at)]
> Sent:	Thursday, March 02, 2000 4:13 PM
> To:	seaint(--nospam--at)
> Subject:	RE: How about a private listservice for
> Roger,
> I find it hard to believe, based on your extensive knowledge and level of
> expectation that you could ever have produced a set of calculations that
> would be anything but perfectly organized. I am not being facetious. It is
> apparent that you have exceedingly high standards and generally this
> beings
> from the very basics.
> I am pleased to say that I learned early in the game that the greater care
> you take in the presentation of your work the less likely you are to have
> plan check corrections.
> This is what surprises me most from some of the submittals I receive. They
> are engineers known to me who have very strong reputations but whose work
> I
> find very difficult to interpret. As I stated, it can take a few hours
> which
> I believe can be avoided in order to work through a calculation package
> and
> set of drawings before the intention of the EOR or designer becomes
> apparent.
> If anything, we should be offering classes or continuing education credit
> in
> presentation of work.
> Another problem I have is when an engineer deviates from the conventional
> methodology to use another approach which, although acceptable, is not
> well
> known to the plan checker. I had this situation when I reviewed plans from
> an engineer outside my area who I knew well. It was a retrofit of a URM
> building. Rather than follow the typical UCBC methodology, he chose to
> design by ultimate strength method. This is perfectly fine, but I had been
> designing retrofits of over 300 buildings in my time and never used this
> approach. I must conceded that I ended spending hours reviewing his
> package
> to learn what he had done and see that he achieved the same end as I would
> with a working stress method.
> After working as both an independent engineer and a contract plan checker
> I
> quickly gained respect for the hardship a plan checker faces every time he
> opens a new set of drawings and calc's to review. In many ways it is very
> much like a puzzle that takes time for all the pieces to fall into place.
> Going back to the issue of presentation, how do other plan checkers deal
> with calculations which are not supported by sufficient commentary or hand
> justifications? If it is not an commercial software, are you justified to
> ask for sample calculations to show the accuracy of the program? Or, do
> you
> assume that the engineer is responsible for getting the math right and we
> as
> plan reviewer only interested in assuring the detailing is correct?
> Opinions????
> Dennis