Need a book? Engineering books recommendations...

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

Re: Technology and Structural Engineering

[Subject Prev][Subject Next][Thread Prev][Thread Next]
        You finally prodded me into a response on this topic!
        I worked on my first 25 storey building in the 1960s.  At that time the big problem was to find a reasonable way to share lateral loads between shear walls and the building frames.
        You talk of recent graduates designing 100 storey buildings with 100 load cases; they can only do this with canned software!  I was talking with an engineer recently (I believe his name is Dr. Leroy Empkin, or something similar, and he sells GTSTRUDL software) and he was telling me that some of the universities he visits teach the use of software packages as a substitute for actual structural engineering and analysis theory.  I wonder how many, if any, of these recent graduates could actually design a transfer beam to carry an apartment building over a parking structure!
        I suspect that some of the "engineers" actually using the software have no idea how it works.  They have no idea whether the program automatically adjusts the "end offsets" to top justify the floor beams or whether they should do it manually or even whether it's important.  They just use the program and the design must be right because "the computer did it".  In a way this is nothing more or less than a variation of plan stamping and it scares the hell out of me.
        On the positive side the modern technology allows us to do things that were completely impractical a few years ago.  Just as an example, about 1990 I had some structures to design that required extremely refined analyses.  The structures were four column towers twenty feet square with columns four feet square and about sixty or eighty feet tall constructed of precast concrete.  There were three structures of varying heights in all.  The horizontal elements were also four feet square and twelve to sixteen feet apart.  The lateral loading consisted of wind plus temperature loading from large diameter process piping.  The programs available were an earlier version of STAAD (without P-delta capability) and Lotus 123.  The complicating considerations: extremely small deflection limit; P-delta analysis was required; and it was necessary to determine analytically which parts of which elements were cracked and which were uncracked and to include this in the analyses.  The procedure in 1990 was run a STAAD analysis, export the output to Lotus 123 to calculate the P-delta loads and the new section properties to be imported back into STAAD for the next iteration, and repeat as required.  A few years earlier, say 1980, such an analysis would not have been practical due to scheduling constraints and a different type of structure would have been designed.  A few years later and one of the 100 new graduates could probably have done it with minimal supervision.
        To conclude, I think the changes in technology are tremendous but there are serious problems associated with the rate of change.  One of the major problems is widespread distribution of software by persons who have no responsibility for its use (but benefit financially by its sale) to persons who have extremely limited knowledge of how it actually works and what it actually does but who actually believe that it (the software) can replace the experience and judgment of a qualified professional engineer.
        There!  I'll stop ranting now.
H. Daryl Richardson
----- Original Message -----
Sent: Tuesday, August 30, 2005 10:50 PM
Subject: Re: Technology and Structural Engineering

I realize that I'm the only person on this list who remembers what engineering was like in the 1960s, but the mechanics of producing calcs., specs., and drawings were SO much more difficult, and the physical quality was so much lower.  I emphasize PHYSICAL quality because I DO NOT mean engineering quality.  I mean illegible calculations (not spread sheets, etc.), actual cut-and-paste and then type and mimeograph specifications, and "Xerox" things one page at a time, and erasing your way through the tracing.  Hail, I even knew architects who still used ink on linen!  Now we have recent graduates who can design 100 story high rises over night, complete with color-coded working drawings, having analyzed 100 load combinations, but the welds fail in an earthquake.   Oops!  Supposedly the guy who designed the Seattle Space Needle just stuck 3 of the heaviest WF shapes made into a triangle for each leg and checked it with his slide rule ... and it's still there! 

Enuf reminiscing. 

Wouldn't go back for the world.  :)

Ralph Hueston Kratz, S.E.
Richmond CA USA

In a message dated 8/30/05 9:35:21 PM, chrisw(--nospam--at) writes:
That said, I'm not all that certain that the technology doesn't seem to
have made across-the-board improvements in engineering productivity,
certainly not in the manufacturing areas where I work. I've heard a lot
of claims, usually by software developers, but I don't see as much
overall impact in particularly improved design methodology, lower costs
or shorter design lead times.