Simplicity Approaching Functional Perfection 19

[jheidt2543]

“The highest attainment in design is a simplicity approaching functional perfection.”
- Linton E. Grinter, Ph.D., C.E.

“Nothing discredits the usefulness of theory as a practical design tool so much as the use of theoretical toys. It is often true that theory tends to become an end in itself instead of a tool for practical use. The literature is full of formulas, graphs, and mathematical studies that are of interest mainly because of their intricacy. This criticism is in no way intended to discredit sound analytical studies, however complex. Mathematics should neither be avoided nor displayed.”
- Linton E. Grinter, Ph.D., C.E.
Vice President and Dean of the Graduate School,
Illinois Institute of Technology in
Design of Modern Steel Structures
The MacMillan Company, 1941, p. 3

I wonder what Professor Grinter would say about structural design today, some 70 years after making the above statement? He saw the future and didn’t know it! The problem today, as I see it, is the compounding of the “theoretical toys” mixing with the evolution of extremely complex building codes that change every 3 to 5 years.

I'm just wondering what others have to say about Professor Grinter’s comment and what, if anything we can do about it?

 

[rowingengineer]

Come on people it isn’t that bad. “The young blokes don’t have any feel; the old blokes don’t help enough” . Well guess what the only thing you can change is yourself and changing the world starts in your own backyard, I’ve started with a grad manual and some technical articles, most of the guys of this website are helping, what are you doing?.

At last count I think there are over 129 million books published .1% of them structural, this makes for a lot of books to reference or digest. While a book may not get you there, it will get you close to the mark if it is the right book. Then there is 100’s of thousands of building to review in your capital city.

Hell if you can’t get a good idea of how a steel building goes together by just walking around a few warehouses, there is something wrong with your eyes. Have you talked to the local steel fabricator, he could probably guess the sizes better than you could. There are so many different places you can get the gold from not just your mentor.

Just today I called a guy from bremick about some 14g screws, found out that a tight thread can screw thru thicker material were as a course thread is better for thin material.

If a young bloke comes and asks me a basic question, sure I don’t say P*SS off, but I do ask what he has done before coming to my desk? If he says this is his first stop, I tell him to get back on the train and stop by once he has opened a book or read a technical article or look at some previous designs. You got to do the leg work.


ANY FOOL CAN DESIGN A STRUCTURE. IT TAKES AN ENGINEER TO DESIGN A CONNECTION.”

 

[SAIL3]

In my opinion, there is no doubt that the code system, as it stands today, is seriously flawed/broken.
As a2mfk pointed out, the amount of time required to chase down insignificant details, to try and make sense or engineering logic out of the opaque reasoning often found in the codes today...all for a 4% gain??.Besides being a waste of time and resources, it is also inherently dangerous. The more time an engineer spends chasing down insignificant details, the less time he has to concentrate and address the really important elements in the design.
Even the ASCE code which I always believed was probably among the best written codes out there is beginning to catch the dreaded disease as evidenced in there treatment of wind loading.
It would really be helpful, if this site could provide a section where the codes could be rated by practicing engineers and perhaps the opportunity to sign a petition to rein in this code sytem that has run amuck.
At least it would provide some feedback to the code committees.This whining amongst ourselves may make us feel good but,in reality, is not very effective.
I feel better already.

 

[ash060]

rowingengineer

What you say is what I did. I have dozens and dozens of textbooks and reference material, and without them I would not be capable of doing many of the things that I do everyday.

Also, I appreciate all the excellent and concise responses that are posted on the forum that help me get out of pickles. Engineers like Ron, JAE, hokie66, and many others help me resolve problems. I am glad such a website exists and that excellent experienced engineers contribute to it.

 

[Qshake]

SAIL3 - What you note about the potential to rate the codes is a great idea. Perhaps this can be accomplished in a more formal way.

I can tell you this: the folks at AISC to read and reply to the AISC forum here and have made some comments regarding their material. However, if you visit that forum you'll see the inherent downside to this sort of open forum endless critizism.

What is needed regarding the codes is more practioners to partake in the process. That begins by getting involved, really involved, in ASCE, NCSEA, etc. I've seen the process, seen how it works and believe that the engineers involved really believe what they are doing is helpful to the practicing engineer. However more input is needed.

And yes, we all know how little time we have.

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.

 

[SAIL3]

Just about everyone's performance is evaluated in form or another and it provides a valuable function by providing feedback and thusly, the opportunity to improve.
This rating could be performed in a professional and not an emotional manner, ie:
Say a rating from 1 to 10 for the following catagories:
clarity
engineering logic
continuity
usefulness
revision cycles
etc.
The survey should have the cability to accumulate the average rating in each catagory.
This is meant to be a constructive mechanism and not a license to rant which is the only option available at the moment. We are all professionals here and, more often than not, will perform in a professional manner, given the opportunity.

A section on this site providing this type of survey could also open up this site to more engineers as we spread the word amongst our collegues and engineering contacts and encourage them to take this survey.
Instead of becoming frustrated and floundering around in the this maze, it might actually lead to some needed changes in the manner of which the codes are written and published.

 

[hetgen]

So who is at fault?
Most consulting companies are owned by old experienced engineers, the economy could be bad out there with fierce computations from Asia, but they are the one employing cheap inexperienced engineer with a super gadget who can perform faster at minimal cost. Input – run analysis – output…

The worst will happen once those experienced retire; structures will be sealed with a license agreement that states that the creator is not liable for any damages that may result from the use of their product? Alike to RISA / CSi products.

 

[rowingengineer]

Being someone that uses codes of many different countries, my opinion would be that the American codes are not complex and are reasonably logical. Personally I love having the more complex methods for analyses especially wind loads. I live in the 4%, if there wasn’t the 4% I wouldn’t make a living. I am waiting for the wind model code to be published just so I don’t have to remember so many different wind coeff, and when that happens it is likely the code will become more complex.
I agree with Qshake, get out there and join a committee or review panel, make comments when drafts are published. Yes it takes time, but it also takes time to complain.

I can’t see any real reason for having a survey for codes on this site; this site is international that is a lot of codes that most people wouldn’t have even heard of. Just think there are over a 100 counties with different codes (if you don’t cover the euro countries with the euro code just yet), each of these will generally have a loading, concrete and steel code as a bare minimum. The codes are generally based off the American, British, FIB, Australian (wind only) or German codes. Most only adopt about 80% of the original code, with 20% local content or changes to safety factors ect. Making for an interesting read each time you pick up a new code. And we are only talking bout structural codes.
Ash060,
Having all those tests books and looking to the big guys for help makes you more aware than an engineer who has been spoon feed, and hopefully you will never have to use my favourite saying "that is how we have always done it"


ANY FOOL CAN DESIGN A STRUCTURE. IT TAKES AN ENGINEER TO DESIGN A CONNECTION.”

 

[jheidt2543]

I agree with Sail3's thought of a code rating system, but I also agree with rowingengineer that this is NOT the place for it. I really believe that should be a function of our engineering societies. They have the technical framework for this and could easly poll their membership with the survey.

 

[csd72]

Ash060,

Agree 100%, the current status is due to my peers and their seniors who have been pushing the universities for ready cooked engineers that can be plugged straight into the latest analysis machines.

The same people have no respect for the value of their feloow engineers experience and therefore push all the experienced engineers into management as the only viable career path.

Thus you have junior engineers who have not had enough time to develop and Experienced engineers that are pulled away from the tasks they are expert at and deprived of any capacity to mentor their juniors.

When machines were invented, workers wages went up as they could produce so much more in the same time. When the same thing happened with engineers and computer analysis we passed the whole saving onto the client not even holding enough back to allow for quality control.

 

[BigH]

Guess I got in this late - great topic Jim. Sometimes I wonder:

a) why do we need such complexity for "run of the mill" projects? Doing bearing capacity equations in LRFD that are meaningless in that the serviceability limits govern the applied safe pressure.
b) Many engineers (typically young - and we have seen them in the various forums on this site) who don't even know what the underlying basis for a particular variable is in a computer programme ("Hey, guys, what is the "E" value?)
c) Use of 4th order analyses with their complexities and still using first order data for input (as a soils engineer I am talking the SPT "N" valule, for instance - a great piece of data if properly used) - or the design strength of concrete vs the actual strength in the field - and how that affects the design behaviour.
d) Why does ASTM and others force you to buy a whole book of standards (look at the D series) each year when they change only 4 or 5 specifications - look at all the trees we lose! It is all pushed by money . . . (look at the cost of the 1 PDH hour from ASCE at $349 per pop).
e) I was fortunate to have one of Canada's pioneering geotechnical engineers as a mentor - saw him go to a site with a failing slope (oops, a slope undergoing a somewhat vigorous distress) and he gave a "thumb in front of the eye" solution - then the engineers in the office after a large investigation programme and many hours of analysis came up with the same design (or very darn close to it).
f) many engineers are forgetting their history . . .

In the end, to add to the quotes that Jim started with, I particularly like one by H.Q. Golder (yes of Golder's fame) who was quite a reknowned geotechnical engineer in the UK before heading to the Dominion:

“For the engineer, … , there are many possible answers, all of which are compromises of truth and time, for the engineer must have an answer now; his answer must be sufficient for a given purpose, even if not true. For this reason an engineer must make assumptions he knows to be not strictly correct – but which will enable him to arrive at an answer which is sufficiently true for the immediate purpose.”(1948)

 

[271828]

Just a couple of comments. I teach the basic analysis and steel courses at a university. Believe me, we try to teach the basics and behavior. These kids start out in their junior year barely knowing the difference between an applied moment and a bending moment, to put the challenge into perspective.

As for computer usage, if I had a quarter for every time I discussed verification of computer output, I could buy a steak dinner or three. I even have several horror stories that they hear over and over again.

Here's some irony for you guys to consider LOL. We hire an excellent local engineer with about 25-30 years of experience to teach one of our classes on building overall behavior. The idea is to bring real-world experience and insight. Know what he does? Spends the entire semester teaching the students how to use a couple of computer programs. No behavior. No insight. Four months of computer program tutorial. We're trying to figure out how to "steer" him another direction, but people get testy when their teaching is commented upon. He doesn't understand the MASSIVE difference between training and education. He's just thinking about the skills he wants his new engineers to have on Day 1.

 

[Teguci]

Disagree - mathematics displayed through a structure can be very edifying. The elegant shape of a suspension bridge which follows the moment diagram's shape is an example. Even a masonry cathedral with flying buttresses illustrating the thrust from a vaulted ceiling is enjoyable to experience. I understand the need to economize and simplify for most projects, but simplicity as an end to itself I would have to disagree with.

 

[epitome1170]

My two cents: (most of which has been said)

I am a fairly young engineer (30) that started out in railroad bridges so had very little experience with buildings and codes until I switched companies five years ago. When I did, I was lucky enough to be under the supervision of someone that was not a "computer person" (I am... in fact I love computers). He could work with the analysis programs (and spreadhsheets), but at the time preferred to things by hand a lot more. So as we dove into buildings and he began training me, he never allowed me to use spreadsheets or analysis software. Obviously, this took more time, but it developed that "gut feel" for loadings and capacities of members. After I had done a half dozen or so buildings was I only then able to use spreadsheets, but for the first few times I used them I had to do it by hand as well in order to verify its accuracy to myself. Similarly, the same process was taken for analysis software.

Now obviously, this created a lot of inefficiencies, but it allowed me to build a fantastic foundation for my knowledge. Now I use the computer more than hand calculations, but at the same time there have been numerous occasions that I did not click a button here or switch something on in the program and not realize it until I look at the member sizes and loadings. Without that background, a "new, fresh faced" engineer would not have known the inaccuracies. Computer programs are dumb, they are only as smart as the person inputting the information. They are extremely valuable though to a business in order to become more efficient.

At this point in my career, when I train a new engineer, I make them undergo the same type of hand calculations that I went through.

 

[dhengr]

Epitome1170.... I think you were really lucky to have found the mentor you did, and that that is the right, although somewhat inefficient, way to do it. But, the boss will gain that back many times over, because you can both trust your structural common sense, gut feeling, and experience; and you will certainly be a better engineer for having taken this tack. And, you are both probably better off for what you can teach each other, or for the way you can blend your strengths. Good for you, by all means, pass it on.

271828..... Your story is an indication of the sad state of the educational system today. The students need a much better grounding in Strength of Materials, Statics, Physics, etc. before they get to your class, and they really need 5 years instead of 4 to get it all, with the last 3 digging much deeper into structures and basic concepts, and with the computer only being used to do the arithmetic, after they understand the classical methods. Then, finally appreciating how the computer sets us free to do the fun part of engineering instead of the complex drudgery, once we understand structures, but not as a replacement or substitute for that understanding. I can kinda understand the engineer/teacher story too, but with all my rants to the contrary, I don’t think I would like teaching the way he did. If you take your stories and Epitome’s story together and marry them; you take them for the first 4-5 years, then the final year they pay me the same tuition they’ve been paying you, so they can learn what Epitome claims to have acquired.

 

[epitome1170]

dhengr-

"claims to have aquired"? Hmm...

The "problem" with the educational system for structural engineers is that they do not seem ever really address how a building is designed from start to finish. I went to an albeit small, but well known (and highly ranked) civil/structural engineering university for my B.S. and M.S. and in all of the courses I took I only remember two that ever even touched on code loads and how they are derived. Typically, you were given the loads and then could determine the capacities.

There is some valuable insight lost on how the loads are actually derived along with the fact that a student when coming out of college has so much knowledge, but (quite frankly) does not need a majority of it except as a background. Although valuable, the knowledge in college does not always translate to the "real" world applications.

 

[jheidt2543]

Teguci:

I don't think we are not on the same wavelength, or at least not understanding Grinter's statement in the same way. First, I agree, “mathematics displayed through a structure can be very edifying”, as the structures you mentioned clearly show.

"Art without engineering is dreaming;
Engineering without art is calculating."
- Unknown

In my understanding, "simplicity approaching functional perfection" does not mean trivial designs, rather avoidance of complexity for complexity’s sake. BigH notes the same thing regarding soils engineering, how do you justify using complex mathematical formulas with field data that is accurate to only one place?

I think the point of all this is that the engineering tools, specifications or codes should match the precision of the data we are using and the project we are working on. Rewriting building codes every three or five years for small but complex changes is foolish and expensive – there are a lot of people making a good living doing so. There are times when I think some in our profession have lost sight of what it is we are trying to accomplish. Just because we have the fancy tools and complex equations does not mean they are required for every instance – engineering judgment should still prevail. As I noted above, a lot of engineering works were done with a slideruler.

Another thought provoking quote:

“Engineers tend to develop tools for the pleasure of developing tools, not to induce nature to yield its secrets. It so happens that some of these tools bring us more knowledge; because of the silent evidence effect, we forget to consider tools that accomplished nothing but keeping engineers off the streets. Tools lead to unexpected discoveries, which themselves lead to other unexpected discoveries. But rarely do our tools seem to work as intended; it is only the engineer’s gusto and love for the building of toys and machines that contribute to the augmentation of our knowledge. Knowledge does not progress from tools designed to verify or help theories, but rather the opposite.”

-“The Black Swan, The Impact of the Highly Improbable”
By Nassim Nicholas Taleb, Random House Paperback, 2010 p. 169

 

[jtx]

As I sit here, suffering from insomnia on a weekend, I offer these thoughts....

During school, I tried to take every possible structural engineering course that I could in order to prepare me for work the real world. Did it help? I'll let you know when I retire. Fortunate for me that the only time that I used a computer for analysis in school was during a structural anlaysis class. We used Risa 2D for a whopping one class and in one homework assignment. The rest of the time was spend on hand methods (moment-area method, conjugate beam method, slope-deflection method, moment-distribution, approximate methods,etc). I also took and FEA course to give me a better understanding on what that program is doing with all those lines and dots. It gave me the tools to do a "reality check" when I analyze something by computer.

Now, after saying that, I see 2 problems. First, is the education system today. With constranints on budget and other factors, modern day universities are in a push to try to graduate a student in less time then that would be actually needed to adequately educate a new engineer. In addtition to that, I look back on all of those general ed course that I took to be a "well rounded student." 5 years out and glad to report that I haven't used anything from those courses. Why did I need to take courses that I don't need and take the additional structural courses that I could have used? I don't know, but at least I am a "well rounded person" (cough cough). Same with the other engineering courses I took. I haven't done anything related to thermodynamics, fluids, water resources, or transportation since college.

Second is the on the job. We are constantly being pushed to get work out the door faster and faster. Design-build, IPD-BIM, 3D, etc, etc, etc. I think that, because we are pushed for less and less analysis and design time, we are beginning to rely on computers more and more. Maybe having blind faith in the analysis if you are really crunched. Just hoping that everything "looks right." It's just the sign of the times.

Begin Rant. Don't get me wrong, as I am not using either of these as an excuse. If you want to get a feel for a structure, take 2 minutes look at how your structure is defined and loaded. Look at a shear-moment diagram. Look at your deflections. These tell you everything about how your structure is behaving. If you get a moment diagram that look like one for a fixed condition and you should be getting one for a simply supported condtition, then that should tell you something is up and you need to check it out. There are countless references that are out there to help you with the design and detatiling of a structure. If you are stumped then there is nothing wrong with asking for help. I agree mentoring provides a young engineer with priceless insight, but come on there are basic things that we all were taught to help us get that feel. And in time, that feel will get better. It just takes a bit of effort.

End rant...please excuse me as the womens World Cup between is on ESPN right now. That should help to put me to sleep.

 

[mtu1972]

Love these comments. My BS degree was in 1972 and I took all of the available structures options. A poor job market led me to take 5 terms of graduate classes, mostly structural, math and one ME course. Left without the MS degree because of a promise to my wife.

Still using punch cards and main frames, we had a class in Matrix Analysis that the Professor taught along with a Moment Distribution textbook (Gere) as they were both stiffness methods. Moment Distribution became the primary way I looked at indeterminate structures.

All of those extra classes set me a little above my contemporaries, many who had only the basic analysis, steel and concrete design classes.

I had tremendous mentors in my first two jobs. So even today, that fundamental knowledge and experience gets my first run member sizes in the ballpark. I do try to pass on any insight I have gained to the younger engineers in our department.

Got re-introduced to computers in the mid '80's and did not become somewhat proficient until I started working on a Masters degree in 2000. I rely on younger engineers to help me get the most out of the computer.

My observations about the Codes and the analysis methods are similar to many of those expressed above. Are buildings really better with diagonal wind loads, wind loads with torsional components, DAM, etc.? Seismic here in the upper mid-west where it could maybe happen but probably won't? I doubt it.



gjc

 

[IDS]

My observations about the Codes and the analysis methods are similar to many of those expressed above. Are buildings really better with diagonal wind loads, wind loads with torsional components, DAM, etc.? Seismic here in the upper mid-west where it could maybe happen but probably won't? I doubt it.

On the seismic question the answer is no if the earthquake does not happen and yes if the earthquake does happen.

There is no doubt that recently designed buildings are much more robust to earthquake loading (or extreme wind loading for that matter) than typical older buildings.

Einstein used to say that things should be simplified as far as possible, but no further. Experience has shown that in some respects old codes were too simplified to give reliable results in some circumstances, or the simplified methods required excessive conservatism.

It's possible that modern codes have gone too far the other way, but we can be certain that achieving the optimum balance between scientific rigour and simplicity is not just a matter of returning to the Good Old Days.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[271828]

Now, after saying that, I see 2 problems. First, is the education system today. With constranints on budget and other factors, modern day universities are in a push to try to graduate a student in less time then that would be actually needed to adequately educate a new engineer. In addtition to that, I look back on all of those general ed course that I took to be a "well rounded student." 5 years out and glad to report that I haven't used anything from those courses. Why did I need to take courses that I don't need and take the additional structural courses that I could have used? I don't know, but at least I am a "well rounded person" (cough cough). Same with the other engineering courses I took. I haven't done anything related to thermodynamics, fluids, water resources, or transportation since college.

As a professor, I totally agree. Hate to say it, but this is going to get worse. There are rumblings “from above” to squeeze the curriculum and simultaneously increase general education courses. We’re to the point of wondering whether or not the students can take basic analysis, steel, and concrete at the undergrad level. It’s becoming impossible.

In reality, what you’re arguing for is a structural engineering degree instead of being a part of civil. I’ve been saying that for years. I’m with you: In 10 years of industry experience before becoming a professor, I never had even the first use for surveying, fluids, transportation, environmental, etc. Could’ve sure used an extra class or two in mechanics of materials, vibrations, concrete, wood, and steel, though, but there’s no room.

As a strange aside, I actually see more use in the general education classes than the non-structural civil classes. I didn’t until I crossed 40, but I very often find myself wishing I knew more about biology, history, and a slew of other subjects.