Letters to the Editor

July 2013

Foundations for Metal Building Systems

The article "Foundations for Metal Building Systems" in the July 2013 issue reminded me of a case from a couple of decades ago. A geotech friend called with a question about a steel gable-frame building that had suffered significant distress. The foundation had been designed (by someone else) as a a hairpin and slab tie system, and it probably would have worked. However, some salesman convinced the contractor that random fiber reinforcement was a good as welded wire reinforcement, and this substitution was made without input from the designers. The result was huge cracks at the ends of the hairpins, significant lateral spread of the column bases, and significant vertical deflection of the rood structure. I don't know what the ultimate disposition of the case was.

Bill Gamble


January 2013

Structural Failures

As a long time Structural Engineer, I read STRUCTURE magazine regularly, and appreciate the selection of articles.

Upon seeing the cover of the January 2013 issue, I was compelled to share some of my experience.

I believe that a significant number of failures can be attributed to the structural design engineer for not having, or insisting on, the opportunity to inspect his/her design in the field before it is covered up or hidden from view. The person who designed the reinforcing, for example, can tell more quickly than almost anyone if it is the same size and placement as he intended.

In the old days, when engineers were a sub-contract to the architect, there was no budget for the engineer’s on-site inspection. I used to go out on my own, without pay, to inspect my design. As often as 50 percent of the time, I found things that needed to be corrected which I documented in a report that went to my Partners, the Owner, the Contractor and the City Inspector.

I do not know whether these instances were ignorance or fraud, intentional or accidental. In some cases, failure would have happened if it had not been discovered and corrected.

I’ve heard all types of excuses, including:

  • “They weren’t needed.” (50% of the plug welds on a steel deck)
  • “I never had to do it that way before.”
  • “I’ve always done it that way.”
  • “I couldn’t understand why it was necessary.”
  • “We have never had to do engineered drawings before, and I don’t know how to read them.”
  • “My supervisor said we could save money if we omitted every other bar (or spaced them farther apart).”

Chapter 17 of the International Building Code allows us to “invite ourselves” to the field to “Observe’ our design. On the front page of my set of plans, I have a separate section entitled “Structural Observation”; here is where I list the items I wish to look at before they are covered up or hidden from view. It’s worth the trip, education, and good public relations.

Sincerely,
James D. Leach, S.E.
California Registered Structural Engineer


July 2012

Knowledge, Rationality, and Judgment

The InFocus article on "Knowledge, Rationality, and Judgment" (July 2012) explains the importance of practical judgment in the engineering profession based on virtue ethics cultivated and possessed by engineers. It reminded me of the following sentence that I read few weeks ago in a piece titled "One Virtue at a Time, Please" in The New York Review of Books (June 21, 2012) regarding Truth, Beauty, and Goodness Reframed: Educating for the Virtues in the Twenty-First Century, by Howard Gardner:

"Nonetheless, Gardner is firmly on Keats's side in wanting us, in our efforts to educate the young and ourselves, to take beauty seriously, to cultivate our aesthetic sensibilities, and to learn how to form intelligent judgments about works of art of all sorts."

The engineering profession also needs to identify and cultivate our own virtues, and to learn how to form intelligent judgments about the role of technology in our civic life, so that we extend its benefits equitably to all of mankind, but do so sustainably, respecting the environmental constraints of our finite earth.

The bimonthly InFocus articles on virtue ethics concepts applied to reframing engineering ethics in the twenty-first century are timely and much needed. The technical rationality developed over two hundred years of technological revolution so distorts our notions of knowledge and judgement that we need to reframe both in the digital age, which otherwise promises an even greater stranglehold of technical rationality on the engineering profession.

Sincerely,
Ashvin A. Shah, P.E., F.ASCE


March 2012

License Engineers and Certify Disciplines

In response to Timothy A. Lynch’s opinion in March’s Structural Forum, I agree with everything that Mr. Lynch said. I am licensed in several states and my area of expertise is civil engineering with a specialty in water systems and water-retaining structures. It is frustrating that, in the States of Illinois and Hawaii, I cannot perform the design of simple reinforced concrete structures because I am not a licensed SE. However, I know architects that can legally perform the services even though they don’t have a clue about water-retaining structures.

I developed my expertise through a career of specialty work and I would naturally limit my services to that expertise. I have no desire to attempt services outside of that expertise and I believe that the vast majority of other professional engineers work within their discipline. I support certain limitations, such as requiring an SE for critical structures such as hospitals, schools, or buildings over 13 stories high, but preventing me from performing a service that I handle on a daily basis is ridiculous.

Thank you,

David J. Peterson, P.E., SWD
President/CEO
Watershape Consulting, Inc.


February 2012

Responding to Forces of Nature

STRUCTURE magazine’s Editorial in the February 2012 issue raised important questions for consideration by the structural engineering profession. It makes a plea for structural engineers to become involved in reducing losses from natural disasters, by investigating building failures, strengthening building codes and working with jurisdictions, etc. While I agree with this premise, the call to action is overly broad and does not acknowledge what we already do in these areas. Indeed, structural engineers are already heavily involved in these activities. However, despite our best efforts, annual US economic losses from natural hazards continue to increase. The editorial also does not provide evidence why greater involvement would be more effective in reducing natural hazard losses.

I disagree with the author's premise that structural engineers have a “natural tendency after any major event (is) to call to strengthen the codes.” Our role as engineers has been to create and advance scientific knowledge on building design through careful study and effort in order to predict building failures under extreme loads.

For example, many engineers are involved in long-term tasks to learn from last year’s tornado losses. In 2011, non-engineered residential structures sustained enormous damage during tornado outbreaks that caused over $22 billion in damage, and the second highest number of annual tornado fatalities on record. Code changes would make little difference in the performance of those existing structures, unless the structural resilience of existing homes was also improved. STRUCTURE magazine (July 2011) reported on the Tuscaloosa tornado damage survey and ongoing analysis of the data. However, it is likely that costly tornado disasters will occur again, and we should admit that more engineering (i.e. better codes or enforcement alone) is not the answer.

According to a 2007 study by the US Nuclear Regulatory Commission, there is a very small chance of a tornado strike within the contiguous United States (probability of 1 in 100,000 per year of a tornado with estimated wind speeds of 160 mph or greater). However, the consequences of that event upon a community can be catastrophic, and should be factored into our design; currently this is not the case.

Natural disasters result from a complex interaction of the physical hazardous event, with the vulnerability of society, its infrastructure, economy and environment. That vulnerability is determined largely by human behavior and actions, and so it is within our control to reduce it. Ultimately, people decide where and in what type of structure they will live. However, any action to reduce disaster risks and damage must hinge on a community’s collective decision, based on its risk perceptions, which go beyond engineering.

The public may be unaware (or in denial) as to the structural inadequacies of the majority of their existing homes. Currently, very few homes are actually engineered. Yes, structural engineers must be part of the conversation (and education) with the public, but it is primarily the public who must decide the level of risk that is adequate for them, and establish acceptable quality of structures they are willing to pay for. The message to our membership should be to understand the latest thinking on vulnerability (social and economic), and be able to discuss risk reduction and structural design issues. In this way we could best advise the public regarding its vulnerability to natural hazard risks, and what should (could) be done to reduce it.

David O. Prevatt, Ph.D., PE (Massachusetts)
Assistant Professor of Civil Engineering
Engineering School of Sustainable Infrastructure and Environment
University of Florida


November 2011

Alfred Pancoast Boller

The "Great Achievements" article by Frank Griggs on Alfred Pancoast Boller in the November issue of STRUCTURE magazine states, “In 1909, Bolller and Hodge were appointed engineers for the Port Authority of New York and New Jersey.” This does not appear to be correct. The Port Authority (originally: The Port of New York Authority) was created on April 30, 1921, nine years after Mr. Boller’s death. The Hudson River crossings studied during the period from about 1900 to 1927 were presumably conducted by the New Jersey Interstate Bridge and Tunnel Commission and the New York State Bridge and Tunnel Commission, which were not part of nor rolled over into the Port Authority. Boller, Hodge & Baird was a consultant to the New York State Bridge and Tunnel Commission in 1913 that recommended a bridge at 57th Street in Manhattan to New Jersey. This plan was rejected primarily because of its substantially higher cost as compared to a tunnel (see The New York Times, 04/22/13). The Holland Tunnel project was advanced instead; it opened in 1927, and the Port Authority took over its operation in 1930.

Joseph Kelly


June 2011

STEM

I enjoyed your article, “STEM”, in STRUCTURE, June 2011. It is a subject that I have given some thought to, and I believe it to be timely. In my mind there are several facets to the problem, some appear addressable and others do not. As modern technology developed, it has become more subtle and difficult to comprehend. For example: a nuclear power plant is far too complex for most of us to comprehend, whereas the functioning of a waterwheel is quite comprehensible. Electricity makes sense with some study but solar cells require another level of knowledge, as do solid state devices and modern astrophysics. Most of us acknowledge their existence and enabling attributes with little comprehension or full appreciation

In fact, most modern technology is beyond the ken of nearly all primary and secondary teachers. Thus, it is nearly impossible to teach or appreciate such technology in primary or secondary schools. This void in education leads directly to a lack of appreciation and interest by youth as well as adults.

When the technologies of chemistry and electricity were in their infancies in England, the "Royal Society" and other organizations gave presentations by such notables as Robert Boyle in chemistry and Michael Faraday in electricity. These lectures and demonstrations were extremely popular with the general public. Of course, they were forming the scientific-technological foundation for the Industrial Revolution that followed in England.

If present-day professors in the various technological areas could be enticed to give, pro bono, exciting lectures with demonstrations in their field of endeavor, the public might well be interested. Of course, technological luminaries from industry might present similar lectures. These professionals are usually adept at presenting complicated matters to non-technical people when writing proposals and making presentations to boards of directors. The professional societies are the obvious initiators for such efforts, much as the Royal Society did over 200 years ago.

Youth should see that science and math are exciting and not all bad, and that one can find a challenging and financially rewarding career in science or technology.

Frankly, many educators teach, or at least imply, that to aspire to make money is simply greed and that technology pollutes the earth.

A deeper understanding of God's creation and having the ability to use it for the betterment of our civilization is not all bad, but is indeed something to appreciate and to which one may aspire.

Dann H. Hall
Coopersburg, PA


June 2011

Dear STRUCTURE Editorial Board,

Each month I receive numerous trade magazines and journals representing a variety of organizations in the Civil Engineering field. In many of these cases, the lifespan for these publications is barely long enough to breeze across my desk before they end up in the trash can. Occasionally, I will browse the headlines and articles to check for interesting stories but I am rarely successful in finding something to catch my eye.

However, my subscription to STRUCTURE is unquestionably the exception. Each month I look forward to receiving it and carefully comb through every article. I even maintain a library of past editions which I frequently reference back to from time to time. I thoroughly enjoy every aspect of reading your magazine and always appreciate the Structural Practices, Opinion, and Feature articles included in every copy. Your publication has allowed me to grow in my career and I am without a doubt a better structural engineer as a result of your efforts.

Thank you for your hard work and dedication to this wonderful publication. I just received my latest edition and have already carved out a few minutes today to jump into the articles!

Yours Respectfully,
Michael Howell, P.E.
Structural Engineer
Austin Brockenbrough & Associates, LLP


March 2011

Bridge Sustainability

In the March 2011 issue of STRUCTURE, author Mark D. Webster points out that the production and fabrication of one ton of steel results in approximately one ton of carbon emissions released into the atmosphere. In other words, the carbon emissions are about equal to the amount of steel. This indicates a real need to reduce the amount of steel manufactured. In my work as a structural engineer for the City of Chicago, Bridge Division, I found two ways to increase the useful life of steel bridges, which reduces the need for new steel and thus reduces carbon emissions:

1) On bascule bridges we try to make the deck as light as possible to reduce the counter weight, foundation loads and operating power. The obvious solution was to use an open grid deck, which is much lighter than a concrete deck and could still easily carry heavy truck loads. Several years passed with no signs of trouble. Then, during a routine inspection, it was found that the flanges of the small beams supporting the open grid deck were badly corroded. Debris from the tires of thousands of vehicles crossing the bridge had landed on the flanges and retained water from rain and melting snow, causing the flanges to deteriorate. Is there any way to prevent this situation? The answer is to use an orthotropic steel deck. Closed ribs can be designed to be airtight, precluding corrosion of the inside of the ribs. An excellent book on the subject is Design Manual for Orthotropic Steel Plate Deck Bridges, published by the American Institute of Steel Construction. It contains design information and discusses practical considerations, such as corrosion prevention.

2) The use of salt to melt snow has caused deterioration of various steel parts of both fixed and movable bridges. The solution to this problem is to find deicing agents that have no corrosive components and use them on bridges and bridge approaches. Away from the bridges, the standard deicing chemicals can be used.

Peter Kocsis, SE, PE


February 2011

Fazlur Rahman Kahn

Your article on Dr. Fazlur Rahman, the great structural engineer, was very inspiring. I have always felt proud and honored of his Pakistani heritage. Your article was beautifully written, except for the small part about the political history on the creation of Bangladesh (East Pakistan). According to a book written by Pakistani military commander (General A. A. K. Niazi), the separation movement in East Pakistan was fully funded and supported by India. This has also been confirmed by Indian author Mr. Ashok Raina, in his book Inside RAW.

The separation of East Pakistan was a tragic period in the history of Pakistan. At its creation in 1947, the union of East and West Pakistan was considered unnatural by many international experts. These two East and West units of Pakistan were separated and located thousands of miles away from each other.

Historically, the people of that region (South East Asia) have lived together for centuries. In 1947, the British Empire divided this region into India and Pakistan. Later, Pakistan was split and Bangladesh was created. However, the people from that region are still connected above and beyond these boundaries. As a Pakistani American, I feel optimistic that one day the people of that region will put aside all their differences and will live in complete peace and harmony, just like people of the Americas and Europe.

Ahsan M. Sheikh, P.E., S.E.


February 2011

Fazlur Rahman Kahn

As past-President/Director of SEAOSC (and past SEAOC Director) it was my great pleasure to host you when you spoke at one of our dinner meetings in Los Angeles a couple of years ago. I have always enjoyed your articles and I liked your book Engineering Legends (I have a personally autographed copy thanks to my wife who also attended that dinner event). I almost always agree with your stated viewpoints about our profession in general and on civil/structural engineering education in particular. As for the great genius Dr. Fazlur Rahman Khan, I have always felt proud and honored that he was a South-Asian American like me and in fact got his start in the profession in Pakistan like I did after I graduated from the California Institute of Technology, way back then. Your article about Dr. Khan (STRUCTURE Magazine – Feb 2011 issue) was beautifully written, except for the small part about the political history of the region where his roots lie. I would like to offer some corrections and background facts about that important subject if I may.

It is true that, in 1947 (not 1971 as was written in the article), the Imperial British Colony of India attained its independence and was partitioned into two sovereign countries, India and Pakistan, generally along the lines of the wishes of the people who resided in those areas of the South Asian Subcontinent. The people of East Bengal chose (and struggled) to become part of Pakistan even though there were 1,500 miles of India in between the two wings of the newly formed country. Post 1947, the relationship between Pakistan and India soured, tragically, and was exploited by some vicious and deceptive political/ethnic groups in the region, including East and West Pakistanis and Indians, as well as by the protagonists of the Cold War which was at its zenith in the 1960s and 1970s. Pakistani government/political policy in East Pakistan often exacerbated the situation. 1971 was the year that another war broke out between India (supported by the USSR) and Pakistan (supported by the USA) leading to an invasion of East Pakistan by the Indian army, resulting in the breakup of the country into Bangladesh (“East Pakistan”) and Pakistan (“West Pakistan”).

It was a dark and painful chapter in the history of the Subcontinen, for which all sides must share blame. As a teenager living in Karachi, Pakistan during the 1971 war, I experienced firsthand the terror of bombs being dropped on my civilian residential neighborhood night after night. Even then I realized that government forces on our side had also committed atrocities in East Pakistan. It is my belief that for most of us the scars from that awful experience have healed. In general Pakistanis, Indians and Bangladeshis have moved on to discover that they have more in common than what divides them, and more to gain from respectful cooperation rather than hubristic confrontation. As is often the case, however, there continue to be some selfish and narrow-minded (and pretty powerful) detractors on all sides who would have us believe otherwise. It is not common for us engineers to delve into politics and history, especially in articles related to our great profession, but I thought it important to clarify this point in your otherwise stellar article about the great Fazlur Khan.

Saif M. Hussain, MS, P.E., S.E., SECB, LEED® AP


Author Response

Messrs. Ahsan Sheikh and Saif Hussain offer clarification and insight into the history of, and past conflicts between, Pakistan and India – as well as between East and West Pakistan. Mr. Sheikh additionally lists two books providing further information for anyone interested and Mr. Hussain stresses that in 1947 India and Pakistan were, indeed, partitioned into two sovereign countries, India and Pakistan, as was stated in my article.

It was during the 1971 conflict between West Pakistan and East Pakistan (Bangladesh) – with interference from and/or instigation by India – that that Dr. Fazlur Khan founded the Bangladesh Emergency Welfare Appeal, a Chicago-based organization, to help the people in his homeland. It was a noble, gallant and caring effort.

It has long been my contention that engineering does not occur in a vacuum, nor do engineers perform their work in a vacuum. Nor are engineers merely technical experts. They are also real “honest-to-goodness” people with human traits and feelings. To tell the story of an engineer and make him (or her) come across as a person, you need to tell the whole story. Dr. Khan was an especially well-rounded individual with many facets to his personality, and accomplishments in his career. To tell his story without mentioning his humanitarian and community leadership activities – and how he fit into history and the life and times he lived in – would be short changing his greatness and what he was all about. He was much more than a mere technologist. He was an iconic engineer with a very broad perspective on life.

Richard Weingardt


July 2010

Small Firm Experience

I totally concur with your comments about BIM and LEED in your July 2010 Editorial. I currently provide structural engineering consultation, analysis, and design in the light construction sector. I have Architect associates that I partner with that still deliver their services utilizing hand drafting. When I first started my own practice, I purchased CAD software. However, I soon found that it was better for me to focus on engineering business and to instead sub-out drafting to a CAD consultant who delivered the drafting service cheaper, faster, and better then I could.

I do not want to sound like a dinosaur, but BIM and LEED do not represent an important development in the business that I am in. Only one of my architectural associates is pursuing BIM – at a high cost and with little profitable return. More are pursuing LEED, but none have reported to me about it being profitable for them, just time consuming and costly. These two emerging areas of the A/E industry are certainly important, but not to everyone.

Peter Cloudas, PE
Isidro-Cloudas LLC
Stamford, CT 06907


Author Response

I would like to thank Mr. Cloudas for his support of my comments on BIM and LEED. I believe that we need to follow the money to see who is really benefitting from these unsolicited additions to our project costs.

I also believe that, with respect to BIM, clients want a free system to run their facility. Although shrouded in the fog of these services being “beneficial to mankind”, there is a cost to everything of value. I trust that clients can be willing to pay those extra costs to the firms providing that additional end product.

Structural engineering firms have been delivering their services quite efficiently for a very long time (since the advent of the pencil!) without BIM or a LEED program. I am sure they would be happy to continue in the same manner until such time that clients make it worth their while to incorporate such a drastic change.

John Mercer, PE


July 2010

Learn, Adapt and Change…

John Mercer’s editorial was thought provoking. I agree that firms need to assess their survival potential, but to describe this problem as something we need to weather until things get back to normal is misguided. Engineers must learn to adapt to a continually changing environment. What we need is not how to increase billings but rather how to learn, adapt and change.

Classifying firms as Finders, Minders, and Grinders is disturbing. How would you feel if your role was just to work harder and grind it out? This attitude contributes to the turnover in firms and the dearth of experienced engineers looking after the technical work.

The fact that the Grinders “typically include entry level engineering staff” should be of concern both for its impact on the quality of the work and on productivity. Too often senior staff is busy acting as Finders and Minders, and putting out fires, while entry level engineers are left to work through their problems. Much time is wasted. Less than desirable solutions are the result, often because there is not time to do it over again.

Technical oversight is often a problem because senior staff is overworked. In addition, because of their focus on finding and minding roles, senior staff has difficulty keeping up to date with new codes and new software. Thus, senior staff may be unable to fully manage junior staff that is doing the work.

If a firm wants to take advantage of opportunities to make improvements, they need to look beyond costs and revenues. Financial data is an important enabler but cannot substitute for vision and leadership.

We need to look at why firm are selected and how they can provide value to their clients. All too often engineers are seen as necessary evils and are given repeat business because they have not screwed up too bad on the last project. If we think that the answer is to work harder or more efficiently, we are in denial.

More importantly, firms need to understand how to implement and maintain constructive change. It is easy for management to issue a new policy memo, but all too often the change doesn’t persist. Firms need to learn how to discuss issues and build consensus with staff. Many firms have office standards which are routinely ignored. We need to adopt a culture of continual improvement where we learn from experience.

IT, BIM, and LEED will continue to be problems until we learn to manage them proactively. We need to learn how to talk about the issues, develop consensus, and implement change.

Firms that adopt a culture of continual improvement, and that can implement and sustain change, will have an advantage and will have the tools to survive and grow. Firms that continue as before will find it difficult to survive in the long run.

Mark Gilligan S.E.


Author Response

Thank you, Mr. Gilligan, for your response to my Editorial. My comments describing differences between principals, project managers, and new hires seem to have struck a chord. In my personal experience, however, these classifications ring true in many firms.

I could not agree more with his comment on the issue of technical oversight through overworked senior staff. Firms often take on more work than they can handle, leaving gaps in quality, or pushing the responsibility for project design downward onto the least experienced staff. I think the result is seen in higher cost projects or other undesirable outcome. The most undesirable, of course, being called to task in a lawsuit.

The importance of continuing education becomes apparent when principals cannot keep up with the technical changes in design codes. Once the limits of senior staff members are reached, the firm needs to make an intentional decision to meter back the workload. Using Mr. Gilligan’s words, senior staff needs to learn, adapt, and change. One way is by becoming active participants in one or more of our three organizations; CASE, SEI, and NCSEA.

My comments on BIM and LEED are targeted at the unsolicited added burden of time, money and education necessary beyond what firms had been adequately providing in their service mix. If the costs of these additional services are passed on to the client, management of these would become very simple. At the moment, the clients’ appetites are larger than their budgets.

Fluctuating economies require firms to continually learn, adapt, and change. The only change that I’ve personally seen is firms cutting staff and expenses wherever they can. The impact of that type of “change” is not only on the firm, but on the engineers and their families.

In keeping with his comment for firms to be looking at continual improvement, I assume Mr. Gilligan has ideas that can be implemented industry wide. I have openings on CASE committees that would allow him to join with other like minded structural engineers to improve our business practices and working environment for all. Do you have good ideas as well? If so, join Mr. Gilligan and others in debating the issues and crafting solutions.

John Mercer, PE


September 2010

Structural Contributions to LEED

A misrepresentation of structural steel’s recycled content appears in the article Structural Contributions to LEED, published in the September issue of STRUCTURE magazine. The article states:

Credit MR 4.1 and MR 4.2 – Most structural steel shapes are made from 97% recycled material. Recycled content in steel plate is about 65%. HSS sections are typically not made with recycled steel and should be avoided on LEED projects.

In actuality, the current industry average for hot-rolled structural shapes is 93.3%; all U.S.-made hot-rolled structural shapes are produced in electric-arc furnaces (EAFs) using steel scrap as the primary feedstock.

Plate is produced in either EAFs or basic oxygen furnaces (BOFs). The average recycled content for plate made via the EAF process is 93.3% and 32.7% for plate from the BOF process. No plate has an average recycled content of 65%.

HSS is a manufactured product made from coil steel. If the coil steel originated in an EAF process, the average recycled content of the HSS is, again, 93.3%; if the coil originated in a BOF process, the average recycled content of the HSS is 32.7%. As such, the statement that HSS is not typically made with recycled steel is incorrect. All HSS has significant recycled content.

In both the case of plate and HSS, the steel supplier can track the product back to the producer and determine the production process. If the material cannot be traced but is known to be of domestic origin, the BOF value can be used as a default. For non-domestic material, USGBC allows a default recycled content value of 25% to be used.

The statement that HSS is not made with recycled steel and that HSS should be avoided on LEED projects is damaging to HSS producers, and may result in decisions made by structural engineers that could increase, rather than reduce, the environmental impact of a building project.

Geoff Weisenberger
Director of Industry Sustainability
American Institute of Steel Construction


August 2010

Is Roof Eave Blocking Required To Transmit Wind/Seismic Forces?

Registered and/or Licensed Engineers, in all states of the US, are obligated to provide for the health, safety and welfare of the general public. With this responsibility, in our opinion, blocking is required between all roof rafters, all roof trusses, all floor joists and all floor trusses to transfer roof and floor horizontal force diaphragm loads to the designated shear wall resisting elements indicated on the appropriate contract structural drawing.

To advocate that the International Building Code (IBC) is not precise and allows the omission of blocking, together with the violation or omission of the continuous load path from the roof diaphragm and/or floor diaphragms to the foundation, is a violation of the Building Code. Blocking and the continuous load paths are to be installed with a rational analysis in accordance with established Principles of Engineering Mechanics. Based upon our review of the current Building Code in effect in California, and presumably across the country, no such deletion is allowed.

In fact, the Building Code unequivocally references blocking between roof structural members, floor structural members and the principle of a continuous load path in numerous Building Code Regulations. Failure to install roof and/or floor blocking and to provide for a continuous load path will place the general public in life hazard safety situations for all designated Building Code lateral force conditions.

Metal hardware installation should not be an alternative for lateral force transfer loads as suggested in the subject article.

Arnold Bookbinder, Structural Engineer
Arnold Bookbinder & Associates


Author Response

My article attempted to arrest the growing practice of omitting eave blocking by reminding design engineers that although eave blocking is not prescriptively required by the ICBO, basic engineering mechanics and metal connector manufacturers' specifications require its installation.

It was my intent to present a balanced analysis of all existing prescriptive and analytical requirements, but at no time did I advocate omitting the use of eave blocking.

Felix Martin, S.E.


July 2010

Engineers Are From Aristotle

Jon Schmidt draws interesting connections between Aristotle's metaphysics and the practice of structural engineering. In a specific example about change, Schmidt mentions that a steel billet has the potential to become a wide-flange beam in the future. True. But how many structural engineers really think of our profession in terms of material, matter, and the process of becoming? More so than in the past, I think, we often lose sight of the physical reality of our designs. I am sometimes asked basic questions by non-engineers about home construction, materials science, or how historic structures were built – and in response, I mumble something about moment diagrams. I’m embarrassed that I call myself a structural engineer in training (I'm unlicensed so far) without knowing the answers to these basic questions. I admit that it’s my own shortcoming, but it seems to be a common problem among younger structural engineers.

Many philosophers disdain or even deny the existence of the physical world, but Aristotle is one of few who value it ("In all things of nature there is something of the marvelous"). As engineers, we should be encouraged by Aristotle, and remember to educate ourselves not only in analysis and design but also in field practices, materials science, and other practical considerations.

A further note on Aristotle's teleological views: Schmidt says that Aristotle viewed "teleology as something that is present throughout the universe, not just confined to human endeavors." But I don't think this is true: non-living things do not have ends toward which they aim. Philosopher Martha Nussbaum writes that "...Aristotle neither applies teleology to non-living natural bodies nor gives any evidence of believing in a universal teleology of nature." (Aristotle's de Motu Animalium, 60)

It can be difficult for engineers and philosophers to find common ground, so thanks to Jon Schmidt for illuminating this connection. Also thanks for explaining the wonderful concept of eudaimonia or "human flourishing," a valuable idea of happiness and a vast improvement over the passive contentment or the momentary joy we wrongly call "happiness."

Leigh Arber


Author Response

I appreciate the feedback, although I stand by my statement about Aristotle’s view of teleology. Non-living natural things do have ends toward which they aim, albeit not consciously (obviously) or in quite the same way as living things or artifacts (e.g., as functions). In fact, that is the only way that our ordinary concepts of (efficient) cause and effect ultimately make any sense; there is something in the essence (formal cause) of a thing that directs it toward producing certain effects (final cause), and not others or none at all. In other words, every natural substance acts for the sake of ends that are determined by its own nature.

Jon Schmidt


November 2009

“Podium” Slabs

Thank you to Mr. Russillo for the informative article on “Podium” slabs (Nov. 2009). These structures were popular in California in the 1960s, even before the advent of post-tensioning. I am glad they have finally found their way to the Northeast.

However, let me register a small critique of the very title of this work. I believe “podium” is a misuse of the term. According to Webster, a podium is either a low wall of some sort, a continuous bench in a room, or a low platform for the conductor of an orchestra. I do not see anything like our structures in these definitions.

I suggest that we call them what they are: “platform slabs”.

Richard Martter, P.E.
Senior Structural Engineer / Founder / President Emeritus
Strand Systems Engineering, Inc.


Author Response

It is agreed that care be taken to ensure the proper use of words in order to provide the reader with the correct picture of what we are describing. The term podium was used (although the thesaurus also listed synonyms such as pedestal and platform) since it matched the term used in the Post-tensioning Manual, sixth edition by PTI which was referenced if the reader desired further information. In any case, this allows us the opportunity to reflect on the line from Shakespeare's Romeo and Juliet:

"What's in a name? That which we call a rose
By any other name would smell as sweet;"

Michael A. Russillo, P.E.
Senior Manager Special Products
Barker Steel LLC


March 2010

Welding Inspection and the New Chapter N of AISC 360-10

This letter is in response to the article “Quality Time” in the March 2010 issue of Modern Steel Construction (MSC). The Structural Engineers Association of California’s Construction Quality Assurance Committee would like engineers to be able to consider a different point of view than that presented in the MSC article, and to be aware that their own responsibility to specify the frequency of welding special inspection may be increasing.

The 2010 AISC Specification for Structural Steel Buildings (AISC360-10) has recently been completed, including a new Chapter N, Quality Control and Quality Assurance. Chapter N provides a complete and comprehensive system of quality control, quality assurance and non-destructive testing for structural steel buildings and for steel elements of composite members.

A proposal to replace the structural steel special inspection requirements of the International Building Code (IBC) Chapter 17, with the quality assurance provisions of AISC 360 Chapter N, has been accepted by the IBC Structural Committee, and will be considered again at the Final Action Hearings in May of 2010.

The Structural Engineers Association of California (SEAOC), Construction Quality Assurance Committee (CQA), is concerned that the quality assurance provisions of Chapter N, as applied to welding special inspection of multipass fillet welds and groove welds, represent a substantial decrease in special inspection over what is in the IBC now.

We feel that it is important for structural engineers to be aware of these issues, to consider how welding special inspection is typically handled in their region, and to be prepared to augment the inspection requirements for their projects, if necessary.

Current IBC Special Inspection

Chapter 17 of IBC 2009 has inspection and testing provisions for all construction, and additional inspection and testing for high-seismic construction. IBC 2009 refers welding inspection and testing for high-seismic to AISC 341, which has provisions very similar to AISC 360-10, Chapter N. IBC 2009 references AWS D1.1 for welding inspection of all construction.

The level of special inspection effort required by the building codes (IBC, BOCA’s NBC, UBC) has long been described by the use of the terms “continuous” and “periodic”. The code’s distinction between continuous and periodic inspection, and the decision as to which work items the terms are applied, has always been based on whether the important aspects of the work are fully available for observation at completion (periodic is appropriate) or whether there are stages during the progress of the work or aspects of the work process that need to be inspected before continuing or during the process (continuous is required.)

Welding inspection is “periodic” except for CJP and PJP groove welds, multi pass fillet welds, fillet welds greater than 5/16”, and plug and slot welds, all of which require “continuous” inspection.

The distinction between continuous inspection and periodic inspection for welding can best be illustrated by the following excerpt from IBC 2009, Section 1704.3, Exception 2 (this is essentially a description of “periodic” welding inspection):

“The special inspector need not be continuously present during welding of the following items, provided the materials, welding procedures and qualifications of welders are verified prior to the start of the work, periodic inspections are made of the work in progress, and a visual inspection of all welds is made prior to completion or prior to shipment of shop welding.”

Continuous inspection is described as “full-time observation” in the IBC. There is, however, quite a bit of variation in how this requirement is interpreted and enforced by engineers, testing and inspection agencies, and building officials across the country.

AISC 360-10 Chapter N Special Inspection

Chapter N takes a different approach. It presents an excellent breakdown of welding inspection tasks, and identifies each task as an Observe task or a Perform task for both the fabricator/erector’s Quality Control (QC) inspector, and for the project owner’s Quality Assurance (QA) inspector (the special inspector). Observe and Perform are defined as follows:

  • O – Observe these items on a random basis. Operations need not be delayed pending these inspections.
  • P – Perform these tasks for each welded joint.

The Observe designation is applied to almost all of the “before” welding tasks, and to all of the “during” welding tasks. The Perform designation is applied to all of the “after” welding tasks. Observe is comparable to “periodic” in the IBC and provides the same broad latitude regarding the frequency of inspection activities.

The terms “continuous” and “periodic” are not used, and there is no distinction made between types of welds, or whether a weld is single-pass or multi-pass.

The fabricator/erector’s quality control organization and function is detailed and the requirements are made explicit. Although AWS D1.1 has always required that visual inspection be made by QC of all welds, these provisions in Chapter N should help ensure that the QC inspections are made.

AISC has noted that the concept behind the use of the Observe level of inspection is that of random sampling, and has suggested that this use of the Observe level of inspection gives the inspector the flexibility to provide the inspections needed. It is fair to say that this corresponds to current practice for “periodic” inspection, and to some extent for “continuous” inspection, given the wide range of interpretation and enforcement of the “full-time” observations required by the IBC.

SEAOC CQA Concerns

The Observe level of inspection, when applied to virtually all of the “before” and “during” welding tasks, represents a significant decrease in special inspection for multipass fillet welds and all groove welds when compared to the continuous inspection requirements of 2009 IBC. Although AISC 360-10 was developed by a consensus process, we do not feel that sufficient reliability studies were performed to justify such a decrease in quality assurance welding inspection. Improved quality control may help, and this may contribute to overall quality. Unfortunately the project owner has no control over the fabricator/erector’s quality control organization, and unless the special inspector is verifying the work, cannot be assured of the desired level of quality.

IBC references AWS D1.1 for all welding inspection. AWS D1.1, Section 6.5.2, Scope of Examinations, states:

The Inspector shall, at suitable intervals, observe joint preparation, assembly practice, the welding techniques, and performance of each welder, welding operator, and tack welder to ensure that the applicable requirements of this code are met.

We have maintained that the building code has defined these “suitable intervals” by the use of the terms “continuous” and “periodic” and has assigned the more stringent interval (continuous) to those welds where it is suitable (multi-pass fillet welds and all groove welds).

Under Chapter N, once the inspector has verified the materials, WPSs, welder qualifications and skills, etc, at the beginning of a project, complete penetration groove weld joints could be started (fit-up and root pass) and completed (filler passes) without any of the steps being observed by the welding inspector (either QA or QC), for any particular weld. This would not be an abuse by the welding inspector – the inspector would be simply following the intent of Chapter N. Our contention is that this represents a substantial decrease in scrutiny over the continuous inspection currently required for this type of weld, regardless of how loosely one interprets the term “continuous.”

Our concerns are not driven by objection to change. SEAOC CQA, which is comprised of both design engineers and engineers in charge of testing and inspection laboratories, has considered input from testing and inspection laboratories who report significant differences in the workmanship of fabricators and erectors, both AISC-certified and non-certified, when inspectors are actively inspecting the work.

Engineers Take Notice

AISC has also indicated that the intent of not defining a time component for the Observe level of inspection is so that the engineer can feel free to define the frequency of inspection. It is true that the Statement of Special Inspection required by IBC Chapter 17 is intended to be prepared by the engineer (or at least the structural items should be), and that the extent and frequency of each test or inspection is to be detailed in the statement. Thus the engineer is already empowered to determine these frequencies. However, under the 2009 IBC, an engineer who is not steeped in quality assurance issues (as we in CQA are) is able (and likely) to default to the “continuous” or “periodic” requirement already in the code. Once those terms are gone, if the engineer does not further detail the frequency of Observe level inspections in the project specifications and the Statement of Special Inspections, the default frequency will apparently be up to the welding inspector.

Our concern of course is that the building code is intended to be a minimum standard, and placing the quality assurance provisions of Chapter N into the building code is a reduction in that minimum standard.

SEAOC’s Construction Quality Assurance Committee and the local California member organization’s committees have worked to elevate the engineer’s understanding of the quality assurance requirements in the code, and to emphasize the importance of the engineer’s participation in construction quality issues. Engineers take notice – your responsibilities for detailing welding inspection activities may be increasing.

Art Dell, Chair
SEAOC Construction Quality Assurance Committee


February 2010

In reading Barry Arnold’s article “The Failure of the Five E’s” in the February 2010 issue of Structure, I am obliged to provide additional information concerning the “Examination” element of his article.

The current licensure process for many U.S. states and territories does not mandate that individuals must take an NCEES examination in the same discipline in which they were awarded their engineering degree. Many people end up in fields of engineering other than their degree based upon interest, job availability, economics, etc. In such cases, once licensed as a professional engineer, these individual are allowed to practice in any discipline of engineering for which they possess minimum competence based upon a combination of education and experience. Practicing outside one’s area of competence is not only a potential danger to the general public, but an action that is not tolerated by state boards and which can result in an individual’s license being revoked.

For many years, NCEES has offered both a Structural I and a Structural II exam. These exams, like all NCEES exams, are created after studies are conducted to find out what practicing professionals in the field believe are the skills and knowledge necessary to practice this discipline of engineering. It is from these surveys of professional engineers working in the specific discipline of engineering that specifications are determined for the knowledge areas that need to be tested. A template with these specifications is then developed and used to write items for the exam. All of the efforts to evaluate the skills and knowledge needed, the design, and the regular maintenance of the exam are done under the authority and oversight of a dedicated group of professional engineers who represent practitioners and educators in the field.

For some years, many states with issues of high seismicity have required candidates who intend to practice structural engineering to successfully complete the NCEES Structural II exam. Often times, there have been additional state-specific examinations that would also be required for anyone who wanted to practice structural engineering in a specific jurisdiction. To bring more consistency to the examination process, the member boards of NCEES have voted to eliminate the Structural I and II exams and to provide a new structural exam with additional rigor to address the concerns of specific state boards. Effective with the April 2011 exam administration, NCEES will offer a new structural exam that is 16 hours in length. The exam will be divided into two distinct parts, vertical and lateral forces. Candidates must receive acceptable results on both portions of the exam in order to be considered for licensure as a professional engineer.

In designing each examination, NCEES adheres to national testing standards with the goal of providing a test that measures whether an individual has the minimum level of knowledge and skills needed to practice and be in responsible charge in a manner that will protect the general health, safety, and welfare of the public. We constantly evaluate the performance of each item offered on an exam and monitor the performance of candidates to ensure that the item is performing as anticipated and that we are measuring those elements that have been deemed as necessary for the practice of the profession.

Irrespective of how structural engineers may be licensed in the future, there will be the need for a psychometrically and legally defensible examination. NCEES is confident that those examinations currently exist and are used to the best interest of the general public.

Jerry T. Carter
NCEES Executive Director


April 2007

I have been loosely following the education committee and surrounding arguments, but recently came across the NCSEA Basic Education recommended curriculum and it provoked a few thoughts. I will say that I am generally in agreement that the typical bachelor’s degree recipient is underprepared to enter structural engineering, I would caution developing a stringent curriculum. If you force undergraduates to take more narrowly focused coursework, they will be denied the opportunity of taking important classes that aren’t strictly engineering courses. Even if you require a master’s degree the recommended curriculum could preclude some valuable engineering electives. I received a master’s degree and had many classes available outside of this curriculum (stability, bridge design, forensic engineering, reliability, not to mention research and thesis, etc) that would not be options if the recommended curriculum were followed. This is why I would be very careful in dictating an exact curriculum.

In increasing the basic education requirements, one must also consider the market for these services. There are many engineers with a bachelor's degree who are quite capable of performing certain design work and they get compensated based on their expertise. Likewise, engineers with advanced degrees are in high demand for more complex tasks and get compensated accordingly. Increasing the required level of education will increase the cost to enter the field without necessarily increasing the benefit in terms of consultant fees and salaries. I’ve read comparisons to the law, medical, and business professions as an impetus to increase the preparedness of structural engineers, but the flip side of that is that all 3 post-graduate degrees (MD, JD, MBA) are all extremely expensive and are often cost-prohibitive. It is becoming increasingly difficult to justify the mammoth cost of going to medical school or business school when the end benefit (salary) isn’t growing fast enough to keep up with the cost.

I believe that either increased licensing requirements either through experience or examination and continuing education is the best way to advance the profession.

Sincerely,

Adam Johnson, P.E., LEED AP
Associate
WALTER P MOORE


November 2009

Mr. Hatem & Tuller,

I enjoyed reading your recent article in the November Issue of STRUCTURE. It was a well written article.

I noticed that you used the example of the design of stairs a number of times in the article to illustrate a point relative to design delegation. You may be interested to know that in the case of cast-in-place concrete stairs and exposed monumental stairs (structural steel or concrete) it is common practice for the Structural Engineer of Record (SER) to take responsibility for the direct design of the structure.

You may also be interested to know that the primary reason that a SER will typically delegate the design of structural steel stairs to the fabricator is because there is a multitude of different ways to frame out any given steel stair. As a result most SER’s avoid detailing one particular method over another because 9 times out of 10 the fabricator will want to design and detail the stair using a different framing method from that shown by the SER. Therefore, as a general rule, to avoid the wasted energy of designing and then reviewing an alternate method of framing, the design of stairs are delegated by the SER to others. To avoid change order requests for beams designed by the SER that directly support the stairs (i.e. the beams located around the perimeter of the stair opening) it is also not uncommon for the SER to make conservative, redundant loading assumptions that cover a multitude of different potential stair framing support scenarios.

Somewhat similar logic has also been used when it comes to delegating the design of structural steel connections, however, the primary reason this is commonly done in the industry is because the minuscule fee (typically 1/2% of the estimated construction cost) that most Architects allow their structural consulting engineers to charge barely covers the cost of the structural design of the main frames, components and foundations, much less the steel connections. For years the AISC tried to push structural design professionals into providing the connection design as a part of the bid documents, however, recently the AISC finally accepted this common practice.

It was also interesting that your article mentioned peer reviews in the last paragraph. As a result of my own professional experience (both as the reviewer and the reviewed) I have taken a keen interest in peer reviews and have published a paper on the subject (see reprint in January 2007 STRUCTURE Magazine: http://structuremag.org/Archives/2007-1/p18-19C-ProfIssuesProjectSpecificPeerReviewGuidelinesJan-07.pdf) and a follow up article on the topic (see reprint in June 2007 STRUCTURE Magazine: http://structuremag.org/Archives/2007-6/C-Sforum-Peer-Review-Stuart.pdf). I am currently working with CASE (ACEC - Council of American Structural Engineers) towards the development of a guideline for project specific peer reviews that can be used for peer reviews that commonly occur outside the realm of legislated, mandated peer reviews in states such as Connecticut and Massachusetts or cities such as Chicago.

D. Matthew Stuart, P.E., S.E., F.ASCE, SECB


June 2008

I read your interesting article, published in Architecture Week. These systems are still in use in Spain, Cuba, Mexico and other Latin American countries.

As an architect in Cuba in 1987-1997 I was in charge of building a factory for the production of precast pre-tensioned beams (viguetas – small beams) for one of these structural systems. The factory included also a facility to produce the concrete block infill tiles. They are called bovedillas in Spanish, due to the fact that structurally they act as small vaults - bovedas. The equipment had been bought in Spain (Cataluña) by the Cuban Ministry of Construction Materials, for which I was working as an employee. (There is no private practice of architecture in Cuba). Once the factory was finished, we started using it in the construction of dwellings and small commercial/office buildings.

In Spain the system is very popular and used for residential and light commercial work, with spans of up to 21 feet, and cantilevered up to 9 feet. The Spanish Normas Basicas de Edificacion of that time used to have a lot of detail about this type of construction.

As for the older types of systems, working in Cuba in restoring older buildings gave us a lot of experience on them too. They were very popular from the 1910’s to the 1930’s, when cast in place concrete replace this kind of systems. Still in the 1950’s a Cuban architect re-introduced a similar type of construction system, with precast concrete beams and clay tile arched units. This system was used until the late 60’s with the trade name PEPSA.

In Mexico several systems of this type are used. You can see an example in http://www.losaryd.com.mx/sistema.htm (Mexico, with infill tiles made of polystyrene), http://vigatec-eirl.com/index.html (Peru), etc.

I hope this e-mail would be of interest to you. If you are interested in some more details about this subject, please do not hesitate to contact me.


Julio Guillén


Viguetas PEPSA:


Typical Spanish “vigueta”:


In Mexico:



May 2009

To the Editor:


I was interested in the InFocus article by Jon Schmidt in the May 2009 issue, entitled “The Nature of Theory and Design”. My background is in structural engineering and engineering mechanics. I have not read the book referenced in the article, so I am commenting only on the article itself. I would like to focus on the second and third paragraphs.

They are a little vague to me, but I suppose the point is that theoretical analysis cannot predict exact stress-strain response and “how structures behave” refers to observations of real structures, either by experimental testing or by studying actual failures and successes. I suppose that “bridge the gap” refers to making theory so good that it includes the design details. What I find missing here is that there are certain insights into structural behavior - particularly in complex arrangements of components, sensitivity characteristics, and effects of complex loadings - that can only be obtained by theoretical analysis. With this, I offer the following comments regarding my understanding of theory and practice.

Theory of any field represents the fundamental knowledge of that field, the collection of all information that is considered to be known about the field. It is open-ended so that more information can be added as new information is found, and it is general in nature, making statements about a large class of systems. It seems to me that physics, which is mature and deep in knowledge, is the most advanced science with regard to theoretical mathematical models. Some of these theories have been formalized into axiomatic systems, which are about as high as you can go; e.g., theory of particle mechanics and theory of rigid body mechanics. The reason I bring this up is because theoretical models, no matter how sophisticated, are never intended to encompass details like those required for structural design, where everything has to be translated into a physical detail made to fit into a unique complex physical system. Design is uniquely dependent on the details for that specific structure. Theories speak to generality, to a class of systems, but they can still furnish related information for specific problems. In fact, all three of the independent models, “one for materials, one for individual components and their arrangement, and one for the loads,” are underlain by theory, though this is not necessarily obvious to the designer.

Prediction (specific deduction) is the function of any theory even though it is not exact. It is only as good as the assumptions (axioms) on which the theory is based. Improved assumptions based on additional information make it better, but it will always be approximate without design details. It doesn’t matter whether steel as a material is linearly elastic or not; it is a logical first approximation to a solution to a complex problem for “small” deformations. If this does not provide sufficient accuracy, then nonlinear effects need to be included, and theory provides the limits of the assumptions and where to go from there when these limits are exceeded. The exact location of joints does not matter if the system is not sensitive to their location. Wind loads can be assumed to have any distribution you choose. The designer has to fill in the difficult details so that everything works together in a safe structure, but theory provides limits, guidelines, and framework for all of this work. I guess I am saying that I do not know how you separate theory from design on any level except within the individual’s focus when dealing with details.

I guess I agree with the statement that theory and design have distinct objectives and cannot be merged by making theory more exact. That is not the function of theory. But they certainly have an interaction that is essential for understanding complex structural behavior and providing an approximation to system response under a specified load for the designer to use. Keep those safety factors intact.

Thank you for your contributions to an interesting and important field.

Sincerely,

Arnold E. Somers, Jr.


January '09

Commentary from David Shepherd pertaining to the article "Tools for Reducing Carbon Emission due to Cement Consumption"

My thanks to Dr. Kumar Mehta for helping to broaden the perspective on concrete in a sustainable context. (Tools for Reducing Carbon Emission due to Cement Consumption, Structure Magazine, Jan. 2009) In the not so distant past, I have read similar articles that seemed to imply that adding fly-ash was all it took to make concrete “green.” Sustainability encompasses a much wider range of perspective, and advances the concept of balance in business decisions across environmental, economic and social factors.

As the largest manufactured product in the world, the concrete industry has a big footprint, and an equally large responsibility - to make our products better and to help our customers optimize the value in the applications we offer. Dr. Mehta notes multiple strategies to improve the materials aspect of concrete through recycled content, efficiency in mix design, and durability. But let’s not think we can stop there.

Dramatic improvements to reducing the ecological impact of the built environment will require evolutionary and revolutionary changes in how we design, construct and use the structures we create. Life-cycle assessment studies reveal that the operation of a facility contributes 85% to 95% of the environmental impact over the total life cycle of the building. Understanding this puts into perspective the real value that the design community can offer by specifying durable, energy efficient structures and envelopes. Consider the synergy derived from integration of daylighting and lighting controls for optimized illuminating performance. Significant savings can be achieved. Now imagine a design process where the structural and mechanical engineers work in concert, integrating the structure, envelope and HVAC system; storing energy in the mass, tightening exterior losses, and reducing floor space and structural loads through downsizing HVAC equipment and ductwork.

Also recognize that while climate change and greenhouse gasses are at the forefront of concerns today, sustainable development extends to indoor and outdoor air quality, land use and bio-diversity, water and material resources, urban design and density, and energy dependency among other issues. Dr. Mehta also briefly mentioned designing for disassembly, an excellent application for pre-cast concrete components and other modular products. This notion dovetails with the value provided by concrete as a highly durable material, further reducing the need for replacement materials, landfill space, or energy to re-process products.

Climate change, a struggling economy reliant on foreign energy, and global competitiveness are all compelling reasons to re-visit how we do things. Regardless of your perspective on global warming, sustainable development provides a framework for the challenges we face.


December '08

Commentary from Donald C. McElfresh pertaining to Matthew Stuart's articles "Antiquated Structural Systems Series"

Dear Ms. Sloat:

I am not familiar with Mr. Casper and his working with Ken Bondy, of Atlas, during and after 1966 “when T. Y. Lin’s load balancing method was introduced to our profession…”

However, when I joined T. Y. Lin & Associates in November 1962, we used as our base design reference T. Y.'s book “Design of Prestressed Concrete Structures (1955). In Chapter 11, Slabs, Page 329, he speaks of Two-Way and Simple Flat Slabs, where he refers to “The only basis for their design is the design of reinforced-concrete two-way slabs, moment coefficients for which are available from building codes on reinforced concrete.” This reference is, of course, known to all of us from ACI 318.

In the June 1963 issue of the Journal of the American Concrete Institute, was published “Load-Balancing Method for Design and Analysis of Prestressed Concrete Structures” by T. Y. Lin. Internally, within the company it was already using load balancing from the day I started with TYLA.

Also in 1963 was published T. Y.'s 2nd Edition of “Design of Prestressed Concrete Structures.” Chapter 11, Page 339 was titled “load-balancing method.” Starting with Page 356, “11-5 Two-Dimensional Load Balancing”, T. Y. discusses tendon placement distribution in the middle and column bands. Later on Page 386, under “12-4 Flat Slabs, Theoretical Considerations” T. Y. states that “some theoretical problems till deserve further investigation…B. The proper distribution of the cables along the column and the middle strips. This can be studied by either the elastic theory for plates or the balanced-load explained in Chapter 11.”

Within the four TYLA offices (Van Nuys, CA; Chicago, IL; Dallas, TX; New York City, NY), and the T. Y. Lin International office (San Francisco, CA), during the early 1960’s, there were many inter-office discussions about two-way post-tensioned concrete slabs, column-middle band post-tensioning distribution ratios, possible problems, and how to create the best possible designs.

Possibly the above will shed some additional “light” on an “Antiquated Structural System.”

Sincerely,

Donald C. McElfresh, S.E., P.E.


July '08

Commentary from Mr. Alfred Commins pertaining to Mr. Ronald Nelson's article "Another View About Shear Wall Hold-down Systems"

All five people who wrote, contributed to or reviewed the article are competitors of Commins Manufacturing with their own agendas and bias. The engineers are presented as if they are independent. They are not independent.

The Author, Mr. Rawn Nelson, is a paid consultant to Zone Four, a supplier of a competitive system that incorporates a ratcheting take-up device. Mr. Edward Chin is a principal of Earthbound Corporation, a supplier of a ratcheting take-up device. Mr. Rick Fine and Mr. William Nelson are paid consultants to Earthbound Corporation. Mr. Steven E. Pryor, S.E., is an employee of Simpson Strong-Tie.

I have three degrees, including one in Mechanical Engineering Technology. I am not a P.E., but I consult with qualified structural engineers as needed for specific tasks. I have a skill set and design experience substantially different from those of the typical structural engineer. I have been designing structural hardware for over 30 years. Products I have designed include the HDA, PAHD, HPAHD, Strap Ties, the Simpson StongWall, double shear nailing, the SDS screw and many others, I have performed over 300 full-size shear wall tests, many to ICC-ES AC 130.

I have designed and tested many of the devices described in the article. These devices met the criteria in effect at the time they were introduced but may not work as expected during an event such as an extreme windstorm or earthquake.

Promised Lateral Capacity

The lateral capacity of shear resisting panels, commonly called shear walls, is defined in the code and can be traced back to ASTM E72. This testing protocol includes tying one end of the shear wall down with a plate and positioning a rod on either side. This hold-down is extremely stiff and very reliable. This testing is for rating the shear resisting capacity of sheathing, nailing and studs. The hold-down connection is to be designed by the engineer and may be a hold-down, building weight or a combination of the two. “Unsafe” is a claim made by Mr. Nelson. I do not know if these walls are “Unsafe,” but I do know that they are not providing the expected lateral performance.

Mr. Nelson is providing a red herring by taking my comment about the four factors required for shear walls to perform out of context. The article concerns tying the shear wall to the foundation or floor below with hold-downs. Other factors such as foundations or shear wall construction are not considered in this article, only vertical connections such as straps, hold-downs and rod systems, and their contribution to the performance of the shear wall.

I am a great fan of shear resisting panels, and in awe of the simplicity and performance that these shear walls provide. The purpose of these articles is to look at the contribution to the performance of shear walls that the hold-down provides and how it might impact the capacity of the whole.

Are Shear Walls Needed?

Mr. Nelson’s comment here is also out of context; just the opposite is true. Many engineers look at STRENGTH only but do not look at STRETCH, SHRINKAGE and RELIABILITY. The point is that we need stretch control, shrinkage compensation AND reliability in addition to strength. If elements are missing, the shear wall may not perform as expected.

Shear Wall Failures

1. Splitting of sill plates

Mr. Nelson is partially right. Sill plate splitting can occur at any load, but a tight, stiff hold-down can substantially raise the load at which the splitting will happen, sometimes by a factor of 5 or 10. Conversely, a “flexible” hold-down will allow the wall to rotate at a relatively low load, which burdens the bottom plate with a twisting load and can fail the shear wall at a load substantially below its rated capacity.

2. Splitting of vertical wood studs

Splitting of vertical wood studs is not due to wood compression. This splitting is commonly due to bolted eccentric holdowns. Reduced wood section coupled with an eccentrically loaded bolt will tend to split the post. As stated, the loads may be due to wind or seismic loading. But, if bolted hold-downs work so well, why are they tested on a steel jig and used with a 2½ safety factor and a further adjustment for wood? A test on wood would be more realistic.

The looseness associated with commonly available bolted hold-downs is one source of shear wall rotation. Keep in mind that the framer is allowed to drill the hole 1/16 inch oversize. The oversize bolt hole is additive to all other sources of stretch or looseness.

3. Nail pull-through, bending or breaking

The load at the corners that will cause the failure to precipitate is much lower with “loose" or "flexible” hold-downs than with tight, rigid hold-downs. An analogy is to compare a shear wall to a phone book when torn apart by a strong man. This is a commonly known parlor trick. When you try to tear a complete phone book, you can’t do it. But the strong man will bend the book in such a way that pages are loaded sequentially, and he easily tears the book apart. In a similar manner, a flexible hold-down will load some corner nails more than others and the wall will fail sequentially, nail by nail. A tight hold-down will tend to load the nails all at the same time, and the wall will perform much better.

During the research that led to the Simpson Strong-Wall, I discovered that the only way to "push” the wall capacity was to address each and every failure and failure mode as it occurred. The Simpson Strong-Wall has 2½ times the stiffness and strength of a similar wall (at least with a 4-foot x 8-foot wall). This cannot be achieved except with every item properly performing.

Section 3.5.3.1 in AC155 specifies a maximum displacement equal to 0.250 inch. Some manufacturers specify a design capacity at a maximum displacement of 0.282 inches. This 0.282 inch is additive to other system deflections. Uplift deflection is a function of all the component deflections added together. A rod system may include rod stretch, bearing plate crushing, shrinkage compensator deflection, hold-down deflection, and shrinkage. Shrinkage introduces uplift deflection without load. It adds to system deflection and must be included.

AC316 rates shrinkage compensators, but there is a problem. Some shrinkage compensators introduce substantial looseness or backlash. Backlash is a function of rod pitch and internal looseness. I have measured backlash at up to 0.190 inches. What good does it do to introduce a product that eliminates ¼ inch of shrinkage looseness, then adds back 0.190 inch?

Shear Wall Hold-down Checklist

In the last three years, I have had the opportunity to review hold-down systems for some 500 projects. Stretch limits are specified on perhaps one out of 100 projects. When I see stretch limits, I see only rod stretch and not total system stretch; of the 500 jobs, only one specified system stretch. Most engineers do not appear to include bearing plate crushing, and do not look at shrinkage compensator deflection

1/8-Inch Stretch and Loose Shear Walls

The ⅛-inch deflection limit is a suggestion for what is needed to allow shear walls to perform to their potential. This number should include all tension items including rod, bearing plates, shrinkage compensators, hold-downs, etc. It does not include couplers, nuts and washers. There appears to be a threshold where shear walls held by tight connections perform. We do not know exactly where this threshold is, so the ⅛-inch limit is suggested as a starting point
To quote the report: Report of a Testing Program of Light-Framed Walls with Wood Sheathed-Shear Panels

“Groups 31 and 32 investigated the effects of free movement at the hold-down anchorages. The free movement allowed was 0.2 inches for Group 31 and 0.4 inches for Group 32. The effects were not significantly different for the free movement allowed. (i.e. 0.2 and 0.4 performed in a similar manner). The Nominal Strength in shear was reduced to about 60 percent of the mean for similar panels. A similar effect on Elastic Shear Stiffness was found. The displacements at the YLS and SLS were relatively unchanged. The SLS shear was unchanged.” (emphasis added)

Several years ago, I was researching the performance of wood shear panels for Simpson Strong Tie. During one test, I loosened a bolt ¼ inch to mirror the effects of shrinkage. The result was a reduction in lateral capacity of about 40%. My experience is congruent with the results of the testing cited above.

Building Settling and Shrinkage

While called a “Shrinkage Table”, this table also includes settling due to misalignment of studs and other assembly issues. By definition, the shrinkage table overestimates settling by assuming worst-case conditions. Most tables look at averages. If averages are used, half the time the shrinkage will be underestimated, and half the time it will be overestimated. Perhaps half of our connections would be loose by using averages, so I took the safe way out. Normal shrinkage estimates assume a degree of precision that is only achieved in the laboratory. The table is not meant to usurp the authority of the engineer; it is only meant as a simple guide

Shear Wall Designs

Hopefully both groups will include elongation from all sources that contribute to uplift movement including, but not limited to, rod elongation, plate compression (and bending), shrinkage compensator deflection and backlash, hold-down deflection, etc. My experience with designers is that they tend to use an average, rather than a worst-case condition. Loose connections cannot be good.

Rod Elongation

The requirement by AISC that nominal area be used sets up an internal conflict. Since the net area, as Mr. Nelson states, can be only 75% of the nominal area (actually varies from 74% to 79% depending on the diameter and thread pitch), the rod can be overstressed, then to use the actual area for stretch introduces another variable. In this case Mr. Nelson is correct, but we see some manufacturers using one number and others using another. The City of San Francisco in its Administrative Bulletin AB-084 (AB) on the subject specifies using the nominal area. This area needs clarification. Should ICC-ES ever provide an Acceptance Criteria for Complete Tie-Down Systems, this is an area that needs clarification.

Strap, Hold-down, or Continuous Tie-down Systems and System Type Take-up Devices

The article did not suggest that connections be designed to resist compression loads. If connections are degraded by compression, such as buckling of straps or rods, such degradation must be considered. If not considered, then just a few wind or seismic cycles can destroy the connection.

I have encountered a sketch showing a strap combined with a rod system. If the rod system has an allowance for shrinkage, why is a strap used at the top? Is there no shrinkage at the top of the building?

I checked with my sources at APA, and they stated that they have not tested a stacked hold-down system. As manager of research and development at Simpson Strong-Tie until 1997, I do not remember ever testing such a condition. Quite frankly, I never even considered it until three or four years ago. Please remember that the performance of the shear wall is at question here, not necessarily the hold-down itself

Shrinkage Control Devices

Devices that comply with AC316 may be a good addition, unless they introduce extra looseness. Just as all engineers are not the same, even if they studied the same material and passed the same examination, all products are not the same, even if they pass the same Acceptance Criteria. Just as some hold-downs are superior to others, some shrinkage compensators are superior to others. There are differences. Some differences are critical to system performance.

AC316 is now in its fourth revision. It started as an acceptance criteria for screw type devices. It has been expanded to include ratcheting devices.

Evaluating agencies make errors, and it may take a while to correct them. In 2002, I noticed an error in a code report. I notified the manufacturer and ICBO (now ICC-ES). The manufacturer has corrected the error, but even today ICC-ES still publishes the same erroneous detail. Once an error is memorialized, it is extremely difficult to delete.

Every device in every photograph in the March 2008 article is incorrectly installed. I have played no favorites. The problem is that moving elements tend to freeze up during installation, during movement or over time unless extraordinary precautions are taken.

Backlash is a term not found in AC316, but it should be. Backlash, as defined by the American Heritage Dictionary, Second College Edition (1982), is “The play resulting from loose connections between gears or other mechanical elements.” Backlash is a term used by mechanical engineers and is not commonly used or understood by structural engineers.

In rod systems, backlash is the looseness caused by the thread pitch (0.090 inches for a 5/8-inch 11-tpi rod to 0.143 inches for a 1¼-inch 7-tpi rod) plus the movement of internal parts. For comparison, screw devices can have backlash as low as 0.001 inches.

The take-up deflection of 0.012 inches is an estimate based on a series of load-deflection tests. For most purposes, it is not significant.

There are at least five different rod ratcheting devices. There may be specific elements on one variant that are not included on the others. For example, the number of ratcheting elements varies from two to four in the samples that I have examined. Every device that I have inspected includes at least one spring.

According to the American Heritage Dictionary, 2nd edition, page 1178, a spring is a flexible elastic object used to store mechanical energy. The device that Mr. Nelson is associated with includes a spring clip that encloses four jaws and pulls the jaws together so that they can engage the rod. The circular clip is a spring in the classical sense.

In 1996, I discovered rod-ratcheting devices and believed that they would solve shrinkage problems. I set up a cyclic test with a wood shear wall and tested the devices in a system according to AC130 (actually a predecessor to AC130). After several cycles, the system released. An examination of the rod showed a stripping of the threads. The device failed the rod, but the device itself was still functional. While this was a rod failure, it was caused by a concentrated load from the device. Failure load was below the design load of the system.

At the time, I ignored the reason for the failure. Since then, I have tested other rod ratcheting devices and failed the rod in a similar manner. My conclusion: the devices “cross thread”. What is Cross Threading? The definition is to screw together two threaded pieces without aligning the threads correctly. The drawing below shows the problem and a correctly installed part.

Ratchet devices will do the same thing during advancement, and load one segment at a time. This overloads the thread on the rod. The devices and the rod do not stay in line with one another. A building will rock back and forth during an event, and the rod and device will also rock. During the rocking, one segment will ratchet onto the rod, but other segments will be out-of-square to the rod and thus not engage. One of the ratchet segments will lock onto the rod and carry the entire load. The overloaded rod segment will fail.

Of course, the point is that structural engineers are not trained in this area and may not understand the mechanics of moving parts. To my knowledge, none of the people involved with Mr. Nelson's article and none of the engineers at ICC-ES are mechanical engineers. Clearly this is outside their area of expertise.

The photograph above shows a rod with deformed threads. This rod was tested in a ratcheting take-up device, according to AC316. I stopped the test when the threads began to yield.

My experience with rotating devices (of which this is one) is that they tend to bind if the elements are not in line. The drive shaft of a car has universal joints to prevent binding. The prop shaft in a boat either has a universal joint or is aligned carefully to prevent binding. From a practical perspective, binding appears to be a problem, and I am not the only one who sees it this way. A competitor reviewed the product, prepared a nine-page analysis and presented it to ICC-ES in October 2007.

Conclusion

As a solution to this controversy, I suggest that all parties selling rod systems and hold-downs have an independent laboratory test complete shear wall systems under identical conditions. All interested parties, including ICC-ES, could witness the tests and share information.

January '08

I read Gerard Feldmann’s article Non-Destructive Testing of Reinforced Concrete that appeared on page 13 of the January issue with great interest, since for the past few years I have been heavily involved in the development of an instrument to perform that type of testing. The article is extremely informative and very comprehensive.

An additional method could be added to the many mentioned by Mr. Feldmann, ultrasound echo (UE). UE finds delaminated areas of concrete and has been shown to accurately identify delaminations when tested on a concrete slab purposely built with such defects. . It utilizes two ultrasonic probes; one transmits a stress wave field into the concrete specimen, and the other receives a signal corresponding to the natural response of the concrete specimen in the time domain. A Fast Fourier Transform (FFT) is then applied to the received response so it can be examined in the Frequency Domain.

When testing a floor with the UE method, the frequency response of an area of satisfactory quality will be uniquely determined by the floor thickness (thickness mode of vibration). In this situation the response sensed by the receiver is from wave reflections occurring at the floor to subgrade interface and at the floor surface. Testing at a known thickness to calibrate the concrete wave speed is not required, as the wave speed is calibrated by measurement of the surface P-wave. At an area with a shallow delamination, the response will shift to a lower frequency and be determined by the existence of delamination (flexural mode of vibration).

 

Ed Pristov

Inspection Instruments, Inc.

 

Author's Response:

Thanks for the compliment on the article.

Yes, I have heard of the test method you mention, but did not include it in the article due to space limitations and my lack of experience with the method. It seems I should look into it a bit more.

Thanks again,

Jerry Feldmann

 

September '07

Thank you for using dwgs from the HAER collection for the West Baden Springs Hotel article in the September 2007 issue. I was HAER principal architect during the summer of 1973 when the West Baden Springs Hotel was recorded. The Historic Landmarks Foundation of Indiana and the Indiana Historical Society were cosponsors. It was the second of a two-summer project to record a selection of engineering and industrial sites in Indiana.

These were heady and exciting years for HAER (Est. 1969) as we were defining the new field of industrial archeology and engineering heritage by recording engineering phenomenon and industrial processes with measured and interpretive drawings, historical data, and large-format photography. Though I retired nearly four years ago, the program thrives under the capable leadership of Rich O’Connor.

SIA, the Society for Industrial Archeology, is in the process of scanning issues of IA: The Journal of the Society for Industrial Archeology, with The History Cooperative. Unfortunately, Vol. 25, No. 1, which is a theme issue on the first 30yrs of the HAER program, has not been scanned. You can find the journal in any descent university library or it can be ordered from SIA, www.siahq.org. See “HAER: 30 Years of Recording Our Engineering Heritage,” IA: The Journal of the Society for Industrial Archeology, Vol. 25, No.1, 1999.

 

Eric DeLony

Chief (Retired)

Historic American Engineering Record

National Park Service

pontist@comcast.net

 

September '07

While I think that Larry Muir's article in the July 2007 issue, 5 Common Myths of Steel Design Debunked, raises excellent points, I must take issue with the net section discussion listed under Myth #1.  Many members can be connected with an effective net section much greater than 75%, even when heavy truss chord splices are made with 4 rows of bolts (double gage) across the flanges, by simply using one or two "lead-in" rows of bolts in the flanges before the web bolts are started.

In addition, there are many instances where, due to different member actual depths, significant shimming is required between members and connection material.  Take for example a heavy truss tension chord splice where the center segment is larger than the next outboard segment.  If the outer member size is dictated by an arbitrary 75% net section capacity, a much larger (and more expensive) section will result.   This is especially important in our current market state of high-cost raw steel.

So, while I am 100% in agreement with Larry's general concept that less material weight does not always equate to less cost, this is one instance where I believe that it does.  I have witnessed engineers attempting to limit member stresses to accommodate the arbitrary AISC listed values without thinking about how the members may be connected and determining whether the full member capacity could indeed be used.  Experienced designers working with competent fabricators need to make this type of decision considering many other criteria specific to the project.

Regarding Myth #5, Larry is 100% right on:  Owners should only hire top-notch, experienced fabricators and erectors.  Unfortunately, not all such firms have the professional credentials, capabilities and competence of Cives, so the Owners do not always have a choice … too bad for us designers.

Sincerely,

W. Steven Hofmeister, P.E., S.E.

Thornton Tomasetti

Kansas City, Missouri

 

Author's Response:

Steve brings up a good point. I did not intend that an engineer should always increase the member size to accommodate the reduced area due to bolt holes, but rather that the effects of the connection design should be considered on a case by case basis in the main member design. Least weight does not always equal least cost, but sometimes it can.

If the decision is made to reinforce a member rather than increase the member size, the reinforcing and associated welding should be indicated on the engineer’s drawings so that it can be accounted for in the bid. Having the engineer and fabricator work as a team early in the project often allows these issues to be more effectively addressed in the design phase, and can lead to increased economy for all parties.

Steve’s comment about the conservatism inherent in the 75% reduction reflected in the AISC Manual is also valid, and the Manual Committee is currently working to revise these tables to be more accurate and useful to the designer.

Larry S. Muir, P.E.

Cives Steel Corporation

Roswell, Georgia

 

August '07

Dear Mr. Weingardt,

I just wanted you to know how much I enjoyed your profile of Timoshenko in the August 2007 STRUCTURE magazine (funny how those back issues pile up). I still refer to Timoshenko's books, and had the pleasure of learning under his co-author Jim Gere, but never knew the turbulent story of Timoshenko's life, or of his wit. Thank you for filling that gap in my technical education!

Leonard Martin Joseph, P.E., S.E.

Senior Vice President

Thornton Tomasetti

Irvine, CA

 

August '07

Thank you for the article in the August 2007 “Great Achievement” series of STRUCTURE ® about the “father of engineering mechanics,” Stephen P. Tymoshenko.

Among his many contributions to science worth noting was the creation of the Ukrainian Academy of Sciences in Kyiv in 1918, of which he was a founding member. A Ukrainian commemorative postage stamp of 1998 illustrating Mr. Weingardt’s biographical sketch witnesses reverence of his illustrious personality by Ukrainians at home and all over the world.

Roman Wolchuk

Fellow, ASCE    

 

August '07

I would like to comment on the gable end article in your August 2007 issue. Gable end trusses are not allowed in Miami Dade County, and for a good reason. They leave the wall under them basically unbraced, unless the top of the wall under them has a structural element that can span the length of the wall and brace it for out-of-plane wind forces.

For taller walls (ever more present as homes get more and more expensive), there is basically no wood-framed bracing system that can be designed to transfer those forces back in the remaining of the roof structure unless fairly large members and bolted connections are utilized.

A gable end wall should have continuous framing from the foundation all the way to the top of the wall, where roof sheathing and blocking can resist the forces. If the walls are masonry, the cells should be reinforced as such. If they are wood framed, they should have studs continuous from the foundation to the roof sheathing.

Eugenio M. Santiago, P.E.

Chief Building Official

Key Biscayne, Florida

esantiago@keybiscayne.fl.gov

 

Response

Thank you for your comments regarding the article on Wood Truss Gable End Frames. The scope of this article was to provide basic and additional design considerations for gable end frames (i.e., vertical and lateral load considerations).  

Page 55 of the article provides information related to gable end frames under lateral loads acting parallel and perpendicular to their plane. In addition, your comments concerning continuous framing are addressed on page 55 and 56; citing section 2304.3.4 of the FBC with regards to Gable End Wall Bracing.  Both Jim and I conducted several claim investigations in Florida related to hurricane and tornado damages, and are familiar with building code requirements in Broward and Dade County.

Figures 6, 7, 8, 10 and 11 (full article) deal with gable end frames resisting lateral loads, stress and/or deflection concentrations and recommended details for gable end walls and/or frames. Figure 11 is taken from SSTD-10-99 (Standard for Hurricane Resistant Residential Construction), which is also provided in AF&PA’s WFCM (Wood Frame Construction Manual).

(The full article can be viewed in STRUCTURE’s online archives.)

Agron Gjinolli, P.E.
WTCA

Agjinolli@qualtim.com

 

July '07

The article The Case for an Engineer of Record for a Metal Building System in the March 2007 issue of STRUCTURE magazine contains good recommendations for Owners who are considering Metal Building Systems.

One item that is questionable from the Engineer of Record (EOR) perspective is the section on “Inspection Services”. For most other projects, the EOR agrees to “observe” the structure during construction to ascertain substantial compliance with the Contract Documents that the EOR prepared. According to the IBC, “inspections” are performed by the Special Inspector hired by the Owner.

It is an abdication of responsibility that the metal building manufacturer does not observe the metal building system during construction. The location of the manufacturer is not a valid excuse. The metal building steel erector is generally under the same contract as the metal building designer. The design, erection details, drawing quality, construction conformance with the drawings, and construction quality are all the responsibility of the metal building designer working with the erector.

While it is preferable to have the EOR review the Order Document, that is not always done. Metal building manufacturers should adhere to the current industry practice of inspection by the Special Inspector and site observation by the designer, rather than shift such responsibility to the EOR.

Sincerely,

Lawrence R. Chute, P.E.

DESAI/NASR CONSULTING ENGINEERS, INC.

 

Mr. Chute brings up an interesting point about whether a metal building manufacturer should be responsible for the observation or inspection of a metal building system during construction. We feel that the “special inspector” hired by the building owner should be someone who is independent from the erector/building manufacturer. The metal building manufacturer has a conflict of interest on top of the logistical problems associated with performing the inspection, because the erector is commonly the customer of the manufacturer. Some manufacturers do include onsite inspection as part of their contract, especially with regard to accepting warranty responsibilities on large or complex projects, but if a manufacturer does not include this service, it should not be considered an abdication of responsibility.

W. Lee Shoemaker, P.E., Ph.D.

 

July '07

Metal Building Systems and the EOR

Dr. Shoemaker's article on metal building system structures in the March 2007 Structural Forum column presents a comprehensive overview of the metal building system world.  Metal building systems provide very efficient structures by keeping material costs at a minimum.  As a structural engineer who has been responsible for the erection engineering of these structures, I have found that low material cost can require extreme efforts to afford safe erection of a structure.  Girders and rigid frames that are adequate, when all intermediate framing members are installed, can require elaborate analysis and special bracing to allow their erection.  Anyone entertaining the use of a metal building system structure needs to be aware that the Metal Building Systems Manual excludes structure erection design from the manufacturer's responsibility unless specified otherwise by the purchaser.  Metal building systems can provide durable, low-cost structures; their design should include a method for getting them erected, safely.

Alan D. Fisher, P.E.

Manager, Construction Structures Group

Cianbro Corporation

 

Mr. Fisher is absolutely correct that metal buildings, like other types of construction, require careful planning for the safe sequencing of erection. Metal building erectors are no different from other steel erectors that are typically responsible for determining the best erection sequence and method based on their available equipment and experience. If the metal building erector requires additional special engineering to help determine the erection procedure, such as Mr. Fisher’s expertise, it seems more appropriate that the knowledge of the site-specific conditions be handled by someone other than the metal building manufacturer, because of logistical and practical constraints. Local design and oversight would seem to be the most realistic method to achieve the goal of safe erection.

W. Lee Shoemaker, P.E., Ph.D.

 

July '07

Communicating with CAD

The author hit the nail squarely on the head with the InFocus column Cad-How It Has Changed the Way We Think (STRUCTURE ®, April 2007).  I passed it to several others here, who immediately had the same reaction. Good job!!

Every one of your points were accurate: the lack of x-refs being bound to the transmitted files,  the lack of dimensions and the Architect’s attitude – “just scale the cad drawing” and when you do it has some funky fraction at the end.  All that stuff is the norm around here and a source of everyday frustration.  Thanks for letting us know we are not alone in this gripe.   

Stephen M Rudner PE
Robert Darvas Associates PC
440 South Main Street
Ann Arbor, Mi. 48104
(734) 761-8713 ext 12
Fax (734) 761-5236

 

July '07

NCEES Model Law

I was reading the article titled “Is Four Years Enough?” in the April 2007 issue of the STURCTURE ® magazine and felt that some readers might be lead to a wrongful conclusion based upon a couple of sentences contained in the article.

The articles states “ASCE drafted a model registration law for consideration by the National Council of Examiners for Engineering and Surveying (NCEES). The model law incorporates the above education requirements.” My concern is that action taken last year by NCEES to amend its Model Law to require a Bachelor's Degree plus 30 additional credits would be attributed to a similar study that has been under review by ASCE during recent years.

NCEES was fully aware of ASCE’s review of this matter, and the development of a body of knowledge on what one might need to know in order to be qualified for the professional practice of engineering. NCEES has studied this matter separate and apart from ASCE, which included work by two distinct and separate NCEES task forces. After several years of study and review by these task forces, a motion was made and ultimately approved by NCEES to amend its Model Law to require, effective 2015, that candidates for the Principles and Practice exam must have a Bachelor's Degree in Engineering plus 30 additional credits. The NCEES Uniform Procedures and Legislative Guidelines (UPLG) Committee has been tasked this year with defining what constitutes 30 additional credits as satisfactory for pursuing licensure.

The point of clarification is that the action taken be NCEES during their 2006 Annual Meeting was not as the result of the ASCE study, but as the result of NCEES’ on study and findings.

Jerry Carter

Associate Executive Director

National Council of Examiners for Engineering and Surveying

 

May '07

The debate between thermally Restrained versus thermally Unrestrained assemblies, has existed for over 30 years. Industry groups, such as the American Iron and Steel Institute (AISI) have sponsored studies, which lead to the conclusion that all structural steel frames, independent of the level of restraint, can be classified as Restrained assemblies. Unfortunately, the conclusions drawn from these studies do not take into account the effect of the structural frame on the overall fire protection package.

Restrained Assembly beams generally exhibit significant deflections before the required hourly rating is met. A building’s fire protection package, in addition to the sprayed cementitious fireproofing, includes other important features, such as compartmentation, suppression systems and detection systems. While a Restrained structure, in the right situation, may transfer loads without collapse, excessive deflections will likely compromise the compartmentation offered by firestopping and smoke damper systems and jeopardize the effectiveness of sprinkler systems. Non-functioning compartmentation and suppression permit the quick spread of smoke and fire to areas outside of the point of origin.

Actual fire events have proven the effectiveness of Unrestrained classification. The Occidental Tower fire in Los Angeles (Nov 1976), State Office Building fire in Olympia Washington (Oct 1983), First Interstate Bank fire in Los Angeles (May 1988), and Union Bank Building fire in San Francisco are all examples of structures protected with cementitious fireproofing applied to Unrestrained thicknesses. In all cases, the damage to the structural steel was minor and the buildings were open for business shortly after the fire event.

Some failed fire protection packages resulting from Restrained classification or sprinkler tradeoffs are: The McCormack Place fire in Chicago (Jan 1967), One New York Plaza fire (Aug 1970), K Mart Distribution Center fire in Falls Township, PA (Jun 1982), and One Meridian Plaza fire in Philadelphia (Feb 1991). All sustained structural damage beyond repair resulting in major financial losses.

In short, Spray-on Fireproofing works and Unrestrained protection affords the best protection for structural steel framed buildings. It insures full compliance with building codes in all jurisdictions, without assuming the liability associated with designating a building as thermally restrained. The cost difference between fireproofing a building to Unrestrained versus a Restrained classification is generally less than 1% of the overall cost of the building. This amount can prove to be the difference between saving a structure with minimal loss of life and a catastrophic disaster.

Michael Giardinelli

W.R. Grace & Company

Author, Fireproofing Steel Structures (STRUCTURE, February 2007)

 

May '07

CASE and the members of the CASE Fire Protection Committee found the article, Fireproofing Steel Structures (STRUCTURE®, February 2007), to be potentially misleading. The premise of the article, that structural steel assemblies should be considered to be thermally unrestrained, is not supported by engineering data. There has been significant engineering research over the past 30 years that suggests that structural steel assemblies behave as thermally restrained in almost all instances.

  • Restrained Fire Resistance Ratings in Structural Steel Buildings by Gewain and Troup (see reference above) states that most common types of steel-framed construction are classified as thermally restrained.

  • Appendix X3 of ASTM E119 lists the few instances where individual steel beams and girders, or steel framed floor and roof assemblies, are classified as unrestrained.

  • AISC Design Guide No. 19, Fire Resistance of Structural Steel Framing, clearly indicates that the position suggested in the article is incorrect.

It requires considerably more sprayed fire-resistive material to achieve an unrestrained fire rating, rather than a restrained fire rating. T his appears to be a conservative approach that could present an economic burden to the project. These conditions must be carefully evaluated by the designer.

CASE Executive Committee

Edward W. Pence, Jr., P.E., S.E., F.ASCE, Chair 2006/07

 

May '07

Much effort has been spent disseminating the research sponsored by the American Iron and Steel Institute (AISI) that confirmed the performance of steel framing and the use of restrained ratings in the selection of fire protection.  The conclusion of that research remains valid – steel-framed structures can be considered thermally restrained.

The test data supporting this conclusion are documented in Restrained Fire Resistance Ratings in Structural Steel Buildings (Gewain and Troup; AISC Engineering Journal, 2nd Quarter 2001).  Furthermore, AISC Design Guide No. 19, Fire Resistance of Structural Steel Framing, contradicts the position suggested in Fireproofing Steel Structures (STRUCTURE® magazine, February 2007).

The issue of restrained vs. unrestrained construction is unique to the United States.  It has been a source of confusion since the concept’s introduction in 1970.  To assist the design professional in determining this parameter, AISC has collected information demonstrating that steel framed construction qualifies for a restrained classification and makes it available so that the provisions of section 703.2.3 of the International Building Code can be satisfied.  This being the case, the opinion expressed in the article serves to perpetuate unnecessary questions that have already been answered repeatedly.

John L. Ruddy

Director of Building Design, AISC

 

March '07

I am writing in response to a portion of Jon Schmidt’s InFocus article in the January 2007 issue of STRUCTURE ®. Jon, stated, “I cannot help but wonder if the ‘commodization’ of engineering services is inevitable if we do not significantly raise the bar of entry into our profession. The moves toward specialty certification and, eventually, separate licensure are certainly steps in the right direction, but may not be enough in the end.” Perhaps I misunderstand the intent, but it seems to me that we are considering professional licensure as a means of job protection. We exist professionally to facilitate the construction of structures — safe structures. If additional licensure or educational requirements are needed to assure safe design, very well. If we begin, however, to consider these methods as some form of trade protection tariff, we are amiss.

Our age is a turbulent one. The ability to communicate globally has altered and is still altering our business fundamentally. This article touched not only on the educational future of engineering, but also on our own fears of our professional future. In a free market, when we perform useful, needed services, we are and will be compensated. If there are quicker, easier or cheaper means of providing those same services, the market will shift the demand to those other sources. If the services performed there are inferior, our services will command more demand and/or money, or at least we hope so!

Perhaps the real goal is not a free market goal. Perhaps the real goal also has an aim to create a union of sorts. I don't know, but I think our intentions ought to be clear at least to ourselves.

Bret Wickham

Contra Costa County, California

 

I appreciate Mr. Wick's comments and do not necessarily disagree with them. When I talk about “raising the bar”, I do not mean making it more difficult per se, but rather making it more rigorous. Protecting the public is, indeed, the primary intent — not protecting our turf.

Response from Jon Schmidt, P.E., SECB

 

Feb '07

Mr. Rouis’ December 2006 article, A Better Base, was well written and informative.  I offer the attached Pin Wheel Isolation Joint Detail as an alternate to his Figure 7a. The Pin Wheel method of simultaneously isolating the column as a part of the installation of sawn control joints is used very often in warehouse distribution facilities in which the slab is cast after the erection of the steel.

An alternate to Mr. Rouis’ Figure 1 is the Pocket Form Isolator. Information on this pre-manufactured item can be found at www.isolationpocket.com.

D. Matthew Stuart, P.E., S.E., F.ASCE, SECB

Schoor DePalma Engineers and Consultants

mstuart@schoordepalma.com

Design Element Design Element

  • Post this page to del.icio.us
  • Digg this page
  • Post this page to Technorati
  • Post this page to Blinklist
  • Post this page to Furl
  • Post this page to Reddit