File Name: 2008v03_q_a_full.pdf
File Size: 495.58 KB
File Type: Application/pdf
Last Modified: 8 years
Status: Available
Last checked: 1 days ago!
This Document Has Been Certified by a Professional
100% customizable
Language: English
We recommend downloading this file onto your computer
QUALITY ASSURANCE FOR STRUCTURAL ENGINEERING FIRMS CLIFFORD SCHWINGER Clifford SchwingerClifford Schwinger, P.E. is Vice President and Quality Assurance Manager at The Harman Group. Mr. Schwingerreceived his BSCE degree from Lehigh University and has been with The Harman Group for 22 years. He’s on theAISC Manuals and Textbooks Committee
ABSTRACTChanges have occurred in the structural engineering profession over the past twenty years which have created a needfor engineering firms to implement formal in-house quality assurance programs. This paper discusses thecomponents of a model QA program and reviews procedures, tips, techniques and strategies for conducting in-housequality assurance reviews on structural drawings with a focus specific to structural steel building structures
1 INTRODUCTIONThe structural engineering profession has undergone dramatic changes over the past twenty years. With fast-trackconstruction, computerized design, complex building codes and younger engineers taking on more responsibilityearlier in their careers, the need for structural engineering firms to have a comprehensive in-house QualityAssurance program has never been greater. Adoption of a comprehensive Quality Assurance program will result inbetter design, high quality contract documents, fewer RFI’s and change orders during construction, a better productfor clients and increased profitability for engineering firms
THE QUALITY ASSURANCE PROGRAMA Quality Assurance program is a defined set of procedures and standards used to facilitate design and to facilitatedocumentation of that design. Implementation of a QA program results in: – Better design – Better drawings – More efficient design process – Fewer mistakes – Fewer RFI’s and Change Orders – Increase client satisfaction – Enhanced reputation – Increased profitPrior to 1990 the concept of formal QA programs was virtually unheard of within the profession. Quality wasassured by relying on the experience, skill, continual oversight and expertise of trained engineers, structuraldesigners and drafters. Structural design was a linear process and contract documents were usually not issued for biduntil the design and the drawings were 100% complete. Formal QA programs, where they existed, consistedprimarily of a senior engineer being assigned as the “go to” person for answering technical questions. That engineerwould also review the drawings before the project went out for bid – providing a second set of eyes on the contractdocuments in order to catch mistakes. Such a QA program, consisting of a “technical guru” and a single QA reviewdoes not work today
Today a comprehensive QA program requires the following components: – Training for young engineers – Design standards – Drafting and CAD standards – Project delivery system – Knowledge base – Involvement of the QA Manager and QA reviews 2 Training for Young EngineersPrior to the use of computers, young engineers working in design offices typically spent the first several years oftheir careers doing repetitive manual calculations. Most new engineers also spent “time on the board” learning theart of structural drafting under the guidance of experienced engineers and senior drafters. The training of a youngengineer was a gradual process. As experience was gained, more responsibility was delegated - reviewing shopdrawings, developing details and eventually coordinating projects with architects and answering questions fromcontractors. Computers have eliminated most laborious manual calculations and while they have greatly increasedproductivity, computers have also altered the informal training phase that all new engineers go through. Youngengineers today are faced with the challenge of taking on much more responsibility early in their careers. Furtherchallenging a young engineer’s transition into the profession are complex building codes, the details of which areusually not learned in school and the lack of any knowledge of structural drafting, a skill which is just as valuabletoday as it was years ago. The ability to convey one’s ideas to paper for interpretation by others will always be anessential skill. For moderate to large-sized engineering firms, the solution to this problem is establishment of aformal in-house training program
Training for young engineers should consist of in-house lunchtime training seminars covering the full spectrumstructural engineering topics that are pertinent to the type of work performed by the firm. Because the goal of thetraining program is to pass on the combined knowledge of the senior staff, the list of topics for these seminars islong. Passing knowledge includes not just interpretation of codes, standards and design procedures, but also adiscussion of practical applications and lessons learned. A short listing of typical seminars includes: AISC 360-05 Braced frames IBC 2006 Moment frames Dead, Live & Snow load Trusses Wind loads Joists Wind Tunnel Studies Metal deck Seismic loads Slabs on metal deck Site Specific Seismic Analysis Floor and roof diaphragms Load Paths 101 Window washing davits Reviewing shop drawings Elevators and escalators Connection design Facade systems Member design Post-installed anchors Stability Expansion joints Braced frames Slide bearing connections Vibration Concrete mix design Coordination issues with MEP Slabs-on-grade Stairs and monumental stairs Masonry design Structural drafting Wood design Framing plans How to perform a self-QA review How to draw details Lessons Learned Foundation design Communication skills Concrete design Legal and liability issuesThese seminars are best conducted once or twice per week. While some topics can be covered in a single session,others, such as structural steel connection design, can take several sessions to fully cover
Seminars focus on actual application of the principles discussed and are interspersed with lessons learned,discussion of common mistakes, examples of manual calculations and tips and techniques for verifying the accuracy 3 of computer analysis and design. Software limitations and assumptions are reviewed with continual emphasis thatcomputers are tools to be properly used by engineers; the creativity and solutions to structural engineeringchallenges come from the mind and imagination of the engineer, not the computer
Design StandardsDesign standards are comprised of: – Design Guides – Formal design procedures – ChecklistsMedium and large-sized engineering firms must have written formal design procedures, standards andmethodologies in order to produce consistently high quality design and to minimize the risk of errors due tomiscommunication
Office standards must be formally established so that there is no confusion regarding design procedures andmethodologies. Is office policy to use ASD design or to use LRFD design? Is the policy to show beam reactions onframing plans or to require that shear connections be designed for a percentage of the member uniform loadcapacity? Are connections designed by the EOR or is connection design delegated to the steel fabricator’s engineer?Is there a minimum percentage of code wind load below which the wind tunnel wind pressures will not be used?Serious consequences could result if two engineers are working on a project with one showing service level memberreactions on the framing plans and the other showing factored reactions. The purpose of office design standards isto keep everyone on the same page and to provide a roadmap to insure uniformity of design
Design guides are one of the ways that design procedures are set forth. Design guides delineate office policyregarding design procedures and bring together building code and design standards, textbook theory, localconstruction practices, practical applications and lessons learned
Checklists are useful tools both for engineers new to the profession as well as for experienced engineers trying toremember the hundreds of things that go into design and documentation of a building structure. While major itemslike reviewing diaphragm strength and stiffness are well ingrained in a seasoned engineer’s mind, little things likeremembering to coordinate locations of fall protection tiebacks on the roof might occasionally slip by but forreminders provided on checklists
Drafting and CAD StandardsStructural drafting is fast becoming a lost art. Whereas mechanical drawing used to be taught to students in highschool and college, many engineers now arrive in the profession with no training in a skill that is essential forcommunication of their design intent to others. Likewise, most structural drafters have now been replaced by CADoperators who, while proficient in use CAD software, may be lacking in the knowledge and understanding of how tolay out framing plans, draw weld symbols or dimension details. The solution to this problem is to establish draftingand CAD standards, the components of which include: – Standardized drafting procedures – CAD checklists – Typical detail library 4 – “go-by” drawings – Standard block libraryDrafting procedures include information related to rules for laying out framing plans, drawing sections and details,setting up column schedules, etc. Uniformity and consistency within the office requires that everyone draw objectsconsistently on the correct layers and use the same linetypes and linetype scales. While these may seem like trivialissues having no bearing on structural design, they will improve the quality and legibility of a set of structuraldrawings
Checklists include the myriad of things needed to produce complete and legible drawings. They cover things asseemingly minor as making sure north arrows are shown on the framing plans to more important items such asmaking sure that beam reactions are indicated
A comprehensive structural engineering typical detail library will contain over hundreds of typical details
“Go-by” drawings are reference drawings that show examples of how to indicate information on framing plans,schedules, etc. While “go-by” framing plans may have originated from actual projects, they will usually be modifiedover time to include everything that can possibly occur on a framing plan. “Go-by” framing plans for variousstructural systems provide engineers and drafters a single point of reference to see how to properly draw anythingthey will encounter on the plans. The use of “go-by” drawings prevents younger engineers from using previousprojects for learning how to show things on the drawings. While using other projects as a frame of reference is notnecessarily a bad idea, doing so can lead to a gradual divergence of drafting standards in larger firms
A standard block library is essential for increasing productivity and maintaining drawing uniformity. “Blocks” arepre-drawn objects such as bolts, angles, W-shapes, weld symbols, headed studs, section cuts, etc
Project Delivery SystemThe Project Delivery System is a library of forms, checklists, procedures and correspondence templates used foradministratively carrying a project from inception through construction. The PDS is divided into five sections: – Project startup – Schematic design – Design development – Contract documents – Construction administrationThe Project Startup section contains things required at the beginning of a project such as a design criteria formlisting design information such as the applicable building code, design standards, loads, wind, snow and seismicdesign criteria, summary of the structural systems being used and fire ratings required. Correspondence templatesfor letters to the client regarding information needed from the geotechnical consultant and wind tunnel consultant aswell as correspondence templates that summarize presumed design criteria and required “due by” dates to meetschedules, etc. are provided
The Schematic Design, Design Development and Contract Document sections contain checklists and proceduresrelated to the deliverables in each phase of design
5 The Construction Administration section contains meeting agenda templates for the pre-steel detailing meeting, thepre-concrete meeting, meetings with the inspector as well as checklists to be used when reviewing shop drawings
Knowledge BaseThe Knowledge Base (KB) is a searchable electronic database of all knowledge related to structural engineering
The KB contains the notes from training seminars, design guides, design standards, drafting and CAD standards, andinformation on all other topics that engineers may need to access. The primary feature of the KB is that it’s a singlesource for answers to all questions related to structural engineering. When a question or topic comes up for whichthere’s no answer on the KB, that information is added. When problems occur or lessons are learned, the solutions tothose problems and lessons learned are added to the KB
Involvement of the QA Manager and QA ReviewsThe QA manager is senior level engineer who is responsible for establishing and maintaining engineering standardsand for verifying that all design is done in accordance with those standards. The QA manager has the followingresponsibilities: – Establishing and maintaining design and drawing standards – Answering technical questions and getting the answers to those questions onto the KB as appropriate
– Staff training – Maintaining familiarity with all projects during design and providing input and suggestions as required
– Signing off on sections and details prior to them going to the CAD department. (A cursory review and signoff of sections and details by the QA manager is required to catch mistakes before sending sections and detail to the CAD department. Such a review saves time, is informative for the engineer whose details are being critiqued.) – Performing quality assurance reviews on all projects
THE QUALITY ASSURANCE REVIEWQuality Assurance reviews are in-house reviews conducted to verify that all design is performed and documented inconformance with the procedures and standards mandated by the QA program
QA reviews serve two purposes. The primary purpose of QA reviews is to provide redundancy via a second set ofexperienced eyes on the drawings to catch mistakes, errors or omissions. The second purpose is to monitor theeffectiveness of the QA program. If the QA program is working properly and engineers are following the proceduresand utilizing the resources provided therein then problems, mistakes, errors and omissions caught during the reviewshould be minor. While the QA manager is usually the one who performs the reviews, other experienced engineerscan likewise perform the task
Changes in the way contract documents are now issued have altered the way QA reviews are performed. Until tenyears ago a single QA review was performed prior to the contract documents being issued for bid. Fast-trackconstruction scheduling now requires multiple reviews at stages during design. It’s not uncommon to have eight ormore reviews on large projects. While the number varies from project to project, a typical QA review schedule for asteel framed structure on pile foundations might be as follows: 6 – Pile bid – Steel mill order – Foundation concrete bid – Steel Addendum / detailing issue – 100% concrete – 100% structural steel – “Issued for Construction” final reviewMultiple reviews are also a good idea for those projects still delivered via the traditional design-bid-build process
Interim reviews will catch mistakes early when corrections can be easily made
There are two primary goals of QA reviews. The first and most important goal is to review the contract documentsto verify that the structure was properly designed, is efficiently framed and is constructible. The second goal is toverify that the contract documents are complete, well detailed, correct and coordinated. The goal of issuing completeand well detailed contract documents is not just one founded on a desire to reduce RFI’s and change orders – it isone that is essential to insuring structural integrity. Finishing the drawings during construction via the RFI process isa bad idea. Not only do RFI’s frequently lead to change orders, unless senior level experienced engineers are theones answering RFI’s, mistakes can slip through. If the drawings are complete and well detailed before construction,those details will have gone through the scrutiny of the QA review process and the probability of engineeringmistakes being made during the process of answering RFI’s during construction will be greatly reduced
A variety of tactics are employed when performing QA reviews. Those tactics are as follows: – Look at the big picture – Verify load paths – Review framing sizes – Look at connection details (constructability) – Look for mistakes – Look for subtleties – Look at the drawings for constructability – Review for clarity – Look for omissions – Look for “little” little things – Look for the ”big” little things – Verify that the structural drawings match the architectural & MEP drawingsLooking at the Big PictureEngineers immersed in large projects can lose sight of the big picture and miss things that are often immediatelyobvious to someone who was not working on the project. Some common mistakes in this category include: a. Missing or improperly located expansion joints b. Improperly detailed connections at expansion joints (example: uni-directional slide bearing connections locking up the expansion joint at corners. See figure 1.) c. Load path problems (example: braced frames cut off from floor diaphragms; failure to design diaphragms at vertical irregularities in the lateral load force resisting system.) 7 d. Equilibrium of forces not investigated (example: horizontal kick at base and top of sloping columns not considered and connections not detailed. See figure 2.) e. Constructability issues (example: moment connections in both directions at a column where beams are different depths and stiffener plates are specified in both directions) f. Inefficient connections (example: severely skewed joists framing to W shape girders.) g. Connection problems (example: column base plate anchor rods that don’t fit in the piers or are too deep for the footings.) h. Inefficient framing configurations (example: too many pieces; beams framed in wrong direction) i. Inefficient spandrel details (example: too much “gingerbread” framing.) j. Wrong design loads used k. Problems with computer model (examples: problems related to “infinitely rigid” diaphragms; double counting structure self-weight or ignoring self-weight; pushing “reduce live loads” button on computer where live load reductions are not permitted.) l. Using wrong “R” factor (Use R=3 for steel buildings in areas of low seismicity.) m. Failure to consider snow drift n. Failure to consider loads such as folding partition storage pockets, heavy runs of piping, window washing davits, etc
o. Excessive deflections on spandrels or ends of cantilevered beamsFigure 1: Example of improperly detailedslide bearing connections locking up theexpansion joint because they are detailed topermit movement in one direction only
8 Figure 2: Forces must resolve
Verify Load PathsThere must be continuous and realistic load paths from the point at which loads are applied to the structure down tothe foundation. While this may seem so basic as to not warrant discussion, it is a common problem
The most common load path problems are with floor diaphragms - usually because when engineers define floordiaphragms as rigid, computer programs give those diaphragms infinite strength and stiffness, both assumptionsoften being unrealistic
Some examples of problems associated with assumptions of rigid diaphragm: – Lateral loads in braced frames increase from the top to the bottom of the braced frame. When braced frame member forces get smaller on the lower levels (see Figure 3), that’s usually a sign that an “infinitely rigid” floor diaphragm diverted lateral load out of the braced frame and sent the load elsewhere. Relying on a slab-on-metal-deck diaphragm to drag loads out of one braced frame and into another is usually not a good idea
– Figure 4 shows a rectangular building with a shear wall at one end and a moment frame at the other. The computer results indicated that 95% of the north-south lateral load was resisted by the shear wall and only 5% was resisted by the moment frame. This result was due to the combined effects of the floor diaphragms having infinite rigidity and the two east-west shear walls preventing the floor from twisting. The problem with this analysis was that while the floor was probably closer to the building code definition of a rigid diaphragm than it was to a flexible one, it did not have sufficient strength to work as a rigid diaphragm
The lateral loads were manually adjusted and conservatively enveloped to account for a more reasonable lateral load distribution between the shear wall and moment frame
– Figure 5 shows a portion of a floor where the slab-on-metal-deck was input as a rigid diaphragm and the computer then modeled the exterior columns as braced at each floor. In reality, the floor slab did not have sufficient strength to provide P-δ buckling restraint to the columns in the strong axis direction at each floor
9 Compounding the problem, the computer did not consider bending and shear in the diaphragm from wind loads on the exterior wall because the diaphragm was infinitely stiff and infinitely strong. This framing was repeated for several floors. The computer sized the columns as W14x90’s. Manual calculations showed that the columns had to be W14x211’s
– Figure 6 shows a brace frame adjacent to an exterior stair that’s substantially cut off from the floor diaphragm. While the computer model sees a connection to the floor diaphragm, the slab engagement to the diaphragm is not sufficient to get the diaphragm load into the braced frame. A drag strut is required
– Figure 7 shows an in-plane vertical discontinuity in a braced frame. A drag strut is needed to transfer the horizontal load from the base of one braced frame to the top of the adjacent braced frame. The computer will not design the connecting drag strut member for the horizontal load because the rigid diaphragm is assumed by the computer to transfer the load from one braced frame to the other
– Figure 8 shows an out-of-plane offset irregularity in a braced frame. The floor diaphragm must be manually designed to transfer the lateral loads from the braced frame BF2 to the two adjacent braced frames, BF1 and BF3. Diaphragms, even if they are rigid, must be manually designed and detailed to resist the applied shears and moments
Figure 3: Load path problem resulting from infinitely rigid floor diaphragm in computer model diverting load out of brace frameFigure 4: Problem related to infinitelyrigid floor diaphragm directing too muchload to shear wall and not enough to thebraced frame
10 Figure 5: Example of floor problem related tofloor diaphragm not being strong or stiff enough tobrace the columns. Computer model sized columnsas W14x90’s. Actual required size =W14x211
Figure 6: Illustration of floor diaphragm cut offfrom braced frame
Solution: Provide drag strut with appropriateconnections to brace frame
Figure 7: Example of shortcoming of computermodel in computing drag strut axial load
11 Figure 8: :Level 1 floor diaphragm must be designed to transfer BF2 lateral loads to BF1 and BF3
Review Framing SizesReview of member framing sizes is the most basic aspect of a QA review. This task can be daunting on largeprojects with thousands of framing members. Fortunately, there are several tricks that can make review of framingsizes easier
Since most framing is designed by computer, member sizes will usually (but not always) be correct as long as theinput is correct. A global review of floor framing can therefore be performed by verifying that all of the computerinputs (loads, slab thickness, concrete strength, headed stud sizes, etc.) are correct. This is accomplished byreviewing several typical beams, girders and columns. If the beam size, number of studs, camber and reactions onthe typical beam and girder are checked and verified to be correct, a review of those two members alone provides ahigh level of confidence that the other framing on the floor is correct. If several different design loads are used ondifferent areas of the floor, then checks of typical beams and girders in each area should be performed
Showing beam reactions on the framing plans is an invaluable aid when performing a QA review. Seeing reactionsallows the reviewer to quickly compute the total load used for design of the member and verify whether or not thatdesign load is correct. The effort required to show reactions on framing plans is minor. A push of a button in theanalysis/design program will transfer the reactions to the CAD drawings. Showing beam reactions also reduces thecost of the structural framing by allowing steel fabricators to detail connections for the actual reactions versushaving to detail connections based on arbitrary and usually overly conservative percentages of non-compositeuniform load capacity
The combined use of strength design and 50 ksi steel can result in beams that sometimes have excessively high spanto depth (L/d) ratios. Beams with high L/d ratios are susceptible to problems related to deflection and vibration. Arule of thumb for the maximum recommended span of composite beams is to limit beam span (in feet) to 2 x depthof the next larger nominal beam depth (in inches). (Example: Maximum span of a W12 beam = 2 x 14” = 28 feet.)For non-composite beams the maximum span (in feet) should not exceed 2 x beam depth (in inches). (Example:Maximum recommended span of a W12 beam = 2 x 12” = 24 feet). This is a general rule and these maximumrecommended spans can be exceeded if justified by analysis
Quality assurance reviews on structural drawings with a focus specific to structural steel building structures. 2 The Project Delivery System is a library of forms, checklists, procedures and …
Free national guidelines for the structural engineer of record, designed to help corporate/municipal clients the scope of services SEs do and do not provide
Join CASE Today for a Stronger Bottom Line! The Coalition of American Structural Engineers represents more than 200 structural engineering firms dedicated to making structural engineering a fair, profitable, and robust industry.
CASE is the recognized authority in the SE industry and the leading provider of risk management and business practice education and information for structural engineering firms. CASE develops guidelines and contracts that outline best practices, along with extensive risk management tools designed to keep liability in check.