Principal Investigators:

Jerome F. Hajjar Roberto T. Leon Carol K. Shield

The Northridge, California Earthquake: The Effect of Composite Floor Slabs on the Fracture of Steel Moment Frame Connections Sponsors:
National Science Foundation American Institute of Steel Construction SAC Joint Venture Georgia Institute of Technology University of Minnesota

Research Fellows:
Paul M. Bergson Gabriel P. Forcier

Graduate Students:
Michael A. Gustafson Patrick F. Carlson

Undergraduate Students:

Jennifer Soderstrom Grant Wyffels

Topographical map showing the green, yellow, and red tagged buildings in the vicinity of the epicenter (from Holmes, W. T. and Somers P., Earthquake Spectra, EERI, Vol. 11, Supp. C, 1995)

The Earthquake:

January 17, 1994, 4:31 am Moment magnitude Mw: 6.7 Epicenter: Northridge, California, USA 32 km west-northwest of Los Angeles, California Persons dead: 57; Injured: Over 5000

Estimated damage: US $50 billion Condemned structures: 1600 timber; 300 concrete; 10 steel

The Problem:

While only one steel structure collapsed in the Northridge earthquake, hundreds of welded steel moment-resisting connections fractured in both old and new steel buildings. Repair costs ranged from $25,000 to $50,000 per connection. The connections consisted of steel I-girders framing into the flanges of steel wide-flange columns. The girder flanges were welded to the column flange, and the girder web was bolted and/or welded to the column flange with a shear tab. These connections were the primary detail used in unbraced steel frames, and they were designed to yield and deform in a ductile manner during large earthquakes Damage included brittle fractures in the steel I-girder to I-column full penetration welds, the column flanges, and the column webs. Over 90% of the brittle fractures occurred at the bottom girder flanges

Research Objectives:

To determine if structural phenomena, in addition to metallurgical (e.g., welding) phenomena, contributed to the connection failures To determine the effect of the composite concrete floor slab, typically not accounted for in the connection design, on the predominance of bottom flange failures

Experimental Research Program:

Three full-scale interior connection specimens Specimens: one bare steel; one 35% composite; one 55% composite Column was loaded with constant axial tension: P/Py = 0.15 Applied 2-3 cycles of displacement-controlled antisymmetric loading to the girder tips at each level of interstory drift ranging from

0.25% to 3.0%

Schematic of the typical connection details (not to scale)

Schematic of the test setup (not to scale)

Computational Research Program:

3D nonlinear continuum finite element analyses of the specimens to corroborate the experimental results Frame analyses of a 4-story building which suffered severe damage: o Specimen sizes were based on sizes in this building o 3D second-order inelastic transient dynamic analysis with a site-specific earthquake correlated well with the damage pattern, while 2D dynamic analyses and 2D and 3D pushover analyses did not

Results:

All six connections failed prematurely at bottom girder flange Strain, acoustic emission, and analysis results confirmed: o The three connections which retained their bottom backup bar failed brittlely at the bottom girder flange full penetration weld-tocolumn interface at 1.5% drift o The three connections which had their bottom backup bar removed failed with a low cycle fatigue fracture emanating from the root of the bottom girder flange access hole at 3% drift o High strain concentrations developed at these two failure locations, and the column exhibited a triaxial tensile stress state Neutral axis shift in connection region was substantial Composite girders did not achieve AISC calculated strength

Moment-plastic rotation, east girder, 55% composite specimen, exhibiting brittle failure

Moment-plastic rotation, east girder, 35% composite specimen, exhibiting low-cycle fatigue failure

Brittle fracture in 55% composite specimen at interface of bottom girder flange and column flange

Low-cycle fatigue fracture in 35% composite specimen in the girder base metal, starting at the access hole

Strain profiles at peak load points in bare steel girder 13.5 from column face, exhibiting typical Euler-Bernoulli beam theory

Strain profiles at peak load points in 55% composite girder 13.5 from column face, showing substantial neutral axis shift

Strains in 35% composite girder at the top and bottom flanges 1.5 from the column face during the 1% and 1.5% interstory drift cycles, showing a 5 times strain increase long the bottom flange

Acoustic emission results of bare steel specimen, exhibiting brittle fracture across the bottom flange (ordinate) of the west girder at 1.5% interstory drift, and low cycle fatigue failure of the east girder at 3% interstory drift

Conclusions:

Bottom flange strains were up to one order of magnitude higher than top flange strains, identifying a chief cause for the preponderance of bottom flange failures At a minimum, the inherent asymmetry in composite steel connections should be accounted for in design provisions

Publications and Presentations:

Refereed Papers in Journals and Special Publications Hajjar, J. F., Gourley, B. C., O'Sullivan, D. P., and Leon, R. T. (1998). Analysis of Mid-Rise Steel Frame Damaged in Northridge Earthquake, Journal of Performance of Constructed Facilities, American Society of Civil Engineers, Vol. 12, No. 4, pp. 221-231. O'Sullivan, D. P., Hajjar, J. F., and Leon, R. T. (1998). Repairs to Mid-Rise Steel Frame Damaged in Northridge Earthquake, Journal of Performance of Constructed Facilities, American Society of Civil Engineers, Vol. 12, No. 4, pp. 213-220. Hajjar, J. F., Leon, R. T., Gustafson, M. A., and Shield, C. K. (1998). Seismic Response of Composite Moment-Resisting Connections. II. Behavior, Journal of Structural Engineering, American Society of Civil Engineers, Vol. 124, No. 8, August, pp. 877-885. Leon, R. T., Hajjar, J. F., and Gustafson, M. A. (1998). Seismic Response of Composite Moment-Resisting Connections. I. Performance, Journal of Structural Engineering, American Society of Civil Engineers, Vol. 124, No. 8, August, pp. 868-876. Leon, R. T., Hajjar, J. F., and Shield, C. K. (1997). The Effect of Composite Floor Slabs on the Behavior of Steel Moment-Resisting Frames in the Northridge Earthquake, Composite Construction in Steel and Concrete III, Buckner, D. C. and Shahrooz, B. M. (eds.), Engineering Foundation, American Society of Civil Engineers, New York, pp. 735-751.

Papers in Conference Proceedings Hajjar, J. F., Leon, R. T., Shield, C. K., and Dexter, R. J. (1998). Structural Behavior and Weld Joint Performance in Steel MomentResisting Frame Connections, Proceedings of the Third U.S.-Japan Workshop on Steel Fracture Issues, U.S.-Japan Partnership for Advanced Steel Structures, Tokyo, Japan, April 20-22, 1998, Building Research Institute, Tsukuba, Japan, September, pp. 419-428. Leon, R. T., Hajjar, J. F., Shield, C. K., and Gustafson, M. A. (1998). The Effect of Composite Floor Behavior on the Failure of Steel Moment-Resisting Connections, Proceedings of the NEHRP Conference and Workshop on Research on the Northridge, California Earthquake of January 17, 1994, Los Angeles, California, August 20-23, 1997, California Universities for Research in Earthquake

Engineering, Richmond, California, 1998, Volume III-B, Engineering, pp. 629-636. Leon, R. T. and Hajjar, J. F. (1997). Effect of Floor Slabs on Behavior of Full Moment Steel Connections, Building to Last, Proceedings of the American Society of Civil Engineers Structures Congress 97, Kempner, L. and Brown, C. B. (eds.), Portland, Oregon, April 13-16, 1997, ASCE, New York, pp. 772-776. Hajjar, J. F. and Leon, R. T. (1996). Effect of Floor Slabs on the Performance of SMR Connections, Paper No. 656, Proceedings of the Eleventh World Congress on Earthquake Engineering, Sociedad Mexicana de Ingenieria, Sismica, A. C. (ed.), Acapulco, Mexico, June 23-28, 1996, Elsevier Science Publishers, New York. One of only 10% of all papers in the proceedings that were accepted for oral presentation.

Technical Reports Carlson, P. C., Shield, C. K., and Hajjar, J. F. (1998). The Behavior of Earthquake Resistant, Composite Steel Moment Resisting Frame Connections, Structural Engineering Report No. ST-98-4, Department of Civil Engineering, University of Minnesota, Minneapolis, Minnesota, March, 205 pp. Hajjar, J. F., Leon, R. T., Gustafson, M. A., and Shield, C. K. (1998). Full-Scale Cyclic Experiments of Composite Moment-Resisting Frame Connections, Structural Engineering Report No. ST 98-02, Department of Civil Engineering, University of Minnesota, Minneapolis, Minnesota, and Report No. SEM 98-02, School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia, March, 426 pp. Leon, R. T., Hajjar, J. F., and Gustafson, M. A. (1998). Seismic Response of Composite Moment-Resisting Connections. I. Performance, Report No. UMSI 98/42, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, Minnesota, March, 45 pp. Hajjar, J. F., Leon, R. T., Gustafson, M. A., and Shield, C. K. (1998). Seismic Response of Composite Moment-Resisting Connections. II. Behavior, Report No. UMSI 98/33, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, Minnesota, February, 40 pp.

Leon, R. T., Hajjar, J. F., and Shield, C. K. (1996). The Effect of Composite Floor Slabs on the Behavior of Steel Moment-Resisting Frames in the Northridge Earthquake, Report No. UMSI 96/70, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, Minnesota, April, 15 pp. Hajjar, J. F., O'Sullivan, D. P., Leon, R. T., and Gourley, B. C. (1995). Evaluation of the Damage to the Borax Corporate Headquarters Building as a Result of the Northridge Earthquake, Report No. SAC 95-07, SAC (Structural Engineers Association of California, Applied Technology Council, and California Universities for Research in Earthquake Engineering) Joint Venture, Sacramento, California, December, 76 pp.

Presentations Hajjar, J. F. (1998). The 1994 Northridge, California Earthquake: Causes of Damage to Steel Moment-Resisting Frame Connections, Department of Architecture, Massachusetts Institute of Technology, Cambridge, Massachusetts, November 23, 1998. Hajjar, J. F. (1997). Effect of Composite Floor Slabs on the Behavior of Steel Moment-Resisting Frame Connections in the Northridge Earthquake, International Association for Bridge and Structural Engineering Working Commission 2 on Steel Structures, Composite Construction Conventional and Innovative, International Association for Bridge and Structural Engineering, Innsbruck, Austria, September 16, 1997. Leon, R. T. (1996). Effect of Floor Slabs on the Performance of SMR Connections, Earthquake Engineering Research Institute Annual Meeting, Los Angeles, California, February 8-10, 1996. Leon, R. T. (1995). Effect of Floor Slabs on the Behavior of Steel Moment-Frame Connections in the Northridge Earthquake, presented by co-author R. T. Leon at the SAC Joint Venture Phase II planning meeting, Earthquake Engineering Research Center, University of California, Berkeley, California, November 10-11, 1995. Leon, R. T. (1995). Effect of Floor Slabs on the Behavior of Steel Moment-Frame Connections in the Northridge Earthquake: Analytical Studies, presented by co-author R. T. Leon at the SAC Joint Venture Phase I coordination meeting, Earthquake Engineering Research Center, University of California, Berkeley, California, June 8-9, 1995.