National Academies Press: OpenBook
« Previous: MULTIPLE-FAMILY DOMESTIC DWELLINGS
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 16
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 17
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 18
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 19
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 20
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 21
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 22
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 23
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 24
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 25
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 26
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 27
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 28
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 29
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 30
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 31
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 32
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 33
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 34
Suggested Citation:"PUBLIC BUILDINGS." National Research Council. 1985. Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984. Washington, DC: The National Academies Press. doi: 10.17226/19291.
×
Page 35

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 PUBLIC BUILDINGS A number of buildings to which the public had access were damaged by the South Carolina tornadoes. They can be classifed as load-bearing masonry, steel-framed, and a hybrid system of steel and masonry. LOAD-BEARING MASONRY On the whole, buildings constructed of unreinforced load-bearing masonry performed extremely poorly. A dance academy housed in an old brick building in Newberry col- lapsed entirely (Figure 20). Fortunately, the occupants sought shelter under a stairway and survived with minor injuries. A short distance away an auto parts store consisting of l8-ft-high hollow masonry walls faced with brick and supporting heavy steel roof trusses collapsed, killing the manager. Figure 2l shows the structural arrangement after the debris had been cleared. Also in Newberry, St. Luke's Episcopal Church, which is listed in the register of historic monuments, suffered severe roof damage and the collapse of one load-bearing masonry wall. In Winnsboro a number of retail stores and workshops made of unreinforced load-bearing concrete masonry units either collapsed entirely or suffered severe damage. An example is shown in Figure 22. One wing of a private school, the Richard Winn Academy, of similar construction also suffered severe damage, mostly from the wind but possibly also from school buses reportedly seen flying through the air during the tornado. The construction of this part of the school fol- lowed the normal practice of using unreinforced walls and short roof anchors. Had the tornado struck earlier, or had after-school activities not been canceled due to the bad weather, serious injury and loss of life might have resulted. There was no safe refuge area in this section of the school, and a lightweight annex building was completely carried away (Figures 23, 24, and 25). A church also collapsed in Winnsboro (Figure 26). The method of construction again followed the common local practice regarding slender- ness of the walls, the positioning of pilasters, and the connection of the roof to the walls. The complete inadequacy of such roof-wall con- nections in resisting uplift forces is clearly shown in Figure 27. The plane of weakness is simply transferred from the roof-wall connection to l6

l7 FIGURE 20 Dance academy in Newberry. FIGURE 2l Auto parts store in Newberry.

l8 FIGURE 22 Masonry retail store in Winnsboro. FIGURE 23 Richard Winn Academy in Winnsboro.

l9 FIGURE 24 Classroom at Richard Winn Academy. FIGURE 25 Classroom at Richard Winn Academy.

20 FIGURE 26 Church in Winnsboro. FIGURE 27 Detail of church showing roof-wall connection.

2l the block-block connection in the row below. The uplift forces not balanced by the weight of the roof must be resisted by the tensile strength of the mortar. The walls of the church also collapsed easily when the resisting diaphragm of the roof was removed. The building was unoccupied at the time, but the consequences of a tornado striking this church during a service are obvious. STEEL-FRAMED BUILDINGS Buildings consisting of moment-resisting steel frames appear to have performed well. Although the cladding and secondary members such as purlins were often severely damaged, the buildings maintained their structural forms and would have provided safe refuge during the storm. Figures 28 and 29 show the Fairfield County Community Center in Winns- boro. Contrast the damage to this structure with the severe damage to the masonry building next to the center shown in Figure 22. Nearby, the gymnasium of the Richard Winn Academy also suffered damage only to its cladding (Figure 30). Figures 3l and 32 show typical damage suffered by retail stores of metal construction, the first in Winnsboro, the second in Bennettsville. HYBRID STEEL AND MASONRY CONSTRUCTION One of the most disturbing examples of damage was the failure of the Northwood Village Shopping Center in Bennettsville. This structure was designed by a firm of architects and engineers and constructed in an area where the Standard Building Code had been adopted. Although no plans of the structure had been lodged with the Building Inspection Department, a set was obtained from another source, and an analysis was made of the building system. A wind tunnel study was also conducted on a model of the structure, as described later in this chapter. During the course of the storm, a tornado over l mile in diameter appeared to have passed over the shopping center, producing a horizontal velocity gradient across the front of the building. The wind appeared to have blown primarily onto the front of the building, causing the most damage on the west end. Figure 33 shows an overall plan of the shopping center. The struc- tural system consisted essentially of a built-up roof system overlaying rigid insulation and a 1.5-in.-deep corrugated metal deck of 22 gage galvanized steel. The metal deck was spot welded to bar joists (metal truss rafters) that spanned between girders or between girders and ma- sonry walls. Figure 34 shows a roof framing plan. The girders them- selves were supported on 5-in.-diameter columns or by masonry walls. The front wall of the building consisted of glass or 4-in. brick with an 8-in. backup wall of hollow concrete masonry units. The side, rear, and partition walls consisted of l2-in. hollow concrete masonry units. A bond beam was provided on the front and rear walls, and the masonry units were filled with concrete for a few courses where girders or bar

22 FIGURE 28 Fairfield County Community Center in Winnsboro. FIGURE 29 Detail of Fair- field County Community Center.

23 FIGURE 30 Gymnasium of Richard Winn Academy. FIGURE 3l Steel-framed retail store in Winnsboro.

24 FIGURE 32 Steel-framed retail store in Bennettsville. REFERENCE NORTH ALL DIM NSIWT, IN FEET HEY (T) DEPARTKNT STORE: 216.0 x 150.0 ("M LOCAL STORE: 74.0 x 80.0 (7) DRUG STORE: 65.0 x 130.0 (T) SUPER MARKET: 130.0 x 140.0 (7) LOCAL STORE: 75.0 x 80.0 i — 75.0 •! [\ n 1 1 ] 150. D rTi W (7) © w © L. 1 f T !.( i 1 12.0 FIGURE 33 Plan of Northwood Village Shopping Center in Bennettsville.

25 ALL DIMENSIONS IN FEET BAR JOISTS . GIRDERS INTERNAL COLUMNS EXTERNAL BRICK AND STEEL COLUMNS 12 INCH HOLLOW CONCRETE MASONRY HULLS 4 INCH FACE BRICK - 8 INCH CONCRETE BACK UP HYTHE GLASS WITH GIRDER OVER HORIZONTAL BRIDGING RUNS PARALLEL TO GIRDERS 562.0 216.0 130.0 75.0 150.0 I 80.0 i 12.0 12.0 FIGURE 34 Framing plan of Northwood Village Shopping Center. joists were supported by walls without bond beams. Control joints were provided in the walls at intervals of between 20 and 30 ft. Horizontal truss reinforcement was provided in alternate courses. No vertical reinforcement was provided. The walls ranged in height from l6 to 20 ft. Figure 35 shows the general extent of the damage. The department store at the west end was reduced to a pile of twisted steel and ma- sonry. The front wall had been blown in, the side and back walls had been blown out, and the roof framing system had collapsed. Next to the department store, the small unoccupied local store lost all its supporting walls, but the center line of columns remained up- right (Figure 36). In the drug store the front and side walls collapsed, as did the front half of the roof (Figure 37). The girders had been joined at a point that would normally carry little moment. With the collapse of the side wall, this became a highly stressed area and the joint failed (Figure 38). In the adjacent supermarket the front wall failed, and the resulting pressurization of the interior and the suction on the roof appeared to have lifted the first row of bar joists and wrapped them around the first line of girders (Figures 39 and 40). The local hardware store at the extreme east end, not shown in Figure 35, suffered relatively minor damage. The front wall, consisting mainly of glass, failed, and some metal decking was removed from the roof, but the roof framing remained intact (Figure 4l). This was also the only part of the shopping center where there was any trace of the

26 FIGURE 35 Aerial view of damage to Northwood Village Shopping Center. (Photograph courtesy The State.) FIGURE 36 Surviving framing in local store.

27 FIGURE 37 Roof collapse in drug store. FIGURE 38 Failure of beam joint.

28 FIGURE 39 Failure of front wall in supermarket. , FIGURE 40 Failure of roof in supermarket.

29 FIGURE 4l Surviving roof system in hardware store. front canopy (Figure 42). Although the wind speeds were undoubtedly lower at this end of the building, the orientation of the roof members in this building and the unoccupied store differed from that of the three larger stores. The main girders ran from front to back in the smaller stores and from side to side in the larger ones. To learn more about the performance of the structure, a detailed analysis was made of the department store (Desai, l984). The results of this analysis are summarized below. In checking the roof system, several points became evident. The bar joists were quite adequate for the design gravity loads provided the decking supplied lateral restraint to the compression flanges of the bar joists. The girders were capable of carrying the design gravity loads provided the bar joists supplied lateral restraint to the compression flange of the girders. These girders were apparently designed on the assumption of there being a pin connection between the girders and the columns. Figure 43 shows that this connection was so meager that this was a reasonable assumption. Indeed, the columns could only carry the required load if they were pin-ended and fully restrained against lat- eral movement. Although this restraint was initially provided by the bar joists and girders, it was ultimately provided by slender unrein- forced masonry walls. These masonry walls were also required to act in other ways. They had to carry vertical loads imposed by the bar joists and girders.

30 FIGURE 42 Surviving canopy at east end of the shopping center, These loads would have been well distributed to the front and back walls, where bond beams were provided (Figures 44 and 45). On the side walls, where no bond beam existed, the load distribution would have been limited (Figure 46). In any case, the roof was extremely light and the vertical loads carried by the walls were small. This had a detrimental effect on the ability of the walls to perform another function, that of resisting the wind loads. In this respect, they had to be capable of transferring the load either directly to the ground or to the roof. Here the metal decking acting as a shear diaphragm would have transfer- red the loads to the side walls. Functioning as very deep shear walls, these walls would finally have transferred the rest of the wind load to the ground. Despite the important role played by the walls, little attention appears to have been paid to their design. They were apparently sized to meet the absolute maximum ratio of unsupported distance to thickness permitted by the Standard Building Code in effect at the time of con- struction—l8—with the roof being deemed to provide support. However, according to the code, if a roof is used to provide lateral support, the maximum horizontal distance between supports should not exceed 75 times the thickness of the wall, which in this instance would be 75 ft. The side wall of the department store stretched for l50 ft without support and contained several control joints. The front wall stretched for 2l6 ft.

3l VARIES .5.0 FT (TYPICAL) J BUILT UP ROOF RIGID INSULATION 1.5 INCH THICK METAL DECK ^\\\\\\\\\\\\\\\\\\\\\\W . •'• •. \:.•:}.^y^.i•':^:'^. , •y'•fe^v.".", BOIl 1*1 CHORD OF JOIST EXTENDED TO NEAREST COLUMN FOR BRACING COLUMN CAP PLATE 7.5 INCH x 12.5 INCH x 0.5 INCH STANDARD BOLTED BEAM CONNECTION _ GIRDER M-SECTION 5 INCH DIAMETER "STANDARD STEEL PIPE b im.H DIAMETER PIPE ~— — - 11 IK ICH x 7 INCH x 0.75 INCH BASE H 0.75 INCH DIAMETER ANCHOR B PLATE OLT 2 INCH NON SHRINK A GROUT \ y — MIT , \ / / // ,• 4 IN( H CONCRETE SLAB VARIES fcsssssg ^ 2 •: •».• •.•••'..•»'.•,••" ',•.••: t \if'f'-, OVARIES ^ L 1 >iii i/; 1.0 FT /,» _.'-" •'•>»':;li §|^ 6 K BARS , • 6 15 BARS 5.0 FT (TYPICAL) FIGURE 43 Typical beam-column connection. The requirements concerning unsupported length are intended to avoid instability, but the walls should have been capable of resisting all the vertical and horizontal loads required by the code. An engineering analysis of the walls indicated that in all instances the critical com- bination of dead, live, and wind loads specified by the code exceeded the allowable strength of the wall. On the other hand, the tempered glass used in the front wall had a capacity far in excess of that required by the code. It is surprising, therefore, that eyewitnesses reported that the glass failed first, creating an opening sufficiently large that occupants and contents of the building were blown to the rear. Shortly after the glass failed, the roof sheeting was stripped off, followed almost immediately by the collapse of the walls and the roof framing system.

32 BUILT-UP ROOF RIGID INSULATION 13.0 FT 12 INCH HOLLOW CONCRETE MSONRY UNITS 10.0 FT 4 INCH FACE BRICK AND 8 INCH CONCRETE MASONRY UNITS FINISHED FLOOR BARS • (.0 FT O.C. 2 K BARS CONTINUOUS (N.T.S.) e^INCtf 12 INCH 6 INCH FIGURE 44 Detail of front wall (as designed). Rescue operations had disturbed the collapsed front wall of the department store, but in other stores evidence was found that the glazing system, including the frame, had been stripped from the surrounding masonry. This indicated that failure may have been initiated by poor wall to window frame connections. Another factor that may have contributed to the failure was the fact that the canopy had not been built as shown in Figure 44, but had been replaced by inferior wood framing as shown in Figure 42. Without the restraint and shelter provided by this canopy, the front wall of the department store possessed very little wind resistance. To determine how the failure of certain components influenced the wind loads on the remaining components, a study was conducted in the structural engineering wind tunnel at Clemson University using a l to 288 scale model of the shopping center and a boundary layer appropriate for the local terrain (Wright, l984). Table l presents the significant conclusions of this study, showing the wind velocities, assumed normal to the front face of the building, likely to cause failure of the components. These speeds were determined from the mean net pressure coefficients obtained in the wind tunnel tests at the stage of damage indicated in the table. When the building was intact, no account was taken of a possible internal pressure in excess of the external ambient pressure.

33 TABLE l Wind Speeds (in Miles per Hour) Expected to Cause Failure of Components of the Department Store at the Northwood Village Shopping Center Building Condition and Component Assumed Factor of Safety Against Collapse Building Undamaged Front wall with canopy 86 (79) l2l (ll2) l48 (l37) Front wall without canopy 89 (27) l26 (38) l54 (46) Canopy 99 (— ) l40 (— ) l72 (— ) Side wall l47 (8l) 207 (ll4) 254 (l40) Rear wall l50 (89) 2l3 (l26) 26l (l55) Front third of roof l73 (— ) 245 (— ) 300 ( — ) Rear two thirds of roof l3l (— ) l86 (— ) 227 (— ) Front Wall Collapsed Side wall l00 (55) l4l (78) l73 (95) Rear wall 90 (54) l27 (76) l56 (93) Front third of roof 76 (— ) l07 (—) l3l (--) Rear two thirds of roof 80 (--) ll3 ( — ) l39 ( — ) Roof Removed Side wall, top-supported l48 (82) 2l0 (ll6) 257 (l42) Side wall, free-standing 96 (57) l35 (80) l66 (98) Rear wall, top-supported 74 (4l) l05 (58) l27 (7l) Rear wall, free-standing 48 (28) 68 (40) 83 (49) NOTE: The first figure in each entry refers to components designed to meet the pressures specified by the Standard Building Code with the factor of safety indicated. The second figure in parentheses is based on the allowable strength of the components as determined from the design drawings by Desai (l984). One would normally expect a factor of safety against collapse of 3 in masonry construction and 2 in steel construction. With these factors of safety, components designed to meet the pressures specified in the Southern Standard Building Code should not have experienced serious damage until the wind speed reached nearly l50 mph. If at that stage the front wall had failed, failure of the roof sheeting would soon follow. This would probably have left the exterior side wall free standing, but the rear wall might still have received support from the bar joists. Irrespective of the type of support, the failure of the walls would be inevitable, and the subsequent collapse of the unre- strained girder and bar-joist system would be likely. Regarding the failure of the actual structure, estimates could only

34 BUILT UP ROOF RIGID INSULATION .5 INCH METAL DECK BAR JOIST 4 INCH X 6 INCH DOM SPOUT 12 INCH CONCRETE MASONRY UNIT 4 INCH CONCRETE SLAB ON GRADE REINFORCEKNT 0.00 FT ,, •.. ?.- •• •.. '• :• ••••••.•^ IH 4 1 / . . 1-2 ASPHI I1 33 FT £> c. c^ 2 •£> ?t V III =•111 *;."•'»,; • »•»';• 2* 2 15 BARS CONTINUOUS 13 BARS 9 6.0 O.C. !6 INCHT12 INCH T6 INCHl FIGURE 45 Detail of rear wall. be made of the strength of the walls. No data were available concerning the ability of the roof sheeting to resist uplift forces or the wind resistance of the wood canopy, it is highly unlikely, however, that they would have met the code design requirements with a factor of safety of more than 2, and the roof sheeting is known to have failed before the walls. Surprisingly, despite the inadequate design of the walls, if the canopy provided shelter and support to the front wall, and if the glaz- ing system remained intact, the structure should not have been damaged until the wind reached nearly l40 mph. However, failure of the canopy or glazing system at any speed in excess of l00 mph would probably have caused the complete collapse of the structure. That is the basic weak- ness of this form of construction. In a conventional steel-framed structure, the failure of cladding elements or glazing will relieve the wind load but will not significantly reduce the capacity of the struc- tural system to carry the loads. In this hybrid steel and masonry sys- tem, relief of the wind loads through failure of the roof is accompanied

12 INCH CONCRETE MASONRY UNITS OR 4 INCH FACE BRICK AND 8 INCH CONCRETE MASONRY UNITS 5L _ HORIZONTAL BRIDGING i.lHiHK W-SECTION BAR JOIST FINISHED FLOOR 10 INCH nn |j / 0.00•FT .• • : .• • ;•<-•"••.:•'.. ;'.^•r J ;j -1.40 FT •. ' •'.' «^ — 2 K BARS COITIN U4 L 6 INCH r!2 INCH I I 13 BARS AT 6.0 FT O.C. 6 INCHl FIGURE 46 Detail of side wall. by a loss of restraint to the walls and a consequent reduction in struc- tural capacity far greater than the reduction in load. Since these walls also provide the lateral bracing for the vertical load-bearing system, the conditions for a progressive collapse are established as soon as the roof diaphragm fails, it is particularly disturbing that the diaphragm's integrity, and ultimately that of the whole structure, depends upon such minor details as the window frame to masonry connec- tions, the design of the walkway canopy, and the welding of the metal decking to the bar joists.

Next: GENERAL COMMENTS »
Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984 Get This Book
×
 Building Damage in South Carolina Caused by the Tornadoes of March 28, 1984
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!