U.S. Department of Labor
Occupational Safety and Health Administration
Department of Labor
Alexis M. Herman, Secretary
Occupational Safety and Health Administration
Charles N. Jeffress, Assistant Secretary
Directorate of Construction
Russell B. Swanson, Director
Office of Engineering Services
This report was prepared by
Scott Jin, Ph.D., P.E.
On December 16, 1999 at approximately 2:05 PM, a 44 ft tall reinforcing steel cage weighing 80 tons fell over and killed one of two workers working near the top of the cage. The wind at the time of the incident was reported to be about 20 to 25 mph. The deceased worker was caught between the collapsed reinforcing cage and the concrete footing surface. The other worker was thrown into an open space uninjured. At the time of the incident, the two workers were installing and tying the last portion of the horizontal reinforcements to the east face of the steel cage. The reinforcing steel cage was installed for the concrete Column 2 of Bridge Pier WB-12 on the west side of the James River. The accident occurred at the western most segment of the proposed Interstate Highway 1-895 in the southeast suburb of Richmond, Virginia.
Personnel from the Virginia Occupational Safety and Health (VOSH) arrived at the scene within an hour of the incident. The Directorate of Construction, OSHA, Washington, DC, was requested by VOSH to provide assistance in the technical assessment of the collapse and in determining of the cause of the incident.
The OSHA investigation began soon after the incident and included: interviewing witnesses; inspecting the collapsed steel cage and observing failures; taking photographs and related field measurements; reviewing construction drawings; and performing an engineering analysis. Based on the results of the investigation, OSHA concludes that:
The following is a chronological list of major events during the investigation.
On December 16, 1999 within an hour of the incident, personnel from the Virginia Occupational Safety and Health (VOSH) arrived at the scene to start the investigation by interviewing witnesses and taking videos and photographs.
On December 17, 1999 the same personnel from the VOSH re-visited the incident site to continue interviewing witnesses and to take additional videos and photographs.
On December 22, 1999, at the request of the VOSH, two engineers from the Federal OSHA Directorate of Construction (DOC), accompanied by the Senior Compliance Safety and Health Officer of the VOSH, visited the incident site to observe the surroundings, inspect the collapsed steel cage, take photographs and field measurements, and attend a meeting with the General Contractor, Bridge Contractor, Reinforcing Steel Erector and others. In this visit, DOC received the following information:
On December 28, 1999, after reviewing of the above materials, the same personnel of Federal OSHA and VOSH re-visited the incident site to inspect the collapsed steel cage again, take additional photographs and measurements, and conduct a meeting with the Reinforcing Steel Erector. In the second visit, DOC received the following additional information:
On March 1, 1999 DOC received the wind speed records of December 12, 1999 through December 17, 1999 at the Richmond International Airport from the National Climatic Data Center. The wind speed at the incident site is assumed to be the same as that at the Richmond International Airport.
Throughout the course of the investigation, the Directorate of Construction worked together with the personnel from VOSH. The Senior Compliance Safety and Health Officer, Paul J. Trabosh, made significant contributions to this investigation.
Dinesh Shah, Structural Engineer of the Office of Engineering Services, took the field measurements, performed the computer analyses and checked the manual calculations of the draft report. Mohammed Ayub, Director of the Office of Engineering Services, reviewed the draft report and made valuable suggestions.
The proposed $324 million, 8.8-mile Interstate Highway 1-895, also known as Pocahontas Parkway, is owned by the Virginia Department of Transportation. It will link the existing Virginia State Route 150, also called the Chippenham Parkway, and Interstate Highway 11 295 near the Richmond International Airport. As shown in Figure 3-1, the proposed 1-895 is located in the southeast suburb of Richmond, Virginia. The project in question is located i at the western most segment of the proposed highway, where it intersects the existing Interstate Highway 1-95. Figure 3-2 presents the architectural rendering of the 1-895 and 1-95 interchange over the James River and the approximate location where the incident occurred.
The photograph on the cover of this report shows the collapsed steel cage. This photograph was taken from the east side of the James River. Figures 3-3 and 3-4 present the site condition around the west (land) side of the concrete footing. Based on the field identification, the collapsed steel cage is at the location of Column 2 of Bridge Pier WB-12 as shown in Figure 3-5. From this figure, the concrete footing is supported by twenty drilled shafts. The Bridge Pier WB-12 is supported on the north half of the footing. The two adjacent 90 ft tall concrete columns in construction, as shown in Figures 3-3 and 3-4, are not related to the incident. They are Bridge Pier EB-13 columns and are supported on the south half of the footing.
The design arrangement of the steel reinforcements for the concrete column of the bridge pier is presented Figure 3-6. As indicated in this figure, the bottom portion of the steel cage is embedded in the concrete footing. Based on the information from Reinforcing Steel Erector (RSE), when the incident occurred, the footing concrete was approximately two months old and had achieved full strength.
On December 16, 1999 at approximately 2:05 PM, under the high winds, a 44 ft tall reinforcing steel cage collapsed and killed one of the two workers working near the top of the cage. The deceased worker was caught between the collapsed steel reinforcements and the concrete footing surface. The other worker was thrown clear into the open space between the east side of the concrete footing and the steel sheet pile wall along the west edge of the river. At the time of the incident, the two workers were installing and tie wiring the last portion of the horizontal reinforcements, also called "candy canes", to the east face of the steel cage. The wind speed at the time was about 20 to 25 miles per hour, based on the readings from the adjacent tower crane.
Based on the discussion with the RSE, in the middle of October 1999, the bottom level of the reinforcements of the steel cage was erected on top of the drilled shaft foundations and was laterally supported by a HP steel frame. The concrete for the footing was cast around October 20, 1999. When the footing concrete had achieved most foist strength, the HP steel frame was removed (around November 4, 1999). At this stage, the height of the steel cage was approximately 30 ft above the concrete footing.
The vertical extension of the steel cage was started on or about December 9, 1999, seven weeks after the footing was cast. In this extension, 20 ft long vertical reinforcements were tie wired to the bottom reinforcements with a lapping length of approximately 6.5 ft. As a result, the height of the steel cage was increased to 43.5 ft. After the installation of the skeleton of the four circular cells, it was found that the partially extended steel cage was out of plumb. The top of the steel cage was reported to be leaning about 6 to 8 in. to the west and 3 to 4 in. to the north. To maintain the verticality of the steel cage, for the proper installation of the remaining vertical reinforcements, two guy wires to the east and one guy wire to the south were installed near the top of the steel cage. In addition, two additional guy wires were also installed to the west. As indicated by the RSE, the five installed guy wires were only intended to maintain the plumbness of the steel cage. These guy wires were not designed as a lateral support to the steel cage.
The condition of the collapsed steel cage from its bottom to top is presented in Figures 3-7 through 3-10. As indicated in these figures, all reinforcements in the steel cage fell and collapsed as a group. In addition, no significant relative movements were observed between the vertical and horizontal reinforcements at their intersections. However, the horizontal reinforcements were severely rotated with respect to the bent vertical reinforcements at the intersections. This observed behavior was consistent with the information from the RSE, that the tie wires were only intended to maintain the reinforcements in position at the intersections. The tie wires were not intended to provide rigid connections between the horizontal and vertical reinforcements. In addition, as per the field observations, tie wires were generally installed at every third intersection, instead of every other intersection as was the normal practice. According to the RSE, the reduced tie wired intersection was approved by the project design engineer due to the congestion of the reinforcements.
Based on the field counts, at the time of the incident, there were 114 horizontal circular ties installed on the steel cage. Given the vertical spacing for the ties as 3.94 in. as defined in Figure 3-6, the ties were installed up to a height of 37.4 ft. Mechanical couplers were used for the bottom 60 ties of each cell to a height of 20 ft from the top of the concrete footing. These circular rings would provide some constraints on the direction of the lateral deflection of the vertical reinforcements during the collapse. A metal concrete form with the timber work platform was observed under the collapsed cage. In accordance with the SRE, the concrete form was not connected to the steel cage. However, it did influence the collapsed shape of the steel cage, due to its presence.
The condition of the top of the steel cage at its south end is presented Figures 3-11 and 3-12. These photographs were taken from the open space between the east side of the concrete footing and the steel sheet pile wall which the second worker fell into. The upper end of the southwest, south and southeast guy wires were also shown in Figure 3-11. A close-up view of the upper end of the southeast guy wire is presented in Figure 3-12.
The resting condition of the steel cage after the collapse is presented Figures 3-13 through 3-15. As a result of the collapse, a significant amount of tie wires were broken and fell on top of the concrete footing as shown in Figure 3-15. The last section of the extension for the west face of the steel cage, as shown in Figure 3-16, was not installed prior to the collapse. This section consisted of 23 vertical reinforcements and 42 horizontal reinforcements. In addition, according to the RSE, the lower lapping end of the outer vertical reinforcements coincided with the upper end of the un-extended inner vertical reinforcements at 23.5 ft as shown in Figure 3-14. This value was consistent with the vertical space counts as indicated in Figure 3-16.
The condition of the northwest, southwest and south guy wires after the collapse of the steel cage is presented in Figures 3-17 through 3-22. All of the three wires were field measured as 1/4 inch in diameter. Figure 3-17 shows the lower connection point of the northwest guy wire at the north end of the steel cage for Column 1 of the same pier. Figure 3-18 presents the damaged condition and the size measurement of the northwest guy wire. Since the northwest guy wire was damaged during the collapse and was removed from the steel cage, the location of its upper end could not be field identified.
Figure 3-19 shows the lower connection point of the southwest guy wire at the south end of the steel cage for Column 1 of the same pier. The lower and upper ends of the southwest guy wire are presented in Figures 3-20 and 3-21, respectively. Since no damages were observed on the guy wire and no latches were present on any of the hooks, it is believed that the guy wire system was disengaged prior to the collapse of the steel cage.
The upper end of the south guy wire is already identified in Figures 3-10 and 3-11. Figure 3-22 presents the middle portion of the south guy wire, it had a hook without latch. The lower end of this guy wire was anchored to the concrete footing. Since the south guy wire was installed approximately along the major axis of the steel cage, perpendicular to the direction of the collapse, it was neither stretched nor damaged. The last two (northeast and southeast) guy wires on the collapsed (east) side of the steel cage were not tensioned, therefore, they were not damaged.
To determine the cause of the collapse, the as-installed condition of the steel cage and its associated guy wires immediately prior to the collapse was re-constructed in Chapter 3.0. The actual wind speed on the day of the incident from the National Climatic Data Center (NCDC) was used in the engineering analysis. In this analysis, the yield strength of structural elements, instead of the allowable stress, was used. In addition, no load factors were applied in the failure analysis. All of this information is included in Appendix A.
The basic premise for the analysis was based on the following observations on the collapsed steel cage:
Considerations were given to examine whether the horizontal circular ties provided any rigidity to the vertical reinforcing bars to resist lateral loads. If it did, the steel cage would have a very high degree of resistance against any lateral load because of a many fold increase in its section modulus. On the other hand, if the ties did not provide any rigidity to the vertical reinforcing steel to form a closed rigid frame, the lateral loads will be resisted by individual reinforcing bars, highly susceptible to flexure and lateral displacements.
Our field inspections revealed that the circular ties were connected to the vertical reinforcements with nominal tie wires and were incapable of providing any stiffness to the intersection. The angle of 90 degrees between the intersecting reinforcements was easily compromised under the lateral load as shown in different photographs previously discussed. So, it was decided to consider the bending stiffness of individual reinforcements instead of the entire steel cage as one element. However, the lateral loads will be equally distributed to all the vertical reinforcements because they were interconnected and were of the same size. The presence of the circular ties resulted in equal deflection of all vertical reinforcements, as observed in the field.
Given the above, two main assumptions were made:
The above assumptions along with other detailed assumptions are listed in Section A-1 of Appendix A.
The total weight and the vertical location of the centroid of the steel cage were calculated in Tables A-1 and A-2, respectively. The weight of the steel cage was estimated as 80 tons. The centroid of the steel cage was estimated as 18.9 ft above the concrete footing. All of this information is included in Section A-2.
The height of the representative vertical reinforcement was calculated as 39.2 ft in Section A-3-1. It was the average height of all (442) installed vertical reinforcements. The wind area for the representative reinforcement was calculated from the projected area of all installed vertical reinforcements including their lapping areas and divided by the total number of the vertical reinforcements. This information is included in Section A-3-1. Due to the potential shielding effects, other projected areas of all exterior and interior horizontal reinforcements were not considered. The wind areas used in the analysis was conservatively assumed, since the eliminated wind areas of the horizontal ties were more than the shielded areas of the vertical bars. In addition, the wind exposed areas of the horizontal ties would increase as the steel cage started to bend.
The wind coefficients for the representative reinforcement was selected based on the requirements of Reference 1. However, the Important Factor (I) of 1.0 was used to simulate the actual wind condition in the analysis. All of this information is included in Section A-3-3. Since the steel cage was classified as a flexible structure, the Gust Response Factor (G) was calculated by a rational analysis that incorporates the dynamic properties of the steel cage. This information is presented in Section A-3-4. The typical wind load on the representative reinforcement was calculated as 0.620 plf in Section A-3-5.
Although the northwest and southwest guy wires were not intended to resist any wind load (from the west direction) on the steel cage, the strength contribution of these two wires should be considered in the failure analysis. However, since the southwest guy wire was disengaged at its lower end without any damages prior to the collapse of the steel cage as documented in Figures 3-19 and 3-20. Thus, its contribution was not considered in the analysis. As presented in Section A-4-1, the wind induced tension on the northwest guy wire was calculated as 10,544 pounds, which is exceeding its nominal strength (breaking load) of 7,000 pound, based on Reference 3, Page 95. Thus, the northwest guy wire would fail. This guy wire actually failed as shown in Figure 3-18.
The wind induced stress and horizontal deflection of the representative reinforcement after the failure of the guy wires is calculated in Section A-4-2. Please note that the calculated stress and deflection of the representative reinforcement were also representing those of the steel cage. Since the steel cage weighed 80 tons, the gravity (PΔ) effect on its deflected shape shall be considered as per Reference 4, Page 6-41. The PΔ analysis was also performed on the representative reinforcement with the share weight (363 pounds) of the steel cage in Section A-4-3. The resulting bending stress of the representative reinforcement was 263.8 ksi, which was much greater than its yield strength of 60 ksi. Thus, the representative reinforcement yielded and the steel cage collapsed.
Based on the above analysis, under a 20 mph wind, the steel cage could not laterally support itself; with or without the installed guy wires. Thus, the Reinforcing Steel Erector (RSE) did not provide adequate lateral support to prevent overturning and to prevent collapse of the reinforcing steel cage for a bridge pier per the requirement of the OSHA Standard 1926.703 (d) (1). In addition, this requirement is further emphasized in ANSI/ASCE 7-88 Paragraph 6.1.1- Adequate temporary bracing shall be provided to resist wind loading on structural components and structural assemblages during the erection and construction phases.
In accordance with the records from the NCDC, the peak winds in the five days prior to the incident (from Sunday, December 12, 1999 through Thursday, December 16, 1999), were 9, 11, 17, 17 and 26 mph, respectively. Thus, the peak winds were in an increasing trend toward December 16, 1999 the day of the incident.
Based on the RSE, the installation of the upper extension of the steel cage started on or about December 9, 1999 with the addition of the skeleton of the four circular cells. The remaining vertical reinforcements would be installed to the skeleton cells on top of the steel cage. The weight and the projected wind area of the skeleton cell were about one·fifth of those for the fully installed cell. Thus, as the installation continued, both the weight of the steel cage and the wind loads on the steel cage would increase until it could not support itself and collapsed on December 16, 1999. In addition, from the NCDC records, the direction of the peak winds on December 16, was 260 degrees (from west), which almost coincides with the direction of the collapse of the steel cage. Therefore, the collapse of the unbraced steel cage was triggered by the effect of high winds.
Based on the result of the investigation, the Occupational Safety and Health Administration concludes that:
A-1 Basic Assumptions
A-2 Estimation of the Total Weight and Vertical Location of the Centroid of the Reinforcing Steel Cage at the Time of the Incident
A-3 Estimation of the Wind Load on a Representative Vertical Reinforcement at the Time of the Incident
A-3-1 The average height of the representative reinforcement
A-3-2 The projected area normal to the wind for the representative reinforcement
A-3-3 The associated wind coefficients for the representative reinforcement
A-3-4 The gust response factor of the representative reinforcement
A-3-5 The typical wind load on the representative reinforcement
A-4 Estimation of the Horizontal Deflection and Flexural Stresses on a Representative Vertical Reinforcement at the Time of the Incident
A-4-1 The tensile force on the guy wire due to the wind load
A-4-2 The wind induced stresses and horizontal deflection on the representative reinforcement after the failure of the guy wires
A-4-3 The second order (PΔ) effects on the flexural stresses of the representative reinforcement
Based on the field observation, the vertical reinforcements above the concrete footing were the only structural elements that support the weight of the steel cage. After the breaking of the guy wires, these reinforcements were the only structural elements to resist the wind load applied to the steel cage. Based on the deflected pattern of the steel cage after the collapse, the horizontal reinforcements did not remain in its original orientation with the vertical reinforcements, i.e., the tie wires were not capable to prevent the rotation of the horizontal reinforcements from the associated vertical reinforcements. In addition, the tie wires were only used in one-third of the reinforcement intersections, the remaining intersections were free. Thus, each vertical reinforcement of the steel cage bent individually, not as a structural frame (cage) with fixed joints. However, due to the numerous closed horizontal circular ties, all vertical reinforcements within the circular ties of the steel cage fell together as a group. The following assumptions are made for the engineering analysis:
The number, the size, the length and the location of the reinforcements in the steel cage are estimated based on the following documents:
Based on the above information, it was determined that there were 442, Number 11, vertical reinforcements installed at the time of the incident. For the lower portion of the reinforcements from the concrete footing, there were 242 outer reinforcements with a height of 30.0 ft, and 200 inner reinforcements with a height of 23.5 ft. In the upper extension of the reinforcements, 219 outer and 177 inner vertical reinforcements with a length of 20 ft were installed. The only missing part of the upper extension was the west face section as shown in Figure 3-16 of the report.
The total number of the exterior horizontal reinforcements (ties) installed at the time of the incident was field counted as 114 layers. This number was also confirmed with the actual height of the vertical reinforcements, based on the vertical spacing between horizontal ties for different elevation as given in Item 1 above.
The number, the size, the length and the location of the interior horizontal reinforcements for each layer were determined based on the Items 1 and 2 above. Please note that since the upper extension on the west face of the steel cage was not installed at the time of the incident, the associated interior reinforcements (PB 1301) to be installed from this face were not considered in the weight calculation.
The total weight of the steel cage at the time of the incident is calculated and tabulated in Table A-1. In addition, the vertical location of the center of gravity (centroid) of the steel cage is also calculated in Table A-2.
S T A A D - III
Proprietary Program of
Research Engineers, Inc.
Date= S: 9: 56000
Time= S: 9:49
Build No. 1007.01.02
USER ID: U. S. Dept of Labor
NUMBER OF JOINTS/MEMBER+ELEMENTS/SUPPORTS= 2/ 1/ 1
ORIGINAL/FINAL BAND-WIDTH = 1/ 1
TOTAL PRIMARY LOAD CASES = 3, TOTAL DEGREES OF FREEDOM 3
SIZE OF STIFFNESS MATRIX = 9 DOUBLE PREC. WORDS
REQRD/AVAIL. DISK SPACE = 12.00/ 1509.5 ME, EXMEM = 1966.2 ME
++ Processing Element Stiffness Matrix. 8: 9: 57
++ Processing Global Stiffness Matrix. 8: 9:57
++ Processing Triangular Factorization. 8: 9: 57
++ Calculating Joint Displacements. 8: 9: 57 ++ Adjusting Displacements 8: 9: 57
++ Adjusting Displacements 8: 9: 57
++ Calculating Member Forces. 8: 9: 57
|JOINT DISPLACEMENT (INCH RADIANS) STRUCTURE TYPE PLANE|
|SUPPORT REACTIONS -UNIT POUN FEET STRUCTURE TYPE = PLANE|
|MEMBER END FORCES STRUCTURE TYPE = PLANE
ALL UNITS ARE -- POUN FEET
END OF LATEST ANALYSIS RESULT
END OF STAAD-III
DATE= MAY 1,2000 TIME= 8: 9:57
For questions on STAAO-III, contact:
Research Engineers, Inc at Build No. 1007.01.02
West Coast: Ph- (714) 974-2500 Fax- (714) 921-2543
East Coast: Ph- (978) 688-3626 Fax- (978) 685-7230
KEY REFERENCE MATERIALS
|Unedited Surface Weather Observations
|Station Elev. (ft)
|Station (Type, Name, State)
ASOS Richmond, VA
|Remarks and Supplemental coded DA|
|s||0054||200||8||10||CLR||07||04||A2993||SLP: 39 T00720044 10089 20044 56010 (DAP|
|s||0154||200||8||10||CLR||07||04||A2993||SLP: 39 T00670039 (DAP)|
|s||0254||210||7||10||CLR||06||03||A2993||SLP: 141 P00610033 (DAP)|
|s||0354||210||6||10||CLR||05||03||A2993||SLP: 141 P00900328 53002 (DAP)|
|s||0454||260||4||10||Ten055||05||04||A2995||SLP: 148 000560039 (DAP)|
|s||0554||280||5||9||Ten055||05||03||A2997||SLP: 154 000500033 (DAP)|
|s||0656||300||4||10||BKS060 BKS250||05||03||A2999||SLP: 152 T00500033 10072 20044 53020 (KB)|
|s||0756||280||3||8||BKS055 BKS250||05||01||A3002||SLP: 172 T00500011 (KB)|
|s||0854||250||5||10||F3M065 SCT250||07||01||A3005||SLP: 180 T00671011 (KB)|
|s||0954||260||12||10||FS0250||09||01||A3006||SLP: 185 T008981011 51024 (KB)|
|s||1054||300||14||13||10||FEW250||09||04||A3006||SLP: 185 T00941039 (KB)|
|s||1154||300||12||10||FEW250||10||06||A3004||SLP: 178 T01001056 (KB)|
|s||1254||270||13||59||10||FEW250||11||07||A3002||SLP: 170 T01061067 10106 20044 58014 (KB)|
|s||1354||240||14||20||10||FEW250||11||07||A3002||SLP: 170 SML SC NK T01061067 (KB)|
|s||1454||260||16||20||10||FEW250||11||08||A3001||SLP: 172 T01061078 (CES)|
|s||1554||260||13||10||FEW250||11||08||A3002||SLP: 172 T0161078 (CES)|
|s||1554||260||13||10||FEW250||11||08||A3002||SLP: 172 T0161078 (CES)|
|s||1654||250||10||10||FEW250||09||06||A3005||SLP: 181 T009410B3 (CES)|
|s||1754||250||10||10||CLR||09||08||A3008||SLP: 193 T00891083 (CRS)|
|s||1854||280||5||10||CLR||07||08||A3011||SLP: 202 T00721083 10106 20072 53030 (CES)|
|s||1954||260||4||10||CLR||07||08||A3013||SLP: 207 T00671078 (CES)|
|s||2054||260||5||10||CLR||06||09||A3013||SLP: 211 T00561089 (CES)|
|s||2154||220||5||10||CLR||03||07||A3016||SLP: 219 T003310B3 (DAP)|
|s||2254||260||4||10||CLR||03||08||A3016||SLP: 219 T003310B3 (DAP)|
|s||2354||250||5||10||CLR||03||08||A3016||SLP: 220 T00281078 401050006 (DAP)|
|Summary of The Day (Midnight to Midnight)|
|Peak Winds||Fastest 2-mon wind||Skycover|
|Remarks, Notes and Miscellaneous Phenomena 65:||Time Checked-clock Correct to the nearest minute at:|
*Accessibility Assistance: Contact OSHA's Directorate of Construction at (202) 693-2020 for assistance accessing PDF materials.
All other documents, that are not PDF materials or formatted for the web, are available as Microsoft Office® formats and videos and are noted accordingly. If additional assistance is needed with reading, reviewing or accessing these documents or any figures and illustrations, please also contact OSHA's Directorate of Construction at (202) 693-2020.
**eBooks - EPUB is the most common format for e-Books. If you use a Sony Reader, a Nook, or an iPad you can download the EPUB file format. If you use a Kindle, you can download the MOBI file format.Back to Top
The Department of Labor does not endorse, takes no responsibility for, and exercises no control over the linked organization or its views, or contents, nor does it vouch for the accuracy or accessibility of the information contained on the destination server. The Department of Labor also cannot authorize the use of copyrighted materials contained in linked Web sites. Users must request such authorization from the sponsor of the linked Web site. Thank you for visiting our site. Please click the button below to continue.