U.S. Department of Labor
Occupational Safety and Health Administration
Directorate of Construction
This report was prepared by
Mohammad Ayub, P.E., S.E.
Tesfaye B. Guttema, Ph.D., P.E.
Directorate of Construction
Occupational Safety and Health Administration
Office of Engineering Services
Directorate of Construction
This report prepared by
Mohammad Ayub, PE, SE, and
Tesfaye B. Guttema, PhD, PE
On September 6, 2011, the Division of Occupational Safety and Health (TOSH) in the Department of Labor & Workforce Development of the State of Tennessee asked the U.S. Occupational Safety and Health Administration, in Washington, DC to provide technical assistance in the investigation of the April 5, 2011 incident at Gatlinburg, TN where two workers were killed. The incident involved the structural failure of a concrete wall at the Gatlinburg wastewater treatment plant.
A structural engineer from the Directorate of Construction (DOC), U.S. Occupational Safety and Health Administration, Washington, DC visited the incident site of the Wastewater Treatment Plant at Gatlinburg, TN on September 14, 2011. He inspected the fallen concrete wall, the failed connections of the intersecting walls, the splicing couplers, and the equalization basin structure.
TOSH provided original engineering drawings of the sewage plant to DOC. They also provided nine couplers recovered by the City of Gatlinburg from the site for our examination. The City of Gatlinburg also took core samples of the concrete.
In 1992, the City of Gatlinburg (City) retained Flynt Engineering Company (Flynt) of 2125 University Avenue, Knoxville, TN to prepare engineering plans for "Modifications to Wastewater Treatment Plant". The treatment plant is located at 1025 Banner Road, Gatlinburg, TN. The plans were dated 1992 but the construction did not start until 1994, and was completed in 1996. Crowder Construction company of North Carolina was to be the general contractor. The plans included the construction of a new equalization basin, a 124 ft. long by 64 ft. wide, and 30 ft. high cast in place open concrete structure. A few feet east of the basin was a Flow Control Room, a small one-story structure to regulate the flow from the basin. During the construction, the City retained Flynt to supervise construction to ensure compliance with drawings and specifications. Flynt and Crowder are both now out of business.
The flow chart of the treatment plant is shown in Fig.1. The equalization basin (basin) is the first recipient of the sewage waste and storm drain water in the treatment plant. Though the plant is essentially meant to treat sewage, an undetermined amount of storm water is inevitably present. The volume of wastewater differs depending upon the weather. In dry weather, the volume could be as low as 2-3 million gallons per day, and in wet weather, it could rise to up to 9 million gallons per day. The plant was designed to treat up to 9 million gallons of wastewater per day. The depth of wastewater in the basin could vary from between 3 to 26 feet. During normal operations, the water is approximately 10 ft. high in the basin. The maximum depth of the water could be 30 ft. before it overflows. There are no confirmed reports that the water level ever reached 30 ft. in the past.
On the morning of April 5, 2011, the 18'' thick, 30 ft. high concrete east wall of the basin suddenly separated from the rest of the basin structure, and fell to the ground in an eastward direction. During the collapse, the control valve room situated a few feet from the east wall was crushed, killing two employees inside the control room. It is estimated that the water in the basin at the time of the incident was in the range of 26-to-30 feet high. Apparently, during previous wet seasons, the water would reach as high as 26 ft.
The structural failure of the east wall was unusual in that the east wall neatly separated from the three orthogonal intersecting walls and overturned away from the basin by pulling away dowels from the footings and the far intersecting walls on the north and the south. The structural drawings Nos. 29 thru 33 prepared by Flynt in 1992 provided details of the proposed construction of the basin. The east wall, approximately 124 ft. long, was 18" thick, reinforced with #9 rebars at 12'' o.c. each face horizontally, and #6 rebars at 6" o.c. each face vertically, see Fig. C-3. The project specifications called for the concrete to be 4,000 psi, and for the rebars to conform to ASTM A615, Grade 60. It is understood that the testing of the concrete cores obtained by the City subsequent to the incident indicated the compressive strength to be higher than 4,000 psi. At the bottom, the east wall was dowelled to the footings with one #8 rebars at 6" o.c., at mid-depth see Fig. C-3. On the two far sides, the horizontal reinforcements of the east wall were dowelled into the north and the south walls. There were three orthogonal walls intersecting with the east walls at approximately 20 ft., 40 ft., 40 ft. and 20 ft. from each end. The east wall was designed to be dowelled to the intersecting walls by # 4 rebars at 12" o.c. each face horizontally, see Fig.C-4.
The inspection of the structure after the incident revealed that the contractor cast the walls in a manner that provided a cold joint between the east walls and the intersecting walls. The separation of the east wall occurred at this cold joint, see Fig. 2. Both faces of the joint were observed to be exceptionally smooth and lacked any bondage between the two pours, see Fig. 2. Instead of providing two layers of horizontal dowels of #4 rebars at 12" o.c. from the intersecting walls to the east wall, the contractor provided #5 rebars each face in the east wall and #5 rebars each face in the intersecting walls. The rebars were threaded into a coupler, thus providing continuity between the east wall and the intersecting walls, see Fig. 3. Instead of providing dowels consisting of #4 rebars at 12" o.c, each face the contractor provided #5 rebars each face at 12" o.c, an increase of 50% over what was required by the drawings.
Field measurement of the couplers indicated the following dimensions:
Outside diameter measured: 0.90" corresponding to actual 7/8" (0.875")
Coupler thickness: 0.19" corresponding to actual 3/16" (0.1875")
The above dimensions closely matched with the coupler D-50 with the product code 77100 manufactured by Dayton Superior., see Figs. C-6 & C-7. For a rebar of ASTM Grade 60, the maximum tensile strength based upon yield strength is 0.31 x 60 = 18.6 kips. Increasing by 125%, the coupler must have a strength of 1.25 x 18.6 = 23 kips. The coupler is made of ASTM A-108 having an ultimate tensile value of 65 ksi which gives a tensile force of 65 x 0.404 = 26 kips greater than the required value of 23 kips.
The coupler required a minimum threaded length of 7/8" which provides a spacing of 1/4" between the two rebars threaded in opposite directions. Site inspection revealed that the spacing between the bars was greater than 1/4", indicating that the bars were not threaded up to the required lengths. It must be noted, however, that only a few couplers could be inspected at the site.
Although the dowels provided by the contractor were 50% greater than those required, the couplers were continuously exposed to acidic wastewater due to seepage across the smooth cold joint. Thus, the couplers were subject to corrosion, reducing their effectiveness and compromising their structural integrity. Photographs of some of the recovered couplers in Appendix B indicate the extent of corrosion of the couplers.
The DOC's investigation included:
The reinforced concrete flow equalization basin at the wastewater treatment plant in the city of Gatlinburg, TN was constructed to regulate the wastewater flow rate to the primary treatment system during peak flow.
The function of a wastewater treatment plant is to improve the quality of wastewater by removing suspended organic and inorganic solids and other materials before discharging it into a waterway (Ref. 8). In treating wastewater, the rate at which the wastewater arrives at the treatment process might vary dramatically during the day, so it is convenient to equalize the flow before feeding it to the various treatment steps. The incoming wastewater flow is regulated prior to being directed to the subsequent treatment systems by a flow equalization basin. The excess sewage stored in the equalization basin is allowed to flow to the primary system for treatment when the incoming flow to the plant subsides. Excess wastewater flow during the peak flow is forced to the equalization basin by the automatic positioning butterfly valve. Therefore, the flow equalization basin helps to allow only a predetermined steady flow rate to flow to the primary treatment system.
The equalization basin was designed by Flynt Engineering of Knoxville in 1992 and built by Charlotte, N.C.-based Crowder Construction Co. in 1995/1996. According to the report from the Division of Occupational Safety and Health in the Department of Labor & Workforce Development of the State of Tennessee, the facility has been operated since 1996 by Veolia Water North America Operating Services under contract with the City of Gatlinburg.
The flow equalization basin at the wastewater treatment plant was an environmental engineering structure with five interior baffle walls. It is a rectangular shaped basin in cross-section. The dimensions of the flow equalization basin were approximately 124 ft. long, 64 ft. wide and 30 ft. deep. The thickness of the external walls was 18" and the five interior baffle walls were each 12" in thick. The thickness of the bottom slab of the basin was 15" and the top slab was 8" thick. The flow equalization basin had a maximum storage capacity of approximately 1.5 million gallons. The level of the raw sewage in the flow equalization basin during the collapse of the east wall was estimated to be in the range of 26-to-30 ft.
The structural design process for a flow equalization basin generally involves consideration of the following loads (Refs. 2, 3 & 8):
The primary load that is considered in the design of a basin is the hydrostatic pressure acting on the walls of the basin (Refs. 2, 3, 8 & 9). The hydrostatic pressure on the walls is assumed to have a triangular distribution. For rectangular-shaped basins that are intended to be monolithically constructed, the design is based on a full continuity between the walls. The basin was designed assuming that the walls were to be supported on three sides. These walls are considered as plates with varying boundary conditions (edge supports) depending on the construction details provided by the designer. The walls modeled as plates and subjected to hydrostatic pressure due to the wastewater will develop either a two-way or one-way action to support the applied loads, depending on the ratio of their spans and their edge support conditions (Ref. 8, 9, & 10 ).
Wastewater treatment plant components experience corrosion during their operation. The components of a flow equalization basin subject to corrosion include reinforced concrete walls, piping, ladders, mechanical and electrical equipment and other components used to construct the basin (Ref. 11). Raw sewage is a source of hydrogen sulfide that is released from the surface of the wastewater, enters the atmosphere and then is oxidized on the surface of the wastewater treatment plant. The oxidation of hydrogen sulfide results in the production of sulfuric acid that leads to the corrosion of metallic components of wastewater treatment plants (Ref.11).
Corrosion of reinforced concrete walls and other components is a major problem facing wastewater treatment plants. Non-watertight walls and cold joints provided during construction of walls of flow equalization basins are known to accelerate the corrosion of steel bars and rebar couplers (Ref. 8).
We conducted a structural investigation of the basin in conjunction with our field observations and review of the documents made available to us.
The technical specifications prepared by Flynt Engineering Company, Knoxville stated that (see Division 3- Concrete in Ref. 6):
The connection between the east wall and the baffle walls of the flow equalization basin was not constructed monolithically as required by the structural drawings. Instead of providing horizontal dowels of #4 rebars each face at 12" o.c. (see Fig. C-4) from the intersecting walls to the east wall, the contractor provided #5 rebars each face in the east wall and # 5 rebars each face in the intersecting walls. The rebars were threaded into a coupler, thus providing continuity between the east wall and the intersecting walls.
There was no available document to identify the manufacturer of the couplers used; but the widely used Dayton Superior D-50 DBR Coupler (for # 5 rebars-product code 77100 with 5/8"-11 UNC thread and having an outer diameter of 7/8" and a length of 2") closely matched the geometric properties of the couplers (see Fig C-6).
The thread engagement length specified by the manufacturer for # 5 rebars (product code 77100) was 7/8", but our inspection showed that the full engagement length specified by the manufacturer was not followed to connect the rebars to the couplers in some locations.
In conjunction with the field observations, a structural analysis of the basin was performed to review the structure as designed. We used both the finite element method and hand calculations to determine the force, moment, and displacement distributions for the walls.
The structural computer program STAAD.Pro V8i (Ref. 13) was used for our investigation. A finite element method was used to model the east and north walls. The east and north walls were modeled using quadratic plate finite elements.
We considered different boundary conditions at the wall joints and wall-to-base slab joint, to obtain the distribution of the moments and reactions to the walls. This analysis technique helped to capture the moment and reaction distributions by accounting for the flexibility of the walls at the wall-to-wall and wall-to-bottom slab joints.
We assumed for our analysis that the walls were to be free at the top. The 8" slab with one layer of rebars at the top of the basin was not accounted for.
The following assumptions were made for our structural analysis:
We considered in our structural analysis the level of wastewater above the base slab in the flow equalization basin to be 26 & 30 ft. and the boundary conditions for the walls to be either fixed or hinged. The various scenarios that were considered in our structural analysis were summarized in Tables 1 and 2, below:
Table 1. Boundary conditions and depth of wastewater considered for the east wall.
|East Wall level of
wastewater @ ft.
|Support at north
and south wall
|Moment Contour shown in Appendix A|
|3||26||hinged||fixed||No support||Figs. A-3 & A-4|
|4||30||hinged||fixed||No support||Figs. A-5 & A-6|
Table 2. Boundary conditions and depth of wastewater considered for the north wall.
|North Wall level of
wastewater @ ft.
|Support at east
and west wall
|Moment Contour shown in Appendix A|
The dowel reinforcements provided to transfer the moment from the east walls to the base slab as per the structural drawings were # 8 dowels @ 6'' o.c. (see Figs. 4 & 5). The moment capacity of these dowels was computed to be 61.8 ft-kips/ft.
The maximum positive moment for the east wall in the vertical direction was determined to be 40.3 ft-kips/ft (Fig. A-2). The reinforcement provided in the design was # 6 rebars @ 6" o.c., each face. The moment capacity of the reinforcement was computed to be 65.98 ft-kips/ft. Therefore, the design of the east wall was found to be satisfactory, provided that the east wall is supported by the three intersecting walls.
The maximum positive moment for the north wall in the horizontal direction was determined to be 93.1 ft-kips/ft (Fig. A-10). The reinforcement provided in the design was # 9 rebars @ 6" o.c, each face. The moment capacity of the reinforcement in the horizontal direction was computed to be 139.6 ft-kips/ft. Therefore, the design of the north wall was also found to be satisfactory.
If the flow equalization basin was built as per the structural drawings in a monolithic manner, the cold joint at the intersection of the east wall and the interior orthogonal walls would have been eliminated. In this case the interior orthogonal walls would have acted integrally with the east wall. Since the interior orthogonal walls were cast with rebar couplers connecting the rebars of the interior orthogonal walls to that of the east wall, the connections were modeled as hinged connections. The reaction and the flexural moments were computed under this condition.
If the depth of the water is considered to be 30 ft., the maximum reaction of the east wall at the intersecting walls was determined to be 41 kips. If the couplers are assumed to be in their original condition with the required engagement of # 5 rebars, each coupler has a maximum tensile capacity of 26 kips or 52 kips for two couplers. However, the couplers had undergone severe corrosion over a number of years due to which their capacities were significantly reduced. Post-incident inspection indicated that the couplers fractured and failed in tension due to overstress. If due to corrosive damage one set of coupler failed, the adjoining coupler would be subject to an even higher load, thus creating a chain reaction. All couplers examined at the site were observed to have suffered extensive corrosion damage (see Appendix B).
With the loss of the supports of the intermediate baffle walls, the east wall then spanned a distance of approximately 124 ft. between the north and south walls. At the hydrostatic pressure of 26 ft. of water, the maximum flexural moments in the horizontal direction and vertical directions were determined to be95 and 74 ft-kips/ft, respectively, without considering any load factors. The maximum moment capacities of the east wall in the horizontal and vertical directions were computed to be 73 ft-kips/ft (30% overstress) and 65 ft-kips/ft (13% overstress), respectively, without considering any capacity reduction factors. The east wall was, therefore, subjected to forces beyond its capacity. If the water is considered to be 30 ft. high in the basin, then the flexural moments in the horizontal and vertical directions would be 146 ft-kips/ft. (100% overstress) and 97 ft-kips/ft (50% overstress). The outward maximum displacement of the wall was computed to be approximately 10".
If the level of water in the basin was in fact 26 ft. and not 30 ft., then the east wall having lost the support of the intersecting walls would be overstressed by 30% and 13% in the horizontal and vertical directions, respectively. The magnitude of overstress, though undesirable, is not considered to be catastrophic. What became catastrophic was the inadequate connection of the east wall at the north and south walls, see Figs. 6, 7 & 8. The east wall was dowelled into the north and south walls by # 9 rebars which required a development length of 21" with a 90 degree hook. Inspection of the failed connection of the east wall at the north and south walls indicated that the development lengths of the # 9 rebars were grossly deficient.
Based upon our investigation, we concluded that:
*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.