U.S. Department of Labor
Occupational Safety and Health Administration
Directorate of Construction
This report was prepared by
Scott Jin, Ph.D., P.E.
Mohammad Ayub, P.E., S.E.
Directorate of Construction
Occupational Safety and Health Administration
On September 7, 2011, at approximately 11:00 a.m., a 500-ton Liebherr mobile crane suddenly collapsed at the National Cathedral in Washington, D.C., amid thunderstorms and heavy rain. The crane's telescopic boom was 152 ft. and the attached lattice jib was 276 ft. long. The crane had been working on securing the cathedral building after recent earthquake damage. The crane tipped, overturned and fell its full length along the South Road, on the south side of the building. It damaged three parked vehicles and a historic Herb Cottage. The collapse was of a catastrophic nature but fortunately did not result in any fatalities. The crane operator sustained some non-life threatening injuries.
Personnel from the Baltimore/Washington, DC Area Office (BWAO) of the Occupational Safety and Health Administration (OSHA) arrived at the scene within hours of the incident. The OSHA investigation began soon after the collapse and included interviewing witnesses, taking photographs and requesting technical information from the general contractor. On the day of the incident, the OSHA Regional Administrator for Region III asked the Directorate of Construction (DOC), in OSHA's National Office in Washington, DC, to provide engineering assistance in assessing the collapse and in determining the cause of the incident. A structural engineer and a safety and health specialist from DOC visited the incident site on September 12, 2011 to examine the collapse, inspect the damages and discuss the collapse with the general contractor, the crane owner and employees at the site.
The DOC's investigation included:
The National Cathedral was built, mainly of Indiana limestone, using traditional stone masonry techniques. The cathedral is located at 3101 Wisconsin Avenue, NW, Washington, DC (See Figures 1 and 2) and is a popular attraction for both tourists and worshippers. Large stone pinnacles and other non-structural decorative elements of the cathedral sustained extensive damage during the 5.8 magnitude earthquake of August 23, 2011. The decorative stones and pinnacles were not fastened to their supports using modern techniques based on current seismic knowledge and practice.
The cathedral was anxious to undertake repairs to restore its former elegance. The cathedral retained the Universal Builders Supply, Inc. (UBS) of Cheverly, MD to stabilize the loose and hanging pieces and repair the earthquake-damaged areas. UBS was the general contractor. It decided to build a scaffold system on the roof of the center tower of the cathedral to reach to the top ornamental and decorative pieces damaged by the earthquake. As the roof of the center tower was approximately 320 feet high (Figure 3), UBS needed a crane to transport various scaffold pieces and structural beams to the roof level to build the scaffold system. For this reason, UBS retained the Crane Service Company (CSC) of Upper Marlboro, MD, to assemble a crane at the site to hoist the scaffold components and the steel support beams to the roof of the center tower of the cathedral building.
CSC deployed a telescopic mobile Liebherr crane at the site, identified as LTM 1400-7.1, Serial Number 072180. It was a seven-axle mobile crane equipped with a telescopic boom and a lattice jib (Figures 4 and 5). This crane had a maximum lifting capacity of 880,000 pounds (880 kips) at 10 ft. radius, a maximum hoist height of 400 ft. and a maximum reach of 300 ft. This crane was designed and manufactured by Liebherr-Werk Ehingen GmbH of Germany. It was a new crane acquired by CSC in June 2011. This crane was load tested and certified on July 25, 2011, by Martin Enterprize, Inc. of Chesterfield, VA.
CSC mobilized the Liebherr crane to the National Cathedral site on Sunday, September 4, 2011. The crane carrier was parked on the South Road on the backside of the main cathedral building at the designated location. The front of the carrier was toward the east and the zero degree mark on the fixed portion of the turntable was approximately towards the west. The four outriggers were extended and the pads were lowered onto the pavement. The southwest (rear right) outrigger pad happened to be near a catch basin, 15 ft. away. The size of the basin was 6 ft. long, 1½ ft. wide and 3 ft. in depth. It was a masonry structure constructed over 100 years ago. It took the entire day to assemble the crane to the intended operating mode. It included placing the counterweights on the counterweight frame, assembling the lattice jib and extending the telescopic boom.
The crane was put to work on Monday, September 5. It made a few lifts mostly in the 3,200 pounds range. In the afternoon, the crane operator and others observed a crack on the asphalt pavement at the rear right outrigger pad.
The next day, September 6, the cracking of the asphalt continued and caused a ¼ inch settlement of the rear right outrigger pad. CSC placed two 32 ft. long, 4 ft. wide and 12 in. thick timber mats side by side on the asphalt; and added an 8 ft. by 9 ft. steel pad between the mats and the rear right outrigger pad (Figures 6 and 7 indicate the pad after the collapse). The purpose of the mats and the steel pad was to uniformly spread the load over a larger area and to minimize any additional settlement. Later in the afternoon, when the wind subsided, the crane successfully made one lift of scaffold cross braces. No other activity took place and the crane was set for the night.
On Wednesday, September 7, after performing the daily inspection, the crane operator set the 152 ft. long telescopic boom to 82 degrees. The attached lattice jib was 276 feet long. A lift was made carrying 3,200 pounds of scaffold components without any problem. The next load was steel I beams that weighed 8,600 pounds. The load was rigged on the northwest and delivered on the north side approximately 320 ft. high to the roof of the tower. As the load was being released on the roof, it began to rain intensely. The load was released and the operator raised up the hook to clear the cathedral tower. At this time, lightning was snapping around the top of the jib, obscuring its view from the operator's cab. The operator then swung the boom and jib to the west (the rear of the carrier), and luffing the jib down to a lower angle to let the storm pass by. At this time, the operator noted that the telescopic boom was at approximately 68 degrees and the jib was at approximately 18 to 20 degrees. Later, however, the crane operator recalled that the jib was in fact at near zero degrees to the horizontal. After a few minutes, at around 11:00 a.m., the operator felt sudden vibration of the crane. Suddenly the boom began to go down. The boom was falling, then, the counterweight was rising until it almost stood vertical at which time it rotated towards the south. The counterweights hit the pavement and the timber mats supporting the outrigger, and pierced through the pavement. The north counterweights were separated and fell off. The counterweights on the south side remained connected.
From the field examination of the collapsed crane:
Based on the examination of the collapsed crane elements at the incident site and at the CSC storage yard, the crane was in the TYSN operating mode with TY guys and a TN adaptor installed before the collapse. At the time of the collapse, the length of the telescopic boom was 152 ft. and the length of the attached lattice jib was 276 ft. (See the photograph on the cover of this report). The total installed counterweight was 154,300 pounds (70 metric tons). It included the base plate (15 tons), the center plate (5 tons), three counterweights (25 tons) on the south side of the base plate, and identical weights (25 tons) on the north side of the base plate. All tons refer to the metric tons (2,200 pounds). It was further confirmed with the CSC personnel at the storage yard that the counterweight with its frame was fully extended and Winch 3 was mounted at the rear end of the frame. Thus, the code of the applicable Liebherr load chart was 1596. Both CSC and the crane manufacturer were using this load chart (with Code Number 1596) for the control and operation of the crane.
The LICCON computer system is a system for controlling and monitoring mobile cranes. The abbreviation "LICCON" stands for LIebherr Computed CONtrolling. Some key functions of the computer controlled operation system from the crane operation manual are summarized below. Note that the sentences in italics are directly taken from the operation manual.
Cranes usually fail either due to structural overstress at high angles of booms or due to instability at lower angles and the large radius of the boom. In this investigation, we are dealing with a case of instability that occurred at the large radius of the boom and near horizontal angle of the jib.
There are two factors related to the stability of cranes, i.e., the overturning moments and the stabilizing moments. If the overturning moment is less than the stabilizing moment, then the crane is stable. However, if the overturning moment exceeds the stabilizing moment, the unbalanced moment will cause the crane to rotate, tip-over and fail.
The stability of the crane was analyzed based on the magnitude of the stabilizing and overturning moments. The analysis was conducted to determine the overturning moment at the time of the collapse considering the orientation of the crane, boom and the jib at the time of failure. As will be discussed later, the boom was at an angle of approximately 63 degrees to the horizontal, and the jib was nearly horizontal. The boom and the jib were positioned approximately along the zero degrees of the turntable (toward the west direction). The wind speed at the time of the collapse and the settlement of one of the rear outriggers were also considered in the analysis.
The basic dimensions of the carrier and the outriggers were taken from the Liebherr manual and verified in the field (See Figure 15). The weights and center of gravity of the carrier, various components of the superstructure, and counterweights are indicated in Figure 16.
The following items were considered in the analysis:
|Crane Member||Individual Element||Individual Weight||Total Weight||Centroid to Tipping Axis*|
|Carrier||---||---||87,100 lbs.||-23.3 ft.|
|Superstructure||---||---||45,300 lbs.||-19.9 ft.|
|Counterweight + Frame + Winch 3||Base Plate
10 Tons Wt. x 4
5 Tons Wt. x 2
Frame + Winch 3
|Telescopic Boom Assembly||Base Section
½ Luff. Cylinder
|Lattice Jib Assembly||G9
2 x G4
2 x G6
2 x 1.86 tons
2 x 1.00 tons
|Load Block Assembly||headache ball + Web Sling + Nylon Ropes||---||1,760 lbs.
* Centroid to the tipping axis is the horizontal distance from the center of gravity of the crane members to the tipping axis at the time of the collapse.
The weights of some of the above items were taken from the operating instructions for the crane (See Figure 17).
At the time of the collapse, the entire crane including all elements mentioned above weighed approximately 200 tons. This weight compared very closely to the total weight of 203 tons derived from the shipping packing lists prepared by Liebherr.
It was determined that when the telescopic boom reached an angle of 63 degrees to the horizontal, and when the jib was nearly horizontal, the overturning moment was 8,000,000 ft-pounds. The stabilizing moment at this position of the crane was computed to be 7,980,000 ft-pounds, and hence the failure. At the time of the collapse, the crane had a working radius of 344 feet which was well beyond the allowable radius provided in the load chart, discussed later.
Wind and settlement of one of the outriggers adversely impacted the stabilizing moment but their contributions were minimal. At the time of the collapse, wind speed was recorded as 15 mph. An average pressure of 0.6 psf was estimated for the exposed elements of the crane. The wind resulted in an additional overturning moment of 100,000 ft-pounds, i.e., 1.3% of the total overturning moment. The impact of the settlement of one of the rear outriggers was also considered. The settlement was assumed to be 6 inches at the rear right outrigger. This settlement resulted in an additional overturning moment of 108,000 ft-pounds, i.e., 1.4% of the total overturning moment. As mentioned the impact of the wind and settlement were minimal. If the crane was operated within the bounds of the load chart, wind and the settlement of one of the outrigger pads would not have caused the collapse of the crane.
The crane manual consists of several volumes. There are numerous load charts provided in the manual. The appropriate load chart is selected based upon the length of the telescoping boom, the length of the luffing boom, the magnitude of the counterweights, and the number of axles of the mobile crane carrier. The load chart for the crane which collapsed indicated the following:
The crane's code number was determined to be 1596 (See Figure 18), and the applicability of this load chart is not in dispute.
The load chart essentially provides the maximum permissible load that can be hoisted at various working radii. Working radius is defined as the horizontal distance from the center of the turntable to the vertical axis of the load being hoisted. Furthermore, the angles of the telescopic boom at which permissible loads are provided at different radii are also given.
Of particular interest to this investigation are seven columns in the load chart, 1 through 7 beginning from the left. Column 1 provides various working radii ranging from 115 feet to 300 feet. The next column provides permissible loads for the radii given in Column 1, but this applies to a telescopic boom length of 118 feet which is not the case here. Column 3 provides permissible loads at radii ranging from 120 feet to 285 feet, applicable to a telescopic boom length of 152 ft. which was the boom length for this crane. This set of permissible loads must also comply with another constraint which is that the telescopic boom must also have an angle of 82 degrees. So, if the 152 ft. long telescopic boom is at an angle of 82 degrees, then loads ranging from 15,600 pounds at a 120 ft. radius to 5,800 pounds at a radius of 285 ft. could be hoisted. Column 4 is applicable to a 118 ft. long boom. Column 5 is applicable to a 152 ft. long telescopic boom, and provides permissible loads at different radii, but the telescopic boom must have an angle of 75 degrees. For example, the crane could hoist a load of 11,600 pounds at a radius of 170 feet, or a load of 2,500 pounds at a radius of 260 feet, provided the boom has an angle of 75 degrees. Column 6 is applicable to a 118 ft. long boom. Column 7, although applicable for a 152 ft. long boom, does not provide any permissible load. The jib angle could vary depending upon the site demands provided that the working radius is not exceeded.
It is interesting to note that the load chart permits a load of 5,800 pounds at a radius of 285 ft. but only permits a load of 2,500 pounds at a smaller radius of 260 ft. This is because a greater load is permitted due to the steeper angle of 82 degrees, and a smaller load is permitted at a shallower angle of 75 degrees. It is well-recognized in the industry that interpolation is not permitted in the crane load charts. If the actual lifting radius lies between the two tabulated radii, then the proper practice is to use the larger radius. If the boom angle lies between the tabulated angles, then the lower angle is used to determine the safe load-carrying capacity.
In this load chart, only two angles for the telescopic boom are provided, i.e., 82 and 75 degrees. So, the boom can only be operated between angles of 82 and 75 degrees. If the boom is at an angle greater than 82 degrees or at an angle lower than 75 degrees, then the crane would be in violation of the load chart, and a failure could be imminent. Similarly, at a boom angle of 75 degrees, if the radius of the load is greater than 260 ft., the load chart is violated and a failure could be imminent. In the present case, the boom was at an angle of 63 degrees at a radius of 344 ft., both exceeding the maximum permissible values. Although the crane was not hoisting any load, the weights of the headache ball (load block) and other accessories were enough to topple the crane.
Our structural analysis, independent of the load chart, confirmed that the crane with the boom angle of 63 degrees, and the jib at nearly horizontal angle will overturn even if there is no load at the hook other than headache ball and the weight of wire ropes, etc. The computations for overturning and stabilizing moments are included in Appendix A.
The crane operator's cab is equipped with a computer that records operations of the crane at intervals of five seconds. This program is called Liccon Lift Analyzer II. The computer-stored information is generally referred to as a black box. The program is proprietary and requires special software to download data recorded during the actual operation of the crane. After several weeks of negotiations between the interested parties, CSC purchased the special software required to download the data. The data recorded in the last few minutes before the collapse provided insight into the actual configuration, i.e., the angle of the boom, the angle of the jib, the orientation of the crane platform, the load at the time of the collapse, the utilization ratio, the outrigger reactions, etc. It also graphically plotted the vertical and horizontal angles of the boom and the jib at different times (See Figure 19).
A typical example is provided below.
Data was downloaded for:
The computer time does not correspond to Eastern Daylight Time (EDT) nor does it correspond to German time. It is understood that the computer time is 4:54 hours ahead of US EDT. The data provided the movements of the crane boom and the jib at various times of the day of the incident, in particular the last 16 minutes before the collapse. The following information is extracted from the data logger, and tabulated below showing the orientation of the crane platform, and the vertical angles of the boom and the jib.
|Computer Time||U.S. EDT AM||Horizontal Angle of the Platform||Vertical Angle of the Boom||Vertical Angle of the Jib|
|15:38||10:44||-96 deg||81.2 deg||51.1 deg|
|15:39||10:45||-96 deg||81.2 deg||51.1 deg|
|15:42||10:48||-90 deg||81 deg||60 deg|
|15:44||10:50||-50 deg||81 deg||63 deg|
|15:46 *||10:52||-4 deg||81 deg||23 deg|
|15:48||10:54||-4 deg||81 deg||23 deg|
|15:49||10:55||-4 deg||81 deg||23 deg|
|15:50||10:56||-4 deg||81 deg||16 deg|
|15:52||10:58||-4 deg||70 deg||4 deg|
|15:54||11:00||-4 deg||81 deg||0 deg|
|15:54:19||11:00:19||-4 deg||64.3 deg||-2.4 deg|
* Figures in bold indicate that the crane was operating outside of the safe zone, as per Liebherr Load Chart.
The crane collapsed at approximately 11:00 a.m., corresponding to the second last row (highlighted) in the above table. The table provides information for the last 16 minutes before the collapse, providing data on the movements of the crane. It begins at 10:44 a.m. at which time the crane was positioned towards the north over the cathedral roof as the crane was unloading the cantilevered steel beam. At this time, the boom was at an angle of approximately 81 degrees to the horizontal, an angle which the crane maintained until approximately four minutes before the collapse even during its counterclockwise rotation towards the west. At 10:44 a.m., the jib was at an angle of approximately 51 degrees. The crane was then well within the load chart.
After unloading the material, rain started, and the operator swung the crane counterclockwise to the west without carrying any load in about 6 minutes until 10:50 a.m. The jib was raised to approximately 63 degrees. The crane continued to be within the load chart. However, at approximately 10:52 a.m. the crane jib was then suddenly lowered to 23 degrees, a drop of 40 degrees in the course of one minute. The crane was then operating outside the load table because our analysis indicated that the radius was approximately 273 ft., which is greater than the maximum permissible radius of 260 ft. This is confirmed by the sudden increase of the recorded utilization values from 43% to 32,767% (See Figure 21). The utilization value is the ratio of the real (actual) load over the maximum permissible load. Any utilization value above 100% indicates the crane is operating beyond the load chart. It is believed that at this time the crane was automatically shut off, and no movement of the boom or the jib could occur unless the bypass switch was activated overriding the shut off.
For the next three minutes, no movement was recorded either in the boom angle or the jib angle, which would indicate that the computer had actually shut off the operation. But, at 10:56 a.m., the operation started again. Is it possible that in the last three minutes, the bypass switch was activated? We do not know because the crane operator denies engaging the bypass switch. At 10:56 a.m., the jib was recorded to be lowered to 16 degrees, while the boom was maintained at approximately 81 degrees. At 10:58 a.m., the boom was lowered to approximately 70 degrees, and the jib was further lowered to 4 degrees. At 11:00 a.m., the boom was recorded to be at approximately 64 degrees with the jib at nearly horizontal. In fact, the recorded support reactions from the data-logger show that the front left pad had just separated (See Figure 20). The collapse followed immediately.
Our independent analysis indicated that the approximate angle of the boom when the overturning and stabilizing moment would be equal would be approximately 63 degrees with the jib being horizontal. This matches closely with the recorded angle of 64 degrees.
At 10:52 a.m. and thereafter, the crane was being operated beyond the load chart, and failure could be expected anytime. The reasons of how the crane became operational again after being stationary for three minutes are not clear. An obvious reason could be that the crane operator engaged the bypass switch overriding the shutoff. Another reason could be a malfunction of the shutoff.
It must be noted that certain information contained in the print out of the data logger was inaccurate in regard to the working radius, the telescopic boom length, the total support forces and the permissible load immediately prior to the collapse (see Figure 22) downloaded from the data-logger. It is uncertain whether these discrepancies occurred due to flaws in the special software used to download the data or in the crane software itself.
The crane operator indicated a scenario of a thunderstorm with heavy rain and lightning near the tip of the jib. His visibility was very limited and he could not see the tip of the jib. He argues that because it takes approximately 20 minutes to telescope the boom inward due to the 12-part pulley system, he decided not to telescope the boom inward. Instead, he lowered the jib to near zero degrees and also lowered the boom to approximately 68 degrees to avoid lightning and thunderstorm. Then, he was waiting for the storm to pass by. As he was waiting, the crane began to overturn. Our analysis indicates that if he had maintained the boom at 81 degrees and the jib at 22 degrees, this incident would not have happened, despite the raging storm.
(For the complete Appendix, see PDF*)
A-2 Weight and Center of Gravity of the Crane Members Provided by the Crane Manufacturer for the Recovery Operation
A-3 Estimation of the Total Weight of the Crane at the Time of the Collapse
A-4 Estimation of the Total Resisting Moment in the Direction of the Collapse
A-5 Estimation of the Weight and the Center of Gravity of the Telescopic Boom Assembly
A-6 Estimation of the Weight and the Center of Gravity of the Lattice Jib Assembly
A-7 Estimation of the Overturning Moment Due to the Telescopic Boom Assembly, Lattice Jib Assembly and Hook Block
A-8 Estimation of the Overturning Moment Due to the Wind Force
A-9 Estimation of the Overturning Moment Due to the Settlement of the Rear Right Support Pad
A-10 Estimation of the Total Overturning Moment in the Direction of the Collapse
*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.