Dynamic Properties of Stay Cables on The Penobscot Narrows Bridge

CHAPTER 3. EXPERIMENTAL PROCEDURES

THE TEST BRIDGE

The Penobscot Narrows Bridge is a 2,120 ft (646 m) long cable-stayed bridge that carries U.S. Route 1 over the Penobscot River connecting Verona Island to Prospect in Maine. A map of the bridge site and the surrounding area is provided below in figure 10. This cast-in-place concrete structure was an emergency replacement for the Waldo-Hancock Bridge, an aging steel suspension bridge built at the same site in 1931. The "owner facilitated design/build project" spanned a mere 42 months from conception to completion; construction was started in 2004 and the bridge was opened to traffic on December 30, 2006.

1 mi = 1.61 km
Figure 10. Illustration. Map of the Penobscot Narrows Bridge site.

The bridge, with its east-southeast orientation, has a main span of 1,161 ft (354 m) and side spans of 479.5 ft (146.2 m) each, as shown in figure 11. The span cross section is a single trapezoidal concrete box that is 12.8 ft (3.9 m) deep with a 57.5-ft (17.5-m)-wide deck. The deck configuration, as shown in figure 12, consists of a roadway with one 12-ft (3.65-m) lane in each traffic direction flanked by a 7-ft (2.1-m) shoulder on the right and 1-ft (0.3-m) shoulder on the left. A raised 14.3-ft (4.4-m) median, containing the cable anchor blocks, separates the two travel lanes.

Source: MaineDOT.
Figure 11. Illustration. Bridge plan and elevation.

1 ft = 0.305 m, 1 inch = 2.54 cm
Source: MaineDOT.
Figure 12. Illustration. Bridge main span cross-section.

The bridge spans are supported by two 430-ft (131.1-m) tall concrete pylons, each containing 40 stay cables. The pylons are tapered with height in the direction of traffic and are constant width transverse to the direction of traffic. A unique feature of the west pylon, Pier 1, is the public observatory, accessible by elevator, located at its top. Clad in glass, the bridge observatory is the only such feature in the United States and the highest public bridge observatory in the world. A photo of the observatory is shown in figure 13.

Figure 13. Photo. Bridge observatory.

The Penobscot Narrows Bridge has four centerline fans of stay cables between the bridge spans and pylons. The two fans on the outer side spans are essentially symmetrical, as are the two inner fans supporting the main span. A photo detailing the cable fan configuration is shown in figure 14. Chord lengths of the cables range from 119 to 594 ft (36.4 to 181.1 m) and inclination angles range from 22.1 to 52.8 degrees. The cables in the inner fans are roughly 13 percent longer than their counterparts in the outer fans. The fans are labeled A, B, C, and D running from the west to east end of the bridge. The cables are numbered 1-20 in each fan, with 20 being the longest. In this report, a cable will be referenced by its number, followed by its fan letter, e.g., cable 19A.

Figure 14. Photo. Bridge cable fan configuration.

The stay cables consist of 41 to 72 steel wire strands loosely encased in an ungrouted high-density polyethylene (HDPE) pipe. The HDPE pipe has a constant diameter of 15.75 inches (400 mm) for all cables, despite the varying number of strands for the different cable lengths, and includes a double helical fillet on its surface. The 0.6 inches (1.52 cm) diameter 7-wire strands are epoxy coated and filled. These strands are attached to anchor blocks at deck level and are continuous, passing through cradles, at each pylon. Each strand acts independently, allowing for removal, inspection, and replacement. The Penobscot Narrows Bridge is one of the first few bridges to incorporate the use of a cable-stay cradle system, eliminating the need for anchorages in the pylons and allowing for increased diameters of stays and, therefore, increased span lengths.⁽⁴⁾ In June 2007, six strands in three stays were removed and replaced with carbon fiber strands to enable monitoring of the carbon fiber material in this special application. Each cradle assembly consists of a galvanized steel outer sheath with an outside diameter of 16 inches (488 mm) and individual stainless steel sleeves with 1 inch (2.54 cm) outside diameter for each strand. At the upper end of each stay cable, the HDPE pipe is attached to the cradle end through a bolted flange.

A galvanized steel guide sleeve, or guide pipe, extends from the concrete anchor block at the lower end of each stay cable. A galvanized steel anti-vandalism pipe is attached to the end of the guide sleeve through a bolted flange to extend protection from damage in this region of the cable. An HDPE "connection sleeve," with 17.72 inches (450 mm) outside diameter and no helical fillet, is used to make the transition from the anti-vandalism pipe to the primary HDPE pipe, and the joint between steel and HDPE is sealed with a neoprene boot and steel bands. No internal neoprene bushings or internal dampers are utilized at either end of the stay cables. A diagram detailing the numerous transitional elements of the lower stay cables is shown in figure 15, while figure 16 contains a photo of these elements on several cables towards the west end of the bridge.

1 ft = 0.305 m, 1 inch = 2.54 cm
Figure 15. Illustration. Bridge stay cable assembly details.

Figure 16. Photo. Detail of lower end of bridge cables.

External viscous dampers are installed in pairs near the lower end of all stay cables. The axes of the dampers in each pair are separated by an angle of less than 45 degrees to provide damping for out-of-plane as well as in-plane vibrations. Each damper-to-stay connection is located 12.33 ft (3.76 m) vertically above the bridge deck; therefore, its position up the length of the cable is a function of β, the cable inclination angle. For the range of inclination angles on this bridge, the dampers are positioned between 6.4 and 16.9 percent of the cable length from the lower anchorage. The dampers are attached to the HDPE connection sleeve using a galvanized steel bracket fabricated in two halves for ease of installation. A multi-sleeve, or cheese, block is installed inside the HDPE connection sleeve at the damper location to ensure that the individual steel strands and pipe move as a unit. The block consists of six HDPE cheese plates separated by five rubber cheese plates. Each damper rests atop a steel extension pipe that is solidly anchored to the top slab of the raised median.

SET-UP AND PROCEDURES

The cable-stays on the Penobscot Narrows Bridge were tested by staff from the FHWA Aerodynamics Lab in two phases, both before and after the installation of dampers. Phase 1 testing of the cable-stays was conducted in early December 2006 during construction, and phase 2 testing occurred in September 2007 with traffic on the bridge.

Data from the vibration testing was obtained by attaching two accelerometer boxes to each stay cable, the first box being from 19 to 21 percent up the length of the cable and the second being from 6 to 9 percent up the length of the cable. The lower box, or enclosure, was positioned close to the connection point for the dampers and the upper box was placed as high as possible on the cable and in a position that would avoid vibration nodes for the modes of interest. Figure 17 shows the accelerometers mounted on top of a stay cable, with the second one sitting further up the cable in the distance.

Figure 17. Photo. Accelerometers mounted on stay cable.

Multiple boxes ensured that useable data would be collected even in case of a malfunctioning sensor or if a sensor was inadvertently stationed at a frequency node. The accelerometers measured data from all three axes, although only data from the in-plane direction were required for analysis. The in-plane direction corresponds to the vertical plane of cables, which was measured as the z-direction on the accelerometer. Data from the accelerometers were recorded using a portable data acquisition system that was also connected to a vane-type anemometer to measure the wind direction and speed present during the test. In general, the anemometer was placed in the vicinity of the cables being tested and thus was moved about the bridge. The data acquisition system was connected to the bridge for power and is shown in figure 18. The scan frequency used for both the accelerometers and the anemometer was 100 Hz.

Figure 18. Photo. Data acquisition system.

The cables were manually excited in the vertical plane with a "pull" rope, while a spotter checked to make sure the proper amplitudes and modes were achieved. Positioned with a good view along the longitudinal axis of the cable under test, the spotter would signal the "pullers" to synchronize the pulling action with the cable motions. This approach enabled efficient excitation, vibrations primarily in the first vertical mode, and peak amplitudes of one cable diameter or more. When the cable reached a sufficient excitation, the rope was released, allowing the cable to freely oscillate and the vibrations to decay. The data acquisition system would be triggered to start recording before the excitation was started then continued recording until the decay subsided and only random traffic or wind-induced vibrations remained.

In addition to the data recorded from the sensors, general test notes were taken regarding sensor locations, weather conditions, quality of the vibration modes achieved, traffic problems, interference from construction activities, and any difficulties encountered while exciting the cable. Figure 19 shows a diagram of a stay cable illustrating the distance each accelerometer was placed from the anchor block, while table 1 contains a summary of the distances used during testing.

Figure 19. Illustration. Location of accelerometers during testing.

Table 1 . Summary of accelerometer locations.