Semisequicentennial Transportation Conference Proceedings
May 1996, Iowa State University, Ames, Iowa

Dynamic Field Performance of Timber Bridges

Douglas L. Wood, Terry J. Wipf, and Michael A. Ritter

D.L. Wood and T.J. Wipf,
Department of Civil and Construction Engineering,
Iowa State University,
Ames, Iowa 50011.

M.A. Ritter,
Forest Products Research Lab,
Madison, Wisconsin.

The dynamic response of three glued laminated (glulam) timber bridges has been determined from field test results using a heavily loaded truck. Bridge deflections and accelerations were measured for various vehicle speeds at the bridge midspan and recorded using a high speed data acquisition system and a dynamic amplification factor (DAF) computed. These tests were part of a field testing program to be performed as part of a larger research study that will include analytical research as well. The experimental data described in this paper will be used to validate analytical models. For brevity, only one span of one bridge will be discussed here. The overall purpose of the larger study will be to determine the dynamic behavior of glulam timber bridges so that reliable design specifications can be developed. Key words: timber, bridge, dynamic, glued laminated, girder.


Wood has been used as a bridge material in the United States for hundreds of years. Despite the exclusive use of wood bridges during much of the 19th century, the 20th century brought a significant decline in the percentage of wood bridges relative to those of other materials. At the present time, approximately 10 percent of the bridges listed in the National Bridge Inventory are wood (1). Recently, there has been a renewed interest in wood as a bridge material and several national programs have been implemented to further develop wood bridge systems. As a result of the Timber Bridge Initiative and the Intermodal Surface Transportation Efficiency Act, passed by Congress in 1988 and 1991, respectively, funding has been made available for timber bridge research (2). A portion of this research is aimed at refining and developing design criteria for wood bridge systems. This project to investigate the dynamic characteristics of wood bridges is part of that program and involves a cooperative research study between Iowa State University, the USDA Forest Service, Forest Products Laboratory, and the Federal Highway Administration. The first phase of the project addressed the dynamic performance of stress-laminated timber bridge decks (3). The second phase of the project is to assess the dynamic characteristics of glulam timber girder bridges.

Glulam timber girder bridges typically consist of a series of longitudinal glulam beams which support transverse glulam deck panels (Figure 1). The girders are available in standard nominal widths ranging from 4 to 16 in. (100 to 400 mm) with girder depth limited only by transportation and pressure treating restrictions. Deck panels are usually 5 to 6I in. (127 to 171 mm) thick, 4 ft (1.2 m) wide, and are continuous across the bridge width. Lateral support and alignment of the girders is provided by transverse bracing at the bearings and at intermediate locations along the span. Glulam girder bridges are feasible for spans ranging from 20 to 140 ft (6 to 43 m), although most are in the span range of 25 to 80 ft (7 to 24 m).

GENERAL RESEARCH PROGRAM

Static and dynamic tests were performed on three bridges of which one is presented in this paper. Vertical deflections were measured for several vehicle velocities for two different road approach roughnesses. The dynamic deflection data were compared to static deflections to quantify a dynamic amplification factor (DAF) for each test. The field tests were designed to observe bridge deflections and accelerations along with the vertical accelerations of the test vehicle.

DESCRIPTION OF BRIDGES AND VEHICLE

Three bridges were tested in Alabama; Mud Creek Bridge, Wittson Bridge, and Chambers County Bridge. Two spans were tested on The Wittson Bridge primarily to look at the response of a long span, 102 ft (31 m). The general bridge descriptions are shown in Table 1. For this paper only the Chambers Co. Bridge will be discussed in detail.

TABLE 1 Bridge Descriptions
Bridge Span ft (m) Stringers Lanes
Mud Creek Bridge 41 (12.5) 5 2
Wittson Bridge - Span 1 52 (15.8) 4 1
Wittson Bridge - Span 2 102 (31) 4 1
Chambers Co. Bridge 53 (16.1) 6 2

The Chambers Co. Bridge is a 53.1 ft (16.2 m) long single-span, two-lane bridge. Support for the bridge is provided by 6 glulam timber girders spaced 60 in. (1.5 m) on-center. The girders measure 8L in. (219 mm) wide by 53H in. (1359 mm) deep and are manufactured from visually graded Southern Pine (E = 1,850 to 1,930 ksi) (12.75E7 to 13.31E7 KPa). Deck panels are 5 in. (127 mm) thick, 48 in. (1219 mm) wide and are 29 ft (8.8 m) long to extend continuously across the bridge width. Steel guardrail on timber posts is installed on both sides of the bridge.

Traveling from the south, (the direction traveled by the test vehicle) the approach roadway to the bridge has slight a downward grade that levels 350 ft (106 m) before the bridge. According to the visual observation, the approach road surface roughness conditions could be characterized as good (asphalt pavement). The bridge pavement surface roughness could be characterized as very good. During the testing a depression about 1 in. deep developed in the middle of the immediate approach to the bridge.

The test truck was a three axle dump truck with a steel walking beam rear suspension with 179 in. (4.5 m) between the steering axle and first rear tandem axle and 53 in. (1.35 m) between the rear tandem axles. The axle loads from front axle to rear axle were 14.3 kips (63.6 kN), 24.8 kips (110.3 kN), and 24.8 kips (110.3 kN), respectively.

INSTRUMENTATION

The dynamic response of the bridge was recorded during the passage of the three axle truck traveling at constant velocity. Deflections were measured at midspan and the quarter-span of each girder using Celesco potentiometer transducers (DCPT). Accelerometers were also mounted on several of the girders at midspan and quarterspan. Details of the complete instrumentation can be found in Ritter, et. al. (3).

Acceleration data were also collected on the vehicle simultaneously with the bridge DCPT data. The accelerometers were mounted on the vehicle frame over the rear axles and on the rear tandem axle.

TEST PROCEDURE

The dynamic load behavior of the bridge was evaluated for several vehicle velocities for in situ and artificially rough approach conditions at the bridge entrance. Two different transverse vehicle positions were used: 1) eccentric, with the left wheel line (driver side) 2 ft (0.6 m) to the right of centerline and; 2) concentric, with the axle of the truck centered on the bridge (i.e., straddling the centerline).

In order to obtain a basis by which the dynamic load effects could be compared, crawl tests were performed for each loading position. During these crawl tests the vehicle velocity was approximately 5 mph (8 km/h). Deflections at higher velocities were then obtained with velocities ranging from 10 to 35 mph (16 to 56 km/h). The artificial rough approach condition was simulated using a 2 in. x 4 in. (44 to 89 mm) board placed at the bridge entrance.

DATA PROCESSING

A plot of bridge deflection vs. vehicle position along the bridge (using the vehicle front axle as a reference) was made for each DCPT location at the bridge midspan. The maximum deflection obtained for crawl speed is referred to as dstat. The maximum dynamic deflection is referred to by ddyn.

A dynamic amplification factor (DAF) was computed for each bridge. Each DCPT location was scanned to find the maximum absolute crawl deflection and this data point was then used as the reference point for the calculation of the DAF. As per recommendations by Bahkt and Pinjarkar (4), this approach yields the most useful design information. It should be noted that the data point that had the highest crawl deflection typically also had the highest dynamic response. The DAF was computed as:

where,

DAF = dynamic amplification factor

ddyn = maximum deflection under the vehicle traveling at normal speed

dstat = maximum deflection under the vehicle traveling at crawl speed

DYNAMIC AMPLIFICATION

The plots of dynamic amplification appear in Figure 2. Based on the natural period of 0.155 seconds for the bridge, the speed to satisfy the condition of the pseudo-resonance with passage of the rear tandem axles is 19.4 mph (31.2 km/h). At this speed, the amplification was high for the bump approach for both concentric and eccentric tests. The amplification increased as the speed of the vehicle increased beyond 25 mph (40.2 km/h). The low amplification at the speed of 29.5 mph (47.5 km/h) for the rough approach (bump) occurred due to the upward amplitude of the truck vibration while the truck passed the midspan of the bridge.

BRIDGE AND VEHICLE VIBRATION MODES

Four normal mode frequencies were determined from the free vibration record and compared well with results from a finite element computer model of the bridge. Structural damping was evaluated from the free vibration record and the calculated damping was found to be 5.8% of critical. Frequencies inherent to the vehicle were determined and the frequency of the vehicle body bounce was found to be 2.7 Hz. The frequency of the vehicle axle hop was found to be 10.2 Hz.

FORCED BRIDGE RESPONSE

Frequency content plots of the bridge response and vehicle response were analyzed and three speed intervals with different bridge behavior were identified: low (up to 10 mph) (16 km/h), medium (10 mph to 25 mph) (16 km/h to 40.2 km/h), and high (over 25 mph) (40.2 km/h).

At the low speed interval, the bridge vibrated at a frequency of 2.7 Hz. The vehicle response is dominated by a frequency of 2.6 Hz. At the medium speed interval, the bridge responded in frequencies of 6.9 Hz 10.6 Hz. These frequencies are close to the observed bridge mode frequencies of 6.4 Hz 11.0 Hz. The frequency of 11.0 Hz dominated the response for the bump tests. The frequency of 2.6 Hz appeared in the response along with the normal mode frequencies of 6.4 and 11.0 Hz. The frequency of 2.6 Hz was also present in the vehicle response.

ACKNOWLEDGMENT

The authors would like to thank Dr. Michael Triche of the University of Alabama, Dr. Steve Taylor of Auburn University, and all of the federal, state, county, and city personnel who provided assistance during the field testing. In addition, Jan Dlabola and Jason Carpenter, graduate students at Iowa State University, are acknowledged for their assistance in the research.

REFERENCES
  1. FHWA National Bridge Inventory. Federal Highway Administration, Washington, D.C., 1992.
  2. S.R. Duwadi and M.A. Ritter. Status of Research on Timber Bridges and Related Topics; Research Update. Federal Highway Administration, Turner-Fairbank Research Center, Structures Division, McLean, Virginia, 1994.
  3. M.A. Ritter, D.L. Wood, T.J. Wipf, C. Wijesooriya, and S.R. Duwadi. Dynamic Response of Stress Laminated Deck Bridges. Proceedings of the Fourth International Bridge Engineering Conference, Transportation Research Board, 1995.
  4. B. Bahkt and S.G. Pinjarkar. Review of Dynamic Testing of Highway Bridges. Structural Research Report, SRR-89-01, Ministry of Transportation of Ontario, Downsview, Ontario, 1989.

CTRE is an Iowa State University center, administered by the Institute for Transportation.

Address: 2711 S. Loop Drive, Suite 4700, Ames, IA 50010-8664

Phone: 515-294-8103
FAX: 515-294-0467

Website: www.ctre.iastate.edu/

Iowa State University--Becoming the Best