Design Guidelines for Test Level 3 through Test Level 5 Roadside Barrier Systems Placed on Mechanically Stabilized Earth Retaining Walls (2022)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

39 3 PEAK DYNAMIC LOAD The objectives of this portion of the study are to update the design impact loads for TL-3 impacts and to quantify the design impact loads for TL-4 and TL-5 impacts. The analysis work includes the prediction of impact loads based on measured test data and FE impact analyses using the commercial FE software LS-DYNA (6). 3.1 Impact Load Study for TL-3 Impact The principal objective of this section is to present the updated TL-3 dynamic load. A load of 54 kips is currently recommended in the AASHTO LRFD Bridge Design Specification and is intended to reflect the design impact load for NCHRP Report 350 TL-3 impact conditions. The Chevrolet C2500 pickup truck model (see Figure 3.1-a) is representative of the 2-door, 2000P pickup truck test vehicle in NCHRP Report 350 (4). The Chevrolet Silverado pickup truck model (see Figure 3.1b) represents the four-door, 2270P pickup truck test vehicle specified in MASH (5). (a) Chevrolet C2500 model (b) MASH 2270 Pickup truck model (c ) TL-3 Crash test vehicle Figure 3-1 TL-3 simulation and crash test vehicles. The IS of MASH Test 3-11 is 13% greater than the corresponding test under NCHRP Report 350 due to an increase in vehicle mass from 4,400 lb. (2000 kg) to 5,000 lb. (2270 kg). There are other differences between the two design vehicles that can potentially influence the load imparted to a barrier during an impact. The MASH 2270P vehicle was used in the crash test performed under the previous NCHRP project (see Figure 3.1c). However, the MASH pickup truck

40 model was not available for the previous study and, therefore, MASH TL-3 impact loads were not updated in NCHRP Report 663 (2). 3.1.1 Finite Element Model The MSE wall and TL-3 barrier model used in the previous study is shown in Figure 3-2. The C2500 pickup truck model impacted the 32-in. (0.81-m) high vertical wall barrier at a nominal impact speed and angle of 62.2 mph (100 km/h) and 25 degrees, respectively. The 10-ft (3.05-m) long barrier sections were attached to 30 ft (9.14 m) long and 4.5-ft (1.37 m) wide (measured from the panel roadside face) moment slabs. The BMS system was placed on top of a 10-ft (3.05 m) tall MSE wall with 10 ft (3.05 m) long steel reinforcing strips. The dynamic impact loads were determined based on maximum lateral contact forces (perpendicular to the barrier length) generated by the vehicle impact on a rigid barrier. The 54-kip (240-kN) design impact load for NCHRP Report 350 TL-3 was confirmed through this analysis as shown in the resulting contact force versus time curve (Figure 3-3). Under the current project, the MASH 2270P pickup truck model was used to determine a design impact load for MASH TL-3 impact conditions. Since the wall reinforcement loads were not part of this analysis, a smaller model that includes the barrier and the moment slab system but not the MSE wall was prepared (Figure 3-4). Figure 3-5 shows sequential images of the test and the FE simulation. A comparison of the acceleration and the angular displacements between the TL-3 crash test results and the simulation results is shown in Figure 3-6. Figures 3-5 and 3-6 show that the simulation results with the modified SUT vehicle correlate reasonably well with the test results. The resulting contact force versus time curve is shown in Figure 3-7. The maximum 50- msec. average force obtained from this curve is approximately 70 kips (311 kN) (see Figure 3-6). This represents a 30% increase in the applied impact force between NCHRP Report 350 and MASH. Figure 3-2 Three-dimensional view of MSE wall, barrier, and C25000 Vehicle Model.

41 Figure 3-3 2000P pickup truck impact force on barrier (50-msec. average). (Source: NCHRP Report 663 (2) p.128.) Figure 3-4 Side view of BMS system and MASH TL-3 with the 2270P-pickup truck model.

42 a) Test t=0 sec. b) Simulation t=0 sec. c) Test t=0.086 sec. d) Simulation t=0.086 sec. e) Test t=0.171 sec. f) Simulation t=0.171 sec. Figure 3-5 Sequential images for 2270P-pickup truck test (2) and simulation.

43 a) x-acceleration (Test 475350-1) b) x-acceleration (MASH Pickup Truck Model) c) y-acceleration (Test 475350-1) d) y-acceleration (MASH Pickup Truck Model) e) Angular Displacement (Test 475350-1) f) Angular Displacement (MASH Pickup Truck Model) Figure 3-6 Comparison of acceleration and angular displacements between MASH TL-3 test and simulation.

44 Figure 3-7 MASH TL-3 impact force versus time history for barrier impact (50-msec. average). 3.2 Impact Load Study for TL-4 Impact The principal objective of this section is to estimate the magnitude and distribution of the MASH TL-4 impact load on barriers of different heights. In addition, the distributions of the lateral impact load in the longitudinal and vertical direction are also investigated using FE analyses techniques. The MASH TL-4 impact involves an SUT (10000S) vehicle weighing 22,036 lb. (10,000 kg) impacting the barrier at a speed of 56 mph (90 km/hr) at 15-degree angle. The nominal IS of the impact is 154.7 kip-ft (209.6 kJ). 3.2.1 Analytical Study When an SUT impacts a barrier there are two distinct impacts. The first impact occurs when the front of the vehicle contacts the barrier. The vehicle then begins to yaw or rotate towards the barrier. The second impact occurs when the rear of the vehicle contacts the barrier. This second impact is sometimes referred to as the “back slap.” Normally, the second impact generates the largest impact force. Due to changes in SUT vehicle properties and impact conditions incorporated into MASH, it has been determined that 32-in. (0.81-m) barrier height is no longer adequate for TL-4. This was demonstrated in a MASH TL-4 full-scale crash test of a 32-in. (0.81-m) tall NJ Safety Shape bridge rail (18). In this test, the SUT vehicle rolled over the barrier and failed the structural adequacy criterion of MASH. A subsequent research project used FE analyses to determine that a 36-in. (0.91-m) tall barrier would be adequate to contain and redirect a MASH 10000S test vehicle (19). This result was confirmed in a full-scale crash test of a 36-in. (0.91-m) tall SSTR. In the test, a 22,000-lb. (9,982-kg) SUT was successfully contained and redirected after impacting the barrier at a speed of 57.2 mph (92 km/hr.) and an angle of 16.1 degrees. The measured maximum 50- msec. average lateral acceleration was 4.5 g’s. A first level approximation of the MASH TL-4 impact load was obtained using a mass- spring model in combination with the mathematical model described in NCHRP Report 86 (33). The procedure employs a dimensional analysis to estimate the influence of a change in impact

45 velocity, impact angle, vehicle dimensions, weight and stiffness of the system on the magnitude of the impact load. This approach, which is shown in Eq. (3-1) through Eq. (3-3), was used by TTI researchers to help derive the current impact loads presented in AASHTO LRFD (3) using forces measured with an instrumented rigid wall (13).       == 1 2 1 2 1 2 1 2 ; W W K Kf a a F F (3-1)                           == 22 11 1 2 2 1 2 1 2 1 2 ; sin sin; LA LA V V f a a F F θ θ (3-2)                               = 1 2 1 2 22 11 1 2 2 1 2 12 W W K K LA LA Sin Sin V VxFF θ θ (3-3) where F = impact load a= lateral acceleration V= impact velocity of the vehicle θ = impact angle of the vehicle AL = distance from the front of the vehicle to its center of mass K = stiffness of the system (vehicle and the barrier) W= mass of the vehicle A series of assumptions must be considered in order to apply this mathematical model. For example, the lateral and longitudinal accelerations are considered constants, the vertical rotation of the vehicle is neglected, the vehicle is not snagged by the barrier, the center of mass moves with the entire mass of the vehicle, the forces generated between the vehicle tires and the roadway surface are neglected, and the lateral force is represented by a sine wave distribution (33). The subscript 1 refers to an NCHRP Report 350 TL-4 test vehicle and the subscript 2 refers to a MASH TL-4 test vehicle. The ratio of A1L1/A2L2 was assumed to be equal to 1 because the dimensions of the SUT test vehicle did not change. The ratio K2/K1 accounts for the relative stiffness of a 10000S vehicle impacting a 36-in. (0.91-m) tall barrier compared to an 8000S test vehicle impacting a 32-in. (0.81-m) tall barrier. Since the two vehicles and the two barrier are considered to be the same material, the difference in stiffness between the two impacts can be only associated with the change in height of the barrier and resulting change in contact area. Therefore, K2/K1 can be written as h2/h1. Then, Eq. (3-3) was used to update the 54-kips (240-kN) impact load from NCHRP Report 350 to MASH TL-4 impact conditions. The result shows a MASH TL-4 impact load of 80.3 kips (357.5 kN). This increase in load reflects the changes in impact speed, vehicle weight, and barrier height. Another way of estimating the impact load of single body vehicles is to use Newton’s second law of motion. In this procedure, the total mass of the vehicle is used and multiplied by the lateral vehicle acceleration measured at the CG of the vehicle. Using this approach, the impact load of the successful MASH TL-4 test (19) can be estimated using Eq. (3-4):

46 ( ) 22050 4.5 32.2 99 32.2impact total lat F m a kips= × = × × = (3-4) These methods can be used to approximate the lateral impact force transmitted to a barrier when the impacting vehicle is a single body. However, they cannot provide information regarding longitudinal distribution or resultant height of the lateral load, nor the impact load in the longitudinal and vertical direction. 3.2.2 Finite Element Analyses for MASH TL-4 Impact The complex nonlinear interaction that occurs during the collision between a heavy vehicle and a longitudinal barrier is difficult to analyze using simplified analysis techniques. Therefore, an explicit nonlinear FE analysis was conducted to capture the impact force generated during the collision of a MASH 10000S vehicle model into rigid barriers of different heights. The numerical simulations were performed using the commercially available FE software LS-DYNA (6). The variation and magnitude of the lateral, longitudinal, and vertical impact forces with barrier height were investigated. a) Validation of the TL- 4 Vehicle Model An SUT vehicle model was developed by the National Crash Analysis Center (NCAC) (37) and further modified by the National Transportation Research Center (NTRC) (38). The Ford F800 Series Truck was developed to meet NCHRP Report 350 (4) specifications for the 8000S test vehicle. The SUT model was modified by TTI researchers to reflect the MASH 10000S test vehicle specifications including changes in mass and ballast height. The ballast height changed from 67 in. (1.7 m) in NCHRP Report 350 to 63 in. (1.25 m) in MASH. MASH specifies a maximum allowable wheelbase and overall length for the SUT vehicle. Thus, variations in these dimensions can exist provided the maximum values are not exceeded. When the minimum recommended barrier height for MASH TL-4 was being investigated, an SUT with a wheel base and overall length on the shorter end of the available range was considered most critical from a stability standpoint. In order to validate the SUT model, the wheel base and overall length were reduced to correspond the SUT vehicle dimensions used in the tests of the 32-in. (0.81 m) tall NJ profile concrete barrier and 36-in. (0.91 m) tall SSTR. Researchers performed validation of the MASH SUT vehicle model using crash test results for the 32-in. (0.81 m) tall NJ profile concrete barrier (18). Figure 3-8 shows sequential images of the test and the FE simulation. The simulation results with the modified SUT vehicle correlate reasonably well with the test results (see Figure 3-9). More detailed information about the modifications to and validation of the MASH SUT vehicle model can be found in reference (19).

47 a) Test t=0 sec. b) Simulation t=0 sec. c) Test t=0.246 sec. d) Simulation t=0.246 sec. e) Test t=0.366 sec. f) Simulation t=0.366 sec. Figure 3-8 Sequential images for SUT test (18) and simulation.

48 b) x-acceleration (Test 476460-1b) b) x-acceleration (MASH SUT Model) c) y-acceleration (Test 476460-1b) d) y-acceleration (MASH SUT Model) e) Angular Displacement (Test 476460-1b) f) Angular Displacement (Model) Figure 3-9 Comparison of acceleration and angular displacements between test 476460-1b (18) and simulation. 0 0.05 0.1 0.15 0.2 0.25 0.3 -6 -4 -2 0 2 4 6 Time, sec x- Ac ce le ra tio n, g , s SAE-60 (Test) 50 msec Ave. (Test) 0 0.05 0.1 0.15 0.2 0.25 0.3 -6 -4 -2 0 2 4 6 Time, sec x- Ac ce le ra tio n, g , s SAE-60 (FE) 50 msec Ave. (FE) 0 0.05 0.1 0.15 0.2 0.25 0.3 -10 -5 0 5 10 Time, sec y- Ac ce le ra tio n, g , s SAE-60 (Test) 50 msec Ave. (Test) 0 0.05 0.1 0.15 0.2 0.25 0.3 -10 -5 0 5 10 Time, sec y- Ac ce le ra tio n, g , s SAE-60(FE) 50 msec Ave. (FE) 0 0.05 0.1 0.15 0.2 0.25 0.3 -20 -10 0 10 20 30 Time, sec R ol l-P itc h- Ya w A ng le , D eg re es Roll Angle (test) Pitch Angle (test) Yaw Angle (test) 0 0.05 0.1 0.15 0.2 0.25 0.3 -20 -10 0 10 20 30 Time, sec R ol l-P itc h- Ya w A ng le , D eg re es Roll Angle (FE) Pitch Angle (FE) Yaw Angle (FE)

49 b) Barrier Height Variation Analyses FE impact simulations were conducted on rigid barriers of different heights: 36 in. (914 mm), 39 in. (991 mm), 42 in. (1067 mm), and a tall rigid wall. The objective of the analyses was to estimate the magnitude, distribution, and location of the dynamic forces associated with a MASH TL-4 impact of an SUT vehicle. The selection of the heights of these barriers is in accordance with current design practice in highway application. The tall rigid wall analysis was conducted to determine the maximum impact force associated with a MASH TL-4 impact into a rigid wall. The analyses were conducted using vertical wall barriers; however, the results should be applicable to other barrier types. Figure 3-10 shows the lower bound model (36 in. (914 mm)) and the upper bound model (tall rigid wall) right before impact and at the time of maximum load. a) MASH TL-4 impact on the 36-in. (0.91 m) tall barrier (t=0 sec.) b) MASH TL-4 impact on the 36-in. (0.91 m) tall barrier (t=0.241 sec.) c) MASH TL-4 impact on the tall vertical wall (t=0 sec.) d) MASH TL-4 impact on the tall vertical wall (t=0.237 sec.) Figure 3-10 MASH TL-4 FE model for the 36-in. (0.91 m) tall barrier and tall vertical wall

50 The distribution of the impact force in the longitudinal and vertical directions was studied for each barrier. The rigid barriers were discretized into multiples segments. The impact force was output over 1 ft (304.8 mm) increments in the longitudinal direction and 6 in. (152.4 mm) increments in the vertical direction. The maximum 50-msec. average impact force in each segment was determined and distributed along each segment length. The LS-DYNA *CONTACT FORCE TRANSDUCER PENALTY (6) was used to capture the total contact forces applied during the impact. The discrete impact load in each section of the barrier was computed and the total load was estimated using Eq. (3-5). The vertical location of the impact force was determined by summing moments of the base of the barrier. Eq. (3-6) was used to determine the resultant height of the total impact load. This calculation was conducted at the time of maximum lateral impact load determined from the time history of the impact load. iimpact FF Σ= (3-5) ( ) impact ii F FZZ = (3-6) where Fimpact = maximum 50-msec. average impact load Fi= discrete impact load in each segment of the barrier Z= vertical location of Fimpact from the base of the barrier, Zi= vertical location of Fi from the base of the barrier. The magnitude and distribution of the MASH TL-4 impact forces obtained from the impact simulations on the different barriers are presented in Figure 3-11 through Figure 3-14. For the specific case of the 36-in. (0.891 m) tall barrier, the maximum 50-msec. average impact force in the lateral (Ft), longitudinal (FL) and vertical (Fv) directions are 67.2 kips (299 kN), 21.6 kips (96.1 kN), and 37.8 kips (168.2 kN), respectively. Figure 3-11(c) shows that Fv is significantly greater than the weight of the vehicle. This is due to the acceleration of the vehicle box as it rolls on top of the barrier during its redirection. A similar analysis was conducted for the other barrier heights. Figure 3-14(a) and Figure 3-14(b) compare the magnitude of Ft for the tall wall using the MASH and the NCHRP 350 test vehicle model. As shown in these figures, the magnitude of Ft increased from 76 kips (338.2 kN) for NCHRP Report 350 impact conditions to 93.3 kips (415.2 kN) for MASH impact conditions. These represent the maximum values of force that can be generated by the respective impact conditions when a full-height wall provides full-height vehicle engagement.

51 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) d) Longitudinal distribution, LL (t=0.241 sec.) e) Transverse distribution and application height He (t=0.241 sec.) Figure 3-11 Results of the MASH TL-4 impact simulation on the 36-in. (0.91 m) tall vertical wall. 0 15 30 45 60 0.0 0.1 0.2 0.3 0.4 0.5 Lo ng itu di na l Fo rc e, k ip s Time, sec. Raw Data 50-msec Ave. 0 15 30 45 60 0.0 0.1 0.2 0.3 0.4 0.5 Ve rti ca l Fo rc e, k ip s Time, sec. Raw Data 50-msec Ave. 0 5 10 15 20 25 0 5 10 15 20 25 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 6 12 18 24 30 36 0 10 20 30 40 50 Ba rri er H ei gh t, in . Ave. Force, kips/ft Resultant Location Approx. Force Dist. 67.2 kips at 25.1 in.

52 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) d) Longitudinal distribution, LL (t=0.109 sec.) e) Transverse distribution and application height He (t=0.109 sec.) Figure 3-12 Results of the MASH TL-4 impact simulation on the 39-in. (0.99 m) tall vertical wall. 0 10 20 30 40 0.0 0.1 0.2 0.3 0.4 0.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec Ave. 0 10 20 30 40 0.0 0.1 0.2 0.3 0.4 0.5 Ve rti ca l Fo rc e, k ip s Time, sec Raw Data 50-msec Ave 0 5 10 15 20 0 5 10 15 20 25 30 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 6 12 18 24 30 36 0 20 40 60 80 Ba rri er H ei gh t, in . Ave. Force, kips/ft Resultant Location Approx. Force Dist. 72.3 kips at 28.7 in.

53 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) d) Longitudinal distribution, LL (t=0.229 sec.) e) Transverse distribution and application height He (t=0.229 sec.) Figure 3-13 Results of the MASH TL-4 impact simulation on the 42-in. (1.07 m) tall vertical wall. 0 10 20 30 40 0.0 0.1 0.2 0.3 0.4 0.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec Ave 0 10 20 30 40 0.0 0.1 0.2 0.3 0.4 0.5 Ve rti ca l F or ce , k ip s Time, sec. Raw Data 50-msec Ave. 0 5 10 15 20 25 0 5 10 15 20 25 30 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 6 12 18 24 30 36 42 0 10 20 30 40 50 Ba rri er H ei gh t, in . Ave. Force, kips/ft Resultant Location Approx. Force Dist. 79.1 kips at 30.2 in.

54 a) MASH lateral impact force (Ft) b) NCHRP 350 Lateral impact force (Ft) c) MASH Longitudinal load (FL) d) MASH longitudinal distribution, LL (t=0.237 sec.) e) MASH transverse distribution and application height He (t=0.237 sec.) Figure 3-14 Results of the NCHRP 350 and MASH TL-4 impact simulation on the tall vertical wall. 0 50 100 150 200 0.0 0.1 0.2 0.3 0.4 0.5 La te ra l F or ce , k ip s Time, sec Raw Data 50-msec Ave. 0 25 50 75 0.0 0.1 0.2 0.3 0.4 0.5 Lo ng itu di na l F or ce , k ip s Time, sec Raw Data 50-msec Ave. 0 5 10 15 0 5 10 15 20 25 30 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 30 60 90 120 150 0 20 40 60 80 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Location 93.3 kips at 45.5 in.

55 A summary of the magnitude, distribution and application of the resultant MASH TL-4 impact loads for the different barriers is presented in Table 3-1. There are three forces involved: Ft is the transverse force which is applied perpendicular to the barrier otherwise referred to as the impact force, FL is the longitudinal force which is applied by friction along the direction of the barrier, and Fv is the vertical force which is applied downward on the top of the barrier. There are also three lengths associated with the results: the length LL over which the lateral load Ft is distributed, though unevenly, in the longitudinal direction, the length Lv over which the lateral load Ft is distributed, though unevenly, in the vertical direction, and the height of the resultant of the peak force He from ground level. Although the impact conditions are the same for all barriers simulated, it is noted that Ft increases as the barrier height increases as shown in Figure 3-15. This is due to the increase in relative stiffness between impacts, which is controlled by the contact area between the vehicle and the barrier (Figure 3-16). Additionally, as the height of the barrier increases, there is less vehicle roll and more mass is engaged in the impact, thereby increasing the impact load. Table 3-1 Summary of magnitude, distribution and application of the MASH TL-4 impact loads. Design Forces and Designations Barrier Height (in) 36 39 42 Tall Ft Transverse (kip) 67.2 72.3 79.1 93.3 FL Longitudinal (kip) 21.6 23.6 26.8 27.5 Fv Vertical (kip) 37.8 32.7 22 N/A LL (ft) 4 5 5 14 He (in) 25.1 28.7 30.2 45.5 N/A= not applicable Figure 3-15 Variation of impact force for different barrier heights for MASH TL-4. 0 25 50 75 100 0 25 50 75 100 125 150 175 50 m se c. A ve . F t, ki ps Barrier Height, in. MASH NCHRP Report 350 50 65 80 32 36 40 44 50 m se c. A ve . F t, ki ps Barrier Height, in.

56 a) 36-in. (0.81 m) tall barrier (front view) b) 36-in. (0.81 m) tall barrier (back view) c) 42-in. (0.81 m) tall barrier (front view) d) 42-in. (1.07 m) tall barrier (back view) Figure 3-16 Comparison of contact area between barriers for MASH TL-4 impact. The forces FL for the 36 in. (914 mm) and the 39-in. (991 mm) tall barriers are controlled by the contact of the front tire and the crushable zone during the front impact. For the 42-in. (1067 mm) tall barrier and the tall rigid wall, FL is controlled by the second impact, which is associated with the contact of the rear tandem axle and the bottom of the box of the SUT vehicle. However, in general, these loads are similar in magnitude and they are not significantly influenced by the change in height of the barrier. The force Fv is highly influenced by the barrier height. The value of Fv decreases as the barrier height increases as shown in Table 3-1. This is associated with a reduction in vertical

57 acceleration of the SUT box onto the top of the barrier due to increased lateral engagement of the box by the barrier and an associated decrease in roll. The influence of the barrier height is also evident in the longitudinal and vertical distribution of the lateral impact load. At a 36 in. (814 mm) barrier height, the box overrides the barrier and only the rear tandem axle contacts the barrier. Therefore, the length of barrier LL over which the load is distributed, though unevenly, in the longitudinal direction and the height of application of the peak force He correspond approximately to the diameter of the tire (3.5 ft (1067 mm)) and the rear axle height (21 in. (533 mm)), respectively. The analyses of the 39 in. (914 mm) and 42-in. (1067 mm) tall barriers show a slightly more distributed peak load (6 ft (1.83 m)) than the 36-in. (814 mm) tall barrier. In addition, the resultant height of the lateral impact load is greater, which increases the dynamic moment imposed on the deck or on the moment slab. The longitudinal distribution of the peak load for the tall wall is controlled by the contact area between the box of the SUT vehicle and the wall. The peak load is distributed over a distance of approximately 14 ft (4.3 m), close to the length of the SUT box. The resultant height of the impact load is also higher (45.5 in. (1.16 m)). 3.3 Impact Loads Study for Test Level 5 Impact The objective of these analyses is to estimate the magnitude, distribution, and location of the dynamic forces associated with a MASH TL-5 impact on rigid barriers. The influence of the height of the barrier on the impact load is also addressed. A MASH TL-5 impact involves a 79,300 lb. (36,000 kg) tractor-van-trailer (36000V) impacting the barrier at a speed of 50 mph (80 km/hr) at an angle of 15 degrees. The nominal IS of the impact is 447.5 kip-ft (606.3 kJ). 3.3.1 Analytical Study Articulated vehicles such as a tractor-van-trailer vehicle typically generate three distinct impacts. The first impact occurs when the front of the tractor impacts the barrier. The redirection of the tractor starts after this impact. The second impact occurs when the rear tandem axles of the tractor and the front of the trailer contact the barrier. This impact sometimes creates the largest impact force depending on the geometry of the system. The third impact occurs when the rear tandem axles of the trailer strike the barrier. This third impact is often associated with the largest impact force. This complex kinematic behavior makes the estimation of the impact load imposed by a tractor-trailer vehicle difficult to estimate using conventional analyses. In 1989, researchers at TTI measured the load associated with an 80,080 lb. (36,334 kg) tractor-van-trailer impacting a 90-in. (2.29 m) tall instrumented rigid wall at 55 mph (84.5 km/hr) and 15.3 degrees (13). The results showed that the first, second and third peak loads were 66 kips (293.6 kN), 176 kips (782.9 kN), and 220 kips (978.6 kN), respectively. The resultant heights of the second and third peak loads were 44 in. (1.12 m) and 70 in. (1.78 m), respectively. These loads were scaled down for application to a 42-in. (1.07 m) tall barrier using a procedure similar to the one described by Eq. (3-3). However, some of the assumptions used to develop Eq. (3-3) do not hold true for articulated vehicles.

58 Alternatively, Eq. (3-4) can be used with caution to approximate the load due to a specific impact. Data collected from previous TL-5 full-scale crash tests indicates that the critical impact load (maximum load/unit length) is generated by contact of the rear tandem axles of the tractor and the front of the trailer. These assumptions are based on acceleration data measured at the rear tandem axles of the tractor and the trailer. Therefore, Eq. (3-7) was applied for the impact of the central axle of the tractor-trailer which is expected to impose the critical impact load to the system. The acceleration data used in these analyzes was collected from accelerometers located at the rear tandem axles of the tractor. The mass was assumed to be the reaction mass at the central axles of the tractor-trailer. latcaimpact amF ×= (3-7) where Fimpact= estimated impact load mca= total reaction mass at the central axles of the tractor-van-trailer alat= lateral acceleration measured at the rear tandem axles of the tractor The results of this analysis, presented in Table 3-2, show a range of impact loads between 108.5 kips (488.3 kN) to 202 kips (899 kN). A relationship between impact load and barrier height is not well defined using this approach. One limitation is the difficulty of estimating the effect of the trailer on the impact load for barriers taller than 42 in. (1.07 m). For barriers taller than 42 in. (1.07 m), the front of the trailer interacts with the barrier at the same time as the rear tandem axle of the tractor. This mobilizes more mass and increases the impact load. As a result, the assumption of using the reaction mass at the central axles of the vehicle is not valid for barriers taller than 42 in. (1.07 m). Therefore, FE analyses were conducted to provide a more in-depth study of TL-5 loading. 3.3.2 Finite Element Analyses for MASH TL-5 Impact FE analyses were conducted to capture the impact loads associated with the collision between a MASH 36000V vehicle model and rigid barriers of different heights. The simulation data was used to determine the peak dynamic load in the lateral, longitudinal and vertical directions. The distribution of the lateral impact load in the longitudinal and vertical directions of the barrier was also studied. a) Validation of the TL-5 Vehicle Model An FE model of a tractor-trailer was recently released by the NTRC (39). The tractor FE model was modified from an existing model developed by the NCAC (37). The modifications included improvements to the element mesh, changes in material properties and their characterization, geometry, suspension components, connections, failure modes, and others.

59 Table 3-2 Computation of impact dynamic forces using the equation of motion Test No (Refer. No.). Test Condition. Weight (lb.), Speed (mph), Angle (deg.) Reaction Weight at the Central axles, mca (lb.) Max. 50- msec. Ave. Lateral Accel., alat (g’s) Barrier Height (in) Barrier Type Computed Force Flat=mca x alat (kips) 4348-2 (24) 80,180 52.8 15 34,030 5.70 42 CMB (NJ) 194.0 4798-13, (25) 80,180 52.1 16.5 30,010 3.1 42 CMB. 108.5 416-1 (22) 80,080 48.4 15 34,170 5.50 50 CMB and Metal Rail 188.0 230-6 (20) 79,770 49.1 15 33,760 5.94 54 Modified TX C202 Bridge Rail 200.5 911-1 (21) 80,120 51.4 15 34,050 5.54 90 Concrete Parapet 188.6 405511-2 (26) 79,286 49.8 14.5 34,239 5.90 42 Concrete Parapet 202.0 CMB= concrete median barrier The NTRC research team developed a new FE semi-trailer model. The new model is considered representative of typical trailers currently seen in service. The FE trailer model developed by the NTRC was based on a 53 ft (16.2 m), dual-tire, tandem axle 2004 Stoughton box trailer (40-41). The tractor semi-trailer FE model used in the analyses reported herein corresponds to the tractor version 10-0520 (day-cab model) and trailer version 10-0521(39-41). The FE model was validated by the NTRC research team using test TL5-CMB-2 (28) conducted by the MwRSF at the University of Nebraska. The overall geometry of the FE model was modified to meet the geometry of the tractor-trailer used in the crash test. The overall length of the tractor and the semi- trailer were 21.2 ft (6.5 m) and 48 ft (14.63 m), respectively. The tractor-trailer FE model has 583 parts with a total of 378,915 elements. The total mass of the empty tractor-trailer FE model is 28,819 lb. (23,098 kg) and it is ballasted to 79,741 lb. (36,170 kg) using concrete median barriers (Figure 3-17). The validation of the tractor-trailer model was conducted by NTRC researchers. Details of the validation results of this model can be found in references (40-41).

60 a) FE tractor model b) FE trailer model c) FE tractor-van-trailer model d) FE tractor-van-trailer and ballast model Figure 3-17 Enhanced FE tractor-trailer model developed by NTRC (40-41). b) Barrier Height Variation Analyses Four barriers were subjected to TL-5 impacts. The heights of the barriers were: 42 in. (1.07 m), 48 in. (1.22 m), 54 in. (1.37 m), and 157.5 in. (4.0 m). The barriers were rigid and selected to cover the range of heights used for previously crash tested TL-5 barriers. The tall rigid wall provided information regarding the maximum impact load associated with a TL-5 impact. Figure 3-18 shows

61 the lower bound model (42 in. (1.07 m)) and the upper bound model (tall rigid wall) right before impact and at the time of maximum load. The procedure followed to capture the impact load in the longitudinal and vertical direction was similar to the procedure described for MASH TL-4. a) Sketch of force transducer location b) MASH TL-5 impact on the 42-in. (1.07 m) barrier (t=0 sec.) c) MASH TL-5 impact on the 42-in. (1.07) barrier (t=0.832 sec.) d) MASH TL-5 impact on the tall rigid wall (t=0 sec.) e) MASH TL-5 impact the tall rigid (t=0.7 sec.) Figure 3-18 MASH TL-5 FE model for the 42-in. (1.07 m) tall barrier and the tall vertical wall. The results of the analyses on the 42-in. (1.07 m) barrier are presented in Figure 3-19. The time history of the lateral impact load indicates that the load associated with the first, second and third impact are 54.6 kips (243 kN), 123 kips (547.3 kN) and 159 kips (707.6 kN), respectively. Downstream Force Transducer Upstream Force Transducer Impact Point (IP)Vehicle Direction Force Transducer @ 2 ft

62 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) d) Longitudinal distribution, LL (t=0.254 sec.) e) Transverse distribution and application He (t=0.254 sec.) Figure 3-19 TL-5 impact force and distribution on the 42-in. (1.07 m) tall vertical barrier. 0 40 80 120 0.0 0.3 0.6 0.9 1.2 1.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec. Ave. 0 50 100 150 200 0.0 0.3 0.6 0.9 1.2 1.5 Ve rti ca l F or ce , k ip s Time, sec. Raw Data 50-msec. Ave. 0 5 10 15 20 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barriers, ft Impact Point Downstream

63 f) Longitudinal distribution, LL (t=0.832 sec.) g) Transverse distribution and application He (t=0.832 sec.) Figure 3-19 TL-5 impact force and distribution on the 42-in. (1.07 m) tall vertical barrier (Continued). Note that the load due to the second impact (123 kips (547.3 kN)) compares very well with the TL-5 design load presented in AASHTO LRFD(3). However, the controlling load is associated with the third impact, which has a magnitude of 159 kips (707.6 kN). The moment imposed by this load on the deck or barrier moment slab is 454 kips-ft (616 kN-m). This moment is similar to the moment imposed by the current AASHTO load (434 kip-ft (589 kN-m)) due to a different resultant load height. The longitudinal distribution length was selected to be 10 ft (3.05 m), which roughly corresponds to the width of the tandem axles. A similar analysis was conducted for the other barrier heights. The results of the 48 in. (1.22 m) and 54-in. (1.37 m) tall barriers and the tall rigid wall are presented in Figure 3-20 through Figure 3-22. The results for the different TL-5 barriers analyzed are summarized in Table 3-3 and Figure 3-23. The dashed lines in Figure 3-23 are not intended to denote a linear behavior of the impact load but are merely traces connecting the results corresponding from the different barrier heights to assist with visualization. As can be seen, the barrier height has a dramatic effect on the peak lateral load. Above a height of 42 in. (1.07 m), the trailer floor engages the barrier, resulting in a significant increase in force applied to the barrier. For example, the lateral load associated with the second impact on the 48-in. (1.22 m) tall barrier increases to 262 kips (1166 kN). This represents a 113% increase over the 123 kips (547.4 kN) lateral load for the 42-in. (1.07 m) tall barrier. This is due to the difference in impact of the front corner of the trailer as shown in Figure 3-24. 0 5 10 15 20 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream

64 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) d) Longitudinal distribution, LL (t=0.183 sec.) e) Transverse distribution and application height He (t=0.183 sec.) Figure 3-20 TL-5 impact force and distribution on the 48-in. (1.22 m) tall vertical barrier. 0 50 100 150 0.0 0.3 0.6 0.9 1.2 1.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec. Ave. 0 50 100 150 0.0 0.3 0.6 0.9 1.2 1.5 Ve rti ca l Fo rc e, k ip s Time, sec. Raw Data 50-msec. Ave. 0 15 30 45 60 0 10 20 30 40 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 12 24 36 48 0 100 200 300 400 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Location Ft=261.8 kips at 42.9 in.

65 f) Longitudinal distribution, LL (t=0.731 sec.) g) Transverse distribution and application height He (t=0.731 sec.) Figure 3-20 TL-5 impact force and distribution on the 48-in. (1.22 m) tall vertical barrier (Continued). a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Vertical impact force (Fv) Figure 3-21 TL-5 impact force and distribution on the 54-in. (1.37 m) tall vertical barrier. 0 5 10 15 20 0 20 40 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 12 24 36 48 0 100 200 300 400 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Location Ft=232.8 kips at 39.7 in. 0 50 100 150 0.0 0.3 0.6 0.9 1.2 1.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec. Ave. 0 50 100 150 0.0 0.3 0.6 0.9 1.2 1.5 Ve rti ca l Fo rc e, k ip s Time, sec. Raw Data 50-msec. Ave.

66 d) Longitudinal distribution, LL (t=0.183 sec.) e) Transverse distribution and application height He (t=0.183 sec.) f) Longitudinal distribution, LL (t=0.183 sec.) g) Transverse distribution and application height He (t=0.183 sec.) Figure 3-21 TL-5 impact force and distribution on the 54-in. (1.37 m) tall vertical barrier (Continued). 0 20 40 60 80 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 10 20 30 40 50 60 0 50 100 150 200 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Force Ft=263.5 kips at 46.6 in. 0 5 10 15 20 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 10 20 30 40 50 60 0 100 200 300 400 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Force Ft=295.5 kips at 48 in.

67 a) Lateral impact force (Ft) b) Longitudinal impact force (FL) c) Longitudinal distribution, LL (t=0.191 sec.) d) Transverse distribution and application height He (t=0.191 sec.) Figure 3-22 TL-5 impact force and distribution on the tall vertical barrier. 0 50 100 150 0.0 0.3 0.6 0.9 1.2 1.5 Lo ng itu di na l F or ce , k ip s Time, sec. Raw Data 50-msec. Ave. 0 20 40 60 80 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 50 100 150 0 100 200 300 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Location Ft=270.4 kips at 51.7 in.

68 e) Longitudinal distribution, LL (t=0.70 sec.) f) Transverse distribution and application height He (t=0.70 sec.) Figure 3-22 TL-5 impact force and distribution on the tall vertical barrier (Continued). Table 3-3 Summary of magnitudes, distributions and applications of dynamic loads for MASH TL-5 impact. Design Forces and Designations Barrier Height (in) 42 48 54 Tall Ft Transverse (kips) (First Impact) 54.6 51.7 53.8 53.7 Ft Transverse (kips) (Second Impact) 123 261.8 263.5 270.4 Ft Transverse (kips) (Third Impact) 159 232.8 295.5 316.6 FL Longitudinal (kips) 73.5 74.6 77.2 72.6 Fv Vertical (kips) 160 108 62.8 N/A LL (ft) (Second Impact) 10 10 10 10 He (in) 34.3 42.9 46.6 51.7 N/A= not applicable 0 5 10 15 20 0 10 20 30 40 50 60 Av e. F or ce , k ip s/f t Distance along the barrier, ft Impact Point Downstream 0 50 100 150 0 100 200 300 Ba rri er H ei gh t, in . Ave. Force, kips/ft Ave. Load/ft Resultant Location Ft=316.6 kips at 78 in.

69 Figure 3-23 Variation of impact force for different barrier heights for MASH TL-5. a) 42-in. (1.07 m) tall barrier (front view) b) 42-in. (1.07 m) tall barrier (back view) c) 48-in. (1.22 m) tall barrier (front view) d) 48-in. (1.22 m) tall barrier (back view) Figure 3-24 Comparison of contact area between barriers for MASH TL-5 impact. 0 100 200 300 400 0 40 80 120 160 50 m se c. Av e. F t , k ip s Barrier Height, in. First Impact Second Impact Third Impact 0 100 200 300 36 42 48 54 60 50 A ve . . F t, k ip s Barrier Height, in.

70 The peak lateral loads associated with the taller barriers are greater than the load measured in the instrumented wall tests conducted by TTI researchers in the 1980’s. The primary reason for this is the difference in the ballast. Many of the early tests conducted with tractor-van-trailers used sand bags and hay bales for ballast. Because the ballast was not rigidly secured to the floor of the trailer, it was able to shift during impact resulting in lower inertia forces on the barrier. While these are still acceptable types of ballast, MASH states that “Ballast should be firmly secured to prevent movement during and after the test” (5). This results in higher impact loads. As shown in Table 3-3, the dynamic load due to the first impact is similar for all barriers. FL, which is controlled by the frictional contact between the tires and the barrier, is also similar in all cases. Similar to the results of the TL-4 study, Fv decreases as the barrier height increases. This is due to a reduction in roll of the tractor-trailer as the barrier height increases. 3.4 Recommendation of Design Impact Loads in Traffic Barriers for MASH TL-4 and TL-5 Impact The FE analyses results were used to define the recommended design impact loads for MASH TL- 4 and TL-5 impacts. The distribution of the critical impact load and the height of resultant load application are also recommended. The recommendations of LL were analyzed by considering the maximum load/unit length as well as the total length of load application. These two criteria were used to study the practical effect on the final design of the barrier (structural and practicality) and used to select the final recommendation of LL. The information is presented in Table 3-4. Table 3-4 Recommended design loads for TL-4 and TL-5 impact Design Forces and Designations TL-4-1 TL-4-2 TL-5-1 TL-5-2 Rail Height, H (in) 36 >36 42 >42 Ft Transverse (kips) 70 80 160 260 FL Longitudinal (kips) 22 27 75 75 Fv Vertical (kips) 38 33 160 108 LL (ft) 4 5 10 10 Lv (in) 18 18 40 40 He (in)(1) 25 30 34 43(2) (1) Vertical height of the resultant load. (2) If barriers taller than 54 in are used, use He=52 in Ft is the transverse force which is applied perpendicular to the barrier otherwise referred to as the impact force, FL is the longitudinal force which is applied by friction along the direction of the barrier, Fv is the vertical force which is applied downward on the top of the barrier, LL is the length over which the load is distributed, though unevenly, in the longitudinal direction, Lv is the length over which the load is distributed, though unevenly, in the vertical direction, He is the height of application of the peak force. The recommendations for the MASH TL-4 impacts accounts for changes in impact condition, vehicle properties, and barrier height. The final recommendation for MASH TL-4 has been split into two categories (TL-4-1 and TL-4-2) to recognize the effect of barrier height on the

71 magnitude of the impact load. The recommendations for MASH TL-5 account for the highest of the three impact loads imposed by the tractor-trailer on a barrier of a given height. Also, the final recommendation for MASH TL-5 has been divided into two categories (TL-5-1 and TL-5-2) to recognize the effect of barrier height on the magnitude of the impact load.