EVALUATION OF TENNESSEE BRIDGE RAIL DESIGNS

Transcription

1 TEXAS TRANSPORTATON NSTTUTE EVALUATON OF TENNESSEE BRDGE RAL DESGNS by Roger P. Bligh Engineering Research Associate King K. Mak Research Engineer and T. J. Hirsch Research Engineer Research Report RF Research Study No. RF 7199 "Crash Testing and Evaluation Tennessee Bridge Rail and Guardrail Designs" Sponsored by Tennessee Department of Transportation Nashville, Tennessee May 1994 Texas Transportation nstitute THE TEXAS A&M UNVERSTY SYSTEM COLLEGE STATON, TEXAS

2 APPROXMATE CONVERSONS TO S UNTS APPROXMATE CONVERSONS FROM S UNTS Symbol When You Know Multiply By To Find Symbol Symbol When You Know Multiply By To Find Symbol LENGTH LENGTH in inches 25.4 millimeters mm mm millimeters inches in ft feet meters m m meters 3.28 feet ft yd yards meters m m meters 1.09 yards yd mi miles 1.61 kilometers km km kilometers miles mi AREA AREA in 2 square inches square millimeters mm 2 mm2 square millimeters square inches in 2 ft2 square feet square meters m 2 m2 square meters square feet ft2 yrj1 square yards square meters m 2 m2 square meters square yards ac ac acres hectares ha ha hectares 2.47 acres mi2 mi 2 square miles 2.59 square kilometers km 2 km2 square kilometers square miles VOLUME VOLUME f oz fluidounces milliliters ml ml milliliters fluid ounces f oz gal gallons liters liters gallons gal ft3 cubic leet cubic meters m 3 m 3 cubic meters cubic feet ft3... ydl cubic yards cubic meters m 3 m 3 cubic meters cubic yards ydl NOTE: Volumes greater than 1000 shall be shown in m 3 MASS oz ounces grams g g grams ounces oz b pounds kilograms' kg kg kilograms pounds b T short tons (2000 b) megagram s Mg Mg megagrams short tons (2000 b) T MASS TEMPERATURE (exact) TEMPERATURE (exact) OF Fahrenheit 5(F-32)/9 Celcius C C Celcius 1.8C +32 Fahrenheit OF temperature or (F-32)/l.8 temperature temperature temperature LLUMNATON LLUMNATON fc foot-candles lux x lux foot-candles fc f foot-lamberts candelalm 2 cdlm 2 cdlm2 candelalm foot-lamberts f FORCE and PRESSURE or STRESS FORCE and PRESSURE or STRESS bl poundforce 4.45 newtons N N newtons poundforee bf psi poundforee per 6.89 kilopascals kpa kpa kilo pascals poundforee per psi square inch square inch S is the symbol for the nternational System of Units. Appropriate (Revised August 1992) rounding should be made to comply with Section 4 of ASTM E380.

3 DSCLAMER The contents of this report reflect the views of the authors who are solely responsible for the opinions, findings, and conclusions presented herein. The contents do not necessarily reflect the official views or policies of the Tennessee Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification or regulation. KEYWORDS Bridge Railings, Traffic Barriers, Highway Safety, Crash Test, Safety Shape ACKNOWLEDGMENTS This study was sponsored by the Tennessee Department of Transportation, in cooperation with the Federal Highway Administration. Mr. Peter Falkenberg, Certification Acceptance Engineer, was the Project Engineer for the Tennessee Department of Transportation and his guidance and support are deeply appreciated. The study was overseen by a steering committee consisting ofmr. Edward P. Wasserman, Engineering Director, Structures Division, Mr. Clifford F. Stewart, Design Engineering Supervisor, Special Design Office, and Mr. Falkenberg of the Tennessee Department of Transportation and Mr. Clarence E. Bennett, Bridge and Safety Management Engineer of the Federal Highway Administration- Tennessee Division. The authors are very appreciative of their comments and suggestions. 111

4 EXECUTVE SUMMARY The objective of this study was to analyze and evaluate the impact performance of various bridge rail, guardrail, transition, and end treatment designs currently in use by the Tennessee Department of Transportation (TDOT). The project was divided into two phases. Phase involved the evaluation of the various bridge rail, guardrail, transition, and end treatment designs through theoretical analyses and computer simulation. Where appropriate, modifications to existing designs were recommended. Phase involved full-scale crash testing of the existing and modified designs which were selected for further evaluation. This is Volume of a three volume final report summarizing the results of the analytical evaluation and full-scale crash testing of Tennessee bridge rail designs. Volume of the final report summarizes the crash testing and evaluation of TDOT bridge rail to guardrail transition designs. Volume presents the design, development, crash testing, and evaluation of a slottedrail end terminal design for use with W -beam guardrails on highways with speed limits of 45 mph or less. This report presents a brief analytical analysis and safety evaluation of six (6) Tennessee bridge railings. Metal railings are analyzed using plastic analysis and plastic design procedures. Concrete railings are analyzed by the yield line method and ultimate strength design procedures. Based on the results of these analyses, recommendations were made as to whether certain bridge rails should be crash tested to certain test levels based on strength and geometric considerations. The results of crash tests on an open concrete beam-and-post railing and a combination rail with a concrete safety-shaped parapet and structural tubing are presented. The open concrete beam-and-post bridge rail was found to have adequate strength to contain and redirect a 4,500-lb (2,04-kg) pickup truck. However, the small car severity test failed to meet NCHRP Report 230 criteria when the wheel of the vehicle snagged on the interior face of a concrete post and was pushed back into the occupant compartment. This adverse behavior can be attributed to the 5-in. (38.1-cm) opening between the bridge deck and the lower edge of the concrete beam which allowed the wheel assembly to go underneath the concrete beam and snag on the concrete post. Candidate design modifications to remedy this problem are included in the report for both new construction and retrofit applications. V

5 The concrete safety-shaped parapet with structural tubing bridge rail was only subjected to a small car severity test since it was determined that the structural capacity of the concrete parapet was well in excess of that required for test level (TL) 3 impacts. The concern was over the propensity for small cars to roll over due to the increase height of the lower sloped surface of the concrete parapet. The trajectory of the test vehicle was found to be stable during and after impact, and all occupant risk criteria for the test were below the maximum allowable limits set forth in NCHRP Report 230. v

6 TABLE OF CONTENTS METRC CONVERSON TABLE DSCLAMER KEY WORDS... iii ACKNOWLEDGMENTS EXECUTVE SUMMARY... V. NTRODUCTON STRUCTURAL STRENGTH EVALUATON OF BRDGE RALNGS Metal Rail Concrete Parapet Rail Combination Concrete Parapet and Metal Rail Distribution Length of Vehicle mpact Load BRDGE RAL REQUREMENTS Strength Requirements Desired Geometric Requirements V. ANALYTCAL EVALUATON OF TENNESSEE BRDGE RALNGS Bridge Railing TBR-A (K , Flush Mounted) Bridge Railing TBR-B (K A, Curb Mounted) Bridge Railing TBR-2A (K A) Bridge Railing TBR-2B (K ) Bridge Railing TBR-3 (M ) Bridge Railing TBR-4 (M ) V. CRASH TEST PROCEDURES Description of Test nstallations Bridge Railing TBR-A: Open Concrete Beam-and-Post Bridge Rail (K ) Bridge Railing TBR-2A: Concrete Parapet with Structural Tubing Bridge Rail (K A) Description of Crash Test Procedures Electronic nstrumentation and Data Processing Photographic nstrumentation and Data Processing Test Vehicle Propulsion and Guidance V. FULL-SCALE CRASH TEST RESULTS Open Concrete Beam-and-Post Bridge Rail vi

7 TABLE OF CONTENTS (Continued) Test 1 (Test No ) Test 2 (Test No ) Summary Concrete Parapet with Structural Tubing Bridge Rail (Test No ) V. CONCLUSONS AND RECOMMENDATONS Summary and Conclusions Bridge Railing TBR-A (K ) Bridge Railing TBR-B (K A) Bridge Railing TBR-2A (K A) Bridge Railing TBR-2B (K ) Bridge Railing TBR-3 (M ) Bridge Railing TBR-4 (M ) Recommendations Bridge Railing TBR-1A (K ) Bridge Railing TBR-2A (K A) Bridge Railing TBR-2B (K ) REFERENCES APPENDX A - SUMMARY OF ANALYTCAL CALCULATONS..., A- APPENDX B - NORTH CAROLNA BR-38 TEST AND EVALUATON... B-1 APPENDX C - STANDARD GUARDRAL TESTS AND EVALUATON C-l APPENDX D - SEQUENTAL PHOTOGRAPHS... D-l APPENDX E - ACCELEROMETER TRACES E-l APPENDX F - ANGULAR DSPLACEMENTS... F-1 Vll

8 Figure LST OF FGURES 1 Three Types of Bridge Rails 4 2 Possible Failure Modes for Beam and Post Barriers Yield Line Analysis of Concrete Parapet Wall Yield Line Analysis of Open Concrete Wall or Parapet 8 5 Combination Concrete Wall and Metal Rail Evaluation- mpact at Mid-span of Rail Combination Concrete Wall and Metal Rail Evaluation-mpact at Post Distribution of mpact Load in Collision with Longitudinal Traffic Rail 13 8 Bridge Railings Design Forces, Vertical Location and Horizontal Distribution Length Definition of He Effective Height to Prevent Rollover and Y Height of Resultant Rail Strength Types of Bridge Rails and Critical Dimensions Desired Relationship Between Vehicle-Rail Contact Dimension and Post Setback Distance Potential for Wheel, Bumper or Hood mpacting Post TBR-A TBR-B TBR-2A TBR-2A Geometry and New Jersey CMB TBR-2B TBR TBR Open concrete beam and post bridge rail V111

9 LST OF FGURES (Continued) 21 Concrete parapet with structural tubing bridge rail Tennessee post and beam bridge rail before test Vehicle before test Test vehicle properties (7199-1) Vehicle/rail geometric for test Tennessee post and beam bridge rail after test Vehicle after test Summary of results for test Vehicle before test Test vehicle properties (test ) Vehicle/rail geometric for test Tennessee post & beam bridge rail (after test ) Vehicle after test Summary of results for test Channel retrofit for open beam and post bridge rail Tennessee safety-shape bridge rail before test Vehicle before test Test vehicle properties (7199-6) Vehicle/rail geometric for test Tennessee safety-shape bridge rail after test Vehicle after test : Summary of results for test X

10 LST OF FGURES (Continued) D 1 Sequential photographs for test (overhead and frontal views)... D-2 D2 Sequential photographs for test (behind rail view) D-4 D3 Sequential photographs for test (overhead and frontal views)... D-5 D4 Sequential photographs for test (behind rail view) D-7 D5 Sequential photographs for test (overhead and frontal views)... D-8 D6 Sequential photographs for test (perpendicular view) D-lO E1 Longitudinal accelerometer trace of test E-2 E2 Lateral accelerometer trace of test E-3 E3 Vertical accelerometer trace of test E-4 E4 Longitudinal accelerometer trace (test )... E-5 E5 Lateral accelerometer trace (test )... E-6 E6 Vertical accelerometer trace (test ) E-7 E7 Longitudinal accelerometer trace for test E-8 E8 Lateral accelerometer trace for test E-9 E9 Vertical accelerometer trace for test E-lO Fl Vehicle angular displacements (test ) F-2 F2 Vehicle angular displacements (test ) F-3 F3 Vehicle angular displacements (test ) F-4 x

11 LST OF TABLES 1 Bridge Railing Design Loads, Horizontal Length of Distribution, and Vertical Location for Various Test Levels Test Levels, Vehicles, Weights, Angles and Speeds Xl

12 . NTRODUCTON The objective of this study was to analyze and evaluate the impact performance of various bridge rail, guardrail, transition, and end treatment designs currently in use by the Tennessee Department of Transportation (TDOT). The project was divided into two phases. Phase involved the evaluation of the various bridge rail, guardrail, transition, and end treatment designs through theoretical analyses and computer simulation. Where appropriate, modifications to existing designs were recommended. Phase involved full-scale crash testing of existing and modified designs which were selected for further evaluation. This is Volume of a three volume final report summarizing the results of the analytical evaluation and full-scale crash testing of Tennessee bridge rail designs. This report presents a brief analytical analysis and safety evaluation of six (6) selected Tennessee bridge railings: 1. TBR-A. "STANDARD CONCRETE BRDGE RAL 1985," TDOT standard drawing K TBR-B. "STANDARD CONCRETE BRDGE RAL 1965," TDOT standard drawing K A. (Note: Same as TBR-A, but mounted on 9 in. high by 23 in. wide curb) 3. TBR-2A. "BRDGE RALNG CONCRETE PARAPET WTH STRUCTURAL TUBNG 1972," TDOT standard drawing K A. This concrete parapet has a sloping face on the traffic side similar to, but different from, the New Jersey safety shape. 4. TBR-2B. "BRDGE RALNG - CONCRETE PARAPET WTH STRUCTURAL TUBNG 1967," TDOT standard drawing K This concrete parapet is a vertical wall with the same structural tubing as TBR-2A on top. 5. TBR-3. "BRDGE RAL LESS THAN 750 CURRENT ADT AND 30 MLE DESGN SPEED 1982," TDOT standard drawing M This steel bridge rail can be mounted on a 6 in. to 8 in. high curb or flush with the bridge deck. 6. TBR-4. "STANDARD GUARDRAL ATTACHMENT TO BRDGES 1983," TDOT standard drawing M (Note: This is a guardrail retrofit to "Standard Concrete Handrail Rail Type--1947, 1948, 1953, 1960, 1962," TDOT standard drawings E , E , C-1O-54, C-l1-92, K-15-59, and H ) This retrofit rail can be 1

13 mounted where all posts are new steel posts or where every other post is an existing concrete post. Section describes the procedures for evaluating the structural strength of bridge railings. The required strength and desired geometric requirements of traffic railings are presented in Section. Results of the analytical evaluation of the bridge railings are summarized in Section V. Based on the results of these analyses, recommendations were made as to whether certain bridge rails should be crash tested to certain test levels based on strength and geometric considerations. Two (2) bridge rail systems were selected by TDOT for further evaluation through full-scale crash testing: 1. TBR-A. "STANDARD CONCRETE BRDGE RAL 1985," TDOT standard drawing K This is an open concrete beam-and-post bridge rail. 2. TBR-2A. "BRDGE RALNG CONCRETE PARAPET WTH STRUCTURAL TUBNG 1972," TDOT standard drawing K A. This is a combination rail with a safety-shaped concrete parapet and structural tubing. Section V of this report presents a description of the test installations along with the crash test and data analysis procedures followed. Section V describes the crash test results, and a summary of findings, conclusions, and recommendations are presented in Section V. 2

14 . STRUCTURAL STRENGTH EVALUATON OF BRDGE RALNGS The six bridge railings under evaluation can be grouped into three basic types as shown in Figure 1. These three basic types of bridge railings are as follows: 1. Metal Rail which consists of five basic structural elements that work together to produce the ultimate strength of the system -- rail, post, base plate, anchor bolts, and bridge deck, 2. Concrete Parapet which consists of a reinforced concrete wall mounted on the concrete bridge deck, and 3. Combination Concrete Parapet and Metal Rail which consists of six basic structural elements that work together to produce the ultimate strength of the system -- rail, post, base plate, anchor bolts, concrete wall, and concrete bridge deck. The procedures used for evaluating the structural strength of these bridge railing types S described below. Metal Rail The method of evaluating the ultimate strength of metal rails utilizes plastic analysis and plastic design as illustrated in Figure 2. The maximum stress which can be developed in the rail and post is assumed to be the yield strength of the metal. F or the anchor bolts and base plates, however, a modified plastic analysis method is used whereby the ultimate strength of the metal is used instead of the yield strength. n order to determine the total ultimate vehicle impact load (w) a bridge rail system can resist, several possible failure modes need to be considered. These failure modes are depicted in Figure 2. Failure modes similar to these have been observed in actual crash tests. When a "weak beam-strong post" system is used, single or two span failure modes have been observed. When a "strong beam-strong post" system is used, three span failure modes have been observed. When a "strong beam-weak post" system is used, four or more span failure modes are likely. All possibilities should therefore be considered in the analysis. The equations given in Figure 2 for the ultimate horizontal load capacity (w) satisfy all equations of static equilibrium. A simpler equation, which does not quite satisfy all equations of static equilibrium for forces and moments in the beam, is as follows: 3

15 _-- RAL - W SECTON POST ):!:J ~::-- BASE PLATE SLAB (A) METAL RAL - ancho r bolts, base plate, post, rail and bridge dec k RENFORCED CONC. WALL "" ': BRDGE CONC. SLAB (8) CONCRETE PARAPET- concrete wall and slab. METAL RAL PLATES ~-BASE PL..,. CONC. WALL ':'l ; :.j j..." CONC. SLAB.,-,.- ~. '.' 4.,',...,' (C) COMBNATON CONCRETE PARAPET AND METAL RAL- metal rail, post, base plate, anchor bolt. concrete wall and slab. Figure 1. Three Types of Bridge Rails. 4

16 wj = 8*Mp L-Jj2 T (A) Single Span T Failure Mode Mt ill ~1fttff~ l-~~,mp wj 8*Mp + P = 2L-Jj2 P ' L l Pp Mp Mp Mp Mp (8) Two Span Failure Mode ~ L ~ Pp Pp Mp Mp Mp (c) Three Span Failure Mode L - post spacing or single span Mp = plastic or yield line capacity of rail Pp = ultimate load capacity of a single post wl total ultimate vehicle impact load (horizontal) 1 length of distributed vehicle impact load PLAN VEW Figure 2. Possible Failure Modes for Beam and Post Barriers. 5

17 wi 8M p NL-i/2 + (N-l)P p (1) where N = number of spans in the failure mechanism. Equation (1) was used to analyze the metal rail on top of the combination concrete wall and structural tube bridge railing systems. Concrete Parapet Rail The method of evaluating the ultimate strength of the concrete parapet bridge railing system is the yield line method. The results of such an analysis on a typical concrete parapet are summarized in Figure 3. n this analysis, the total ultimate load (w) is applied at the top of the concrete wall. This is the most critical location and also yields the maximum effective height (H or Y) of loading. The critical length (L) is the length which gives the minimum ultimate total load (w). t is interesting to note that the ultimate load capacity of the concrete parapet is a function of the moment capacity of the beam at the top of the wall (M b ), the moment capacity of the wall (M w )' and the cantilever moment capacity of the wall at the bridge deck (Me). f the bridge deck. is weak, it may control or limit the cantilever moment capacity (Me). However, as indicated by these equations, the total load capacity of the wall can be increased by strengthening the beam and/or wall. This will increase the length (L) and bring more bridge deck into play. A second type of concrete wall or parapet is shown in Figure 4. This wall has openings or gaps of length (G) spaced at regular intervals. This type of wall can also be evaluated by the yield line method as illustrated in of Figure 4. The analysis presented here does not consider impacts near open joints (expansion or contraction) in the concrete walls. Such joints, which frequently occur, can be evaluated by modifying the yield line analysis. The ultimate strength of the concrete parapet at these locations would be approximately one half the values computed here. To reduce the effect of joints, it is recommended that their use be minimized and that continuous reinforcing steel shear keys be used. 6

18 L <-t Mb H Mb (~f + H - L - wl - height of wall, ft critical length of wall failure. ft total ultimate load capacity of wall. kips Mb = ultimate moment capacity of beam at top of wall, kip-ft. Mw = ultimate moment capacity of wall per ft of wall height. kip-ft/ft Mc = ultimate moment capacity of wall cantilever up from bridge deck per ft of length of wall, kip-ft/ft 1 = length of distributed impact load. ft Figure 3. Yield Line Analysis of Concrete Parapet Wall. 7

19 " :. r (w.i)ult. = (L -.L12) + McL(L-G) H ( L - ' /2) L = i i = length of distributed impact load, ft wi. = total ultimate distributed load capacity of wall H L = hei ght of wall, ft critical length of wall failure, ft Mb = ultimate moment capacity of beam at top of wall, kip-ft Me = ultimate moment capacity of wall cantilever up from bridge deck per length of wall, kip-ft 1ft G : enoth of gop or wall opening, ft Figure 4. Yield Line Analysis of Open Concrete Wall or Parapet. 8

20 Combination Concrete Parapet and Metal Rail n order to determine the impact resistance capability of a combination bridge rail, the strength of each of its components is determined as described above. The bending strength of the rail is determined over one span (PJ and two spans (PD. The strength of the post (Pp) on top of the wall is also determined. This may be controlled by the anchor bolts or post section modulus. n addition, the strength of the concrete parapet or wall (Pw) is determined as previously described (Figure 3). Figures 5 and 6 illustrate two possible critical impact points for a combination bridge railing. Each of these impact conditions should be evaluated in order to determine the critical or minimum strength of the railing. Figure 5 depicts a scenario in which the vehicle impact is at midspan of the metal rail. The bending strength of the rail (PJ and the maximum strength of the concrete wall (Pw) act together to yield the maximum resultant strength (R), as shown. The effective height (Y) of this resultant (R) is somewhere between the height of the rail (hj and the height of the wall (hw). Figure 6 illustrates a scenario in which the vehicle impacts the railing at a post. Although this impact condition usually yields a smaller resultant force (R), both cases should be investigated to assure the critical load has been identified. As shown in Figure 6, there are two resisting forces at the centroid of the metal rail, the post strength (P p) and the rail bending strength over two spans (PD. At the top of the concrete wall, a reduced wall capacity (P~) will result since the post load is also partially transmitted to the bridge deck at this point. Consequently, the maximum resultant strength (R) is the sum of the post capacity (Pp), the rail strength (PD and the reduced wall strength (P~). The effective height of this resultant is as - shown in Figure 6. t should also be recognized that a maximum effective height (Y) equal to the centroid rail height (hj could be obtained but at a reduced resultant strength (R) equal to the post capacity (Pp) and rail capacity (PD only. Again, the analysis presented here does not consider impacts near open joints in the concrete wall or parapet. The metal rail will help distribute load across such joints. t is recommended that expansion and contraction joints be minimized and that continuous reinforcing steel or shear keys be used to distribute load across such joints. 9

21 R a i ",,,=, --tf--h--~ R Post---- Woll--- --'T y.,.-.";... : Deck _111 B a PLAN VEW Max. Resultant R a P R + P w Effective height Y PR hr + Pw hw R P R = ultimate capacity of rail over one span p = w ultimate capacity of wall (Figure 3) h w = height of wall hr = hei ght, of ra i Another possibility is Max. Effective Height Y hr M in. Resultant R P R Figure 5. Combination Concrete Wall and Metal Rail Evaluation-mpact at Mid-span of Rail. 10

22 Deck 1S 'l a. -n Pw P LAN VEW Max. Result. R.. ~ + ~+ P~ Effective HeiQht Y " Pp ha + Pn hr + ~~ hw R Pp= ultimate capacity of post ~= ultimate capacity of rail over two spans \V" ultimate capacity of wall (Figure 3 P~= reduced capacity of wall Pw hw - Pp hr since post load must be hw resisted by wall too. M in. Resultant R Pp + p~ Max. Effective Height Y.. hr Figure 6. Combination Concrete Wall and Metal Rail Evaluation-mpact 11

23 Distribution Length of Vehicle mpact Load When a single unit vehicle impacts a longitudinal traffic railing at an angle, there is typically first a front and then a rear impact, as shown in Figure 7. Crash tests have shown that the crush length of sheet metal is about 4 to 5 ft (1.22 to 1.52 m) for impacts involving automobiles and pickup trucks. For purposes of this analysis, the load is assumed to be distributed over a length of 5 ft (1.52 m) for a 60 mph, 25 deg impact with a car. For larger single unit trucks, only the tire may contact the rail. The diameter of the truck tire is about 42 in. (1.07 m). 12

24 w = distributed load, blft or N/m J. = eng th of distributed load, ft or m wi= total impact load, b or N Figure 7. Distribution of mpact Load in Collision with Longitudinal Traffic Rail. 13

25 . BRDGE RAL REQUREMENTS Strength Requirements n order to compare the computed strength of a railing with the impact loads it must resist, it was first necessary to establish the design impact conditions. There are two documents that provide guidance in the selection of an appropriate test matrix for bridge rails: National Cooperative Highway Research Program (NCHRP) Report 230CD and the 1989 AASHTO Guide Specifications for Bridge RailingsW. These two test documents differ considerably in terms of both vehicle type and impact conditions. Service Level 1 ofnchrp Report 230 recommends the use of a 4,500 lb passenger car impacting at 60 mph and 15 deg as a strength test, and an 1,800 b passenger car impacting at 60 mph and 15 deg for geometric evaluation and assessment of occupant risk. Performance Level 1 of the 1989 AASHTO Guide Specifications for Bridge Railings recommends testing with a 5,400 lb pickup truck at 45 mph and 20 deg and an 1,800 lb passenger car at 50 mph and 20 deg for the strength and severity tests, respectively. Service Level 2 of NCHRP Report 230 recommends a strength test with a 4,500 lb passenger car impacting at 60 mph and 25 deg. Performance Level 2 of the 1989 AASHTO Guide Specifications for Bridge Railings requires a strength test with an 18,000 lb single-unit truck impacting the railing at 50 mph and 15 deg. Further complicating the selection of a test matrix was the fact that, at the time this analysis was performed, both of these documents were in the process of being revised under NCHRP Project 22-7, "Update of Roadside Safety Hardware Crash Test Specifications," and NCHRP Project 12-33, "Development of a Comprehensive Bridge Specification and Commentary. " n an effort to use the most current information available, the test levels shown in Table 1 were utilized. Although tentative in nature and subject to future revision, these test levels were under serious consideration by both of the projects mentioned above at the time this analysis was performed. Table 2 presents impact design forces for the test levels listed in Table 1. These impact forces were determined from actual measurements of vehicle impact forces(11) and, at the time of this study, were being considered by Project

26 TABLE 1. Test Levels, Vehicles, Weights, Angles and Speeds. Test Vehicle Description and mpact Angles Test Level (TL) and Test Speed, mph Vehicle Description W (kips) e (deg) TL-l TL-2 TL-3 TL-4 TL-5 TL-6 Small Automobile Pickup Truck or * 60* 60* Sports Wagon Truck Medium Single Unit Truck Van Type Tractor-Trailer Tank Type Tractor-Trailer *These tests should be conducted unless it can be conclusively shown that these tests would be no more severe than the small automobile test and the truck test. TABLE 2. Bridge Railing Design Loads, Horizontal Length of Distribution, and Vertical Location for Various Test Levels (Reference Table 1 and Figure 8). Design Loads FH,FL,FV Test Level Length of Distribution LH, LL, LV Vertical Location TL-l TL-2 TL-3 TL-4 TL-5 TL-6 H andh FH Horizontal (kips) FL Longitudinal (kips) FV Vertical (kips) LH and LL (ft) LV (ft) He (min.) (in.) H(min.) Height of Railing (in.)

27 U sing an appropriate test level, the required impact load (FH = horizontal impact force) can be determined from Table 2 and compared with the calculated strength (w) of a given bridge rail. Figures 8 and 9 illustrate how the design loads, length of distribution, and vertical location, as given in Table 2, are applied to the bridge railing. The longitudinal force (FL) is equal to the coefficient of friction (assumed to be 0.33) times the horizontal force (FH) and is distributed over the same length and applied at the same time as FH. The vertical force (FV) is the weight of the vehicle which may lay on top of the traffic railing and is distributed over the assumed length of the vehicle (LV). This typically occurs after the impact forces (FH and FL) and thus is not applied at the same time. The impact force height (He) was determined from the analysis shown in Figure 9. f the resultant strength of the traffic rail (R) has a height (Y) equal to or higher than the height of the impact force (He)' the vehicle should remain stable and not roll over the rail. Desired Geometric Requirements Crash tests of a large number of bridge railings have indicated certain geometric deficiencies that should be avoided, if possible. Figure 10 shows several geometric dimensions which are known to have a significant effect on the vehicle-rail impact interaction such as height (H), rail-vehicle contact dimension (A or LA), vertical clear opening (C) between horizontal rails or deck, and post setback distance (S). Figure 11 shows the recommended relationship between post set back dimension (S) and the ratio of the rail contact dimension (A) to rail height (H). f the vehicle-rail contact dimension (A) is small, the rail may cut deeply into the vehicle sheet metal and result in snagging of the vehicle on a post if an adequate set back distance is not provided. Figure 12 shows the recommended relationship between the vertical clear opening (C) and the post set back dimension (S). 16

28 R = Resultant Rail Strength ~ Y = Height of Resultant Rail Strength > He FH FV = Weight of vehicle, vertical force not simultaneous with FL and FH FH = Horizontal force of the vehicle. FL = O.33*FH "Friction Force" Applied simultaneously with FH at the same height and over the same length. H = Height of Railing LV = Length of Vehicle or Trailer ht.rjujrr-fai. 4rd. LH and LL = Length over which th71longitudinal load is distributed. R =; r--, R1 y = f R1 *Y1, 1 + R Rn + R2*Y Rn*Yn :t R1 + R Rn,., f H t R y ~ R1 =- R2... He Figure 8. Bridge Railings Design Forces, Vertical Location, and Horizontal Distribution Length.

29 rr Y1 -r- R2 t- R Y or He 8/2. - -_c 1,_GJ--,- FH i W G ANALYSS OF VEHCLE FH * (G - He) = W * (8/2) WHERE: He > G - (W /FH) * (8/2) He = The effective height of rail to prevent roll over. W = Weight of Vehicle G = :eight of Vehicle Center of Gravity B:; Wdth of Vehicle ANALYSS OF RAL R = Resultant Rail Strength Y = Height of Resultant Rail strength R == ~Rj NOTE: R > FH Y > He Figure 9. Qefinition of He Effective Height to Prevent Rollover and Y Height of Resultant Rail Strength. 18

30 NOTE: 1\ ur 1\ 1 +1\2 or 2. 1\ = fuil contact dimension with vehicle y -~_\- H (See Figure J-~2) S = (See Figure ) +' QJ Q. 0 L 0 (L L a 0 3: T H and A Post --!-r Beam or Rail _L H t A3 sj f+ Curb or Parapet r CONCRETE PARAPET CONCRETE RAL COMBNATON - AND METAL RAL CONCRETE A2 -L~r =CC H A1 c On Top Moun METAL OR TMBER RAL COMBNATON - AND METAL RAL CONCRETE METAL Oll TMBER RAL Figure 10. Types of Bridge Rails and Critical Dimensions. 19

31 0.8 l- CJ [jj 0 - z 0 (f) Z w :2 0 - () «- Z 0 ()...J «a: LL 0 0 ~ a: / «W PREFERRED NOT RECOMMENDED 0.0 a S = POST SETBACK DSTANCE (in) Figure 11. Desired Relationship Between Vehicle-Rail Contact Dimension and Post Setback Distance. 20

32 POTENTAL FOR WHEEL. BUMPER OR HOOD MPACTNG POST... c '-' 15 C) z z W et: L5-1 u -1 «u i= et: w > 10 u 5 HGH POTENTAL LOW POTENTAL RALS RAL c o o S = POST S~fBACK DSTANCE (in) Figure 12. Potential for Wheel, Bumper, or Hood mpacting Post. 21

33 V. ANALYTCAL EVALUATON OF TENNESSEE BRDGE RALNGS Results of the analytical evaluation of the six (6) selected Tennessee bridge railings are summarized in this section. A more detailed description of the calculations made in the evaluation are presented in Appendix A. Bridge Railing TBR-A (K , Flush Mounted) Figure 13 presents a brief summary of the geometric and structural characteristics of bridge railing TBR-A as well as the results of the analytical evaluation. Each post can resist a horizontal load of 14.6 kips (64.9 kn) and the beam can resist 29.2 kips (129.9 kn) over one span. The ultimate strength of the bridge rail (w) is 38 kips (169 kn) applied at a height (Y) of 21 in. (S3.3 cm) above the deck as determined by the two-span failure mode shown in Figure 2. This railing should withstand a test level (TL) 3 impact with a 4,SOO-lb (2,041-kg) vehicle impacting at a speed of 60 milh (96.S km/h) and an angle of 2S deg, i.e., a horizontal load (FH) of 43 kips (191 kn) at an impact force height (He) of 23 in. (S8.4 cm). While the strength analysis of the rail is 12 percent smaller than that required, it is believed that the railing would have sufficient structural strength since the analytical procedure is known to be very conservative. t neglects the inertia of the concrete, the increased strength of materials under high strain rates, and the fact that the materials are almost always stronger than the minimum strengths specified. The strengths of the posts and beam in this rail are controlled or limited by their shear capacity. f the spacing of the #2 stirrups or ties were decreased from 6-in. (1S.2-cm) center-tocenter spacing to 3-in. (7.6-cm) spacing, the strength of the bridge railing would be increased to 43.1 kips (191.7 kn). Since the S-in. (38.1 cm) vertical clear opening (C) and post setback distance (S) of 2 in. (S.l cm) does not comply with the recommended values shown in Figure 12, there is some concern regarding the potential for the wheel, bumper or hood of the vehicle to snag on the posts. t is therefore recommended that this bridge railing be crash tested to assess this potential for snaggmg. 22

34 r 12" l l"cl 1 10 ", typ- - POST 8 ft c-c spacing 4-#8 fy = 60 ksi Ties 6" c-c Concrete fd = 3000 psi 2 BEAM 4 - #7 Stirrups fy = 60 ksi ~2",l0"'_~ -1/2 / 3'1 1 #7 6" c-c - -f'cl 1" cl typ - typ - j ~ 2"~ 12" 27" wi = 38 Kip s Y = 21 i n. 4-#6 1 2 " ro- 3"f. 2" \! J V 2" ) t 5" c-c 10" c-c POST STRENGTH 14.6 kips (Shear Controls) BEAM STRENGTH kips (Shear Controls) STRENGTH OF TBR-1A 38 kips Y = 21 in. (Two span failure mode) SL-3 4,500 b/60 mph/25 deg in. Figure 13. TBR-1 A. 23

35 Bridge Railing TBR-B (K A, Curb Mounted) Figure 14 presents a brief summary of the geometric characteristics of bridge railing TBRlB. The structural characteristics of this railing are exactly the same as those of bridge railing TBR - A discussed previously. The ultimate strength of the bridge rail (w) is 38 kips (169 kn) - applied at a height (Y) of 30 in. (76.2 cm) above the pavement. The spacing of the #2 stirrups and ties should be decreased from 6 in. (15.2 cm) to 3 in. (7.6 cm) and the TBR-B should be crash tested for the same reasons given for TBR-A. n fact, the potential for post snagging is greatly increased by the presence of the 9-in. (22.9 cm) curb. The typical bumper of a 4,500-lb (2,241-kg) passenger car is between 11 and 22 in. (27.9 to 55.9 cm) above the pavement and the typical bumper for a 1,800-lb (816-kg) passenger car is between 14 to 19 in. (35.6 to 48.3 cm) above the pavement. n comparison, the opening between the beam of the bridge railing and the curb is 9 to 24 (22.9 to 61.0 cm) in. above the pavement. This would place the bumper of an impacting vehicle in the opening between the beam of the bridge railing and the curb and would greatly increase the potential for snagging at the posts. Bridge Railing TBR-2A (K A) Figure 15 presents a brief summary of the geometric and structural characteristics of bridge railing TBR-2A as well as the results of the analytical evaluation. This bridge railing can resist a horizontal load (w) of 194 kips (863 kn) at a height (Y) of 23.8 in. (60.5 cm) above the deck. The 5 in. (12.7 cm) diameter standard pipe rail on top of the concrete parapet will resist a load of 18.2 kips (81.0 kn) over one span. The strength of the metal post is 20.2 kips (89.8 kn) and is controlled or limited by the structural capacity of the anchor bolts. The 22-in. (55.9 cm) high concrete parapet or wall can resist a load of 198 kips (881 kn) at its top. This bridge rail can easily resist a test level 3 impact, i.e., a 4,500-lb (2,241-kg) vehicle impacting at 60 milh (96.5 km/h) and 25 deg, with an impact force (FH) of 43 kips (191 kn) at an effective height (He) of 23 in. (58.4 cm). Figure 16 compares the geometric shape of this rail with that of the standard New Jersey concrete safety shape. The lower sloping face is 3 in. (7.6 cm) higher than the standard safety shape. This could potentially increase the propensity for minicars, i.e., 1,800 lb (816 kg) or less, 24

36 " " 2"1 10"1 9 "- / 12" CJ = 38 Kips -2"~ 15" 36" Y = 30 in. 9" \ 1 '--_/ STRENGTH OF TBR-1 B = 38 kips Y = 30 in. SL-3 4,500 b/60 mph/25 deg n. Figure 14. TBR-1 B. 25

37 9" -.-l wi kips 37 1/4" #6 6" 22" 13" Y " 1" cl ----' =1 3" fy - 60 ksi g psi 12" c-c 18" ---l-. 3" 1 1/2" Figure 15. TBR-2A, 26

38 2"1.,?~ N.J. 9 Tenn. Tenn. TBR-2A Shape N.J. Shape 12 1/2" f ~~~~1 ~~ 6" 1 9" 13" 1 0" c Q) - 3" -,. Z 3" The N.J. shape has caused "minicars" to at 60 mph and 20 degrees. roll over on impact The TBR-2 shape may be more vulnerable to minicar rollover. Figure 16. TBR-2A Geometry and New Jersey emb. 27

39 to roll over. Therefore, although this rail has sufficient strength, it should be crash tested with a small car to evaluate its geometric effect on vehicle impact behavior. t should be noted that the TBR-2A bridge railing could be upgraded to withstand a test level 4 impact, i.e., an 18,000-lb (8,165 kg) single unit truck impacting at a speed of 50 milh (92.6 km/h) and at an angle of 15 deg, with an impact force (w) of 54 kips (240 kn) at a height - (Y) of 32 in. (81.3 cm). This could be accomplished by increasing the anchor bolts to 718 in. (2.2 cm) diameter and reducing the post spacing to 5 ft (1.52 m) center to center. The load capacity of the rail (w) would be increased to 67 kips (298 kn) at a height (Y) of 34.5 in. (87.6 cm). However, the impact performance of the modified bridge railing should be verified at this test level with a full-scale crash test. Bridge Railing TBR-2B (K ) Figure 17 presents a brief summary of the geometric and structural characteristics oftbr- 2B, along with the results of the analytical evaluation. This rail is similar to bridge railing TBR- 2A discussed previously except that the concrete parapet or wall has a vertical or flat face and is only 21 in. (53.3 cm) high. The metal rail on top is identical to that of bridge railing TBR-2A. This bridge railing can resist a horizontal load (w) of 154 kips (685 kn) at a height (He) of 23.3 in. (59.2 cm). This bridge railing system can easily resist a test level 3 impact, i.e., 4,500-lb (2,241 kg) vehicle impacting at 60 milh (96.5 km/h) and 25 deg, with an impact force (FH) of 43 kips (191 kn) at an effective height (He) of 23 in. (58.4 cm). This rail also easily satisfies the geometric requirements shown in Figures 10, 11, and 12. Furthermore, this bridge railing is very similar to North Carolina bridge rail BR-38 which has been successfully crash tested (see Appendix B). n fact, as shown in Appendix B, TRB-2B is actually stronger and taller than BR-38. For these reasons, this bridge railing should not require any crash testing. t should be noted that bridge railing TBR-2B could be upgraded to withstand a test level 4 impact, i.e., an 18,000-lb (8,165 kg) single unit truck impacting at a speed of 50 milh (92.6 km/h) and at an angle of 15 deg, with an impact force (w) of 54 kips (240 kn) at a height (Y) of 32 in. (81.3 cm). This could be accomplished by increasing the anchor bolts to 718 in. (2.2 cm) diameter and reducing the post spacing to 5 ft (1.52 m) center-to-center. This would increase the load capacity of the rail (w) to 50.2 kips (223.2 kn) at a height (Y) of 32 in. (

40 wi = 154 kips 36 1/4" 6" 4 1 / 4" 21 " Y " 1" cl 10 3/4" SL-3 4,500 b/60 mph/25 deg in. Figure 17. TBR-2B. 29

41 cm). However, the impact performance of the modified bridge railing should be verified at this test level with a full-scale crash test. Bridge Railing TBR-3 (M-Ol-148) Figure 18 presents a brief summary of the geometric and structural characteristics of bridge railing TBR-3 along with the results of the analytical evaluation. This bridge railing is very similar to the standard 12-gage W-Beam guardrail mounted on W6x9 steel posts with a 6 ft-3 in. (1.91 m) center-to-center spacing. The TBR-3 bridge railing has W6x16 steel posts mounted on a 6 in. (15.2 cm) high concrete curb. The analysis indicates that the rail can resist a load (w) of 13.6 kips (60.5 kn) located at a height (Y) of21 in. (53.3 cm). This bridge railing should, therefore, easily resist a test level 1 impact, i.e., 4,500-lb (2,241-kg) vehicle impacting at a speed of 30 milh (48.3 krn/h) and an angle of 25 deg, with a horizontal load of 14 kips (62 kn) at a height of 15 in. (38.1 cm). n an alternate embodiment, bridge rail TBR-3 can be mounted directly on the bridge deck without the 6 in. high curb. n this configuration the post is 6 in. higher and thus the post strength is reduced from 21.8 kips to 15.6 kips. This does not reduce the one span capacity of TBR-3 of 13.6 kips and, hence, it is still capable of redirecting a 4,500 lb vehicle impacting at 30 mph and 25 deg. This rail should not require crash testing at this test level since similar rails have withstood crash test impacts of much higher severity (see Appendix C). Bridge Railing TBR-4 (M ) Figure 19 presents a brief summary of the geometric and structural characteristics of bridge railing TBR-4 as well as the results of the analytical evaluation. This retrofit bridge rail is also similar to, but much stronger than, the standard W-Bearn guardrail (see Appendix C). The TBR-4 retrofit bridge railing incorporates a stronger lo-gage W-bearn mounted on W6x16 steel posts (and existing concrete posts) spaced at 3 ft-l 112 in. (95.3 cm). This TBR-4 bridge railing can easily resist an impact load (w) of 52 kips (231 kn) at a height (Y) of 21 in. (53.3 cm). A variation of this bridge railing is mounted using only the W6x16 steel posts and not the existing concrete posts. When mounted in this fashion, its strength is increased from 52 kips to at least 64 kips. This bridge railing system should, therefore, easily qualify under test level 3, i.e., 4,500-lb (2,241-kg) vehicle impacting at 60 milh (96.5 krn/h) and 25 deg, with an impact 30

42 Post W6x 16 6'-3" c-c Spacing, gao W-Beam f f Ull = 13.6 kips 2'" Base t 11"x11"x3/4" A36 Steel 27" Y = 21 in. 6" 4-7/8" (j) A307 Anchor Bolts '~ 4 1 /2 "----f-_-+---=:--=-va~r_'_':i e~s =-l 3.5" -7.0" W-BEAM STRENGTH kips (one span) POST STRENGTH = 21.8 kips (anchor bolts control) STRENGTH OF TBR in. (one span) SL-1 4,500 b/30 mph/25 deg in. NOTE: The guardrail is located anywhere from flush with the curb to 3.5" set back. Therefore add dimension from face of curb to face of base plate (varies 3.5" to 7.0") and dimension from face of base plate to center of guardrail po s t (4. 5" ). Figure 18. TBR-3, 31

43 V. CRASH TEST PROCEDURES Description of Test nstallations Based on the analysis presented above, two existing TDOT bridge rail designs were selected for further evaluation through a series of full-scale crash tests. The test installation consisted of a 100-ft (30.5-m) long simulated bridge deck with a 3 ft (0.91 m) overhang. The two bridge railings, each approximately 50 ft (15.24 m) in length, were constructed side by side on this simulated bridge deck. Details of the two bridge rail designs are described below. Bridge Railing TBR-1A: Open Concrete Beam-and-Post Bridge Rail CK ) This bridge railing is comprised of a 10 in. x 12 in. (25.4 cm x 30.5 cm) concrete beam cast on 10 in. x 12 in. (25.4 cm x 30.5 cm) concrete posts. The total rail height is 27 in. (68.6 cm), thus creating a S-in. (38.1-cm) opening between the deck and the lower edge of the beam. The posts are offset 2 in. (5.1 cm) back from the face of the beam and are spaced at 8-ft (2.44- m) intervals. A 50-ft (15.24-m) length of bridge rail was constructed using two 25-ft (7.62-m) sections with a 112-in. (1.3-cm) open joint placed at midspan of the installation. Details of the bridge rail are shown in Figure 20. n the Phase analysis, the ultimate strength of the bridge railing was calculated to be 38 kips (169 kn) applied at a height of21 in. (53.3 cm) above the deck which corresponds to midheight of the concrete beam. For a test level (TL) 3 impact, i.e., a 4,500-lb (2,041-kg) vehicle impacting the bridge rail at a nominal speed and angle of 60 mph (96.6 km/h) and 25 deg, the expected impact load on the bridge railing is 43 kips (191 kn). Although the computed strength of the rail is 12 percent smaller than that of the expected impact load, the analytical procedure is known to be very conservative and it was believed that the railing should have sufficient structural strength for a test level 3 impact. However, a crash test was conducted to verify the structural adequacy of the railing. n addition to structural adequacy, there was also concern about the potential for parts of the vehicle, e.g., the wheel, bumper, or hood, to snag on the concrete posts due to the combination of the S-in. (38.1-cm) high opening between the deck and the bottom of the beam and an offset of 2 in. (5.1 cm) between the face of the beam and the posts. Therefore, a crash test was conducted to evaluate occupant risk and severity. 34

44 #7 TENNESSEE 151 8" 6" 7" c-c 8" x 10" 6" overlap 27" 32" #8 3 5" c-c LL--- T~ --112" f- 1~ 30" #8 2 10" c-c #6 ~ 4" x 1/4" \ 12" \ steel strap 5" c-c - 12" Existing concrete deck and beam 'r r "'-- 10" c-c 1 2" _ " Figure 20. Open concrete beam and post bridge rail,

45 Bridge Railing TBR-2A: Concrete Parapet with Structural Tubing Bridge Rail CK A) The base ofthis railing consists of a 22-in. (55.9-cm) high concrete safety-shaped parapet. The face of the parapet has the same slope as the standard New Jersey concrete safety shape, but the sloped face extends 3 in. (7.6 cm) higher than the standard safety shape design. Mounted on top of the parapet is a 5-in. (12.7-cm) diameter standard aluminum pipe supported by cast aluminum posts spaced at 10 ft-6 in. (3.2 m) intervals. A 48 ft (14.6 m) long continuous segment of this barrier was constructed on the simulated bridge deck, next to the open concrete beam-andpost bridge rail installation. Details of the bridge railing design are shown in Figure 21. n the Phase analysis, it was concluded that the structural capacity of this combination bridge rail is far in excess of the impact forces expected from an TL 3 impact, i.e., a 4,500-lb (2,041-kg) vehicle impacting the bridge railing at a nominal speed and angle of 60 mph (96.6 km/h) and 25 deg. However, while the bridge rail was judged to have sufficient strength, there was concern over the effect of the geometry of the parapet on the stability of a small car. Therefore, a severity test was conducted to assess the effect of the bridge railing geometrics on vehicle impact behavior. Details of the crash test procedures used to evaluate these bridge rails are described below. Description of Crash Test Procedures According to guidelines contained in National Cooperative Highway Research Program (NCHRP) Report 230m, two crash tests are recommended for evaluating the impact performance of a bridge rail: 1. Test designation S13. An 1,800-lb (816-kg) vehicle impacting the bridge rail at a nominal speed and angle of 60 mph (96.6 km/h) and 20 deg. This test is a severity test intended to evaluate the potential risk of a hypothetical occupant during the impact event and to investigate the dynamic interactions of a small car with the bridge rail. 2. Test designation 10. A 4,500-lb (2,041-kg) vehicle impacting the bridge rail at a nominal speed and angle of 60 mph (96.6 km/h) and 25 deg. This test is a structural adequacy or strength test intended to examine the strength of the bridge rail and its ability to contain and redirect an impacting vehicle. Note that a pickup truck was used as the test vehicle for this test instead of a passenger car. As mentioned previously, 36

46 10"-1 :: mint max 5" f- 14" 1 12" c-c 10" Rod. 5" <l> 1.0. Aluminum Pipe t 6" 1"\11 holes weld T 2' 14" 12" c-c 6" -9 \ 13[$ / ~,------j13-" " #-5-@-12"c-c 8L /2" 4" X 1/4" :---_-Jl steel strap - l' 2" ~ ~ l' 2"-j 5" c-c 8" L'---"o ;.;.-----t # '- 15" --- f " ~... 10" c-c Figure 21. Concrete safety-shaped parapet with structural tubing bridge rail.

47 NCHRP Report 230 was being updated during the course of this project, and the best available information indicated that a 4,500-lb pickup would be selected as the test vehicle for strength tests. For the concrete beam-and-post bridge rail design, both the structural adequacy and severity tests were conducted since there was concern over both the strength of the bridge rail design and the potential for vehicle snagging on the interior faces of the concrete posts. For the concrete parapet with structural tubing bridge rail design, it was previously determined that the bridge rail has more than sufficient strength for an TL 3 impact and the structural adequacy test was not required. Thus, only the severity test was conducted for this railing to evaluate the effect of the bridge rail geometrics on vehicle stability when impacting the sloped face of the parapet. All crash test and data analysis procedures were conducted in accordance with guidelines set forth in NCHRP Report 230. Brief descriptions of the crash test and data analysis procedures are presented below. Electronic nstrumentation and Data Processing Each test vehicle was instrumented with three solid-state angular rate transducers to measure roll, pitch and yaw rates; a triaxial accelerometer at the vehicle center-of-gravity to measure longitudinal, lateral, and vertical acceleration levels, and a back-up biaxial accelerometer in the rear of the vehicle to measure longitudinal and lateral acceleration levels. The accelerometers were strain gauge type with a linear millivolt output proportional to acceleration. The electronic signals from the accelerometers and transducers were transmitted to a base station by means of constant bandwidth FMlFM telemetry link for recording on magnetic tape and for display on a real-time strip chart. Provision was made for the transmission of calibration signals before and after the test, and an accurate time reference signal was simultaneously recorded with the data. Pressure sensitive contact switches on the bumper were actuated just prior to impact by wooden dowels to indicate the elapsed time over a known distance to provide a measurement of impact velocity. The initial contact also produced an "event" mark on the data record to establish the exact instant of contact with the guardrail system. The multiplex of data channels, transmitted on one radio frequency, was received at a data acquisition station, and demultiplexed into separate tracks of ntermediate Range nstrumentation 38

48 Group (LR.LG.) tape recorders. After the test, the data was played back from the tape machines, filtered with a SAE J211 Class 180 filter, and were digitized using a microcomputer, for analysis and evaluation of impact performance. The digitized data were then processed using two computer programs: DGTZE and PLOTANGLE. Brief descriptions on the functions of these two computer programs are given below. The DGTZE program uses digitized data from vehicle-mounted linear accelerometers to compute occupant/compartment impact velocities, time of occupant/compartment impact after vehicle impact, and the highest lo-msec average ridedown acceleration. The DGTZE program also calculates a vehicle impact velocity and the change in vehicle velocity at the end of a given impulse period. n addition, maximum average accelerations over 50-msec intervals in each of the three directions are computed. Acceleration versus time curves for the longitudinal, lateral, and vertical directions are then plotted from the digitized data of the vehicle-mounted linear accelerometers using a commercially available software package. The PLOT ANGLE program uses the digitized data from the yaw, pitch, and roll rate charts to compute angular displacement in deg at second intervals and then instructs a plotter to draw a reproducible plot of yaw, pitch, and roll versus time. t should be noted that these angular displacements are sequence dependent with the sequence being yaw-pitch-roll for the data presented herein. These displacements are in reference to the vehicle-fixed coordinate system with the initial position and orientation of the vehicle-fixed coordinate system being that which existed at initial impact. Photographic nstrumentation and Data Processing Photographic coverage of each test included three high-speed cameras. One camera was stationed overhead with a field of view perpendicular to the ground and directly over the impact point. A second camera was placed to have a field of view parallel to and aligned with the bridge rail at the downstream end. For the open concrete post-and-beam bridge rail, the third camera was placed behind the bridge rail to document wheel contact on the concrete posts. Since snagging was not a concern for the concrete parapet with structural tubing bridge rail, the third camera was placed perpendicular to the traffic face of the bridge rail to observe vehicle stability during the impact event. A flash bulb activated by pressure sensitive tapeswitches was positioned on the impacting vehicle to indicate the instant of contact with the guardrail system and was 39

49 visible from each camera. The films from these high-speed cameras were analyzed on a computer-linked Motion Analyzer to observe phenomena occurring during the collision and to obtain time-event, displacement, and angular data. A 3/4-in videotape camcorder and still cameras were used for documentary purposes and to record conditions of the test vehicle and guardrail system before and after the test. Test Vehicle Propulsion and Guidance The test vehicles were towed into the guardrail system using a steel cable guidance and reverse tow system. A steel cable for guiding the test vehicles was stretched along the impact path, anchored at each end, and threaded through an attachment to the front wheel of the test vehicle. Another steel cable was connected to the test vehicles, passed around a pulley near the impact point, through a pulley on the tow vehicle, and then anchored to the ground such that the tow vehicle moved away from the test site. A 2 to 1 speed ratio between the test and tow vehicle existed with this system. Just prior to impact with the guardrail system, the test vehicle was released to be free-wheeling and unrestrained. The vehicle remained free-wheeling, i.e., no steering or braking inputs, until the vehicle cleared the immediate area of the test site, at which time brakes on the vehicle were activated to bring the vehicle to a safe and controlled stop. 40

50 V. FULL-SCALE CRASH TEST RESULTS Descriptions of the test results are presented below. Sequential photographs of the individual tests are shown in Appendix D. Accelerometer traces and rate gyro data are presented in Appendices E and F, respectively. Open Concrete Beam-and-Post Bridge Rail Two crash tests were conducted on the open concrete beam-and-post bridge rail. The first crash test involved an 1,800-lb (816-kg) passenger car impacting the bridge rail at a nominal speed and angle of 60 mph (96.6 km/h) and 20 deg. The objective of this test was to evaluate impact severity and assess the potential for wheel snagging on the concrete posts. The second crash test was a strength test which involved a 4,500-lb (2,041-kg) pickup truck impacting the bridge rail at a nominal speed and angle of 60 mph (96.6 km/h) and 25 deg. The purpose of this test was to evaluate the structural adequacy of the beam-and-post bridge rail design. F or each test, the critical impact location was selected so as to maximize the loading and/or the potential for wheel contact on the concrete post. The critical impact points for rigid barriers of this type were previously determined to be 3.5 ft (1.07 m) and 4.3 ft (1.3 m) upstream from the face of the post for the severity and strength tests, respectively.(l) Test 1 (Test No ) Photographs of the open concrete beam-and-post bridge rail test installation prior to the test are shown in Figure 22. A 1988 Yugo, shown in Figure 23, was used as the test vehicle. The test inertia mass or empty weight of the vehicle was 1,800 lb (816 kg). An unrestrained 50th percentile male anthropometric dummy was placed in the driver position of the vehicle. The gross static mass or total weight of the vehicle was 1,965 lb (891 kg). The height to the lower edge of the vehicle bumper was 13.0 in. (33.0 cm) and the height to the upper edge was 18.8 in. (47.6 cm). Other dimensions and information on the vehicle are given in Figure 24. The position of the vehicle relative to the bridge rail prior to impact is shown in Figure 25. The vehicle was directed into the bridge rail using the cable reverse tow and guidance system, and was released to be free-wheeling and unrestrained just prior to impact. 41

51 Figure 22. Telilessee post and beam bridge rail before test

52 Figure 23. Vehicle before test

53 Da te: T es t No.: _7L ~-..J..1 VN: VX1BB1216JK Make: _V..."U4.;;j9u.<.O Mode 1: -----lg.uvu..l~ Year: 1988 Tire Size: 145 SR13 Ply Rating: Ti re dia hee 1 dia r 95 3/4/1 Accelerometers s Bias Ply: 11 Odometer: ~6:..!..7.!:c2~38~ Belted: Radial: -2L Tire Condition: good fa i r X badly worn Height of Rear Accelerometer ~ 25' 1/2/1 Vehicle Geometry - inches c T a 60". b 27" --=-'---- c 84 3/4" d* 54 3/4" e 24 1/2" f g h 30.2/1 i j 30 1/4/1 k 14 1/2" R /2" j m h c e k g m 18 3/4 11 n 2 ] ]311 P r 21 lull s 14 ] L~ 4-wheel weight for c.g. det. if 584 rf 575 ir 319 rr Mass - pounds Curb Test nertial Gross S ta tic ~ote Ml M MT any damage to vehicle prior to test: Crack in windshield (marked}. kd = overall height of vehicle f 322 Engine Type: 4 cyl Gas Eng i n e C D: --=.1~. =-l--.!l=--- Transmission Type: ~~J~\:{X~~k~ or t,1anua 1 FWD or ~).(): or #10< Body Type: 3 door Steering Column Collapse Mechanism: Behind wheel units --Convoluted tube --Cyl i ndri ca 1 mesh units -Elilbedded ba 11 -rwt collapsible --Other energy absorpti on --Unknown Bra kes: Front: d i sc_x_ drum_ Rear: disc drulll_x Figure 24. Test vehicle properties (7199-1). 44

54 Figure 25. Vehicle/rail geometries for test

55 The vehicle impacted the concrete beam and post bridge rail 3.S ft (1.07 m) upstream of the post contiguous to the construction joint located midspan of the installation at 61.1 mph (98.3 km/h) and at an angle of 21.3 deg. Shortly after impact, the concrete beam loaded the upper portion of the right front tire, causing the wheel to rotate about the spindle assembly and fold under the rail. The right front wheel subsequently contacted the interior face of the concrete post. This caused the wheel assembly to be pushed into the fire wall of the vehicle, resulting in significant deformation of the occupant compartment. Shortly thereafter, the rear of the vehicle rotated into the barrier. Coinciding with the rear impact, the dummy impacted the right passenger's door and penetrated beyond the occupant compartment of the vehicle. The vehicle lost contact with the bridge rail at approximately 0.S91 second after impact, traveling at a speed of 4S.3 mph (72.9 km/h) and at an exit angle of 0 degree. After the vehicle exited the installation, the brakes were applied and the vehicle yawed clockwise and came to rest approximately ft (33.8 m) from the initial point of impact. As shown in Figure 26, the bridge rail received only minor cosmetic damage. The vehicle was in contact with the installation for a distance of 14.3 ft (4.4 m). Damage to the test vehicle is shown in Figure 27. There was considerable damage to the right front quarter of the vehicle. The maximum recorded crush was 13.0 in. (33.0 cm) located at the floor pan on the right side of the vehicle. The right front wheel and control arm were severely bent and pushed rearward 17.3 in. (43.8 cm). n addition, the roof, right door, and upper A-pillars were bent. A summary of the test results and other information pertinent to this test are given in Figure 28. Occupant impact velocities were 2S.3 fils (7.7 mls) and fils (8.8 rn/s) in the longitudinal and lateral directions, respectively. The highest lo-msec average occupant ridedown accelerations were -2.S g in the longitudinal direction and -S.9 g in the lateral direction. The maximum SO-msec average accelerations experienced by the vehicle were g in the longitudinal direction and S.6 g in the lateral direction. The change in vehicle velocity at loss of contact with the bridge rail was S.8 mph (2S.4 km/h). n summary, this test was judged to be a failure. Although the occupant risk criteria were all within the maximum allowable values outlined in NCHRP Report 230, the wheel assembly snagged on a concrete post, causing significant deformation of th~ floor pan of the occupant compartment. The vehicle sustained extensive damage while the bridge rail test installation received only minor cosmetic damage. There were. no debris or detached elements. The vehicle 46

56 Figure 26. Tennessee post and beam bridge rail after test

57 Figure 27. Vehicle after test

58 '" "....~, " " " ::" " s s s s Test No Date Test nstallation. nstallation Length Test Vehicle. Vehicle Weight Test nertia /14/92. Tennessee Post & Beam Concrete Bridge Rail FT (15. 2 m) 1988 Yugo 1800 lb (816 kg) Vehicle Damage Classification TAD RFQ-7 CDC RYAWB Maximum Vehicle Crush in (33.0 cm) Maximum Perm Rail Deformation. N.A. mpact Speed mi/h (98.3 km/h) mpact Angle deg Exit Speed mi/h (72.9 km/h) Exit Trajectory deg Vehicle Accelerations (Max sec Avg) Longitudinal g Lateral g Occupant mpact Velocity Longitudinal ft/s (7.7 m/s) Lateral ft/s (8.8 m/s) Occupant Ridedown Accelerations Longitudinal g Lateral g Figure 28. Summary of results for test

59 remained upright and stable throughout the impact sequence and after exiting from the bridge rail. The velocity change of 15.8 mph (25.4 km/h) was higher than the recommended limit of 15.0 mph (24.1 kmlh). However, with an exit angle of 0 deg, the vehicle trajectory did not exhibit any potential for intrusion into the adjacent traffic lanes. Test 2 (Test No ) Although the results of the severity test (test 1) were found to be unacceptable, it was decided to proceed with the evaluation on the structural adequacy ofthe existing system. For this strength test, a 1984 GMC C2500 pick-up, shown in Figure 29, was used as the test vehicle. Test inertia mass of the vehicle was 4,500 lb (2,041 kg). The bumper height of the vehicle ranged from 18.3 in. (46.4 cm) at its lower edge to 27.0 in. (68.6 cm) at its upper edge. Additional dimensions and information pertaining to the vehicle are given in Figure 30. The position of the vehicle relative to the bridge rail is shown in Figure 31. The vehicle impacted the concrete post-and-beam bridge rail 4.3 ft (1.3 m) upstream of the post contiguous to the construction joint at 61.9 mph (99.5 km/h) and at an angle of25.6 deg using a cable reverse tow and guidance system. At approximately second, the left front tire lost contact with the roadway. By second, the vehicle was travelling parallel to the rail at 50.1 mph (80.6 kmlh) and both left tires had lost contact with the roadway. As the vehicle continued down the rail, the left rear tire also briefly lost contact with the roadway. The vehicle exited the rail at approximately second after impact, traveling at a speed of 49.3 mph (79.3 km/h) and at an exit angle of 7.3 deg. The tires of the vehicle regained contact with the roadway upon exiting the rail. The brakes were applied after the vehicle cleared the test installation. The vehicle yawed clockwise and came to rest approximately ft (54.9 m) from the point of impact. The installation received only minor damage as shown in Figure 32. The vehicle was in contact with the installation for a total length of 13.0 ft (4.0 m). A stress crack in the post contiguous to the construction joint was noted after the test. The crack extended from the face of the post around the inside of the post and into the deck. This hairline crack was marked in black for visibility and can be seen in the bottom photograph of Figure 32. Damage to the vehicle is shown in Figure 33. There was considerable damage to the right front quarter of the vehicle. The maximum crush was 20.0 in. (50.8 cm) located at the right front 50

60 Figure 29. Vehicle before test

61 Date: T est No.: ~7...:::1-=-9.::...9-_4,,-- VN: Make: GMC --"''-'-'''----- Model: SERRA 2500 Year: J 984 Tire Size: LT Ply Rating: Bias Ply: Accelerometers GTEC24D8ES Odome te r: --:::r9.u..69;u2;...;)3'-- Belted: Radial: X Tire Condition: good ~ fair badly worn _ p a Vehicle Geometry - inches a 11 b c d* ~ A-:.ce'erometers l... :1 r Tire dia. e f g h i j k t m n P wheel weight for c. g. det..f 1117 rf 1040.r 1172 rr 1171 Mass - pounds Curb Test nertial Gross Sta ti c Ml M MT Note any damage to vehicle prior to test: Crack in windshield (Marked) *d overall heiqht of vehicle Figure 30. Test vehicle properties (test ). 52 r s cyl. Engine Type: Gasoline Engine CO: 4.1 liter Transmission Type: ~utomatic) or Manual FWD or (FwD) or 4WO Body Type: _P~/_P_U Steering Column Collapse Mechanism: Behind wheel units --Convoluted tube --Cyl i ndri ca 1 mesh units -Embedded ba 11 -NOT collapsible --Other energy absorption -Unknown Brakes: Front: disc X drum Rear: disc drum X

62 Figure 31. Vehicle/rail geometric for test

63 Figure 32. Tennessee post & beam bridge rail (after test ). 54

64 Figure 33. Vehicle after test

65 quarter of the vehicle. The right front wheel and control arm were bent and pushed rearward 12.3 in. (31.1 cm). n addition, the frame and right door were bent, and the floor-pan was bent in the vicinity of the door. A summary of the test results and other information pertinent to this test are given in Figure 34. Although not required for the strength test, occupant risk factors were computed for information purposes. n the longitudinal direction, occupant impact velocity was 20.6 fils (6.3 m!s), the highest lo-msec average ridedown acceleration was -5.4 g, and the maximum 50-msec average acceleration was -6.8 g. n the lateral direction, occupant impact velocity was ftls (8.2 m!s), the highest lo-msec average occupant ridedown acceleration was 12.9 g, and the maximum 50-msec average acceleration was 12.1 g. The change in vehicle velocity at loss of contact was 12.6 milh (20.3 km/h). n summary, this test was judged to have met the intent of the performance criteria set forth in NCHRP Report 230. The bridge rail successfully contained and redirected the impacting vehicle. The test vehicle remained upright and stable during the impact period and after leaving the installation and there was no debris from the vehicle or barrier that might present undue hazard to other traffic. Damage to the test vehicle was significant, but not unusual for a test of this severity with a rigid barrier. There was some minor deformation of the floor pan at the base of the A-frame, but this deformation was not judged to be life threatening. Although not a requirement for the structural adequacy test, all occupant risk criteria were below the maximum allowable values recommended in NCHRP Report 230, further indicating that the vehicle was smoothly redirected without experiencing any severe decelerations. Both the exit velocity and exit angle of the vehicle were below the recommended limits set forth in NCHRP Report 230. Summary The open concrete beam-and-post bridge rail performed satisfactorily in the structural adequacy test, but failed in the severity test. Although the occupant risk criteria were all within the maximum allowable values outlined in NCHRP Report 230, the wheel assembly snagged on a concrete post, causing significant deformation of the floor pan of the occupant compartment. This adverse behavior could be attributed to the 5-in. (38.-cm) opening between the bridge deck and the lower edge of the concrete beam. The excessive height of the opening allowed the wheel assembly to go underneath the concrete beam and snag on the concrete post. 56

66 0.000 s s s s Test No Date /28/92. Test nstallation. Tennessee Post & Beam Concrete Bridge Rail. nstallation Length 50.0 FT (15.2 m) Test Vehi cl e GMC C2500 Pick-up Vehicle Weight Test nertia 4500 lb (2041 kg) Vehicle Damage Classification TAD RFQ-7 CDC RYAWB Maximum Vehicle Crush in (50.8 cm) Maximum Perm Rail Deformation. N.A. mpact Speed mi/h (99.5 km/h) mpact Angle deg Exit Speed 49.3 mi/h (79.3 km/h) Exit Trajectory deg Vehicle Accelerations (Max sec Avg) Longitudinal g Lateral g Occupant mpact Velocity Longitudinal ft/s (6.3 m/s) Lateral ft/s (8.2 m/s) Occupant Ridedown Accelerations Longitudinal g Lateral g Figure 34. Summary of results for test

67 The obvious solution to this problem is to reduce the height of the opening between the deck and the bottom of the concrete beam. As shown in Figure 12, for an offset of 2 in. (5.1 cm) between the face of the beam and the posts, a vertical clear opening of 11 in. (27.9 cm) or less is recommended for proper impact performance. For new construction applications, the depth of the concrete beam could be increased from 12 to 16 in. (30.5 to 40.6 cm), thus reducing the height of the opening between the deck and the bottom of the concrete beam to 11 in. (27.9 cm). For retrofit situations, the following two alternative remedies were considered. One alternative is to effectively extend the depth of the beam by attaching a steel channel to the concrete posts immediately below the existing concrete beam. The channel could be attached to the posts using concrete anchor studs. A C4x7.25 or C6x8.2 channel could be used for this purpose and would reduce the opening between the deck and the bottom of the channel to 11 or 9 in. (27.9 or 22.9 cm), respectively. Both sizes of channels should provide acceptable impact performance and the choice of channels would be one of economics and availability. A second alternative would be to place a 4 to 6 in. (10.2 to cm) high curb on the bridge deck flush with the traffic face of the concrete beam. Once again, this would reduce the vertical clear opening between the top of the curb and the bottom of the concrete beam to 11 or 9 in. (27.9 or 22.9 cm), respectively. The curb could be anchored to the deck using anchor studs or dowels and could be cast in place or assembled from precast segments. The alternative of adding a channel to the bottom of the concrete beam to reduce the height of the opening is probably the more attractive of the two alternatives from the standpoints of operations, cost and esthetics and is therefore recommended. The C6x8.2 channel is preferred over the C4x7.25 channel since the C6x8.2 channel is already used by TDOT as rubrails for guardrails and transitions. This would eliminate the need to keep another size of channel in the inventory, thus simplifying the maintenance requirements. There should not be any significant difference in impact performance between these two channel sizes. A schematic drawing showing the addition of the channel beneath the concrete beam and the attachment details is shown in Figure 35. Concrete Parapet with Structural Tubing Bridge Rail (Test No ) For the concrete safety shaped parapet with structural tubing bridge rail, only the severity test was conducted since the structural capacity of the bridge rail was far in excess of that 58

68 B'-()"-----~-B--i'l /'-----" C6x8.2 Channel Rubrail (See Detail D) --- ~ Channel Splice (See Detail B) Elevation L ".- 2" t 5" : 0 4 l/z' 0 0 Lc... -' t:f~2j 5/B" Concrete Anchor Bolt See Detail A ::.":) Section A-A o o 0 -i 4 1/2" - Notes: All Slots 11/16" x2" All Square Holes 11/16" Detail A End Angle 8x6x 1 /2x4 1/2" Detail D C6x8.2 Channel Rubrail -Yr- 10"--j /2C / 3'1 1i2 (» 7' c-c '"" -+ 2r c===_ = P 6" =d <j' S'l 1 Section 8-8 C6x B.2 Rubroil.~ Splice Plate (See Detail C) c_=_~_~ r , ' -----E3-- :0 0: L J ~ =_~_~ =_~_~_~ Note: Use 5/B"xl 1/2" Carriage Bolt and Nut (6 per Splice) Detail 8 Channel Splice -rtr!"i 7i ir!"t"i ii ~~---3~-~] t Detail C Splice Plate Figure 35, Channel retrofit for open beam and post bridge rail.

69 required for a test level (TL) 3 structural adequacy test. The primary concern with this bridge rail design was the effects of the bridge rail geometrics (particularly the sloped face of the concrete parapet which has the same slope, but 3 in. (7.62 cm) higher than the standard New Jersey safety shape) on the stability and impact behavior of the impacting vehicle. Photographs of the concrete safety shaped parapet with structural tubing bridge rail test installation are shown in Figure 36. The test vehicle for this test was a 1988 Yugo shown in Figure 37. The test inertia mass or empty weight of the vehicle was 1,800 lb (817 kg). An unrestrained 50th percentile male anthropometric dummy was placed in the driver position of the vehicle. The gross static mass or total weight of the vehicle was 1,965 lb (891 kg). The height to the lower edge of the vehicle bumper was 13.5 in. (34.3 cm) and the height to the upper edge was 18.8 in. (47.6 cm). Additional dimensions and information pertaining to the test vehicle are given in Figure 38. The position of the vehicle with respect to the bridge rail prior to impact is shown in Figure 39. The vehicle was directed into the bridge rail using the cable reverse tow and guidance system, and was released to be free-wheeling and unrestrained just prior to impact. The vehicle impacted the bridge rail installation 16.7 ft (5.1 m) downstream of the upstream end of the bridge rail at 59.3 mph (95.4 km/h) and at an angle of 20.0 deg. Shortly after impact, the right front tire contacted the sloped face of the concrete parapet, allowing the vehicle to begin climbing the sloped face of the parapet. A short time thereafter, the upper edge of the right front fender contacted the bottom of the structural tubing mounted atop the concrete parapet. As the vehicle continued to redirect, the left front tire lost contact with the roadway. Just prior to the vehicle being parallel to the installation, the dummy impacted the right side of the vehicle in the general vicinity of the right door post and the upper A-pillar. At second, the vehicle was parallel to the rail and momentarily airborne, traveling at a speed of 55.4 mph (89.1 km/h). The vehicle exited the bridge rail at a speed of 50.7 mph (81.5 km/h) and at an exit angle of approximately 4.4 deg. At the time the vehicle lost contact with the bridge rail, only the front tires of the vehicle were in contact with the roadway. After the vehicle cleared the test installation, the brakes were applied and the vehicle yawed clockwise and came to rest approximately ft (57.9 m) from the point of initial impact. As shown in Figure 40, the bridge rail test installation received only cosmetic damage. The vehicle was in contact with the concrete parapet a distance of 14 ft (4.3 m) and with the structural tubing atop the concrete parapet a distance of 11.5 ft (3.5 m). 60

70 -~--=.~~~ Figure 36. Tennessee safety-shape bridge rail before test

71 FiJUre 37. Vehicle before test

72 1211 Date: Test No.: 7i99-6 VN: VX] BA] 2141JK4] 4231 Make: vugo Model: GV Year: --=.=-=...;:= Od ome te r: ---'5"-'8"'-4'-"5'-"'5=-- Tire Size: 145 R13 Ply Rating: Bias Ply: Be lted: Radial: _X_ 1 t a p ~ Tire d i a.----k--'-r~ Hheel dia----l--k-~ Acce 1 erollle ters Tire Condition: good fair L Height of Rear Accelerometer badly worn =26 1/211 --LOll Vehicle Geometry - inches Ton a 60 1/411 b 26 3L4 center 11 c 85 1/411 d* /411 Accelerometers e f 136 ]12 11 g---- k 16":;1.-'----- h j t 31 3/411 m 18 3/4 11 n 2 3/4 11 j m k 9 o 13 1/211 P r 21 1/211 s 14 1L411 4-wheel weight for c.g. det. if 579 rf 563 b c f ir 330 rr 328 Mass - pounds Curb Test nertial Gross Static Note any damage to vehicle prior to test: d overall height of vehic 1p Figure 38. Test vehicle properties (7199-6). Eng i net y p e: -..:..4-.:c~y-=-l Engine CO: 1100 cc Transmission Type: Automatic or Manual FWD or RWO or 4WO Body Type: Hatch Steering Column Collapse Mechanism: Behind wheel units --Convoluted tube -Cylindrical mesh units -Embedded ba 11 -NOT collapsible --Other energy absorption -Unknown Bt-a kes : Front: d i sc_x_ drum_ r~ea r: disc drulll 63

73 Figure 40. Tennessee safety-shape bridge rail after test

74 Damage to the vehicle, shown in Figure 41, was relatively minor for a test of this severity. The maximum crush was 6.5 in. (16.5 cm) at the right front comer of the vehicle. The right front wheel and control arm were bent and pushed rearward 3.0 in. (7.6 cm). n addition, the dashboard, floor-pan, and roof were bent, and the windshield broken. The entire right side of the vehicle was dented and scraped. A summary of the test results and other information pertinent to this test are given in Figure 42. Occupant impact velocities were 14.5 fils (4.4 mls) and fils (8.2 mls) in the longitudinal and lateral directions, respectively. The lateral occupant impact velocity of fils (8.2 mls) was higher than the design value of 20 fils (6.1 mls), but was below the maximum allowable limit of 30.0 ftls (9.1 mls) as outlined in NCHRP Report 230. The highest lo-msec average occupant ride down accelerations were 1.8 g in the longitudinal direction and 7.0 g in the lateral direction. The maximum 50-msec average accelerations experienced by the vehicle were- 8.0 g in the longitudinal direction and 14.4 g in the lateral direction. The change in vehicle velocity at loss of contact was 8.6 mph (26.1 km/h). n summary, this test was judged to be a success. The installation contained and smoothly redirected the test vehicle. There was no debris or detached elements from the vehicle or barrier that might present undue hazard to other traffic. Although the vehicle briefly lost contact with the roadway, it remained upright and stable during the impact sequence and after exiting from the bridge rail. Damage to the bridge rail was cosmetic in nature, and damage to the vehicle was relatively minor for a test of this severity. There was no intrusion and only minimal deformation to the occupant compartment. Both the exit velocity and exit angle of the vehicle were below the recommended limits set forth in NCHRP Report

75 Figure 41. Vehicle after test

76 0.000 s s '"' s s Test No Date /06/92. Test nstallation Tennessee Safety Shape Bridge Rail. nstallation Length 50.0 ft. (15.2 m) Test Vehicle Yugo Vehicle Weight Test nertia Vehicle Damage Classification TAD.... CDC.... Maximum Vehicle Crush. Maximum Perm Rail Deformation lb (816 kg ) 01-RFQ-4 01RFEW2 6.5 in (16.5 cm) N/A Figure 42. Summary of results for test mpact Speed mi/h (95.4 km/h) mpact Angle deg Exit Speed 50.7 mi/h (81.5 km/h) Exit Trajectory deg Vehicle Accelerations (Max sec Avg) Longitudinal g Lateral g Occupant mpact Velocity Longitudinal ft/s (4.4 m/s) Lateral ft/s (8.2 m/s) Occupant Ridedown Accelerations Longitudinal g Lateral g

77 outlined in NCHRP Report 230. The trajectory of the test vehicle was stable during and after impact, and all occupant risk criteria for the test were below the maximum allowable limits set forth in NCHRP Report 230. Bridge Railing TBR-2B CK-38-l54) Analysis indicates that this design meets all of the strength and geometric requirements for test level 3 bridge rail. Furthermore, this bridge rail is very similar to North Carolina bridge railing BR-38 which has already been successfully crash tested. No crash testing was recommended or conducted on this bridge railing. Bridge Railing TBR-3 (M-lOl-148) This bridge rail is very similar to but stronger than the standard l2-gage W-Beam guardrail. This bridge rail easily withstand a test level 1 impact, i.e., impact by a 4,500-lb (2,241 kg) car at a speed of 30 milh (48.2 kmlh) and an angle of 25 deg, as evidenced by similar railings which have withstood much more severe crash tests. Bridge Railing TBR-4 (M ) This bridge railing retrofit is similar to but much stronger than the standard W-Beam guardrail, and analysis indicates that this retrofit design could easily withstand a test level 3 impact, i.e., a 4,500-lb (2,241-kg) car at an impact speed of 60 milh (96.5 kmlh) and an angle bf 25 deg. No crash testing was recommended since similar weaker railings have withstood test level 3 impacts. Recommendations Bridge Railing TBR-A (K ) The impact performance of the open concrete beam-and-post bridge rail can be improved by reducing the vertical clear space between the deck and the bottom of the concrete beam from 15 in. (38.1 cm) to 11 in. (27.9 cm) or less. n new construction applications, an alternative would be to increase the depth of the concrete beam itself from 12 in. (30.5 cm) to 16 in. (40.6 cm), thus effectively reducing the height of the opening to 11 in. (27.9 cm). n retrofit situations, the recommended solution is to effectively extend the depth of the beam by attaching a C6x8.2 70

78 steel channel to the concrete posts immediately below the existing concrete beam. The channel could be attached to the posts using concrete anchor studs. This would reduce the height of the opening to 9 in. (22.9 cm). Details of the retrofit design were previously shown in Figure 35. Typically, a retrofit design will have to be crash tested to evaluate its impact performance. However, it can be argued that crash testing of this retrofit design to the open concrete beam-andpost bridge rail is unnecessary for the following reasons. First, the concrete beam would carry most of the impact force so that the loading on the channel is expected to be minimal. Second, bridge rails of similar design, but with openings of 11 in. (27.9 cm) or less, have previously been crash tested and found to perform satisfactorily. There is no reason to believe that this retrofit design with only a 9-in. (22.9-cm) high opening would perform differently from the other similar bridge rail designs. Thus, a crash test to evaluate this retrofit design is not recommended. Bridge Railing TBR-2A CK A) The current concrete parapet with structural tubing design was found to be acceptable for test level 3 impacts. The researchers believe this railing can be upgraded to test level 4 (i.e., an 18,000-lb (8, 172-kg) single unit truck impacting the bridge rail at a nominal speed and angle of 50 mph (80.5 kmlh) and 15 de g) by increasing the size of the anchor bolts to 7/8 in. (22.2 mm) diameter and reducing the post spacing from 10 ft-6 in. (3.2 m) to 5 ft (1.52 m) center to center. However, it is recommended that the impact performance of the modified test level 4 bridge railing be verified with a full-scale crash test. Bridge Railing TBR-2B CK ) Analysis indicates that this bridge railing can be upgraded to test level 4 (18,000 lb truck) by increasing the anchor bolts to 7/8 in. (22.2 mm) diameter and reducing the post spacing to 5 ft (1.52 m) center to center. However, the impact performance of the modified bridge railing system at test level 4 should be verified with a full-scale crash test. 71

79 REFERENCES 1. Michie, 1. D., "Recommended Procedures for the Safety Performance Evaluation of Highway Appurtenances," NCHRP Report 230, Transportation Research Board, National Research Council, Washington, D.C., Mar Guide Specifications for Bridge Railings, American Association of State Highway and Transportation Officials, Buth, C. E., "Safer Bridge Railings," Vol. 1, 2, 3, and 4, Report No. FHWAlRD-82-0n, Texas Transportation nstitute, Texas A&M University, June Noel, J. S., Hirsch, T. J., and Buth, C. E., "Loads on Bridge Railings," Transportation Research Record 796, Transportation Research Board, Bligh, R. P., Sicking, D. L., and Ross, H. E., Jr., "Development of a Strong Beam Guardrail to Bridge Rail Transition," Transportation Research Record 1198, Transportation Research Board,

80 APPENDX A SUMMARY OF ANALYTCAL CALCULATONS TBR-A, TBR-B, TBR-2A, TBR-2B, TBR-3, TBR-4

81 r 12" l l"cl 1 10 "1 typ- - POST 8 ft c-c spacing 4-#8 fy = 60 ksi Ties 6" c-c Concrete f'; = 3000 psi b = 1 2" d = 8 1 /2" 2 #7 ~2"1.10"1 ~ -1/2 / 3"1 1 BEAM 4 - #7 Stirrups 6" c-c j fy 60 ksi 1" cl ty~ 1- b = 12" d = 8 1/2" " cl typ - ~ 2"f- 12" 27" 1 2 " r- 3"1 1 2" \ l-. 1 j V ~ 2" ) t 5" c-c 10" c-c POST STRENGTH kips (Shear Controls) BEAM STRENGTH kips (Shear Controls) STRENGTH OF TBR-1 A - 38 kips Y = 21 n. (Two span failure mode) SL-3 4,500 b/60 mph/25 deg in. TBR-1A A-2

82 f " " 2"1 1011\911-12" -211~ 15" 36" 9" \! STRENGTH OF TBR-1 B = 38 kips Y = 30 n. SL-3 4,500 b/60 mph/25 deg n. TBR-1B A-3

83 BEAM a = ANALYSS OF TBR-A 60x 2x 0.6 = 2.39".85x 3x 12 al2 = 1.20" arm = d -a12 = 7.3" <f>mn = <f>iyas (d -a2) =.9x 60x 1.2x 7.3 <f>mn = k-in. = 39.4 k-ft = <f>mn <f> Vc = <f> 2 i! bd =.85 x 2x 54.8x 12x 8.5 <f> Vc = 9,500 lb = 9.5 kips <f>vs = <f>a~(~) =.85X2X.05X60X(!:~) = 5.1 kips <f> Vn = <f> Vc + <f> Vs = = 14.6 kips BEAM = 29.2 kips Shear Controls NOTE: S =:; d/2 = 4.25" not 6" Recommend Stirrup Spacing S = 3" c-c <f> Vs = 5.1 (~) = 10.2 kips <f> Vn = 19.7 kips BEAM = 39.4 kips POST a = 60x2x.79 = 3.10 al2 = x3x 12 arm = d -a12 = = 6.95" <f>mn =.9x 60x 2x.79x 6.95 = k-in. = 49.4 k-ft P = <f>mn = = 28.2 kips n 21" 21 <f> Vn of POST = <f> Vn of BEAM = 14.6 kips = P" Shear Controls Post Strength NOTE: Ties are 6" c-c -- should be 3" c-c as in BEAM P n = 19.7 kips if this done A-4

84 TBR-A FALURE MECHANSM (See Fig. 1) (w) 8<1>Mn L-1/2 = 8 x 39.4 = 57.3 kips one span 8' -2.5' BEAM POST TOTAL STRENGTH + STRENGTH = STRENGTH NO. SPANS LENGTH kips kips kips 1 8ft = 29.2 shear 2 16 ft = 38.0 kips bending 3 24 ft = ft NOTE: This is a conservative analysis since a "static" failure-mode analysis is used and inertia resistance of the heavy concrete and high strain rates are neglected. Table indicates a horizontal impact force of 43 kips for a 4,500 lb vehicle at 25 angle and 60 mph (SL-3). When the barrier deflects, the 43 kip force will be reduced. This bridge rail should withstand an SL-3 impact. The front wheel will contact the post. The #2 stirrup and tie spacing in the beam and post should be reduced to 3" c-c which would increase the strength of the bridge rail to 43.1 kips. A-5

85 ft c-c POST = 20.2 kips (controled by anchor bolts) 5" <D Pipe RAL = 18 kips (one span) == o 0:: 2 3/4" 1/2" Base Plate 4-3/4" <D A307 Anchor Bolts N T 6" WALL - TOTAL kips 216 kips to 1'1") N '-... 1'1") """" -:.q. ' ,... 1'1") :c 13" N N '+- '+- W 1" cl 3" 3" 12" c-c 18" /2" fy - 60 ksi f~ psi TBR=2A A-6

86 PPE RAL TBR-2A 5 in. std. steel pipe S = 5.45 in. 3 O.D. = 5.56 in. Z = 6.92 in. 3 F6 = 35ksi <l>mn =.9x 35x 6.92 = 218 k-in. = 18.2 k-ft Span = 10.5 ft max. 8 <> Mn 8x 18.2 (wl) = L-l/ = 18.2 kips Rail Strength One Span POST - STEEL A36 Steel Fy = 36 ksi = 64.5 in. 4 S = 16.1 in. 3 Z = 18.2 in. 3 ALUMNUM (112" Alum.) (7/16" Alum.) Alum T6 Fy= 35 ksi <> Mn =.9 x 36 x 18.2 = K-in. = 49.2 K-ft P = = 49.2 kips moment n 12 in. <> Vn =.85x.55x 36x 8x.3125 <> Vn = 42 kips Shear ANCHOR BOLTS 4-3/4 in. <> A307 Bolts F u = 60 ksi Ult. = (.334 in. 2 x 60) = 20 kips <l>m =.9x20x2x7 in. = 252 k-in. P = 252 = 20.2 kips Post Anchor Bolts Control 12.5" A-7

87 CONCRETE WALL Top Beam a =.SSx 60 = 3.46" a/2 = 1.73".S5x 3x6 <l>mn =.9x.SSx 60 ( ) = k-in. = 32.S k-ft a =.55x60 = 2.16" a/2 = LOS".S5x 3x 6 <l>mn =.9x.55x 60 (10 -LOS) = 265 k-in. = 22.1 k-ft avg. Mb = 27.5 k-ft.33x60 a = =.44" a/2 =.22".S5x 3x 17.5 <l>mn =.9x.33x 60.x 13.S = k-in. =.9x.22x 60x 13.S = k-in k-in. avg. Mw = k-in. = 17.1 k-ft Cantilever.31 x 60 a = =.60S a/2 =.3.S5 x 3 x 12" k-in k-ft <l>me =.9x.31x 60 ( ) = ' = ft ft = Me A-S

88 (w) L = ( ) 22.6 = = 8.6 ft 8x2.75 8x x ( ) = = 198 kips = WALL The 22 in. high shaped concrete parapet should withstand an impact force of at least 198 kips. The 34.S in. high metal rail on top can only withstand an impact force of 18 kips. The total strength of this bridge rail is about 216 kips at a height of 23 in. This bridge rail could easily resist an impact of SL-3 (4,SOO b-60 mph-2s0) with an impact force of 43 kips with an effective height of 23 in. The rail could not withstand an impact of SL-4 (18,000 lb-so mph-so) with a force of S4 kips with effective height of 32 in. The metal rail is too weak at 18 kips. f the anchor bolts in the posts are increased to 7/8 in. diam. and the post spacing reduced to S ft center-tocenter, the metal rail could resist an impact of 48 kips and the bridge rail could resist an SL-4 impact. A-9

89 ft c-c POST = 20.2 kips (controled by anchor bolts) 12" " (j) Standard Pipe 2 3/4" 1 RAL = 18 kips (one span):= o a:::: 1/2" Base Plate 4-3/4" (j) A307 Anchor Bolts WALL = 158 kips N 361/4" 21 " 6" 4 1/4" 1" c 10 3/4" 5" 12" c-c o 3: N -j 3" SL-3 4,500 b/60 mph/25 deg in. TBR-2B A-O

90 CONCRETE WALL Top Beam a = LOS x 60 = 3.64".S5 x 3x7 a/2 = 1.82" <l>mn =.9 [.64 x 60 x ( ) +.44 x 60 x (6-1.82)] =.9 [ ] = k-in. Mb = 30.4 k-ft ~ h=12"=n d1 = 9.5" r d2= ~ t, b = 7" 0# in 2 L 0.44in2 0# in 2 As = 1.08 in 2 r--9"~ d = 7"----l a = a/2 = 0.5".60x.60.S5x3x " <> Mn =.9 x 0.60 x 60 x 6.5" = k-in. Mw = Cantilever 17.5 k-ft a.44x60 = ----.S5 x 3 x 12" =.86" a/2 =.43" jd = 7.07" <l>mn =.9x.44x60x7.07" = k-in. Me = (w) 14.0 k-ftlft L = 2.5' S X 1.75' ( ) 14.0 = 2.5' ' L = 9.86 ft Sx30.4 Sx x9.S S (9.S6-2.5) = (w) = kips WALL STRENGTH 0# in2 b = 14" 0 # in 2 0# in 2 As = 0.60 in 2 r-- 9" j t-- 1- d = 7.5" j b = 12" #60 L 0.44in2 As = 0.44 in 2 ~ A-ll

91 TBR-2B RAL - 1 Span P R = 18.2 kips hr = 33.5" 2 Span P~ = 7.9 kips hr = 33.5" POST - Spacing 10.5' Pp = 20.2 kips hr = 33.5" WALL P w = 158 kips h = 21" w P~ = kips h = 21" w mpact at Mid-Span = P R + P w = R = kips h = 158 x x 33.5 = 22.3" mpact at Post = P p + P~ + P~ = R = kips h = 20.2 x x x 21 = 23.3" P~ = 158 k x21"-20.2 k x33.5" = " A-12

92 Post W6x 16 6'-3" c-c Spacing,---12 gao W-Beam 21" Base t 11"xll"x3/4" A36 Steel 27" 4-7/8" A307 Anchor Bolts 5"min.1-6.5'~ 4 1 /2 "--t---+--=--=vt,'-'a 3.5" -7.0" r~i e=s-..---l 6" W-BEAM STRENGTH = 13.6 kips (one span) POST STRENGTH = 21.8 kips (anchor bolts control) STRENGTH OF TBR in. (one span) SL-1 4,500 b/30 mph/25 deg in. NOTE: The guardrail is located anywhere from flush with the curb to 3.5" set back. Therefore add dimension from face of curb to face of base plate (varies 3.5" to 7.0") and dimension from face of base plate to center of guardrail post (4.5"). TBR-3 A-13

93 ANALYSS OF TBR-3 W-BEAM 12 gao S = 1.37 in. 3 Z = 1.93 in. 3 <l>mp = 0.9 x 50 x 1.93 = 86.9 k-in. = 7.2 k-ft POST W 6 x 16 Zx = in. 3 F y = 36 <l>mn = O.9x 36x 11.7 = 379 k-in. 4>Mp 379. P = -- = -- = 26.6 kips arm ANCHOR BOLTS 7/8" <l>a307 Ult. = 38 kips Anchor Bolt Control <l>mn = O.9x 28 k x 6.5 in. x 2 bolts = k-in. P = <l>mn = = 21.8 kips = POST arm 15 One Span w = 84>Mp 8x7.2 L-/ = BEAM 13.6 kips one span No. Spans Length Beam + Posts Total ft = 13.6 kips ft kips kips The strength of this rail is 13.6 kips and compares very well with the impact force of 14 kips from Table for SL-l 4,500 lb, 30 mph at 25 angle. A-14

94 """ #5 #4 8" c-c g = 3 ksi fy = 40 ksi Concrete Post Strength = 14.5 kips Steel Post Strength = 26.6 kips POST Post Spacing 3'-1 1/2" c-c 12 x 18 Existing Concrete Post 12" ----t-- 9" " max. "-y r er go. W-Beam Post W6x 16 A36 3" max. r-+--+ logo. W-Beam D ~[) 21" 21",-~------~~-----~.-r 6" J -- 8" /8" (j) A307 Anchor Bolts Base t 11"xl1"x3/4" 6" SL-3 4,500 b/60 mph/25 deg in. TBR-4 A-15

95 ANALYSS OF TBR-4 POST W6 x 16 Same as TBR-3 Steel Post P = 26.6 kips ANCHOR BOLTS <> 4-7/8<1> A307 UZt. = 28 kips each Mn = 0.9x 28 k x 2x 8/1 = k-in. P = = 26.9 kips ls/ Post Controls EXSTNG CONCRETE POSTS i = 3 ksi b = 18 1/ d = 10/1 As = 2x.31 =.62 in. 2 iy = 40 ksi.62x40 a = = 0.54" al2 = Sx3x 18 <l>mn = 0.9x.62x40x9.73 = k-in. P = = 14.5 kids Bending ls/.. arm = 9.73/1 <> Vc =.8Sx 2x J3000x 18x 10 = 16.8 kips Shear Bending Controls 10 GA. W-BEAM S = 1.76 in. 3 Z = 2.48 in. 3 <l>mp = 0.9x2.48xSO = k-in. = 9.3 kip-it for 3 1-1~ length (w) = 8 <l>mp 8 x 9.3 = 119 kips Beam one span 2/1 L S-2.S No. Spans Length Beam + Posts = Total ' or 26.6 = 34.3 or ' = 52 kips A-16

96 The strength of this retrofit bridge rail is about 52 kips. Every other post is steel (P=26.6 kips) or existing concrete (P=14.5 kips) so the analysis is a little more complex than the others. The bridge rail TBR-4 should easily resist an impact of SL-3 (4,500 b-60 mph- 25 angle). The curb height was assumed as 6 in. f curbs are higher than this, strength of the posts and bridge rail will increase. A-17

97 APPENDX B NORTH CAROLNA BR-38 TEST AND EVALUATON

98 PagL- No. 4 08/28/90 Bridge Railings That Meet HCHRP 230 Criteria or Perormance Level Criteria Code Bridge Railing Test mpact mpact PE'rOrmance Railing Height Vehicle Speed,ng le Level / inches mph degrees Comments BR-33 OHO BOX BEAM 27 1,980 LB CAR RAL (W-BEAM 4,790 LB CAR BACKED UP WTH BOX BEAM) BR-34 KANSAS CORRAL, 2.7 1,971 LB CAR MODFED (OPEN 4,690 LB CAR CONCRETE BEAM & POST> BR-35 OKLAHOMA 29 1,980 LB CAR MODFED TR-1 4,660 LB CAR BRDGE RAL BR-36 NEBRASKA 32 1,970 LB CAR TUBULAR THRE 4,700 LB CAR BEAM BR-37 OREGON - 2 TUBE 32 1,994 LB CAR MOUNTED RAL 4,640 LB CAR (CURB MOUNTED) A ' J3R-38 NORTH CAROLNA 32 1,990 LB CAR STANDARD 1 4,660 LB CAR BAR METAL RAL 19,920 LB BUS BUS CONTANED, BR-39 CALFORNA TYPE 32 4,54tzl LB CAR (NJ CONCRETE 4,540 LB CAR SAFETY SHAPE) 4,540 LB CAR BR-40 CALFORNA TYPE 39 4,895 LB CAR (NJ SAFETY 4,895 'LB CAR SHAPE WTH 4,895 LB CAR' RAL> 4,895 LB CAR BR-41 CALFORNA TYPE 36 1,850 LB CAR ,530 LB CAR ( SEE-THROUGH, COLLAPSNG RNG) BR-42 NEVADA SAFETY 39 1,911 LB CAR SHAPE PARAPET 4,650 LB CAR ,.000 LB BUS BUT HOLLED ON TS SDE B-2

99 -... "'V,.. ll '"... '..... \.. oor l... _.....,. _... --",...,.. ~~,_ -,., ~_... " _ '-- _J..,,'''''.. "-1... _,. "_1'",...,,..,. _, ""_. A, to W.. ~~ ~ ~~..;~. _F1<O,{T cl.r.v.>.,1lq.!l QF.TA.JL S Of' PO ST CLAMP ~ WJl )'SSCMrA-Y """,11 ""~ OJ.o.n ''-'',,<,,,,.f,u,.),lc'" e-rtu.j(ro e-oj C 111'"00. }")rlo,. y,.;:',1\(jt. ~,U\'~,L""'\ (.O..(.J n ~~?f}.~-:j r....!... ""'., f--_... _!. ",' - _CLAl:1[~,.BARJ:1ETA)L Cl A'iOUfHO f"1!1'\ roc,'r HO...c.AJ...1,,,a.. =~ ':-... :J ~-' ::. ~.. ~ A...,...,.- _.",..._, ". L_.. '... n.. _ 1.."" L 0,.'" _,.. "" _,..,... - r_ " -""'",.. "...., _.., ".-' ,..,... _..., ,,,'..,... _.,.. """ _.. ""..., "-'-''''', NO... '...A "\ J...,-... _,'- -- _.....,-...,. ~-, -"''''1 _,.. ~-,~_... _.. _ , t_j>- _.." _, _1... J. t n, l, ; t.-- "V,....,... ~ V. _"',._..,_~...,., ';-1_...r... -,..,..."' _....., -."... -.L ~... ' "-'\..0 ~.;:::~ :-:-~--:':._-.....,._ ,.. ",,-... _., ~.,~ ,~ '-""-.,,,,,1._- :=: ~ -0--':':::: ::-:..~ " ,~.....L... 1 ~...,." _",..... ~..., ~ -:.:::~ : ~: w_... " 1_" _... ",... """U, 0.-. " _.. " -,,...., -.,-...-".,...,,"'" "'''-J....,. _.~,. _, :.~...':,~.~~": w... _.. _,..,_., 1-." '~ _"t,. "'_.. '" '...,.,-, ," b....,... _,.. _J..,'...,..... \001:_, ,... - U "4.... _ f,... 1,... -, '... '" n_'_... 1 " ---.,....., L...,...".,,. _... _,.. ",M" ::.:.::... :: 7...:::: w...,...1, _ "'" "",,: "_1"" ~'::...--:~,., ,... _,... -, "....._--..., -, _... ". -_1.._.., _ :.;: ~;.:.-~,~ _.1. P:::>ilrl C)JYXN.~ e-!'-'c..vc. 1 ~1.- 1J.J..P.J...,~ J ~ "'" ~- --, _., ~,... om.t~+-c-o)

100 ;::}/L:J ~ ~ / TENNESSEE RF 7166.~~ T_EX_A_S_T_R_A_N_S_PO_R_T_A_T_O_N_'_N_S_n_T_V_TE TRUCTURAL RESEARCH DVSON Area COC9 4G9 Telf'pllone R4~' 4414 ToxAn January 29, 1988 Mr. Charles F. McDevitt HSR-20 Federal Highway Administration Turner-Fairbank Highway Research Center 6300 Georgetown Pike McLean, VA RE: Pooled Funds Bridge Rail Study Contract DTFH-86-C Dear Mr. McDevitt: Transmitted herewith is our comparative analysis of the Texas T4 and North Carolina railing designs. For this analysis we have used the ultimate strength/yield line procedures applied to other railing designs evaluated in this bridge rail study. A summary of results is presented on the fi rst page. Compari sons of strengths of individual structural elements as well as strengths of the total structural systems can be made. Both railing designs consist of concrete parapets with metal beam and post rails mounted on the parapet. t appears that the metal rail and post portion of the Texas T4 ~esign is stronger than that in the North Carolina design and that the parapet portions of the two are virtually the same strength. Computed strength of the Texas T4 design is more than adequate for a performance level 2 strength test. When comparing the entire railing system, both the magnitude and height of the force should be considered and direct one-to-one comparisons are more difficult to make. For example, for an impact at a post, the North Carolina railing is expected to support 72.3 kips at a height not greater than 22.5 in. and the Texas T4 railing will support 66.1 kips at a height of 27 in. or less. Adjustments to allow comparisons at the same height for both rails are rather complex (and doubtfully appropriate) and serve only to cloud the issue. However, one simple way to equate the heights is to maintain a constant bending moment at the base of the parapet and let the height and magnitude vary such that Fh = N. f this technique is applied to the Texas T4 design, it would be expected to resist a force of 79.3 kips at 22.5 in. f the adjustments are made to the North Carolina THE TEXAS A&M UNVERSTY SYSTEM COLLEGE STATON, TEXAS B-4

101 Charles F. McDevitt -2- January 29, 1988 design, it would be expected to resist 60.3 kips at a height of 27 in. Computed strengths for impacts at mid-span of the metal rail indicate that the T4 design is slightly stronger. n summary, the North Carolina railing design has been tested to NCHRP Report 230 requ i rements and has been judged acceptable. The Texas T4 railing has virtually the same geometry and computations indicate that strength of the Texas T4 is about the same or sl ightly higher than the North Carolina railing. The Texas T4 railing would be expected to demonstrate acceptable performance in full-scale crash tests. We trust that this analysis and evaluation meets with your approval and await your comments. EB:pa Sin~~relY:Q ", r~/ J:/l Eu~ene Buth Pfincipal nvestigator B-5

102 Comparison of Texas T4 and North Carolina Bridge Rails North Carolina Texas T4 Rail Element 4 3/4"x4" Semi-Ellipse 6 7/8"x4 112" Semi-Ellipse One Span (PrJ 25.1 kips 35.6 kips Two Span (P'rJ 10.6 kips 16.1 kips Post 6' -6" Spacing 8'-4" Spacing Mp = ZxFy 16.6 kips 30.7 kips Parapet Wall Midspan (P w) 72.8 kips 71.8 kips Post (P'w) 45.1 kips 19.3 kips Resultant for R = 97.9 kips R = kips Midspan h = 21" h = 22.3" Resultant for R = 72.3 kips R = 66.1 kips Post h = 22.5" h = 27" For an impact at midspan, the two railings have approximately the same strength with the height of the resultant force on the T4 slightly higher. F or an impact at a post, the tow rails have approximately the same strength with the height of the resultant force on the T " higher. Both rails would be acceptable for PL-2 if snagging does not occur. B-6

103 B-7 12.%

104 /2/1 / 31. " 2.'h B" E,,-, syz. T yzl'l---t--t-----t-l B-8

105 APPENDX C STANDARD GUARDRAL TESTS AND EVALUATON

106 GUDE FOR SELECTNG, LOCATNG, AND DESGNNG TRAFFC BARRERS (l N 1977 American Association of State Highway and Transportation Officials

107 Table B 1 Continued Tllhlc B 1 Continued " -lj:l ' '! :[}; Melrlc Coover. ion, 1,,1 " t m n. 2~.4 mm l 16'. 0"!501''4'' mph' 1.61 km/hr '.- ~. b 04~4kg 0 U 0 :3()O;m,n 254 JTVn "C"POST 6 ~,. ~~1 Melnc Conver.iont m"" ' 61 km/r.- 'b 0454kg.". -" ~."... ' 6'0' ".~1:: ~~! 48 ", f] U 1 -- ) 1 ) SYSTEM r.411w) G4 (! S) Blocked-Out "w" Beam Wood Post) Blocked Out "\oj" (l!.'drn (Steel Post) BARRER DESCRPTON PosT SP/.tC1NG 6' J" 6' J" POST TYPE 6" x 8" Douglas. Fir ',15)(8.5 steel po<"t 8EAM TYPE Steel "\01" section, 11 GA Steel "W" sect lon, 12 GA 4 OFFSET BRACKETS 6" x 8" x 14" Douglas Fir BlOCk W6:o::8.5x 14" long steel block MOUNTiNGS 5/8" diameter,,,rria9e' bolts S/8" di"meter bolt FOOTNGS None None MPACT.MPlllCT MPACT 1MPACT2So MPACT P RFORMANCE ANGLE' 1\0 ANGLE' 14 0 ANGLf 1 \0 ANGLE ) MPACT CONDTiONS Speed (mph) NO 1(T 68.0 NO TET \6.8) Vehlclt' WeiQh! ( b ) (3813) BARRER DynamiC D.fleeTlon (fl) 1.J)) \) VEHCLE ACCLERAnONS(G',)1 lole rol,d 6./ ) LonOllUdlnQ 6./1 1./ ) Total UNA V UNA V UNAV) VEHCLE TRAJECTORY " EXit AnQle (deq ) A) Roll AnQle (deq ) >S 0 (UNAY) Plch AnO" (oeq ) UNM 0 UNAY) 1S' of "W" 2S' of "W" BARRER DAM AGE section and ')ect ion and 4 posts 3 post'), system 04(/1) 09 Slocked-Out "w" Sedm (Steel "C" Posts) illncke1s~~;1 "~hri~ Bedm" 8ARAE:R DESCRlPnON POST SPACNG 6' 3" &' l" POST TYPE 4 1!3",1 \/8".3/16" "C" steel post "(,xb.5 steel BEAM TYPE Steel "W" section, 11 GA hrle ijeilm, steel OFFSET BRACKETS 4 1/3" x5 5/8"d/16" "C" steel post 3 \1,'6(8.5 and M4d7.2, steel MOUNTNGS S/8" diameter bol t 2 SS" dlameter steel bol ts FOOTNGS None UNAV MPACT MPtl.CT MPACT MPACT MPACT PERFORMANCE ANGLE' 11 0 ANGLE 25 ANGLE 11 ANGLE' 1, MPAC T CONDTONS Speed (mph) NO TET 19, V.hl (; ~ W.Qht (b ) BARRER Oynamlc O.fltc:,jon (1 r.) 1,90 O.,R 1.10 VEHCLE ACCLERATOHS (G', Lol. rol JO 7,9(1 LonQltudinO 3./0 2.')1) 3.90 TolO' UNA\, UNfl' \JNAV VEHCL E TR AJECTORY E:lf An",le (d.q ) UNAV Le~s than Roll An",l. (dtq ) less th,)n Pilch AnQlf (d.g.) BARRER DAMAGE 21' of "W" 1/' 6" of 1/' 6" pf sect ion and thrle beam dnd thrie beam 5 posts 4 posts dnd 4 posts REFERENCES F ELD PERFORMANCE DATA2 NO 19 19, (B) System is similar to G4(lW) except See text for e!tplanhion of d~f~ for smaller posts and block out feresees in data sho... n for 25 and REMARKS si leo System performed well. 28,4 tests. Smooth redirection. H REFERENCES FELD PERFORMANCE DATA2 NO NO Smooth redi.-ection but with some- Smooth fedirecbion, ~6x8.S block 7~a~, hi ~~s ~~ \ ~d~n6! e~o ~ r~~;!~d t~~~m ~~ ~c~~~~t 1: o3!5 us~~s ~ n a~~om~:~i ~. 2 REMARKS Heel sheets, Both systems performed well. UNAV - unovoil obi"!!lom1111.eond o... roq. un1t.. otherwl.. not.d 2" ovolloblt... ummory n App.ndlx 1M,)/,lmul1 p("rmancnt dt>!lection T~~~t<j show that d."~'" section bo3c~-up pldte. ft. in length, must be placed behind rall elements at lntl'rmediate posts {non-splice posts). UNAV - unovoll ablt 'e.omilli.fcond ovtroo. un1... othtrwl.. Mild ~ f o'w'clloblt... lummory n App.ndlx Test show that a "w" section back~up plate. ft. in 1 ength, l1lu~ t be placerj bt'hlnd fail elpulcnts at intermedate posts (non~splice posts),

108 TECHNCAL REPORT STANDARD TTLE PAGE 1. Reporl No. 2 G nl Accession No 3. Reclpienl's Calalog No. FHWA/TX-87/405-2F 4. T rfle and Svbtttl~. ov~rnme _ ----~ ; ;:;_- :::_ USE OF GUARDRALS ON LOW FLL BRDGE LENGTH CULVERTS ~ 5. Report Dote Auqust Performing O'gonl lotion Code 7 Author( s: T. J. Hirsch and Dale Beggs 9. Performing Orgoni zotion Nome and Address Texas Transportation nstitute The Texas A&M University System College Station, Texas Performing Or90nllo.,00 Report No. Research Report 405-2F 10. Work Unit No. 11. Conlrac or Granl No. Study No Sponsoring Agency Nome and Address ~ 13. Type 01 Reporl and Peflod Covered Final _ September 1984 Texas State Department of Highways and Public August 1987 Transportation: Transportation Planning Division 14. Sponsoring Agency Code P. O. Box 5051 A1~tin Tpxa~ Supplemenlary Noles Research performed in cooperation with DOT, FHWA. Research Study Title: Guardrail on Low Fill Bridge Length Culverts 16. Abstract When multiple box culverts span over 20 ft, they are defined by AASHTO as bridge length and thus normally require the use of a full strength rigid bridge rail. The use of a rigid bridge rail creates a transition problem between the flexible metal beam guard fence whicb is commonly used upstream of the bridge rail. t would be safer and more economical to continue the flexible metal beam guard fence across the culvert even when the culvert length is over 20 ft and even when the soil fill depth over the culvert is less than the standard guardrail post embedment depth of 38 in. in Texas. t was believed that more post could be used with a shallow embedment to achieve the desired guardrail strength. A metal beam guard fence design of this type was crash tested in this study and proved to be unsatisfactory. Another concept investigated was to rigidly mount steel guard fence post (with blackout) to the top of the culvert deck when full soil embedment could not be achieved. A design of this type was also crash tested in this study and proved to be satisfactory. 17. Key Words Culverts, Bridge Rails, Guardrails, Longitudinal Barriers, Roadside Barriers. 18. Distribution Statement No restrictions. This document is available to the public through the National Technical nformation Service 5285 Port Royal Road Springfield, Virginia Security Classif. (of this report) 20. Security Classil. (of this page) 21. No. of Pages 22. Price Unclassified Unclassified 59 Form DOT F (8-69) C-4

109 10" 1"x1 1/2" slotted holes 3 1 3/4" f.-3 1/S" '-0" Min. W-beam guardfence 6'-3" post spacing.. ~ (\ ~ Q) ),.. Q) () ex) ~.. 0-0') 0 Q, Gi Q) -0') CJ) )( Q) ~ W6x9x 14" spacer or W6x8.5x14" 3/4" QA307 bolts with washers Length s slab thickness +2" 10"x6"x5/8" A36 steel plate Culvert Slab 1 114" formed or drilled holes, 8"x6"x 1/4" A36 steel plate with 15/16" ~ holes FGURE 12. DETAL OF STEEL GUARDFENCE POST AND ATTACHMENT TO CULVERT SLAB. C-5.

110 BTATC TEST RESULTS Test #1 load 39 Note: 7... f1 n...( :i '--' e 3 0 "" W6x9 steel post No al1ure or bend1ng o baseplates or bolts. Fallure due ent1rely to bending 1n post. Plot of Foro. vs. 0sploo.rn.nf B ~... -= --- ~ 5 ~J 2 1 / r "m 0 o 2. 8 O.fl.otlon ot lood (n.) FGURE 13. STATC LOAD TEST RESULTS FOR GUARD FENCE POST USED N CRASH TEST 3. C-6

111 20 18,-.. en Q..::J ~ >-!:: 12 u <C( 0- <C( u c 10 (') <C( l 0..J - 6 (f) " 18" 27" ~ :---...,::;,.\- Pos!.-+.20; 1. Ohe~~ '''' Post + Soil Cohesionlessl 38" \ SOL FLL DEPTH (tn.) 40 FGURE 14. ANALYSS OF GUARD FENCE POST LOAD CAPACTY FOR VAROUS SOL FLL DEPTHS.

112 0.000 sec sec sec n 00 Test No. Date Face Rail Post Post Spacing Length of Beam Rail Hax. Hax. Vehicle TAE SAE Dynamic Permanent Damage nstallation Deflection 2"05-3 Vehicle 7/8/86 Vehicle Weight 12 gao steel W-shape (w/instr.) W6x9 mpact Speed 1.9 m (6 ft 3 in) mpact Angle 53.3 m (175ft) Exit Speed Exit Angle 0.82 m (2.7 ft) Vehicle Acceleration 0.67 m (2.2 ft) (Hax sec. avg. ) Longitudinal 0RFQ" Transverse 01RFES35 Vertical FGURE 16. SUr~r1ARY OF CRASH TEST Cadillae DeV i 2019 kg (4450 b) 99." kmlh (61.8 mph 25.3 deg kmlh (37.2 mph 15.6 deg g "

113 APPENDX D SEQUENTAL PHOTOGRAPHS

114 0.000 s J= s ~ s s Figure D-1. Sequential photographs for test (overhead and frontal views). D-2

115 0.318 s s s s Figure D-l. Sequential photographs for test (overhead and frontal views continued). D-3

116 ~ - -.,.. - "li i' ~ s s s s s 0.39'3 s s Figure D-2. Sequential photographs for test (behind rail view). D-4

117 0.000 s s s s Figure D-3. Sequential photographs for test (overhead and frontal views). D-5

118 0.273 s s s s Figure D-3. Sequential photographs for test (overhead and frontal views continued), 0-6

119 0.000 s s s s s s s s Figure D-4. Sequential photographs for test (behind rail view). D-7

120 "".c.--~ s s s s Figure D-S. Sequential photographs for test (overhead and frontal views). 0-8

121 --=----- ='='= ~ -~~ ~ ~--~- -~ ~ s s - ~ ----~~-~-~- -== s s Figure D-S. Sequential photographs for test (overhead and frontal views continued). D-9

122 ----""""-==~ -~ ~~==-= ~ - --=--- --=~ =~-~~ - --~ - =-~ :~-~ ~=_~"=o c;;'"-::- ~ ~ ~-= Y' s s ~~--=-:~-~~ ~--=:--~ - --:-~ -=~ _.--~~ ==:;=_ ~_:~~_~OO -- ~ ~ ~ - ~ _~ _~ s s s s ~ ~::=------=- ~ ~- = ~-- - ~ ~ - - ~~-=---=;--= = =:~ ~ s s Figure D-6. Sequential photographs for test (perpendicular view). D-lO