Technical Report Documentation Page
3. Recipient's Catalog No.
5. Rep0rt Oate FEBRUARY 1990
5. Performing Organization Code
8. Performing Organization Repofl No.
10. Work Unit No.
11. Contract or Grant No.
HPR-PL-l(37) ITEM 114
13. Type of Report & Period Covered
CONSTRUCTION REPORT
14. Sponsoring Agency Code
1. ~eporNt O.
FHWA-U4802/8@Q3
15. Supplementary Notes
Prepared in cooperation wRh the U.S. Department of Transportation, Federal Highway Administration
18. Abstract
Arizona's canal network is extensive and necessitates the existence of many short bridges on the highway network.
The necessity for maintaining access to adjacent canal roads dictates that any barrier installation intended to shield
errant vehicles from the brldge rail hazard must fit within the limited space between the bridge end and the canal road.
The available space for such an installation is often less than 35 feet.
Energy Absorption Systems, Inc. (EASI) has developed two similar attenuating end terminals, the SENTRE system
and the TREND system, for use in such limited space applications. EASl has demonstrated that both their TREND and
SENTRE systems meet the dynamic performance requirements set forth in NCHRP-230. The conclusion that these
systems conform to dynamic performance specifications is based on full scale crash testing. The length of time,
however, that these devices have been formally monitored on highways is not sufficient for validating the adequacy of
in-service performance.
The objective of this research effort is to evaluate the in-service performance of the TREND and SENTRE anenuator
systems when installed on appropriate ADOT projects. Two construction projects, both lrmolving canal bridge rail
modification, were selected for test installations. This research effort embraces two separate experimental projects, and
hence two experimental project numbers. The SENTRE system was installed and reported to the FHWA as
Experimental Project Number US802 and theTREND system was installed and reported to the FHWA as Experimental
Project Number AZ-8803.
At the time that the TREND and SENTRE projects were constructed, both systems were classified as Experimental by
the FHWA. Although the SENTRE attenuator has since been upgraded to Operational status, both installations will
continue to be evaluated for the full two year evaluation period specled in the original workplan. Upon completion of
18. Distribution Statement
2. Government Accession No.
4. Title and Subtitle
SEMRE AND TREND ATENUATOR FIELD INSTALLATIONS
7. Author (s)
DOUGLAS J. LATTIN
9. Perlorming Organization Name and Addrew
ARIZONA TRANSPORTATION RESEARCH CENTER
ASU COLLEGE OF ENGINEERING - ERC405
TEMPE, ARIZONA 85287
12. Sponsoring Agency Name and Address
ARIZONA DEPARTMENT OF TRANSPORTATION
206 SOUTH l7TH AVENUE
PHOENIX. ARIZONA 85007
111 APPROXIMATE CONVERSIONS TO SI UNITS 111 APPROXIMATE CONVERSIONS FROM SI UNITS 111
Svmbol Whon You Know U u l l l ~ l vB v To Flnd Symbol
LENGTH
in 25.4 mlI'llnsUas IIM
k . lrsl 0.305 meUes m
2 Y a d 0.914 makes rn
mYes 1.61 klornel,sr krn
AREA
VOLUME
lo. Id -a 29 57 mUliliUss rnl
0.1 paknr 3 785 Wes L
k' c d r 1-1 0 028 mares crbed rn'
Y@ slar yandr 0 765 rnells crbed m'
NO=: V&ms wsllsr lhln I WO L be rhown in ma.
Symbol When You Know Mulllply By To Flnd Symbol
LENGTH
mm rnillimelrsr 0 03s inshes in
rn rnel#os 3.28 tee1 h
rn nhsl,es I 0 9 y~zdr
km klomelrsr 0 621 niYer
Yd
ml
AREA
VOLUME
mL millililcor 0 034
L hrer 0 264
rn' males cthd 35.315
rn' ma1184 c h d 1308
MASS
rquale ineher
square lsel
as,as
rquale miles
I1 01
gal
It'
yd'
MASS
0 ~ 1 ~ s ~ 20 35 glarnr
wndr 0 4% kuqprn~
shod lonr (20W b) 0 907 msglgrarnr
klognms 2 205 -.
pounds
megsglAmS I102
b
rhw) IMr (MOO b) T
TEMPERATURE (enact)
111
PI ' S16 the symbol la lhr hWnalanal Syslem d Uearuaentm . -.. - - - -. . -. . . .. . - - .. . . . - . .. - - . . - - .- - - .. .. .. - . .. . . . . -. .. . . .- -- . . -- (8rurod Apil 1089) . -- . - --
.C Cekius + 32 Fahrenhel 'F
TEMPERATURE (enact)
'F Flhrenhsl 5(F-32y9 CeWm .C
smparr~s ~omperat~.
Ismpcualure ternperat~e
.F 32 911 8 SF
"C 2I C
TABLE OF CONTENTS
INTRODUCTION ................................................................................................................................. 1
Background ................................................................................................................................ 1
Roadside Barrier Elements ................................................................................................. 1
Roadside Barrier Service Requirements ............................................................................ 1
Barrier Classification. .......................................................................................................... 1
Problem Statement ..................................................................................................................... 2
ADOT Standard End Treatments ........................................................................................ 2
Umited Space Available for Barrier System ....................................................................... 3
Alternate End Treatment ...................................................................................................... 3
FHWA Approved on Experimental Basis ........................................................................... 6
Objectives ................................................................................................................................... 6
Field Installation ............................................................................................................ 6
In8ewice Evaluation ........................................................................................................ 6
PROJECT LOCATION AND DESCRIPTION ..................................................................................... 6
AZ-6802: SENTRE Attenuator Field Installation ......................................................................... 6
Accident History .................................................................................................................. 7
AZ-6803: TREND Attenuator Field Installation ........................................................................... 8
Accident History .................................................................................................................. 8
CONSTRUCTION PLANS ................................................................................................................... 8
MATERIALS TESTING ......................................................................................................................... 8
DESIGN ............................................................................................................................................... 8
Requirements ...................................................................................................................... 8
Barrier Design ..................................................................................................................... 9
CONSTRUCTION. .... ............................................................................................................................. 9 General Act~l tyD escription ....................................................................................................... 9
Traffic Control ..................................................................................................................... 9
Concrete Footings .............................................................................................................. 9
SENTRE Translion Segment .............................................................................................. 10
Baseplates .......................................................................................................................... 10
Pand Support Posts ........................................................................................................... 10
Thrie Beam Panels and Sand Containers ......................................................................... 13
SENTRE Splice to Transition Segment .............................................................................. 13
TREND Splice to Concrete Parapet Wall ........................................................................... 14
Redirecting Cable ................................................................................................................ 15
Completed Systems ............................................................................................................ 15
Economics ................................................................................................................................ 17
Construction Costs -Alternates vs . SENTRE and TREND ................................................ 17
Cost Analysis ...................................................................................................................... 17
PREUMINARY EVALUATION .............................................................................................................. 18
Concerns and Potential Problems .............................................................................................. 18
Conclusions ................................................................................................................................ 19
REFERENCES .................................................................................................................................... 20
TABLE OF CONTENTS (continued)
APPENDIX A -CURRENT ADOT STANDARD DETAILS .................................................................... A1
APPENDIX B . EASl TECHNICAL DISCUSSION ............................................................................... B1
APPENDIX C . FHWA APPROVED WORKPLAN ............................................................................... C1
APPENDIX D -CONSTRUCTiON PLANS FOR AZ-8802 (SENTRE) ................................................. Dl
APPENDIX E -CONSTRUCTION PLANS FOR AZg803 (TREND) .................................................... El
LIST OF FIGURES
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
Figure 21
Figure 22
Figure 23
Figure 24
Figure 25
Breakaway Cable Terminal (BCT) ................................................................................. 2
Standard Attenuator Assembly ..................................................................................... 2
Canal Road Access to be Maintained ........................................................................... 4
Full LON Segment is Provided .................................................................................. 4
The SENTRE Attenuator System .................................................................................. 5
The TREND Attenuator System ..................................................................................... 5
AZS802 . SENTRE Attenwtor Installation Site ............................................................ 7
AZ-8803 -TREND Attenuator Installation Ste .............................................................. 7
SENTRE Concrete Footing ........................................................................................... 10
SENTRE Transition Segment ........................................................................................ 10
Drilling Bolt Hole ......................................................................................................... 11
Pouring Epoxy into Bdt Hole ........................................................................................ 12
Base Plate Hardware ..................................................................................................... 12
Hole at Bottom of Post 1 ............................................................................................... 12
Mushroom Bolt Assembly ............................................................................................. 13
Sand Container Configuration ..................................................................................... 13
Special Downstream Splice .......................................................................................... 14
Proper Upstream Splice ................................................................................................ 14
Bracket Attaching TREND to Front of Concrete Parapet Wall .................................... 15
Attachment of TREND Backstrap to Back of Concrete Parapet Wall ......................... 15
Front Anchor of Redirection Cable ............................................................................... 16
Rear Anchor of Redirection Cable ................................................................................ 16
Completed SENTRE ................................................................................................... 16
Completed TREND ........................................................................................................ 17
Panel Lap is a Concern on Departure Sie .................................................................. 19
INTRODUCTION
Background
Roadside Barrier Elements
Roadside barriers are highway appurtenances used to prevent vehicles from impacting rigid
or otherwise dangerous objects located within an established clear zone along a highway border.
These longitudinal barrier elements are designed to intercept and safely redirect errant vehicles
away from all potential hazards. Since roadside barriers will themselves be impacted by errant
vehicles, they are warranted only when their inclusion is expected to reduce the severity of a
potential accident at the installed location.
The total longitudinal barrier length required for shielding roadside hazards is called the
lengthofffeed (LON) and generally consists of a long standard guardrail section with a short
transition section near rigid objects. When the end of a roadside barrier falls within the clear zone
a crashworthy end terminal is installed.
Roadside Barrier Service Requirements
The criteria by which roadside barriers are evaluated are delineated in the National
Cooperative Highway Research Program report numbered 230 (NCHRP-230), RECOMMENDED
PROCEDURES FOR THE SAFETY PERFORMANCE NALUATION OF HIGHWAY APPURTENANCES'.
This report, published in 1981. is an expansion on previous puMication~~.3.~.d5a*6ti ng back to
1972; however, the qualitative performance criteria has changed very little from report to report.
The order of priorities in roadside barrier deslgn is safety first and economics second.
The dynamic performance of a roadside barrier system determines the safety of that system.
The primary measurements of dynamic performance are structural integrity, vehicle deceleration,
and vehicle postimpact trajectory. The structural integrity of a barrier system is determined by its
ability, in terms of strength and geometry only, to restrain vehicles in a predictable and acceptable
manner. Furthermore, vehicle deceleration levels should be minimized in order to reduce the
magnitude of interior compartment impact forces imparted on vehicle occupants. Finally,
redirected vehicles should not have postimpact trajectories in the direction of traffic regardless of
the inliai angle of impact.
While satisfaction of dynamic performance requirements is the first consideration when
selecting a barrier system, the optimal system is the one offering the greatest safety at the least
cost. in addition to the first cost for the actual barrier system, other costs to be considered are
system installation, routine maintenance and damage repair. These costs can be estimated with
varying accuracies to arrive at an economic measure for comparing different systems that satisfy
dynamic performance requirements.
Barrier ClassNication
The AASHTO GUIDE FOR SELECTING, LOCATING, AND DESIGNING TRAFFIC BARRIERS2
outlines three different stages of roadside barrier development. An operational barrier system is
one which has performed satisfactorily in full scale crash tests and has demonstrated satisfactory
in-service performance. Determination of satisfactory performance was somewhat subjective in
the past: however, the standardization of crash testing and performance evaluation provided by
NCHRP-230 has introduced objectivity into barrier classification. An experimental barrier system is
one which has performed satisfactorily in full scale crash tests and shows promise for satisfactory
in-service performance. This classification implies that a barrier system has performed adequately
under controlled conditions and will be classffied as operational only after a historical data base
showing adequate performance in an uncontrolled environment is established. An R & D barrier
system is one for which an insufficient amount of both crash testing and in-servlce performance
evaluation exists. No conclusions, positive or negative, can be drawn about this type of system.
In order to qualify for federal funding on any construction project involving the installation of a
barrier system classified as experimental by the FHWA, the responsible agency must agree to
monitor and report on the performance of that barrier in-service for a prescribed period of time.
The data collected from the performance of that barrier is then used by the FHWA in conjunction
with data from numerous other projects nationwide-to determine whether or not the system should
be upgraded to an operational status. When a barrier system is already classified as operational,
or when the project on which it is installed does not involve federal funds, no formal monitoring is
required by the FHWA
Problem Statement
ADOT Standard End Treatments
Although proprietary systems have been used in the past. ADOT currently has only two
standard end treatment designs. The two systems used are the standard Breakaway Cable
Terminal (BCT) and the standard Anenuator AssemMy. Figures 1 and 2 below show these two
systems. Typical ADOT details for both of these assemMies are contained in Appendix A of this
report.
Figure 1 Breakaway Cable Terminal (BCT)
Figure 2 Standard Anenuator Assembly
The BCT is a guardrail end treatment which provides the end anchorage required for
developing redirective forces when hi laterally as opposed to endon. The standard attenuator
assembly, often used as a median barrier, incorporates standard BCT features while providing
intentional attenuating capability when impacted end-on. The standard BCT was recommended
for field installation in NCHRP report number 129~A. system similar to ADOT's standard attenuator
assembly was Introduced in Transportation Research Record (TRR) report number 4Ei6. These
reports discuss the respective results of crash testing performed on each of these systems in the
early 70's. Both the BCT and the attenuator assembly systems appear to have performed in a
crashworthy manner although the criteria subsequently outlined in NCHRP-230 is not addressed
directly.
Typical details similar to those contained in Appendix A have existed and been used regularly
on ADOT projects since approximately 1974; however, no formal data base has been established
to suppon any conclusions about the in-sewlce performance of either system. Regardless of this
fact, both systems are considered operational when used in conjunction with other appropriate
roadside barrier elements so that dynamic performance requirements are predicted to be satisfied.
Limited Space Available for Barrier System
Arkona's canal network is extensive and necessitates the existence of many short bridges on
the highway network. The necessity to maintain access to utility roads adjacent to the canals
creates a condition requiring a modified approach to bridge rail protection. Figure 3 shows a
typical canal access road and the limited space avalabie for installation of a roadside barrier.
Longitudinal rail dements typically used on canal bridges consist of concrete parapet walls or
other rigid guardrail type barriers. An appropriate barrier configuration for shielding both the rigid
brklge rail and the canal would consist of a several hundred foot long LON guardrali segment with
an increasingly stiff transition segment near the bridge end and a standard BCT on the terminating
end. Figure 4 shows this full LON configuration used on a bridge very simiiar to the type used at a
canal crossing except that in this instance their is no canal road to consider.
The necessity for maintaining canal road access dictates that any barrier installation intended
to shield errant vehicles from the bridge rail hazard must fit within the limited space between the
bridge end and the canal road. The available space for such an installation is often less than 35'.
While this design constraint precludes provision of the full LON segment, the necessity for
providing some type of safety appurtenance meeting the stated dynamic performance
requirements still remains.
One approach to this problem is to forgo attempts at shielding the hazard of the canal itself
and to concentrate on shielding only the rigid bridge rail. A short longitudinal barrier installed
between the bridge and the canal road capable of safely slowing an impacting vehicle while
redirecting it away from the rigid bridge rail, but not into the highway or canal, would be
acceptable.
Alternate End Treatment
According to one safety barrier manufacturer, a device satisfying the requirements as stated
does exist. Energy Absorption Systems, Inc. (EASI) has developed two simiiar barrier end
terminals which function as both a transition section and an attenuating end treatment. These two
end terminals are the TREND which stands for TRansition END treatment, and the SENTRE which
stands for Safety barrier ENd TREatment. The systems are essentially the same except for the
dmerent type of rlgld barrier, or bridge rail in this case, that they are expected to shield. Aside from
the different rigid barrier elements that the two systems tie into, both systems consist of the same
components and behave in the same manner. EASI drawings of the SENTRE and TREND are
shown in Figures 5 and 6 respectively.
Figure 3 Canal Road Access to be Maintained
Figure 4 Full LON Segment is Provided
The physical composition common to both systems is as stated in lierature supplied by
EASi7,8 and summarized below. The SENTRE has fwe nested overlapping thrie beam panels
supported on vertical posts which are positioned on slip bases. For the TREND, there are six of
these panels. Slip bases consist of a top plate welded to the vertical support post and a bonom
plate anchored in a concrete footing. Both the top and bonom plates have open ended slots to
accommodate bolting together in a controlled breakaway manner. Subsequent to end-on vehicle
impact, the support posts wiil successively breakaway at the base level and progress realward
causing the attached fender panels to collapse in a telescoping manner. The first three posts
include sand containers of sufficient mass to dissipate the kinetic energy not removed by the slip
bases as they overcome frictional forces and separate. The first support post is also connected to
a redirecting cable which wiil guide vehicles away from the hard point of the rigid barrler being
shielded. The description provided here is ~dirnentarya nd the manufacture's much more detailed
technical discussion is contained in Appendix B of this report.
Figure 5 The SENTRE Attenuator System
Figure 6 The TREND Anenuator System
5
FHWA Approved on Experimental Basls
EASl prepared two certification reports7.* supporting their pronouncement that both the
TREND and SENTRE systems meet the dynamic performance requirements set forth In NCHRP-
230. Based on the reported full scale crash test results, the FHWA agreed that these systems do
conform to dynamic performance specffications; however, niether of these devices had been
formally monitored for a sufficient period of time to validate the adequacy of in-se~ce
performance. As a result, at the onset of this experimentai effort, the TREND and SENTRE
attenuator systems were approved by the FHWA on an experimental basis only.
Objectives
Field Installation
The objective of this research effort is to evaluate the in-sewlce performance of the TREND
and SEMRE attenuator systems when installed on appropriate ADOT projects. Two construction
projects, both involving canal bridge rail modification, were selected for test installations. The
TREND system was installed on a project where a concrete parapet type bridge rail was
constructed. The SEMRE system was installed on a different project where a tubular thrie beam
bridge rail was constructed.
In-Sewice Evaluation
Installations at both these project locations will be evaluated in accordance with the
procedures for in-service evaluation as presented in NCHRP-230. The systems will be monitored
for a period of two years at which time it is anticipated that a recommendation wll be made
regarding the use of these systems on Arizona highways. The formal FHWA Approved Workpian
developed for this research effort is included in Appendix C of this report. The workplan delineates
the evaluation procedures to be followed.
The reason for field installatbns is not to assess a banier's dynamic performance under
severe impact conditions. In an uncontrolled environment this type of analysis is precluded by the
lwei of monitoring required for a useful quantitative crash evaluation. In the event of an in-service
Impact, the dynamic performance of a barrier can only be assessed in a qualitative manner. The
primary reason for an in-service evaluation period is to alleviate u~nticipatedp roblems and design
deficiencies that may only be manifested during construction or operation. if such deficiencies
exist, affected departments can propose system modifications for improving operations or
lowering costs prior to wide spread implementation.
PROJECT LOCATION AND DESCRIPTION
This research effort embraces two separate experimental projects. and hence two
experimental project numbers, in accordance with the requirements stated in the FederaIAld
Highway Program Manuaia. The SENTRE system was installed and reported to the FHWA as
Experimental Project Number AZ-8802 and the TREND system was installed and reported to the
FHWA as Experlmental Project Number AZ-8803. Figures 7 and 8 show the two project sites prior
to construction.
a-8802: SENTRE Atlenuator Field Installation
Four SENTRE attenuators were Installed on ADOT construction project HES-022-2(33)P,
Wickenburg-Phoenlx Hwy (US 60) which is located at approximately milepost 138.0 on US 60. The
1989 design ADTIO for this two lane highway is 8,309 of which 62% are passenger type vehicles.
Among other safety related activitles, the work on this project included retrofitting a new railing
System on the Beardsley Canal bridge. This involved removal of the existlng concrete curb down
to the bridge deck level and reconstruction of a new curb configuration wide enough and stout
enough to accommodate a tubular thrie beam bridge rail assembly. The existing 40' clear roadway
width was maintained.
Figure 7 AZ4802: SENTRE Anenuator Installation Sne
on US 60 Prior to Construction.
Figure 8 AZ-3: TREND Anenuator Installation Site
on US 89 Prior to Construction.
The total bridge span across the canal is 20.5'. The new tubular thrie beam bridge rail has
four support posts anchored in the new concrete curb. The average distance from each of the
rigid bridge rail ends to its respective canal access road is approximately 37.5' and this is the total
distance available for installation of a protective barrier system. The SENTRE was selected since
this assembly has a 17.5' total length, allowing for an additional 20' transition section to make up
the difference. The transition section consists of the same tubular thrie beam as the bridge rail:
, however. the support posts are anchored in soil rather than concrete. The result is that the rigidity
, of the transitlon sectlon falls between that of the bridge rail and that of the SENTRE. Photos of the
, completed installation are contained later In this report.
Accident History
An accident record listingw for US &nIea r MP 138 spanning the period 1973-1988 was
obtained from the ADOT Traffic Studies Branch. During this period there were 55 reported
accidents. The majority of these were livestock related. Of the 55 accidents. 13 involved some
sort of collision with the existing bridge mil and 6 involved injury. The highway is flat and the view
is unobstructed. Immediately east of this project is a tavern and a convenience store presenting a
potentlal merglng hazard.
m-8803: TREND Attenvator Field Installation
Four TREND attenuators were installed on ADOT construction project F-081-1 (2), Oracle Jct-
Florence Hwy (US 89) which is located at approximately miiepost 132.6 on US 89. The 1989
design ADTIO for this two lane highway passing through the town of Florence is 2.565 of which
57% are passenger type vehicles. Along with milling and overiaying the existing roadway, the work
on this project included retrofiiing a new railing system on the Casa Grande Canal bridge. This
lmrolved removal of the existing concrete curb and attached W-section bridge rail, and
construction of a new concrete parapet wall type bridge rail. The exlstlng 40' dear roadway width
was maintained.
The total length of the new parapet wall is approximately 100'. leaving an average distance of
70' on each side for the canal access rcad and the attenuator. The TREND was selected since this
assemMy has a 21' total length, still leaving a typical canal access road entry way approximately
49' wide after installing the anenuator. Photos of the completed installation are contained later in
this report.
Accident History
An accident record for US 89 near MP 132.6 spanning the period 1973-1988 was
obtained from the ADOT Traffic Studies Branch. During this period there were only 3 reported
accidents. The canal bridge is located just north of the convergence of US 89 and SR 287 and
approximately two tenths of a mile south of the developed portion of Florence. The accident
record listing showed a much larger number of accidents than is cited here; however, all but three
were in the adjacent developed area and not at the canal bridge.
CONSTRUCTION PLANS
Plans for the SENTRE and TREND pmjects are contained respectively in Appendix D and
Appendix E of this report. These plans have been reduced and should not be used for scaling
distances. These copies are not as-builts and do differ slightly from the actual field installations.
The plans for the Beardsley Canal project incorrectly refer to the TREND system. The TREND
system was originally specified for both projects untii realizing that the SENTRE system was
required at the Beardsley Canal bridge due to the thrie beam type bridge raii specified for use on
that bridge.
MATERIALS TESTING
The only material tests applicable on these projects are the concrete footing compressive
strengths and the soil classification of the soil supporting the SENTRE's transition section. The
concrete compressive strength will affect the anchorage of the support post slip base plates and
redirecting cable. The soil properties on the SENTRE project will affect the rigidity of the transition
section. At the time of this report, only the concrete compressive strength test results were
available. The actual 28 day compressive strength for the SENTRE footing concrete was 4829 psi
and for the TREND was 5839 psi.
DESIGN
Requirements
The 1977 AASMO Guide2 was used to determine the LON required on a typical canal bridge.
The design data for the Beardsley Canal project was used. The bridge rail must be protected, but
the required length-of-need is governed by the additional requirement of keeping vehicles out of
the canal should they become enant earlier in their approach to the bridge.
Figure Ill-A-3 of the AASMO Design Guide indicates that for a design speed of Wmph,
obstacles less than 30' from the travelled edge of the madway must be shielded. The travelled
roadway edge at the Beardsley Canal bridge is 8' from the bridge rail which is less than the
required 30' clear zone distance; therefore, a roadside barrier is required. Furthermore, Figure III-A6
of the AASHTO Guide indicates that since the 40' bridge width is less than twice the required
30' clear zone distance, barriers are requlred in both directions on both the leave the approach
ends of the bridge.
Section Ill-E-l of the AASHTO Guide outlines the variables of interest and the method for
calculating the LON dimension. Projected ADT design data, a 60mph design speed, the fact that
no flare is provided, and a minimum allowable unobstructed clear zone of 30' is sufficient
information for determining the required LON dimension. Equation Ill-E-1 of the AASHTO Guide
yields a LON dimension parallel to the roadway of 293'.
As discussed earlier, only 37.5' was miable for a barrier system. Because of this constralnt,
provision of the MI LON segment was exduded and oniy the rigid bridge rail was shielded. Having
protected the most severe hazard, no further safety provisions were considered necessary by the
designers. Additional guardrail could have been specified on the other side of the canal access
roads up to the full LON; however, any projected gains in safety from such action would not have
compensated the added expense.
Barrler Design
The SENTRE and TREND are proprietary barrier systems. ADOT designer's are oniy
responsible for specifying use of these systems and any available options. EASi engineers are
responslble for the actual design of the system. Shop drawings with details applicable to the
specific project are provided by EASl and sealed by a mechanical engineer registered in the state
of California.
CONSTRUCTION
The activities invdved in constructing the SENTRE system are nearly identical to those
invdved in constructing the TREND system. The forthcoming activity description will therefore be
considered typical of both systems. Construction differences that do exist between the two
systems will be described as they occur.
General Activity Description
Traffic Control
The traffic contrd specified on both of these projects was functionally identical. The actual
work area was protected by temporary concrete barrier wall located a few feet away from the new
bridge rail location, thus narrowing the road. The concrete barrier was tapered toward the
shoulder over a 75' distance beyond the approach and leave ends of the bridge. On the approach
side, additional Type II barricades were tapered over a distance of 500' beyond the concrete
barrier from the shoulder stripe to the edge of the pavement. The traffic control plans for each
project, as specified by ADOT's Traffic Design Services, are contained in the construction plans
included in Appendix D and Appendix E of this report.
During construction of the SENTRE systems there was one reported accident which involved
the traffic contrd. According to the accident record listing, a semi-truck ran off the traveiied
roadway and struck the temporary concrete barrier wall. The DPS report attributed this non-injury
accident to driver inattention. The attenuator installation had not been started at the time of the
accident.
Concrete Footings
The attenuators were constructed on concrete footings specified as Class S concrete with
f '~=4000psi. The footings provide anchorage to the redirecting cable and the bottom half of the
. slip bases and are of sufficient mass for preventing overturn. The footings were finished smooth
and level with the roadway surface.
The footing dlmensions for the TREND are shown on both sets of construction plans. The
SENTRE dimensions are not shmn because the TREND was originally specified on both projects
as discussed earlier. The TREND footing is 21' long. 4' wide, and 8" deep except in the last 3' on
the end away from the bridge where the depth is increased to 3' to accommodate embedment of
the redirection cable front anchor. The SENTRE footing is the same except that the length is only
17.5' because this system uses five posts compared to the TREND'S six posts. Both footings have
two mats of longitudinal and transverse reinforcement as shown on the plans.
The cabie front anchor was the only hardware required to be embedded in the footing. The
base plate bolts can be embedded in fresh concrete or epoxied in at a later date. The SENTRE
system also has an option for downstream tensioning. If this option is selected by the instaillng
agency, the footing is thickened near the fourth post and a hook is anchored at that location,
allowing for the introduction of an initial tension to the guardrail beyond the attenuator. This option
was not used on the SENTRE installation featured in this repoR
Figure 9 shows one of the SENTRE concrete footings with Mlly the redirecting cabie front
anchor embedded.
SENTRE Transltion Segment
A 20' transition segment was used befween the SENTRE system and the tubular thrie beam
bridge rail. Swen posts with the same cross section and spacing as those to be used on the
SENTRE were driven in the soil spaced at 3' O.C. in the transition section as shown in Figure 10.
Figure 9 SENTRE Concrete Footing
Figure 10 SENTRE Transition Segment
Base Plates
The SENTRE has five base plates and the TREND has SIXT.h e bottom half of each base plate
is anchored in the concrete footing by sk 7.5' long 3/4" diameter bolts. The layout of the base
plates is critlcal since it determines the alignment of the attenuator and the ease of assembly. The
base plate locations relative to the footing edges and to the new rail systems where not clearly
defined on the project constructions plans; however, this information was shown on the EASl shop
drawings. Base plates were spaced 36" O.C. for the SENTRE and 37.5' O.C. for the TREND. The
attenuator alignment was specified to be straight and not flared on these projects.
The plate locations were layed out on the hardened concrete footings and the bolt hde
locatlons were determined using the plates as templates. Bdt holes with 7/8' diameter were
drilled 6' deep and the debris was blown out of the hdes with air. Washers and hex nuts were then
screwed onto the bolt ends so that an acceptable amount of thread would be left on the bdt ends
when installed. A two component epoxy. which is part of the SENTRE and TREND packages, was
mked and poured into the drilled holes. The bolts were then pushed by hand through their
respective base plate holes into the epoxy flled hdes and allowed to set for two hours. After the
epoxy set the six bolts were tightened with the specified 120 ft-lbs of torque.
Rather than epoxying the bolts into the footing as was done on this project, another option is
to set hook ended bolts in the fresh concrete. This option was not chosen for these projects since
the epoxy method requires less precision and allows for layout modification. The drilling and
epoxying activities do consume a considerable amount of time, however, and an alternate
anchoring method may be considered on future installations.
The drilling and epoxying activities are shown in Figures 11 and 12 respectively.
Panel Support Posh
Prior to attaching the posts, additional hardware is required on the slip bases. On each slip
base four 2.5" iong 3/4" diameter bolts were positioned in the slotted holes with washers and a bolt
keeper plate as shown in Figure 13. The suppon posts also have slotted 1/2" thick plates welded
to their bases. The Wo plates were bolted together with a specified MI ft-lbs of torque. Insuring
proper torque Is impoltant since this determines the amount of energy dissipated by the slip base
breakaway action in the event of a longitudinal collision.
The actual support posts consist of 32" long W6.5~9 A36 steel posts to which an additional 21"
iong W6.5~9 steel blockout is anached with two 3/4" bolts. All posts are interchangable except the
front post which is designated as Post 1, and the next two posts designated as Post 2 and Post 3.
Post 1 has a hole in the bottom to accommodate passage of the redirecting cable as shown in
Figure 14. Post 1, Post 2 and Post 3 ail have holes to attach sand containers. All posts have two
holes for attachment of a backstrap, but these are only used on the TREND system.
Figure 11 Drilling Bolt Hole
11
Figure 12 Pouring Epoxy into Bolt Hole
Figure 13 Base Plate Hardware
Figure 14 Hole at Bottom of Post 1
12
Thrie Beam Panels and Sand Containers
One thrie beam panel was bolted to each post blockout such that adjacent paneis overlap.
The overlap is away from the direction of travel on approach side installations and just the opposite
on corresponding departure side installations. The panels were set such that their top edges were
32 abwe the footlng. Adjacent panels were connected together through horizontal slots by a
mushroom bolt to allow a telescoping action upon longitudinal impact. The mushroom bolt
attachment Is shown in Figure 15.
Sand containers were attached to each of the first three posts. Posts 1 and 2 each support
two 1M) pound containers and Post 3 supports two 150 pound containers Figure 16 shows the
sand container arrangement. Each container was filled to the top with dry sand and the container
Ild was secured. The sand container lids snap tightly shut by hand.
Figure 15 Mushroom Bolt Assembly
Figure 16 Sand Container Configuration
SENTRE Splice to Trimsition Segment
The fiih panel of the SENTRE system must tie into the transition segment. in this case the
transition segment is a tubular thrie beam which becomes singular at the end to accommodate a
splice with another section of thrie beam. In order to accomplish a clean splice. care must be
taken during design and construction to insure that the singular end portion of the transition
section is situated such that its concavity is compatable with the concavity of the SENTRE singular
thrie beam panels.
This special requirement was nd forseen in the construction of this field installation. The
upstream ends of the tubular thrie beam matched up properiy with the SENTRE panels; however,
the downstream ends did not. On the downstream ends the thrie beam panels had their convavity
pointing opposite that of the SENTRE panels This situation required a special splice on the
downstream ends which is best illustrated by Figure 17. For comparison, the proper splice
configuration is shown in Figure 18.
Figure 17 Special Downstream Splice
Figure 18 Proper Upstream Splice
TREND Splice to Concrete Parapet Wall
TheTREND end panel was spliced to the concrete parapet wall by a bracket assembly which
was anchored at the end of the wall with two 314' diameter bolts as shown in Figure 19. The steel
strap that runs along the back of the TREND was also anchored in the concrete parapet wall as
shown in Figure 20.
Figure 19 Bracket Attaching TREND to Front of
Concrete Parapet Wall
Figure 20 Attachment of TREND Backstrap to Back of
Concrete Parapet Wall
Redirecting Cable
A 23 foot steel cable with threaded ends was supplied as part of the system. The cable was
passed through the hole in Post 1 and bolted to the front anchor as shown in Figure 21. The cable
was then extended to the rear anchor location forming an approximate angle of 25" with respect to
the roadway. A hole of approximate dimensions 2' square by 4' deep was then dug at the location
of the free end of the cable and the specfiied reinforcement was placed. The hole was then filled
with concrete into which the rear anchor was imbedded as shown in Figure 22. After the specified
concrete strength was reached, the cable was tensioned by applying the specified 100 ft-lbs of
torque to the tightening bolt.
Completed Systems
The completed systems are shown in Figures 23 and 24. The individual Districts in charge of
maintaining these systems have enhanced their respective systems with reflective stickers and
signs as shown. The time required for a two man crew to install the four SENTRE units on
completed footings was approximately three days. This works out to be 12 manhours per unit.
Construction of the TREND units was faster since a larger crew was used.
Figure 21 Front Anchor of Redirection Cable
Figure 22 Rear Anchor of Redirection Cable
Figure 23 Completed SEMRE
Figure 24 Completed TREND
Economics
Construction Costs -Alternates vs. SENTRE and TREND
In the design stage, a decision was made to concentrate efforts on shielding only the rigid
bridge rails so that the canal roads would not be blocked. If the restriction of maintaining access
to these roads was removed, the full LON guardrail wouid have been provided and the SENTRE
and TREND would not have been necessary. Based on 1987 construction costs published by
ADOT'~, the average cost per foot ot installed W-section guardrail on projects containing more
than 1000' of rail was $10.69/lf. The average cost of an installed BCT based on all projects
involving installation of 4 or less BCTs was $887.05. As previously discussed. the total required
LON on these projects was 293'; therefore, this length of guardrail with one BCT on the end wouid
cost approximately $4019 installed.
Another alternative which conceivably could have been employed if redirection subsequent to
head-on impact was not a primary objective is the standard ADOT anenuator. These devices wiii fit
within the restricted space, however, their performance is different than that of the TREND or
SENTRE. Based on the average of the three low bids on the single ADOT standard attenuator
installed in 1987, the cost per anenuator would be $3558 installed.
The cost per unit for the SENTRE system was $7421 installed and for the TREND was $8600.
The approximate portion of this attributable to the steel and plastic anenuator components was
$2500. The footing volume is approximately 4CY so using a consewative estimate of $100/CY in-place,
the footing material is another $400 bringing the material total to $2900. The result is that
several thousand dollars in excess of actual cost was bid on each attenuator. Since no special
expertise is required for installation of these attenuators and one unit can be installed in an average
of 12 manhours, the differential seems high. This high price is in part the result of including traffic
control costs in the price of the anenuators.
Cost Analysis
Cost analyses are inconclusive until consideration is given to the functional differences
between the TREND and SENTRE type systems and the systems ADOT usually employs. The
TREND and SEMRE should not be compared directly with the BCT and standard ADOT anenuator
on a cost basis because their applications, while similar, are not the same. These new systems
were chosen due to the special circumstances of the projects on which they were installed.
The alternate design which provides the full LON and uses a BCT on the end is clearly
desirable from a first cost standpoint, but the design objectives are not met. The canal road
access is not maintained. Implementation of this alternative would require realignment of the canal
roads and in addition to being expensive, wouid often not even be feasible.
The alternate design which employs the standard ADOT anenuator is less expensive, but
again all design objectives are not met. For vehicie impacts that are not end-on, the standard
attenuator and the TREND and SENTRE systems wili provide similar redirective capability. This
statement may not be true for end-on impacts. Crash testing of systems similar to the ADOT
standard anenuatoP appears to lndlcate that vehides impacting end-on wiil be stopped within the
systems 25' longitudinal dimension at 60mph (59 deceleratlon). However, if the impacting vehicie
has sufficient momentum to continue beyond the entire 25' anenuator length, the vehicle wili
impact the hard point of the bridge mil. The TREND and SENTRE units are designed with a
redirecting caMe intended to turn vehicles impacting end-on away from the hard point. Thls is an
added safety feature which the standard ADOT attenuator does not have.
Clearly the cost analysis is complex. Weighting factors must be assigned by the installing
agency which reilect the reiative Importance of each of the design objectives. The objective of
keeping existing canal roads clear wili take on different levels of importance depending on the
available alternate options in a given situation. The objective of redirecting end-on impacting
vehicles away from the hard point may achieve extremely high priority when considering potential
litigation costs that may result from not providing such an added safety feature.
Maintenance costs wiil also be a consideration In determlning the reiative economy of
systems. Maintenance data for the TREND and SENTRE systems are not available but wiil be
cdiected throughout the length of these projects. Consideration will be given to the availability of
replacement parts in a limited application environment such as this and compared to a projected
full scale implementation situation where parts can be purchased in bulk and stock piled.
PRELIMINARY EVALUATION
Concerns and Potential Problems
The actual construction of both the SENTRE and TREND systems was completed without
difficulty. ProMems and concerns did exist, but were not construction related. The first potentiai
problem was related to the specification of the SENTRE system on the Beardsley Canai bridge. As
previously discussed, the TREND system was incorrectly specified on both projects. The SENTRE
system was later specified on the Beardsley Canal project by way of change order and the
downstream tensioning option of the SENTRE was not included.
Exciusion of the downstream tensioning option does not represent a design deficiency
alhough it did create some confusion. The final determinatlon was made by ADOT's Highway
Plans design engineers that the downstream tensioning option was unnecessary due to the rigidity
of the thrie beam bridge rail system being used on that project. Other situations may warrant use
of the downstream tensioning option. Regardless of this decision, the apparent ability of the
SENTRE to provide adequate end anchorage to the connecting guardrail system will be assessed
in the event of future vehicle impacts.
The second concern which was expressed by ADOT District 1, District 2, District 3 and
Highway Plans had to do with an inherent feature of the SENTRE and TREND systems. In two way
highway applications such as these where anenuators are installed on both the approach and
departure ends of bridges, a potential problem is created. As discussed previously, the thrie beam
fender panels of the system are anached to their posts with a lap so that telescoping of the panels
will occur upon longitudinal impacts. The lap is away from the direction of traffic on the approach
end, but is toward traffic on the departure end. The departure end panel lap is shown in Figure 25.
Having the panels lap toward the trafflc may be hazardous to vehicles barely brushing the
departure end attenuator. Vehicles which may only experience minor damage in a brush by
encounter with an anenuator where the lap is away from traffic have potential for snagging on an
attenuator where the lap is toward traffic. Any effects of this feature will be evaluated on these
projects.
Figure !2!j Panel Lap Is a Concern on Departure Side
Conclusions
Between the tlme that the TREND and SENTRE attenuators were installed and the time that
thls report was prepared a significant change of roadside barrier classification has occured. In
April of 1989 the SENTRE attenuator system status was upgraded from Experimental to
Operational by the FHWA Office of Engineering. Regardless of this change, the Experimental
Project currently at hand will continue and all previous commitments concerning this project will be
satisfied.
As of January 1990 there have been a total of four Impacts with the attenuator systems In
place. All of these impacts have apparently been by heavy construction vehicles working on
adjacent construction projects and are clearly not representative of the type of impacts for which
the TREND and SENTRE systems were designed. However, the resulting damage has required
repair which will facilitate In assessing the maintenance features of these anenuators.
No data or condusions will be presented at thls point in the investigation of the SENTRE and
TREND systems. The evaluation of both of the systems will be carried out in accordance with the
workplan contained in Appendix C of this report. This workplan covers both the SENTRE and
TREND evaluation activiles and has been adhered to thus far.
REFERENCES
Michie, Jarvis D., 'Recommended Procedures for the Safety Performance Evaluation of
Highway Appurtenances.' NCHRP Report 230 (March 1981).
American Association of State Highway and Transportation Officials, 'Guide for Selecting,
Locating, and Designing Traffic Barriers.' (1 977).
Michie, Jawis D., and Bronstad, M.E.. 'Location, Selection, and Maintenance of Highway
Traffic Barriers.' NCHRP Report 118 (1971).
Bronstad. M.E., and Michie, Jarvis D., Recommended Procedures for Vehicle Crash
Testing of Highway Appurtances.' NCHRP Report 153 (1974).
Michie, Jawis 0.. and Bronstad, M.E.. .Guardrail Crash Test Evaluation. New Concepts
and End Designs.'NCHRP Report129 (1972).
Bronstad. M.E.. and Michie, Jawis D., 'Development of a New Median Barrier Terminal."
Transportation Research Record 488 (1974). P. 24-33.
Energy Absorption Systems, Inc., 'TREND (Transition End Treatment) NCHRP 230
Cenification Report.', (December 1985).
Energy Absorption Systems, inc., "SENTRE (Safety Barrier End Treatment) NCHRP 230
Cenification Repoft.', (May 1983).
Federal-Aid Highway Program Manual, Volume 6, Chapter 4, Section 2.
1989 Traffic Design Data, ADOT Materials Pavement Services
Accident Location identification & Surveillance System (ALISS), ADOT Traffic Studies
Branch
Construction Costs 1987, ADOT Contracts and Specifications
APPENDIX A - CURRENT ADOT STANDARD DETAILS
66"Cn"L "0,CJ .-,.&.., "Rr,. L..CL.
N I X
1
liu*r.rf'".",m I-,.., w u a,
P",, n.*n , . ,
- *..,. ."A w ""a .a I,,. . *C,..Y.,~ I.., 0, .."'
t.w 0
nr,m LII
1 m u w..*.,,YI.,. ""r,, .,el_-
I I I W , , M
ONE W4r r.TCZ-IrrT IIO€ W R040141
D.. @.W.#, I ..r*. Per, .,.."' . ,..*,."
hr. I.,,1 D M l
II... M1 8 0.w. I I * C I T"-! h* Om,,# ur
E T * H E
,*.*c I* .r< CJ
L7Sr.a Y msr all* m
twn. ' r m r wcr
WI1( J
OrCE' m50
(PI,'" IIXl
APPENDIX B - EASl TECHNICAL DISCUSSION
TECHNICAL DISCUSSION
TEST CONDITIONS
Test F a c i l i t y
The Energy Absorpt ion S y s ~ ~ nInsc . t e s t f a c i l i t y i s l o c a t e d a t t h e L i n c o l n
A i r p o r t i n Lincoln. California. The t e s t area i s situated on f u l l y asphalted
l e v e l ground and has been cleared of a l l obstructions f o r u n r e s t r i c t e d t r a j e c t o r y
o f t h e v e h i c l e . The s o i l i s composed o f very s t i f f to hard s i l t s and clays and
can be c l a s s i f i e d as a NCHRP 230 type 5-1.
Test A r t i c l e (Design)
The SENTRE (see Figure D-1) has been designed and constructed t o provide
s t r u c t u r a l adequacy minimum occupant r i s k and minimum vehicle t r a j e c t o r y as Set
7 1
f o r t h i n NCHRP 230 Table 3--. "Crash Test Conditions f o r Minimum Matrix".
(Table 1)
The SENTRE i s designed as an end treatment f o r w-beam o r t h r i e beam guard-r
a i l which w i l l r e d i r e c t the nose o f the impacting vehicle away from the
u n y i e l d i n g g u a r d r a i l w h i l e a t the same time dfssipate the energy o f the impacting
vehicle.
The SENTRE consists o f f i v e nested overlapping t h r i e beam fender panels
which telescope rearward i n response t o a l o n g i t u d i n a l impact force and an angled
side cable f o r u r g i n g t h e f i r s t fender panel and post assembly l a t e r a l l y away
from the f i x e d guardrail end. The fender panels and angle side cable f u n c t i o n t o
d i r e c t the nose o f the impacting vehicle away frm the hard p o i n t on the
g u a r d r a i l while a t the same time d i s s i p a t i n g the impact energy o f t h e v e h i c l e .
The fender panels are s l o t t e d and secured together i n a nested fashion by
fasteners which allow the fender panels t o telescope upon t h e a p p l i c a t i o n o f an
a x i a l impact force. The fender panels are supported above the ground on v e r t i c a l
I support posts which are positioned on s l i p bases. These s l i p bases a l l o w t h e
F O F ~ Z t O break away from submerged ground anchors so t h a t the fender panels may
telescupe.
The f i r s t fender panel, o r more s p e c i f i c a l l y i t s v e r t i c a l support post i s
connected t o a r e d i r e c t i o n i n g cable. This cable i s secured t o an anchor located
a t t h e f r o n t o f the u n i t , and a rear anchor located a t a l a t e r a l p o s i t i o n away
' f r o m the guardrail. This cable i s positioned so t h a t when a l o n g i t u d i n a l impact
force i s imposed on the f r o n t o f the system. the cable w i l l urge the f i r s t fender
panel l a t e r a l l y as it telescopes rearward. The l a t e r a l force o f the cable and
f i r s t post i n conjunction w i t h l a t e r a l forces contributed by the subsequent posts
w i l l urge the vehicle away from the hard p o i n t on the g u a r d r a i l .
Test A r t i c l e (Construction)
The SENTRE drawings are shown i n Appendix D. The f o l l o w i n g discussion w i l l
describe how the i n d i v i d u a l components are constructed and assembled i n t o a
working u n i t .
The 52 inch. 10 gauge t h r i e beam fender panels include 32 inch s l o t s and are
secured together by fasteners (mushroom b o l t assembly). These s l o t s allow the
fender panels t o telescope upon the a p p l i c a t i o n o f a l o n g i t u d i n a l force.
The mushroom b o l t assembly i s designed with a shoulder t h a t t r a v e l s i n the
s l o t o f the fender panel. The assembly secures two overlapping fender panels with
a grade 5, f l a t head 3" x 5/8" diameter b o l t which passes through a hole i n the
center o f the mushroom washer and a hole i n the underlying fender panel. The
mushroom b o l t assembly i s constructed so that it does not s o l i d l y clamp the two
fender panels together, b u t r a t h e r secures them i n a p o s i t i o n r e l a t i v e t o One
another with s u f f i c i e n t tolerance t o allow the f i r s t fender panel to telescope
i n t o the second panel. The l o n g i t u d i n a l movement o f the f i r s t fender panel i s
halted when it reaches the end o f the s l o t .
The fender panels are supported above the ground by v e r t i c a l support posts.
The 32" long posts are W 6.5 x 9 steel "I" beams t o which an a d d i t i o n a l 21 inch W
6.5 x 9 "I" beam blockout i s bolted with two 1 112" x 3/4" diameTer b o l t s . The
fender panels are then attached t o the blockout with two 2 ' x 3/J" diameter grade
2 b o l t s . The purpose o f the blockout i s t o prevent automobiles w i t h small wheels
from snagging on the v e r t i c a l support posts o f the SENTRE during a side angle
impact.
The v e r t i c a l support posts are welded t o a 1/2" x 8" x li" steel s l i p base.
The s l i p base assembly includes a top p l a t e and a bottom p l a t e which are secured
to each other. The bottom p l a t e i s attached t o an earth anchor.
The top and bottom s l i p base plates each include four open ended s l o t s which
are designed t o receive 2" x 3/4" diameter b o l t s which secure the plates
together. The plates are large enough so t h a t they w i l l not y i e l d during a
l a t e r a l impact. The s l o t s are open ended so t h a t when a s u f f i c i e n t l o n g i t u d i n a l
impact force i s applied to the v e r t i c a l support post by the impacting vehicle,
the plates w i l l s l i d e apart. To insure t h a t the plates w i l l s l i p apart i n a
predictable manner, they are separated by four 3/4" diameter f l a t washers. The
washers provide a consistant bearing area between the two p l z t e s so t h a t the
force needed to cause the plates t o s l i d e can be c o n t r o l l e d . Testing has shown
t h a t the vehicle sustains acceleration l e v e l s o f 4 t o 5 "Gs" when a torque of 60
f t . - l b s . i s applied t o the four s l i p base b o l t s . 2 2
The v e r t i c a l suoport posts also include a 4" x 4" steel gusset attached from
the v e r t i c a l support post to the top o f the s l i p base p l a t e . This gusset
strengthens the v e r t i c a l support post during r e d i r e c t i v e impacts.
An additional 3" x 6" angle p l a t e i s welded t o the bottom s l i p base t o
provide a ramp and prevent possible snagging on each other as they break away and
move rearward i n response t o a l o n g i t u d i n a l impact force.
The f i r s t v e r t i c a l support post i s s i m i l a r i n construction t o the other
posts except t h a t it contains a 1 3/4" x 2" diameter schedule 80 steel pipe
gromnet. The gromnet i s located 1 1/2" from the top o f the s l i p base and i s
designed t o receive a 1 1/2" diameter threaded steel f i t t i n g which i s swedged to
the end of a 7/8" diameter steel cable. The cable extends from the previously
mentioned f r o n t cable anchor, through the gromnet, t o the rear cable anchor.
The rear anchor i s located on an imaginary l f n e which runs through the center of
the f i r s t v e r t i c a l post a t an angle o f 25 degrees with respect t o the c e n t e r l i n e
of the roadway. The cable forces t h e f i r s t fender panel and v e r t i c a l post t o
move l a t e r a l l y upon the a p p l i c a t i o n o f a l o n g i t u d i n a l impact force.
The f r o n t and rear cable anchors are t y p i c a l l y embedded i n a concrete
foundation measuring 18" diameter by 4 f e e t deep. The f r o n t and rear anchor
consist of a 1" x 3" x 29'' steel bar welded t o a 1/2" x 5" x 7" plate. The
anchors are designed t o be universal and secure each end o f the cable. The f r o n t
cable anchor i s positioned ahead o f the f i r s t v e r t i c a l support post and secures
the c l e v i s end o f the cable using 1 5/8 " diameter p i n and c o t t e r pin. The cable
passes through the gromnet i n the f i r s t v e r t i c a l support post and i s then secured
t o the rear cable anchor by i n s e r t i n g the threaded f i t t i n g on the end o f the
cable through the 1 3/4" diameter hole i n the steel anchor and attaching a washer
and nut. The 1 1/2" nut i s torqued t o approximately 100 f t . - l b s . The cable
aids i n r e d i r e c t i o n i n g vehicles which impact the SENTRE head on. By urging the
f i r s t fender panel l a t e r a l l y the cable imposes a l a t e r a l f o r c e on the fender
panels. The cable i s constructed from 7/8", 6 x 25 IWRC, galvanized, steel cable
and w i l l s t r e t c h 1 to 1 1/2 % o f i t s length upon a p p l i c a t i o n o f a l o n g i t u d i n a l
impact force.
The l a t e r a l force w i l l now be described i n more d e t a i l . Uhen a vehicle
impacts the guardrail end terminal head on, the f i r s t panel i s forced backwards
telescoping i n t o the second panel. As the vehicle continues i t s motion, the
f i r s t v e r t i c a l post impacts a second v e r t i c a l support post causing the top p l a t e
o f the second s l i p base t o disengage. The rearward movement o f the f i r s t panel
stretches the cable u n t i l the cable w i l l not s t r e t c h any f u r t h e r . The cable then
urges the f i r s t panel l a t e r a l l y causing the f i r s t fender panel t o give a small
l a t e r a l impulse t o the nose o f the impacting vehicle. As the f i r s t fender panel
reaches the end of i t s t r a v e l the second fender panel begins t o telescope i n t o
the t h i r d fender panel. The f i r s t fender panel reaches the end of i t s
l o n g i t u d i n a l movement before the second s l i p base breaks free. Each s l i p base
.decelerates and dissipates some of the energy of the impacting vehicle. This
process continues u n t i l a l l the s l i p bases o f the SENTRE have disengaged g i v i n g a
large l a t e r a l force t o the impacting vehicle. The net consequence o f t h i s l a t e r a l
force moves the vehicle away from the hard point.
The SENTRE includes additional mass i n the form o f sand because the s l i p
bases do not remove a s u f f i c i e n t amount of energy t o keep an impacting vehicle
from h i t t i n g the hard point. The sand i s held i n containers which add 200 lbs.
t o the f i r s t and second v e r t i c a l support posts a t a 24 i n . center o f g r a v i t y and
300 lbs. t o the t h i r d v e r t i c a l support post a t a 21 in. center o f g r a v i t y . The
260 lbs. o f sand i s e q u a l l y d i v i d e d i n t o two 100 l b . containers f o r each of the
f i r s t two posts. A 100 1b. container i s placed on each side o f the block out and
v e r t i c a l post so t h a t two 2" x 3/8" diameter b o l t s can be inserted and clamp the
containers t o the v e r t i c a l post assembly. The 300 1b. sand mass i s equally
divided i n t o two 150 lb. sand contaners and attached t o the t h i r d post i n the
same manner as the 100 l b . sand containers. A 1 i d i s included on each container
t o keep moisture from entering the sand. The l i d i s designed w i t h a self locking
feature so no assembly tools such as r i v e t s are required.
A b u l l nose has been included as p a r t o f the design t o a i d i n the aesthetic
appeal. The b u l l nose i s made of gray p l a s t i c and includes a f l a t section on the
nose which may be used to attach r e f l e c t i v e markers.
The SENTRE may be attached t o e i t h e r w-beam or t h r i e beam g u a r d r a i l . Both
guardrail types must include an end anchoring system which extends from a ground
anchor t o t h e section o f g u a r d r a i l imnediately f o l l o w i n g the SENTRE. A
transition panel must be included when installing the SEHTRE on w-beam guardrail.
The transition panel i s connected from the l a s t fender panel of the SENTRE to the
. .
hard point of the length of need guardrail. A 7/8" 6 x 19 .IWRC cable extends from
a concrete deabman, located betweeen posts 4 and 5 , to a cable anchor on the
transition panel. The cable and concrete deadman are strong enough so that the
L.O.N. guardrail develops i t s full tensile strength during a redirect impact.
APPENDIX C - FHWA APPROVED WORKPLAN
AZ-8802 (SENTRE)
AZ-8803 (TREND)
WORKPLAN
1. Evaluate and document the site selection criteria, and design conditions. This will include
expected service requirements of the project and anticipated service life of the device.
The geometric alignment, device location, traffic volume, vehicle operating speeds and
mix, environmental conditions, and soil stratigraphy will be documented.
2. Perform a risk analysis and pre-installation safety evaluation. Traffic and accident data will
be obtained and analyzed for each installation for the period 2 years prior to award of the
construction project. An appropriate risk analysis program will be selected and used to
determine the probability for collision for each device. Local maintenance authorities/ DPS
officers will be interviewed to determine il any unique safety or environmental conditions
are prevalent.
3. Assign reporting procedures and responsibilities. The ATRC will develop field evaluation
forms for use by ADOT construction and maintenance personnel and DPS officers. The
. frequency and content of reporting will be established by the ATRC in conjunction with the
participating personnel.
4. Monitor and document the construction of the devices. The as-built condition of each
device and roadway condition will be documented and an emphasis placed on verifying
that design goals were achieved. Roadway friction testing will be conducted to document
skid properties at the time of construction. Field construction/contractor personnel will be
interviewed for suggested design/ procedural changes and/or improvements.
5. Prepare a construction report documenting the design and construction of the devices.
A construction report will be prepared in accordance with ATRC procedures for reporting
of experimental projects and submitted to the FHWA within 120 days after construction of
the last device.
6. Monitor in-service performance.
* The in-service evaluation will be conducted for 2 years.
* Monthly field inspections will be performed by ADOT maintenance forces to
record "brush hits" and drive away collisions, damage to the appurtenance.
required repairs, routine maintenance, and evidence of near misses. The
availability of replacement pans, level of technical support by the supplier, and
total down time of the appurtenance will be documented. Unique problems such
as vandalism or corrosion will be identified.
+ Reported accidents will be investigated by the ATRC as required. Damage to
appurtenances will be documented and video taped. Accident reporting will be
performed using techniques of the National Accident Sampling System or other
acceptable procedures.
Traffic volume and mix will be obtained annually.
* ATRC will perform scheduled inspections of the installation at 6 months, 1 year,
and 2 years after construction. Interviews with maintenance personnel and DPS
officers will be performed annually.
* Annual maintenance costs will be collected by the ATRC.
7. Evaluate in-service performance. A before and after evaluation will be performed which
evaluates the relative effectiveness of the appurtenances. Specific appurtenance
performance will be evaluated on the basis of three factors: structural adequacy, occupant
risk, and vehicle trajectory after collision. The evaluation criteria are as follows:
a) Structural Adequacy: Measure of geometrical, structural and dynamic properties
of an appurtenance to interact with a selected range of vehicle sizes and impact
conditions in a predictable and acceptable manner. Nonvehicie collision-type
forces such as wind are not included. Criteria:
Acceptable redirection of vehicle.
Controlled penetration of vehicle
Controlled stopping of a selected range of vehicle sizes impacting the Installations
at specified conditions.
Detached elements, fragments, or other debris should not penetrate or show
potential for penetrating the passenger compartment, or present undue hazard to
other traffic.
b) Occupant Risk: Vehicle responses of acceleration and velocity changes. Criteria:
Vehicle remains upright during and after collision although moderate roll, pitching,
and yawing are acceptable.
Minimize velocity change in vehicle. Small cars at both low and high impact
speeds are the critical test.
Minimize vehicle velocity change prior to occupant impact.
c) Vehicle Trajectory: Criteria:
Vehicle trajectory and final stopping position should intrude a minimum distance, if
at all, Into adjacent or opposing traffic.
For longitudinal barrier terminals, vehicle trajectory behind the test article is
acceptable in theory.
The evaluation will determine if the design goals were achieved, identify special problems
that affected appurtenance performance, examine impact the devices exhibited on other
highway conditions, and document the initial cost and annualized maintenance cost.
8. Prepare a final report. A final report detailing the efforts of this study and the conciusions
and recommendations will be prepared. This report wiil be prepared within 90 days of the
completion of the final evaluation in accordance with ATRC procedures for reporting
experimental projects.
APPENDIX D - CONSTRUCTION PLANS FOR AZ-8802 (SENTRE)
. .
x-? t'?
r=.- rl
5gg
2 % .
= - . . --> ...
I..?-?
. .. ...-. .......a. .............. - - - - - . - 3 : z z===: : i. .;?a- -i a.o .i s.- , ; s ~ ; s : e z ? ~ : r r ~ = ~ e e i : ~ z ~ ~-i . ............................ / I . ii_i.___ i _ * ^ ^ i i ^ ^ X ^ * > ^ ^ _ A > ^ A - - d d $ " m - . - ^ ................ ......................................................... ..* . -z x-
APPENDIX E -CONSTRUCTION PLANS FOR AZ-8803 (TREND)
z IJIIIJI*
ZP ===:* E=z92z: %= z m c t c o=~ ?z:e~;? :z:: n n ??exr t:ea r:g z= : = g r g I : e I tees-srzstr z p:r??a Zen =ens
- f yya$'i;i=#i; 2 ? a z i r r d i d d d d d " j;;;;;;; """-""""*; """" ;";;;&A"" """ij ii "" i" . . I----- --- --a-
5 a+ac?gggG'f 2
i
E
5 .....
s Z553Z a X'.FX.XE- - - 2 a z s
- *"
3 a
____^____._""_"_--_---.~-"-~"-"--
S * I 1 S Z L l l l l l S f l l l 3 3 3 1 1 J S 3 f * f S S X S . I
9 ,_".,,;,,- ,. .e.s tr z a . 3 r z !r 5: i a r g ~ r ~ s a ~ r z ~ : z ~ ~ ~ n ~ = : ~ z : : : : c ~ ~ ~ s ~ ~ : I : : : ~ ~ ~ ~ ~ ~ j;;,+; ;,$ i~;did 2: Q; ;;: 4; ~ ~ ~ e + e e + g ~ < < < ~ $ ! z ~ e e ~ r < ~ e < f f < e e. .e . z .~ . ~. ~ e. . e. s ~ ~ e e e < e e """""- A""" :: ""-" """ "G A""""""""""""LLA"AAA""""-"AAA""""-""-L-""""""""-"""""""""
5. --" ---- -- ".-.--- - --- "- i- :;s5E'Es-----""----.r--,------f- -------.--C**::t:l*::
Ei --- ---- 25;? ii -- XI ::;--rr>:::::::::i;>2:ii:~::j::f~jjf::jI5??2:2:>>2: -
i
L Z :.A38
r;;=zEicl ? ;I ,/ . i r
zg' ! I
:=:i . 11 ::; - a g'
,! x:,;;:-; s-: 8 1 ..--- *.%_ _ ! : . : -
I l l ,
i12, i:! 1.2
!?:?-
.,,.a-,-., . . !. I&t-i
! " - ! '
! $l, ;,; ;: s: li
............ . . . . . . - .......... - -- - - . . . .
.. 3rn.I I ' I C , ."b
- .... 1::: -- -- 1 7, - , I H I I I I M * 9,rn.n.U WrnU41
................
. --.. ..-.. - . -- -- ---
..... -~ .. - . . mr*M" ,"," l&., .wwn,r
.A - I
n311- 51 "m,P "3ul m,?"
a 3 w " arm rr, w~mrnwm o7wvrrm n n,,mr wturr on sarr m WUI mwd
n~n!,o! w , ( I .m mvm ' I i l y n ~ w ~1m~ 3~. 7 3 ,f~i ' ~ x m ~ 0n1d o nwa II
3 a t $ 3 ~ ~3 ~ m3m1.m~ m LTM .wnnv.m~ro m n d w ' ~ x n dJ J ~ M )m nf i
WCU 7dOl T*1 W I W I i " 7 l J d XlSVld OBIWU3?I3# X mul IMI8S W-3, W
I1lllnlr
101131*(ivm fiu mlrwnol wnn suw ra w ~ ~ r , , #n mmw n, mm r m s m r mn s
1!8* 4Wnr33dW HNOY"4 r*IW ZLIIBIIISI, OIS yll rn '1Md OmXKS*U is TW6
alv< $",*MI w mw?m ma 1-m n,rn,na, m, nr omnl w Om343 m r q n
A W ~ J N ' u o~?#~n#~rw ~ n 5"-nr 3 53rnra1~na~rmm 1 ~ 1 as n t r 0, wm
311u +E ma( 3 ~ anm mi 01 nznurn a ~ a rn u s rws mlmirm r nm W,VWL n
