Guide for Traffic Signal Preemption near Railroad Grade ... · PDF fileGuide for Traffic...

Technical Report Documentation Page

1. Report No.

FHWA/TX-01/1439-9 2. Government Accession No. 3. Recipient's Catalog No.

5. Report Date

September 2000 4. Title and Subtitle

GUIDE FOR TRAFFIC SIGNAL PREEMPTION NEAR RAILROADGRADE CROSSING 6. Performing Organization Code

7. Author(s)

Steven P. Venglar P.E., Marc S. Jacobson, Srinivasa R. Sunkari, RoelofJ. Engelbrecht, and Thomas Urbanik II P.E.

8. Performing Organization Report No.

Research Report 1439-9

10. Work Unit No. (TRAIS) 9. Performing Organization Name and Address

Texas Transportation InstituteThe Texas A&M University SystemCollege Station, Texas 77843-3135

11. Contract or Grant No.

Project No. 0-1439

13. Type of Report and Period Covered

Guide:September 1999-August 2000

12. Sponsoring Agency Name and Address

Texas Department of TransportationConstruction DivisionResearch and Technology Transfer SectionP. O. Box 5080Austin Texas 78763-5080

14. Sponsoring Agency Code

15. Supplementary Notes

Research performed in cooperation with the Texas Department of Transportation and the U.S. Department ofTransportation, Federal Highway AdministrationResearch Project Title: TxDOT Support of the Texas A&M IVHS Research Center of Excellence16. Abstract

Traffic operations at traffic signals adjacent to railroad grade crossings are very complicated. The currentoperational guidelines are not comprehensive and cause confusion due to inconsistent terminology. Researchconducted by the Texas Transportation Institute (TTI) has identified some of the limitations of the currentpreemption guidelines. These limitations are more apparent in cases where advance preemption is beingused. This guide will assist traffic operation engineers to design safer preemption timings.

17. Key Words

Railroad, Preemption, Advance Preemption, SignalController, Trucks, Grade Crossing.

18. Distribution Statement

No restrictions. This document is available to thepublic through NTIS:National Technical Information Service5285 Port Royal RoadSpringfield, Virginia 22161

19. Security Classif.(of this report)

Unclassified20. Security Classif.(of this page)

Unclassified21. No. of Pages

6422. Price

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

GUIDE FOR TRAFFIC SIGNAL PREEMPTIONNEAR RAILROAD GRADE CROSSING

by

Steven P. Venglar, P.E.Assistant Research EngineerTexas Transportation Institute

Marc S. JacobsonAssistant Research Scientist

Texas Transportation Institute

Srinivasa R. SunkariAssistant Research Scientist


Roelof J. EngelbrechtAssociate Transportation Researcher


and

Thomas Urbanik II, P.E.Associate Director


Project Report 1439-9Project Number 0-1439

Research Project Title: TxDOT Support of the Texas A&M IVHS Research Center ofExcellence

Sponsored by theTexas Department of Transportation

in cooperation with theFederal Highway Administration, U.S. Department of Transportation.

September 2000

TEXAS TRANSPORTATION INSTITUTETexas A&M University System

College Station, Texas 77843-3135

Page v

Acknowledgment

This guide for the Texas Department of Transportation and Federal HighwayAdministration was developed by the Texas Transportation Institute of the Texas A&MUniversity System. Material for this report was obtained from studies funded by varioussources including Intelligent Transportation System Research Center of Excellence,TransLink® Research Program, the Texas Department of Transportation, and theTransportation Research Board’s High Speed Rail IDEA Program.

Preface Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page vi

Disclaimer

No warranty is made by the Texas Department of Transportation, the Federal HighwayAdministration, the United States Department of Transportation, the TransportationResearch Board, the Texas Transportation Institute, or the Texas A&M UniversitySystem as to the accuracy, completeness, reliability, usability, or suitability of thecomputer programs discussed herein, and/or their associated data and documentation.The guide, references to the Rail Exerciser computer program, and references to otherliterature are provided for purposes of training only. The above parties assume noresponsibility for incorrect results or damages resulting from the use of this guide. Theengineer of the responsible agency must ensure that the inputs into any control devicefor railroad preemption are accurate and appropriate for each individual case beinginvestigated. The engineer must test the control devices and input parameters todetermine that they work safely with such timing inputs, and that adequate phase timingand clearances (i.e., minimum green time, yellow change interval, red clearanceinterval) are displayed to motorists under all operating conditions.

Guide for Traffic Signal Preemption near Railroad Grade Crossing Preface

Page vii

PREFACEIn simple terms, the purpose of traffic signal preemptionnear railroad grade crossings is to increase safety atthese intersections by clearing vehicles from the path oftrains. The sequence of events that occur duringpreemption can be compared to a choreographed dancein which each step is dependent upon the previous inorder to make the dance complete.

The goal of this guide is to provide a comprehensivediscussion of the state-of-the-practice in traffic signalpreemption near railroad grade crossings.

Finally, it is important to recognize that EVERY highway-rail intersection is unique. While this guide will providesome example case studies, every situation that anengineer encounters must be carefully studied todetermine the preemption parameters that are appropriatefor a particular intersection. It is our goal that theinformation obtained from this guide will help the engineerin these investigations.

More information about this area can be found on theInternet at:

http://transops.tamu.edu

http://transops.tamu.edu

Guide for Traffic Signal Preemption near-Railroad Grade Crossing Outline

Page ix

GUIDEOUTLINE

Introduction................................................................ 1Background..................................................... 1Definitions ....................................................... 2

On the Rails............................................................... 7Types of Warning Devices .............................. 7Train Detection ............................................... 9Railroad Design Times.................................. 13Interconnection with Traffic Signals .............. 15Traffic Signal Design Times .......................... 15

On the Highway ....................................................... 17Preemption of Traffic Signals........................ 17Traffic Signal Controllers............................... 18Entry into Preemption ................................... 20Preemption Hold Operation .......................... 28Exit from Preemption .................................... 30Preemption Time Requirements ................... 30Design and Operations Concerns................. 32Use of Pre-Signals ........................................ 39

Important Considerations......................................... 41Pedestrians................................................... 41Multiple Track Crossings............................... 42Multiple Clearance Intervals.......................... 43Intersection Design Vehicle .......................... 44Type of Trains............................................... 44Light-Rail Traffic Signal Preemption.............. 45Failure of Preemption Circuit ........................ 46

Recommendations and Conclusion ......................... 47

Sources ................................................................... 49

App endix .................................................................. 51

Guide for Traffic Signal Preemption near-Railroad Grade Crossing Introduction

Page 1

BACKGROUND In 1979, the Institute of Transportation Engineers (ITE)published Preemption of Traffic Signals At or NearRailroad Grade Crossings with Active Warning Devices:A Recommended Practice. This document highlighted theneed for preemption at signalized roadway intersectionsthat were in close proximity to a highway-railroadintersection. It included recommendations for when topreempt and preemption sequences to use.

For 15 years after ITE’s Recommended Practice waspublished, very little changed in the way signalpreemption plans were designed and implemented,despite changes in traffic signal technology and increasesin train speeds and vehicle volumes. Then, in October1995, a commuter train struck a school bus that had itsrear end extending into a crossing. This collision sparkedrenewed concern for traffic signal operations nearrailroad-highway intersections. A Task Force from theU.S. Department of Transportation was appointed toinvestigate such operations. Based on the work of thistask force ITE published a new recommended practice in1997. The new document recommended new methodsfor determining when to preempt traffic signals, discussedsome pitfalls of current train detection methods, andemphasized the need for an open line of communicationusing a standardized vocabulary between those in chargeof the highway traffic signals and railroad personnel.

The goal of this guide is to provide a comprehensivediscussion regarding the state-of-the-practice inpreemption of traffic signals with nearby highway-railintersections. Much of what is presented in this guide hasbeen adapted from ITE’s 1997 Recommended Practice.In addition, the National Cooperative Highway ResearchProgram (NCHRP) Synthesis of Highway Practice 271:Traffic Signal Operations Near Highway-Rail GradeCrossings provided a number of figures used in this guide.

The following section presents a list of definitions that willbe referred to throughout the rest of this guide. The abilityto communicate with personnel involved in all aspects of

��

Introduction Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 2

DEFINITIONS

signal preemption and highway-rail intersections isimperative to improving safety at these locations. Whennecessary, the definitions indicate important differencesbetween railroad and highway terminology.

Active Highway-Rail Grade Crossing WarningDevices/Systems—the railroad flashing light signalswith or without warning gates, together with thenecessary control equipment, used to inform road usersof the approach or presence of trains at highway-railgrade crossings.

Advance Preemption—notification of an approachingtrain is forwarded to the highway traffic signal controllerunit or assembly by railroad equipment for a period oftime prior to activating the railroad active warningdevices. Buffer time should be considered to be zerowhen calculating advance preemption time (APT).

All Red Hold—control mode involving holding all motorvehicles until the train passes through the highway-railgrade crossing.

Approach—a set of lanes accommodating all left-turn,through, and right-turn movements arriving at anintersection from a given direction.

Automatic Flash—a flashing operation resulting frominput from a time switch or system command.

Buffer Time (BT)—Buffer time is discretionary and maybe provided in addition to minimum time (MT) andclearance time (CT) to accommodate minor variations intrain handling. Buffer time is a railroad design elementand should NOT be considered in the traffic signaldesign process.

Clear Storage Distance—the distance available forvehicle storage measured between 2 m (6 feet) fromthe rail nearest the intersection to the intersection stop-bar or the normal stopping point on the highway.

Clearance Time (CT)—For two-quadrant railroad warningdevices, the minimum track clearance distance is thelength along a highway at one or more railroad tracks,measured from the railroad warning device to 6 feetbeyond the track(s) measured perpendicular to the farhighway, as appropriate, to obtain the longer distance.If the minimum track clearance distance exceeds 35 ft.,clearance time is one second for each additional 10 ft.,or portion thereof, over 35 ft. CT may also be added by


Page 3

NOTE: Railroadpersonnel refer to aninterconnect as theopening of the circuitbetween the railroadwarning devices andthe highway trafficsignal when a train isdetected. Trafficengineers refer to thesame circuit openingas a preempt.

the public agency or railroad to account for site-specificneeds. Examples of additional CT include additionaltime for simultaneous preemption and/or additional gatedelay time.

Cycle Length—the time period required for one completesequence of traffic signal indications.

Demand Volume—the traffic volume expected to desireservice past a point or segment of the highway systemat some future time, or the traffic currently arriving ordesiring service past such a point, usually expressed asvehicles per hour.

Design Vehicle—the longest vehicle permitted by statuteof the road authority (state or other) on that roadway.

Equipment Response Time (ERT)—Adjustments shallbe made to provide for control circuit equipmentresponse time. This is a railroad design element andshould NOT be considered in traffic signal design.

External Start—an input, which when energized,normally causes the signal controller to revert to itsprogrammed initialization interval.

Flashing Operations—traffic signal operation modewhere an indication is cycled on and off at a given rate.

Fully Actuated Operation—a type of operation of acontroller unit in which all signal phases are operatedon an actuated basis.

Hold Intervals—the highway traffic signal indication(s)displayed after the track clear intervals during the timethe preemption circuit is active.

Interconnected Signals—traffic signals that areconnected together by some means (hardwire or radiowave), primarily for the purpose of establishing adefinite timing relationship between the signals.

Interconnection—the electrical connection between therailroad active warning system and the traffic signalcontroller assembly for the purpose of preemption.

Internal Preemption—signal controllers capable ofaccommodating preemption without special outsidecontrol processes.

Interval—the part or parts of a signal cycle during whichsignal indications do not change.

Interval Sequence—the order of appearance of signal


Page 4

NOTE: The minimumtrack clearancedistance is highlydependent upon theangle that the railroadtracks cross thehighway and thenumber of tracks at thecrossing.

NOTE: MMU Flash isalso known as conflictflash in controllers withconflict monitorsinstead of a MMU.

NOTE: Thepedestrian clearancetime is the time duringwhich the DON’TWALK signal flashesand depends on theintersection geometry.

indications during successive intervals of a cycle.

Maximum Preemption Time (MPT)—the maximumamount of time needed following initiation of thepreemption sequence for the highway traffic signals tocomplete the timing of the right-of-way transfer time,queue clearance time, and separation time.

Minimum Time (MT)—one of the two components of theminimum warning time (MWT) and is usually equal to20 seconds.

Minimum Track Clearance Distance (MTCD)—thelength along a highway at one or more railroad tracks,measured either from the railroad stop line, warningdevice, or 4 m (12 ft) perpendicular to the trackcenterline, to 2 m (6 ft) beyond the track(s), measuredperpendicular to the far rail, along the centerline or rightedge line of the highway, as appropriate, to obtain thelongest distance.

Minimum Warning Time (MWT) (Through TrainMovement)—The least amount of time active warningdevices shall operate prior to the arrival of the train at arailroad-highway grade crossing. It can be defined asthe sum of MT (usually 20 seconds) and CT.

Malfunction Management Unit (MMU) Flash—a flashingoperation resulting from input from the malfunctionmanagement unit.

Pedestrian Clearance Time—the time provided for apedestrian crossing in a crosswalk, after leaving thecurb or shoulder, to travel to the far side of the farthesttraveled lane or a median.

Phase—the part of the signal cycle allocated to anycombination of traffic movements receiving the right-of-way simultaneously during one or more intervals.

Preemption—the transfer of normal operation of trafficsignals to a special control mode.

Pre-Signal—supplemental highway traffic signal facesoperated as part of the highway intersection trafficsignals, located in a position that controls trafficapproaching the highway-rail grade crossing andsignalized intersection.

Pre-timed Operation—a type of controller unit operationin which cycle length, interval duration, and intervalsequence are predetermined.

Queue Clearance Time—the time required for the designvehicle stopped within the minimum track clearance


Page 5

NOTE: This is thedesign (worst case)condition. In actualoperation, this timemay be zero.

NOTE: In actualoperation, separationtime may increase fromthe design value.

distance to start up and move through the minimumtrack clearance distance.

Railroad Preemption Circuit—a control circuit utilizing asupervised/closed-circuit principle activated by a train’sapproach to a highway-rail grade crossing thatpreempts the operation of a highway traffic signal.

Right-of-Way Transfer Time (RTT)—the maximumamount of time needed for the worst case condition,prior to display of the clear track green interval. RTTcan include any railroad or traffic signal controlequipment time to react to a preemption call, any trafficsignal green, pedestrian walk and clearance, yellowchange and red clearance interval for opposing traffic.

Saturation Flow Rate—the equivalent hourly rate atwhich vehicles can traverse an intersection approachunder prevailing conditions, assuming that the greensignal is available at all times and no lost times areexperienced, in vehicles per hour of green or vehiclesper hour of green per lane.

Semi-Actuated Controller—actuated controller modewith detectors placed only on the side-streetapproaches to give only enough green to service thelow and somewhat predictable side-street trafficdemand.

Separation Time—the component of the maximumpreemption time during which the minimum trackclearance distance is clear of vehicular traffic prior tothe arrival of the train.

Signal Phase—the right-of-way, change, and clearanceintervals in a cycle that are assigned to an independenttraffic movement or combination of movements.

Simultaneous Preemption—notification of anapproaching train is forwarded to the highway trafficsignal controller unit or assembly and railroad activewarning devices at the same time.

Start-up Flash—a flashing operation that may beprogrammed to occur prior to initialization, after electricpower is applied to the signal controller.

Start-up Headway—start-up time between twosuccessive vehicles in a traffic lane as they depart froman intersection, measured from front bumper to frontbumper, in seconds.

Total Warning Time (TWT)—the total warning time maybe determined as the sum of Minimum Warning Time


Page 6

and Buffer Time. TWT is a railroad design element andis used to determine the location of the circuit on thetracks. Traffic engineers should NOT use TWT in thetraffic signal design process.

Total Approach Time (TAT)—the total approach may bedetermined as the sum of TWT, ERT, and APT. It is arailroad design time and should not be used in trafficsignal design.

Track Clearance Green Interval—the time assigned toclear stopped vehicles from the track area on theapproach to the signalized intersection.

Traffic Signal—an electrically powered traffic controldevice, other than a barricade warning light or steadyburning electric lamp, by which traffic is warned ordirected to take some specific action.

Traffic Signal Controller—the part of a controllerassembly that is devoted to the selection and timing ofsignal displays.

Guide for Traffic Signal Preemption near Railroad Grade Crossing On the Rails

Page 7

TYPES OFWARNINGDEVICES

At highway-rail intersection trains always have the right-of-way. These intersections have their own sets ofwarning devices that are used to assure that the right-of-way for the trains is maintained. When highway trafficsignals are located near a highway-rail intersection, thosesignals should work together with the railroad warningdevices to move vehicles from a train’s path. This sectiondiscusses the warning and train detection systems usedby the railroad as they relate to traffic signal preemption.

There are two main types of warning devices used athighway-rail intersections. Passive warning devicesconsist of signs and markings that simply indicate theexistence of a railroad crossing. These devices provideno indication as to a train’s imminent presence at acrossing. It is the motor vehicle driver’s responsibility toensure that a crossing is safe before attempting toproceed. If a train is present the law requires a motorvehicle to yield the right-of-way. Some examples ofpassive devices are shown in Figure 1 below:

��

The advance warningsign indicates that thereis a railroad crossingahead.

The crossbuck is placedat the crossing andindicates that vehiclesshould yield to trains thatare approaching orpresent on the tracks.

The number of tracks sign is placed at thecrossing to indicate the number of tracksthat must be crossed.

FIGURE 1. Examples of Passive Warning Devices

On the Rails Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 8

Traffic signalpreemption shouldconsider the effect ofthis failsafe operationon traffic signaloperations.

Did you know?

The alternating patternof the railroad warninglights is supposed toresemble a rail workerswinging a lantern at arail crossing to stopmotorists.

The second type of warning device used at highway-railintersections is the active warning device. Thesedevices consist of flashing lights, which may beaccompanied by gates and/or bells, that are activated asa train approaches a crossing. These devices alert themotorist that a train is nearing and they must stop at thecrossing. Active warning devices must be present at acrossing in order to preempt any nearby traffic signals.

Active warning devices are designed tobe “failsafe.” “Failsafe” means that ifpower should be lost at a crossing or if thetrain detection system should malfunction,the lights would flash and the gate (ifpresent) would remain lowered. Thismethod of operation is accomplishedusing batteries to provide power to thewarning lights in the event of a powerloss.

When the device detects a train, the pairof warning lights at the crossing willalternately flash red. This flashingindicates that a train is approaching thecrossing and motorists need to come to acomplete stop before proceeding. If notrain is present the warning lights will bedark.

Some railroad crossings are equippedwith gates that accompany the flashingred lights at a crossing. These gates areheld upright by electricity when a train is not present at acrossing. Once a train is detected and the warningdevices are activated, the electric lock is released andgravity pulls the gate to its horizontal position, blockingmovement across the tracks.


Page 9

TRAINDETECTION

DC and ACcircuits

Almost all highway-rail intersections that have activewarning devices use some method of on-track circuits todetect the presence of a train. While it is not critical thatthe traffic signal engineer understands all of theelectronics that make the detection system work, theengineer should be aware of how the type of detectionsystem used determines the amount of warning timeprovided. Before discussing warning times in the nextsection, a brief discussion of the most common types oftrain detection systems used is provided below.

The oldest type of train detection circuit used is the directcurrent (DC) circuit. This circuit is illustrated in Figure 2below.

FIGURE 2. The DC Track Circuit


Page 10

NOTE: This type ofdetection circuit willprovide warning timesthat vary depending onthe speed of the trainthat enters the circuit.For example, if thecircuit length has beendesigned so that 20 sof warning is providedfor train traveling 60mph, then 40 secondsof warning will beprovided for a 30 mphtrain using the samecrossing. If a traintraveling 70 mph usedthe circuit, less than 20seconds of warningtime would beprovided.

Audio FrequencyOverlay Circuits

As Part A of the figure indicates, a direct current (DC)battery is connected to the steel rails, causing the relay atthe other end of the circuit to be energized. Insulatedjoints define the boundaries of the circuit along the rail.

When a train enters the detection circuit, the train’s steelwheels will shunt the circuit causing current not to flow tothe relay, as shown in Part B of Figure 2. The loss ofcurrent at the relay causes it to de-energize and activatethe warning devices at the crossing.

The train detection circuit that the crossing uses actuallyconsists of three zones as shown in Part C of Figure 2.An insulated joint separates each of the zones. As a trainenters the east approach zone, the warning devices areactivated as described above. The devices remain activeas the train crosses the island zone that is located at thecrossing. Once the train leaves the island zone, thewarning devices would deactivate. Trains approachingfrom the west would repeat the same process except thewarning devices would activate as the train entered thewest approach circuit.

The length of the approach zones for a DC circuit is basedon the fastest train speed expected at the crossing.Crossings with higher speed trains will need longerdetection circuits in order to provide an adequate amountof warning time prior to the train’s arrival at the crossing. Itis very important that this detection circuit beextended if faster trains are expected to begin using acrossing.

Alternating current (AC) circuits operate in the samemanner as the DC circuits except that a low frequency ACpower source replaces the DC battery. This type of circuitis most commonly used with electrified tracks (such aslight-rail trains [LRT]).

The audio frequency overlay circuit (AFO) is anothercommon type of train detection circuit used today. Thiscircuit uses a high-frequency AC transmitter and receiverto define a train detection zone. The advantage of thistype of system is that it does not need insulated joints thatmay prevent other systems (such as the train signalsystem) from working properly. The length of the track


Page 11

Motion SensorSystems

ConstantWarning Time

Systems

NOTE: The length ofthe CWT circuit is stilldetermined by thespeed of the fastesttrain that will utilize thecrossing.

circuit is again determined by the speed of the fastesttrain that is expected to use the crossing. This type ofsystem also provides warning times that vary based onthe speed of the train.

Motion sensor systems use similar technology as AFOcircuits to detect the presence of a train. However, motionsensor systems also monitor the track circuit impedanceto determine the direction a train is moving. Thesesensors can also determine if a train that has entered thedetection zone has come to a stop and can deactivate thewarning devices if appropriate.

Constant warning time (CWT) systems build on thetechnology used in motion sensor systems. Unlike thepreviously discussed circuits that provide varying warningtimes dependent on train speed, CWT systems aredesigned to provide nearly the same warning timeregardless of the speed of the train provided the trainspeed does not change. These systems utilize a devicethat uses the measured speed and direction of a train as itenters the detection zone to predict when the train willarrive at the crossing. This device will then delayactivating the lights and gates at the crossing (and signalpreemption circuit) until a set time before the train ispredicted to arrive at a crossing.

Critical Point

While the CWT devices greatly reduce the variability inwarning times provided by different train speeds, they tooprovide a slightly variable amount of warning time. Theactual warning time provided depends on theacceleration/deceleration of trains after initial detectiondue to track environment and other reasons. In otherwords, if a CWT is set to provide the MWT needed at acrossing to successfully preempt the traffic signals, and atrain that enters the circuit increases its speed toward thecrossing, it will arrive earlier than predicted and lesswarning time than the MWT required will be provided.

!


Page 12

After the collision between the school bus and thecommuter train in Illinois, a study of CWT warning timeswas conducted to demonstrate how the actual warningtime could vary. Figure 3 shows the results of this study.

FIGURE 3. Results of Study Investigating CWT Devices

As the Figure 3 chart indicates, when the slower (65 km/h)train uses the crossing, the actual warning time providedequaled the preselected MWT. However, when the fastertrain (110 km/h) used the crossing, the actual timeprovided was as much as 5 seconds less than the pre-selected minimum time. In this case, the difference wasdue to the 4.8 seconds required to acquire the train in thedetection zone (i.e. predict its arrival at the crossing) andactivate the warning devices. The conclusions that canbe drawn from this example are that the actual warningtime provided might be less than the minimum warningtime selected on the CWT device, and that CWT systemshave warning time variability that preemption design mustconsidered. Railroads deal with variability by adding


Page 13

Future TrainDetection

RAILROADDESIGN TIMES

NOTE: T he te rm s used inth is sec tion a re d e fine db y th e AmericanRailway Engineeringand Maintenance ofWay Association andare used to spec ify ho wm uch w arn ing tim e th era ilroa d p rov ides . T heam o un t o f tim e ne ed edfo r tra ffic s ign a lp reem ption is d iscussedin the ne xt cha p te r.

NOTE: T he c lea ra nced is tance d iscussed in th issec tion w ill on ly c lea rveh ic les from thecross ing . It is no tnecessarily the sam ed is tance tha t the tra ffics igna l m us t be des ig ne dto c lea r.

buffer time to their system. Traffic engineers must selectappropriate controller settings to deal with this variability.

New methods of train detection are under developmentand may one day replace the track circuit systems thatare used today. These systems include off-trackdetection methods, and some are focused on moreaccurately predicting the train’s arrival time at thecrossing. In addition, it appears that some of these newermethods may provide a more cost-effective means ofproviding long warning times than the extension ofexisting track circuits. Some of the technologies beingexplored include microwave detection, Doppler radardetection, and tracking using a global positioning system(GPS). However, at the current time such systems arenot widely deployed, and almost all traffic signalpreemption locations that the engineer will encounter willbe using some method of on-track detection.

Federal and state guidelines exist that specify a minimumtime for warning device operation time before train arrivalat a crossing that must be provided at every highway-railintersection equipped with active warning devices. TheCode of Federal Regulations, the Manual on UniformTraffic Control Devices (MUTCD), and the AmericanRailway Engineering and Maintenance of WayAssociation’s Signal Manual (formerly known as theAssociation of American Railroad’s Signal Manual) allstate that the minimum time must be no less than 20seconds.

The railroad industry defines clearance time as additionaltime that must be provided in excess of the 20 secondminimum time in order to establish the minimum warningtime. One second of CT is needed for every 10 feet ofclearance distance greater than 35 feet. The clearancedistance is measured parallel to the highway centerlinefrom the railroad warning device to a point that is nocloser than 6 feet to the farthest track. The combinationof the regulation-mandated minimum time of 20 secondsand the CT is the unique MWT for the crossing.

MWT = 20 second minimum time + CT (if provided)

The train detection systems described in the previous


Page 14

section must be designed to provide at least the MWT(i.e., 20 second minimum time plus any additional CTrequired) for the fastest train that designers expect to usethe crossing. Because of the inherent variability inwarning times (found with either non-constant or CWTtrack circuitry), railroads provide additional warning time toensure that at least the MWT is always provided, includingvariability concerns. This additional warning time isknown as the buffer time. This buffer time is ONLY arailroad component and should NOT be considered in thetraffic signal design.

The total warning time provided by the railroad warningdevice can be computed as follows:

TWT = MWT + BT

In the next chapter we will see how to determine theamount of notification that the traffic signal needs in orderto safely clear the tracks before the train arrives at thecrossing.

If the time needed by the traffic signal exceeds the MWTas calculated, then some additional time is needed. Thisadditional time is generally provided as advancepreemption time, although some may chose to increasethe MWT by increasing CT.

Some additional time must also be provided to account fordelays in the equipment before the warning devices areactually activated. This additional time component isknown as the equipment response time.

The total approach time that the track circuitry must beable to provide is:

TAT = TWT + ERT + APT

The track circuit must be long enough to provide the TATwhen crossed by the fastest train that can use thecrossing.

However, it is important to note that some of thecomponents discussed will NOT be included in the signaldesign process. This discussion has been provided tohelp understand the railroad design process and thevariability in observed times

NOTE: Y ou w ill obse rveTW T , w h ich w ill b egen era lly la rge r than yo urdes ig n M W T . H o w ever,th is is the ra ilroads sa fe tyfac to r to gu aran te e M W T


Page 15

INTERCONNECTIONWITH TRAFFICSIGNALS

TRAFFIC SIGNALDESIGN TIMES

The discussion to this point has focused on detecting atrain approaching the crossing and then activating therailroad warning devices. In addition to turning on theflashing lights and lowering the gate (if present), the traindetection system must also alert the traffic signalcontroller that a train is approaching (so that traffic signalpreemption can begin).

The electrical circuit between the railroad equipment andthe traffic signal controller is known as the interconnectcircuit or the railroad preemption circuit. Like the traindetection circuitry, the interconnect is designed to beclosed (energized) in the absence of a train and open (de-energized) once the system detects a train. In thismanner, the interconnect is also designed to be “failsafe,”since the traffic signal would go into its preemptionsequence upon a loss of power from the railroadequipment. This power loss could be due to equipmentmalfunction, power failure, or accidental cutting of thephysical cable between the highway and rail equipment.The traffic signal engineer should consider the effects ofan extended railroad preempt due to the “fail-safe” natureof the railroad preempt.

For traffic signal design, the design times are:

MT + CT + APT

Guide for Traffic Signal Preemption near-Railroad Grade Crossing On the Highway

Page 17

PREEMPTIONOF TRAFFICSIGNALS

When to Preempt

This chapter focuses on traffic signal operations nearhighway-rail intersections with preemption. Thediscussion begins with an overview of signal preemptionand then discusses various options found in traffic signalcontrollers. Next, the three main stages of signalpreemption are presented. The chapter concludes with adiscussion of the use of pre-signals.

When a traffic signal controller receives a preemptioninput, operation is transferred to a special mode that hasits own set of timings and rules for operation. Thefollowing sources can generate preemption inputs from(listed in the order of increasing priority): drawbridge,railroad, emergency vehicles, and transit vehicles.Priority is determined based on the relative hazard thateach source represents to the motorists. For example, atrain represents a much more serious hazard than anemergency vehicle, which would have the ability to stopquickly and/or swerve to avoid a vehicle at an intersection.The remainder of this discussion will focus on the railroadpreemption of traffic signals near highway-railintersections.

The purpose of traffic signal preemption at highwayintersections near highway-rail intersections is to clearany vehicles that may be in danger of being hit by thetrain before the train arrives at the crossing. While it isillegal for vehicles to stop on the tracks/crossing, manydrivers choose to do so either due to their own impatienceor because they were “caught” by a changing traffic signalas they approached the intersection.

Traffic signals are normally preempted by highway-railroad grade crossings when the queue from thesignalized intersection will normally extend back into thecrossing while that particular approach has the red light.In some cases, traffic backing up from a railroad crossingmay block the signalized intersection. It may be desirableto detect such queues and modify the signal timing orphasing.

��

On the Highway Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 18

TRAFFICSIGNALCONTROLLERS

NEMA

The MUTCD states that when highway-rail gradecrossings are within 200 feet of a signalized intersection,preemption should be considered at that location.Experience, however, indicates that sometimes thisdistance is not long enough. For this reason, the 1997revision of the Recommended Practice by ITE suggestsusing a detailed queuing analysis to determine whethersignal preemption is necessary. The analysis is based onthe traffic volume on the approach crossing the tracks,nearby traffic signal timing, the number of lanes on theapproach, and characteristics of the vehicles using theapproach. Depending on the parameters at a particularintersection, queues could extend well beyond the 200feet suggested by the MUTCD. Preemption should beconsidered at all intersections that would be affected byqueues extending from the railroad grade crossing. Adraft of the upcoming new release of the MUTCDsuggests the queuing study should be performed whenhighway-rail intersections are located within 1000 feet of asignalized intersection.

Most traffic signal controller equipment in use todayconsists of a microprocessor to control the operations atthe intersection. There are two main types of controllersin use: devices built according to NEMA standards, anddevices built according to the Type 170/179/2070specification. The next two sections discuss each ofthese controllers.

A NEMA controller follows the standards set forth by theNational Electrical Manufacturers' Association. The firsttraffic signal standard issued by this group, known as TS-1, did not provide for preemption in the traffic controllerunit itself. Intersections using this type of controller andneeding preemption either made use of an externalpreemption device or a controller that had preemptioncapabilities that a particular manufacturer had added.Since preemption was not included in the standard,preemption capabilities of TS-1 controllers are highlyvariable.

The latest NEMA standard for traffic signal controlequipment, named TS-2, added preemption to therequired capabilities of the traffic signal controller. Thestandard calls for a least six preemption inputs that can


Page 19

Type170/179/2070

respond with a least six unique preemption sequences.According to the standard, Preempt Input 1 and PreemptInput 2 are specified for rail use and have a higher prioritythan the other preemption inputs. By default, Preempt 1has a higher priority over Preempt 2 in the controller, butthis preference can be canceled through programmingoptions.

The user-programmable controller units of the170/179/2070 type offer preemption capabilities that varydepending on the type of software that is being used inthe specified 170/179/2070 controller hardware. Themost common preemption operation provides sixpreemption inputs. Two of those inputs, named RR1 andRR2, are reserved for railroad preemption. These twoinputs have equal priority but vary in their operation duringthe preemption hold interval (discussed in a subsequentsection) while the train is occupying the tracks.

Critical Point

It is important to recognize the preemption capabilities ofsignal controllers will vary from manufacturer tomanufacturer. This is especially true for the older NEMATS-1 style controllers and the different software for theuser-programmable 170/179/2070 type controllers. Forexample, some controllers will allow a delay to bespecified before preemption is initiated once the controllerreceives the preemption input. If this delay were to beunknowingly set, the desired operation may not beobtained. Some controllers will allow minimum greentimes and pedestrian WALK intervals to be shortened forphases conflicting with track clearance when thepreemption signal is received; however, on manycontrollers these intervals cannot be shortened and will beretained in their entirety even after preemption is initiated.As these examples illustrate, it is very important to befamiliar with the preemption operation that each controllerunit being used in the field provides.

Regardless of the type of controller that is being used, theoverall preemption sequence is divided into three stages.The entry into preemption begins as soon as the controllerreceives the preemption input. The second stage is

!


Page 20

ENTRY INTOPREEMPTION

NOTE: S o m e con tro lle run its w ill a llo w a d e la y tobe se t be fo re p re em ptionopera tio n b eg ins once thec ircu it has bee n d e -energ ize d .

Right-of-WayTransfer Time

(RTT)

known as the preemption hold interval(s) and occurs afterthe controller completes its entry sequence. Finally, anexit from preemption occurs when the signal is returned tonormal operation once the train has safely passed throughthe crossing (and the preemption input has been removedfrom the controller). The details of each of the stages willbe discussed in the following sections.

In most cases, the traffic signal controller will enter intopreemption operation as soon as the interconnect circuitfrom the railroad is opened, or de-energized (i.e., a train isdetected and the railroad crossing predictor - or processor– indicates that the time threshold for preemption warningtime has been reached by the train approaching thecrossing). However, some controllers incorporate a lagtime (analogous to equipment response time for traindetection circuits) to verify the continuity of the preemptioncall. The entry into preemption is the most time-criticalportion of the preemption sequence and presents thegreatest hazard to motorists if it is not completed in atimely manner. The two parts that compose the entry intopreemption are the right-of-way transfer time and cleartrack green interval.

Right-of-way transfer time is the maximum amount of timeneeded for the worst case condition, prior to display of theclear track green interval. RTT includes any railroad ortraffic signal control equipment time to react to apreemption call, and any traffic signal green, pedestrianwalk and clearance, yellow change and red clearanceinterval for opposing traffic.

Since preemption can occur at any point in a trafficsignal’s normal cycle of operation, enough time must beprovided to safely terminate any active phase orcombination of active phases at any point in the cycle. Inorder to terminate the active phase safely, minimumvehicle times, vehicle change and clearance times, andpedestrian clearance times must all be considered. Thisis the design (worst case) condition. It is important toremember that in actual operation, this time may be zero.


Page 21

NOTE: S o m e con tro lle run its w ill use the m in im umgreen tim es spec ifie d fo r aphas e ’s no rm a l tim ing p lanif the p re em ptio n m in im umgreen is se t to 0 . O the rcon tro lle rs , ho w ever, w illa llo w a m in im um g ree n tim eo f 0 seconds to b espec ified . Consult withyour signal controllermanufacturer to be sure!

NOTE: T he M U T C D d oesa llo w the track c lea ranc ephas e to go from ye llo wim m ed ia te ly back to g reen(w ithou t d isp la ying a re dind ica tio n ) if it is d isp la y in ga ye llow a t the tim e thepreem ption in pu t isrece ived.

NOTE: W hen p ed es trianpush b u ttons a re use d a tthe s igna lize d in te rsec tion ,it is im portan t to no te th a tthe pe des tr ian c lea ra ncetim e w ill on ly b e a fac to r ifthe pe des tr ian b u tton w aspush ed jus t be fo re th econ tro lle r rece ive d thepreem ption in pu t o r theped es tr ian tim ing is onreca ll due to coo rd ina tio n .H o w ever, it s till m us t beinc lud ed in th e ca lcu la tio no f the tim e ne ed ed fo rp reem ption if the con tro lle ris no t se t to sho rten o r om itthe pe des tr ian c lea ra ncein te rva l.

Minimum Times

Most controllers allow a special minimum green time to bespecified for each phase during preemption. Byspecifying a minimum time that a conflicting phase mustdisplay a green signal, very short green and confusinggreen indications that result when the preemption input isreceived immediately after a conflicting phase has turnedgreen are avoided. The minimum green time specified fora phase during preemption is only timed if the phase hasnot been active for at least the time specified when thepreemption input is received.

Vehicle Clearance Times

Regardless of the minimum green time provided, theMUTCD requires that the necessary yellow changeinterval be provided for every terminating phase. Mostcontroller units use the yellow and all red clearance timesprovided for a phase’s normal timing. However, somecontrollers do allow you to specify a different yellowchange interval and all-red clearance time for eachphase under preemption operations. Take care toensure that the vehicle clearance times are neithershortened nor omitted in an effort to clear the tracks in atimely manner.

Pedestrian Clearance Times

Under normal operation, the pedestrian WALK display isterminated with a flashing DON’T WALK display whoseduration is determined by the amount of time required tosafely cross the street. However, Section 8C-6 of theMUTCD states that, “… because of the relative hazardsinvolved, pedestrian clearances may be abbreviated inorder to provide the track clearance display as early aspossible.” Thus, it is common preemption practice tosimply omit the pedestrian clearance interval (flashingDON’T WALK) and terminate the WALK with a steadyDON’T WALK. While omitting the pedestrian clearancetime (which may be over 20 seconds in some cases) willallow track clearance to begin sooner, pedestrians whomay be in the crosswalk will be placed at risk and befaced with opposing traffic if they are not providedadequate clearance time. Illinois is now placing signs(see left) to warn pedestrians of abbreviated walk times.

W ALK TIM E

SHORTENED

W HEN TRAIN

APPROACHES

CAUTION


Page 22

Clear TrackGreen Once the controller has terminated the conflicting phases,

the process of clearing vehicles from the path of the traincan proceed. Most traffic controllers allow the user tospecify which phase(s) is used to clear the track/crossing.In addition, a clear track green time is specified. It isessential to include a left-turn phase in track greenphases. This phase will have an arrow and will minimizeany confusion to the left turning traffic to quickly clear theintersection. The clear track green must be displayedlong enough to clear all vehicles that might be stoppedwithin the limits of the crossing. This duration is a factorof the design vehicle characteristics for the roadway, thegeometry of the crossing, and the distance between thestop bar (or normal stopping point) at the signalizedintersection and the warning device location (or normalstopping point) at the crossing (on the approach to thesignalized intersection). If there is a significant distancebetween the tracks and the signalized intersection, thegreen time provided must only be sufficient to clear thetracks/crossing and may not clear every vehicle from thestorage area (see Figure 4).

Before Clear Track Green

Before Clear Track Green After Clear Track Green

(b) Tracks far from intersection – entire storage area not cleared.

FIGURE 4. Example of Vehicles That Must be Cleared During the Clear Track Green

After Clear Track Green

(a) Tracks close to intersection – entire storage area cleared.


Page 23

The ITE Recommended Practice also suggests that thetrack clearance green be long enough to allow the signalto remain green for a few seconds after the railroadwarning devices have activated. This longer green wouldprevent the display of a “premature red” indication thatcould trap more vehicles on the tracks/crossing before thetrain arrives at the crossing.

Various methods exist to compute the duration of thetrack clearance green. Many state departments oftransportation use a procedure similar to that used inTexas, which is shown below.

Track Clearance Green based on Clearance Distance, L(Texas DOT)

L Minimum Desirable(feet) Track

ClearanceTrack

Clearance(seconds) (seconds)

25 6 1050 7 1075 9 10100 10 12125 11 14150 12 17175 14 19200 15 21

>200 Asdetermined

Asdetermined

This procedure assumes that all vehicles are through-moving passenger cars, vehicle length is 25 feet, and thatdeparture headways are consistent with Greenshields.The Track Clearance times above are factored up toaccount for trucks and/or left-turning vehicles along thetrack clearance approach. For each truck expected to bewithin the clearance distance, (L), a 1.5 multiplier is used;for each left-turning vehicle, a 1.3 multiplier is used.

More involved procedures for computing the duration ofthe track clearance phase have additional flexibility thatcan account for many design combinations andconsiderations, including many varieties of designvehicles (car, truck, bus, semi-trailer) and clearancedistances greater than 200 feet. Vehicle dynamics can beused to tabulate a host of useful information to aid in thedesign of track clearance green time, including how long it

L (Clearance distance)TrackPhase

RR Warning Device

Edge of the pavement


Page 24

takes a vehicle in queue to start moving, distance traveledwith respect to track clearance green time elapsed, andspeed attained as track clearance green time elapses.Examples of such values for different classes of designvehicle are shown below.

Time to Start Moving Nth Vehicle (in seconds)(Accelerating from a stop)

VehicleQueue

Position,N

AverageThrough

Car

AverageLeft-

TurningCar

AverageSingleUnit

Truck

AverageWB-15Truck

AverageSchool

Bus

1 2.2 2.0 2.5 4.0 2.72 3.4 2.9 3.9 7.0 4.5… … … … … …8 10.5 8.7 12.7 24.9 15.09 11.6 9.7 14.2 27.9 16.710 12.8 10.7 15.6 30.8 18.5

Distance Traveled in Time T (in feet)(Accelerating from a stop)

TimeElapsed,

T(seconds)

AverageThrough

Car

AverageLeft-

TurningCar

AverageSingleUnit

Truck

AverageWB-15Truck

AverageSchool

Bus

0.0 0.0 0.0 0.0 0.0 0.00.5 0.8 0.8 0.6 0.1 0.8… … … … … …4.5 56.3 49.3 42.0 11.8 56.55.0 68.2 59.0 50.9 14.5 68.85.5 81.1 69.3 60.4 17.5 82.1

Speed at Time T (in feet/second)(Accelerating from a stop)

TimeElapsed,

T(seconds)

AverageThrough

Car

AverageLeft-

TurningCar

AverageSingleUnit

Truck

AverageWB-15Truck

AverageSchool

Bus

0.0 0.0 0.0 0.0 0.0 0.00.5 3.2 3.1 2.4 0.6 3.1… … … … … …4.5 22.9 18.9 17.0 5.2 23.55.0 24.8 20.0 18.4 5.7 25.65.5 26.6 21.1 19.6 6.2 27.6

Additional details about vehicle dynamics, includingexpanded tables and source information, can be found inthe Appendix.

When designing and entering values for the trackclearance phase, remember that the clear track green


Page 25

Putting It AllTogether

must be terminated with a yellow change interval and all-red clearance interval (if used in normal signal operations)of sufficient duration.

The steps described above that comprise the entry intopreemption are completed every time a train approachesa highway-rail intersection with a nearby signalizedintersection. These steps can begin at random points inthe signal’s normal cycle and often occur in less time thanit takes to read this page.

To help more clearly demonstrate the order in which thesesteps occur, the following example of a simple, two-phaseintersection near a highway-rail intersection is presented.

Figure 5 shows a schematic of the example intersection.Notice that the signal is a two-phase signal with thenorthbound approach crossing the tracks. This approachis controlled by Phase 2 (∅ 2), which will be the cleartrack phase. Figure 6 shows a timeline of the events thatoccur at the crossing assuming ∅4 has just turned greenwhen the train enters the detection circuit.

As per the timeline in Figure 6, the traffic signal is notifiedof the train’s presence 6 seconds after the train enters thedetection circuit (the 6 second delay is due to theequipment response time of the train detection circuitry).Once the traffic signal is notified, the preemptionsequence begins (after a 1-second traffic signal controllerprocessing delay) by assuring that the current phases areactive for at least the minimum green time.

FIGURE 5. The Example Two-Phase Intersection

B Street

A S

tree

t

∅ 2

∅ 6

∅ 4∅ 4

N

On the H

ighway

Guide for T

raffic Signal P

reemption near R

ailroad Grade C

rossing

Page 26

FIGURE 6. The Time-Line of Events for the Example Intersection


Page 27

NOTE: T he ac tive p hasein th is exa m p le w aste rm ina ted on ly 6seconds (3 secon ds o fg reen a nd 3 secon ds o fye llow a nd a ll red ) a fte rthe p reem ption inp u t w asrece ived . T h is tim e w ou ldno t be e no ug h to a llo w afu ll ped es tr ian c lea rancefo r an av e ragein te rsec tion . If a fu llc lea rance w as ne ed ed ,up to an ad d ition a l 10seconds (o r m ore ) m igh tbe requ ire d b e fo re thec lea r track gree n cou ldbeg in . T h is extra tim ew o u ld m e an th a t the trackc lea r g reen w ou ld n o td isp la y un til 2 seco ndsa fte r the g a tes a re fu llylow ered , an d on ly 4seconds b e fo re the tra ina rr ives a t the cross ing .The effect of pedestrianclearance times onpreem ption cannot beover em phasized!

Next, nine seconds after the train has entered thedetection circuit, the right-of-way transfer interval beginsterminating the active phases that conflict with the cleartrack phase. After an additional 3 seconds, the clear trackgreen begins (just as the railroad warning gates begin tolower). For this particular crossing, the clear track greenlasts 9 seconds. This time is designed to be long enoughto clear the necessary vehicles at the crossing and allowthe signal to remain in clear track green for 2 secondsafter the railroad warning gate is completely lowered. Theclear track green is followed by 5 total seconds of cleartrack yellow change and all-red clearance intervals.Finally, 20 seconds after preemption began (and therailroad warning devices were activated), the train arrivesat the crossing and the preemption hold interval(discussed in the next section) begins.

It is important to recognize that the exact timeline that thissections presented will not always occur. This is due tothe fact that the exact timeline depends on the phase(s)active when the preemption input is received by thecontroller. For example, if the signal was in ∅2 and ∅6when the train entered the detection circuit and hadalready completed its minimum green time, the right-of-way transfer interval would have begun 7 seconds afterthe train was first detected; Therefore the clear trackgreen would end just as the gate completely lowered—2seconds earlier than before.


Page 28

PREEMPTIONHOLDOPERATION

NOTE: The redflashing of the trafficsignals (which indicatethat motorists mayproceed after acomplete stop) cancause some confusionwhen displayed inconjunction with theflashing red railroadsignals (which indicatethat you must stop andwait for a train.) Inaddition, signals willflash ALL RED duringa malfunction when atrain is not necessarilypresent at the crossing.Also, when a controlleris flashing due to amalfunction, it cannotrecognize a call forpreemption.

NOTE: O f course , twoconflic ting m ovem e ntscann ot b o th b e flash ingye llo w . If th is isp rogram m ed in to thecontro lle r, wh en thepreem ptio n ho ld in te rva lin itia te s, the co n flic tm on ito r (o r m a lfu nctionm ana gem ent u n it) wo u ldbe trip ped cau sing th esign a ls to g o on A LLR ED co nflic t flash andpreventin g the co n tro lle rfrom e xiting p reem ptio neven a fte r the tra in hascleared .

The preemption hold operation occurs after the trackclearance intervals have been completed, and while thetrain is occupying the crossing. The preemption holdoperation will remain in effect until the preemption input tothe controller is removed. The purpose of the preemptionhold operation is to allow those movements that do notconflict with the train to proceed through the intersection.Most modern controllers (both NEMA and Type170/179/2070) allow for the following modes of operation:

• FLASHING ALL RED

In this type of operation, all directions will flash RED.This flashing indicates that the motorist must come toa complete stop, and only proceed when it is safe todo so (following the applicable right-of-way rules inthat jurisdiction). Most often under this operation, thepedestrian signals are kept dark. This type ofoperation is used when some movements (such asright turns) can safely take place, but a separatephase or signal indication does not exist to provideonly those movements with a green indication.

• YELLOW / RED FLASH

In this mode of Preemption Hold operation, designatedphases which do not conflict with one another will flashyellow while the remaining phases will flash red.Motorists faced with a flashing yellow signal shouldapproach the intersection with caution, but do not needto stop before proceeding. The approaches with theflashing red signal must come to a complete stop andyield to the other traffic before proceeding. In thismode, the flashing yellow would be used for thosesignal heads where all of the movements controlled bythose heads can safely proceed while the train ispresent. The flashing red signals indicate that some ofthe movements controlled by the particular signalcannot proceed safely. This mode of preemption holdis not used very often in common practice.

• STEADY ALL RED

As its name implies, this mode of operation provides asteady (not flashing) red indication on all approaches.In addition, the solid DON’T WALK is usually


Page 29

displayed. This mode of operation is very uncommonand should be used only when no movements cansafely occur at the intersection (or no separate signalheads or phases exist for those movements that couldproceed).

• LIMITED SERVICE

The most versatile method of preemption hold is thatwhich limits those phases that can be serviced tothose that can proceed safely while the train is presentat the crossing. All phases that control movementsconflicting with the train remain red as the controllercycles through the non-conflicting movements (or, insome cases, only one movement is non-conflicting sothat movement remains green). The controller canalso allow non-conflicting pedestrian indications tocycle with their complementary phases or can be heldon a solid DON’T WALK (the pedestrian heads canalso be dark, but this is not recommended for limitedservice operation.) The illustration below provides anexample of the movements that could be serviced.

• REST IN GREEN

The final mode of preemption hold simply allows theprimary movement(s), which does not conflict with thegrade crossing (i.e., most commonly the two throughphases on the road paralleling the tracks), to rest in itsgreen interval.

B Street

A S

tree

t

Serviceablemovement

Although notrequired, separateleft turn phases onthe SB, EB, and WBapproaches wouldbe desirable for thistype of operation.


Page 30

EXIT FROMPREEMPTION

NOTE: Somecontrollers will ignore asecond preemptioninput if this input isreceived while thecontroller is exitingfrom the firstpreemption. Thissituation could causethe green indication tobe given to preciselythe wrong movementas a second trainapproaches thecrossing. Be sure tocheck with yourcontroller manufactureron how your controllerhandles a secondpreemption call.

PREEM PTIONTIM EREQUIREM ENTS

Traffic SignalControllerLag Time

Right-of-WayTransfer Time

The last step in traffic signal preemption for signalizedintersections near highway-rail intersections is returning tonormal operations once the train has left the crossing.This step occurs immediately after the preemption input isremoved from the controller.

Most controllers in use today allow specific return phasesto be specified. These phases would be serviced firstonce the preemption input has been removed and theactive phases have been terminated with proper changeand clearance intervals. The exit phases chosen dependon the particular intersection. In many cases, the returnphases would be those movements that were held whilethe train was occupying the crossing. However, at someintersections, the blocked movements may have lowvolumes and an immediate service may not be desirablefor coordination or other intersection operations as awhole.

Now that all of the preemption steps have been identified,the time that the traffic signal requires to provide thenecessary track clearance can be determined. The trafficsignal design time (TSDT) is equal to the sum of threecomponent times: traffic signal controller lag time(analogous to equipment response time for railroad trackcircuitry), right of way transfer time (change and clearanceinterval times), and the clear track green time. Thesecomponents will be discussed in detail below.

As with the railroad warning equipment, traffic signalcontrollers will take some time to process the preemptioninput before the preemption sequence begins. Withtoday’s modern controllers, the response time is usuallyno greater than 1 second, but check with your controller’smanufacturer to obtain the correct value; in some cases,the lag time value can be programmed.

The time requirements for the change and clearanceintervals depend on the signal phasing used at theintersection and the intersection geometry. Whenchoosing which phase (or phases) to use in calculatingthe time requirements, the engineer must consider the


Page 31

“worst case” scenario. The “worst case” scenario requiresthe most time to safely terminate the current phasesbefore the clear track interval can begin (i.e., 1 secondafter the start of green for the conflicting phase with thelongest minimum green and pedestrian clearancerequirements). The following items are included in theright-of-way transfer time.

Pedestrian Clearance Time

The pedestrian clearance time that is used to determinepreemption requirements depends on local practiceregarding shortening or omitting the WALK and flashingDON’T WALK indications. If shortened times are set inthe controller, this calculation uses those shortened times.If the full WALK and flashing DON’T WALK times are setin the controller, then those times must be considered forthis calculation (even if pedestrian push buttons arepresent at the intersection).

Minimum Green for the Conflicting Phase

The minimum green that must be displayed for eachphase before it can be terminated is used for this portionof the total time calculation. In some controllers, thisminimum green is the minimum green time that is input fornormal signal control operation. Some jurisdictionsrequire that at least some amount of minimum green beused, even though the MUTCD allows the minimum greento be abbreviated.

Yellow Change and Red Clearance Times

The final component of the change and clearance intervalare the required YELLOW change and ALL REDclearance times for the phase (or phases) that are drivingthe calculation. The MUTCD states that all requiredYELLOW and ALL RED times be provided duringpreemption.

The total right-of-way transfer time is the sum of theyellow change and red clearance times and the larger ofthe pedestrian clearance and minimum green times. Allphasing combinations should be checked to ensure thatthe longest RTT is used.


Page 32

Clear TrackGreen Time

NOTE: If the tracks crossm ore than one ap proach ,bo th a ppro aches w ill h aveto be c lea re d . C a lcu la tea separa te c le a r tracktim e fo r each ap proachand ad d them tog e the r toob ta in the to ta l c lea r tracktim e .

DESIGN ANDOPERATIONSCONCERNS

SimultaneousPreemption

Design Case

As its name implies, the clear track green time is the timerequired to safely clear the tracks/crossing before thetrain’s arrival. Clear track green time includes the timerequired to dissipate the queue that has built up betweenthe signalized intersection and the tracks/crossing, thetime required to move the design vehicle a safe distanceaway from the tracks (i.e., out of the crossing), and aseparation time between the clearing of the last vehicleand the arrival of the train.

These timings can be summed to obtain the TSDT. If theTSDT is more than the total warning time in the rail design(MWT) then additional warning time is required from therailroads. This additional time can be in the form ofadvance preemption time. However, in some cases thisadditional time can also be obtained by increasing the CTand thereby increasing a component of MWT in the raildesign.

In the preemption design process documented in theliterature and used by many state DOTs, the trackclearance green is calculated based on a single scenario.This scenario assumes that the traffic signal controllerreceives the preempt call just after the onset of green onthe approach having a phase conflicting with the trackclearance phase. Any desired minimum green timeand/or pedestrian clearance time, and the necessaryvehicle clearance time, has to be provided for the longestconflicting phase(s) before the initiation of the trackclearance phase. Colloquially, this set of circumstances isreferred to as “worst case” scenario. It is used as thedesign case in standard practice because it requires thelongest time to arrive at the track clearance phase. It isabsolutely essential to note that this set of conditions onlyestablishes only the upper bound of maximum right-of-way transfer time.

Real-world conditions require that multiple considerationsand design “cases” be factored into establishingpreemption timing. The preempt call can come in to thetraffic signal controller at any point in the cycle. Thus,preemption design should consider the impacts of thepreempt call coming in at different points in the cycle,including the scenario where the controller is already inthe track clearance phase when it receives the preempt


Page 33

CrossingWarning Time

Variability(simultaneous

preemption

call. This scenario has been traditionally known as the“best case” scenario and can have a significant impact onpreemption operations.

During simultaneous preemption, railroads are required toprovide at least 20 seconds of minimum time, and theclearance time is added to this value to produce thecrossing's MWT. Traffic signal preemption is designedaccording to the MWT value. As shown below, all actionsof the crossing warning system and the traffic signalcontroller are well coordinated under ideal conditions.

Simultaneous Preemption - Design CaseCrossing Warning Time = Preemption Warning Time = 20 seconds

with CT = 0 Seconds"Worst Case" Scenario, Maximum Right-of-Way Transfer Time

Queue Clearance Time

Lights

Gates

Signal

Queue

Time

Warning Railroad Warning Time (MT = 20 Seconds)

Gates DescendingFixeddelay

Maximum RTT

Train ArrivesAt Crossing

Preempt To ControllerLights Start To Flash

Track Green

Lights Flashing

Gates Down

0 10 20

However, traffic engineers can fall into the trap of makingover simplified assumptions about the nature of availablewarning time. Due to the time required for theirequipment to measure and estimate the train speed andto allow for some buffer time, railroads usually providelonger warning times than the 20-second absoluteminimum. This extra warning time is not guaranteed and,hence, is not typically relied upon by the traffic engineer.But the engineer should be aware of this extra time, as itmay have some implications for preempt operations.


Page 34

Preemption andCrossing

Warning TimeVariability

(simultaneouspreemption)

Simultaneous Preemption with Rail VariabilityCrossing Warning Time = Preemption Warning Time > 20 seconds

"Worst Case" Scenario, Maximum Right-of-Way Transfer Time

(variable) Minimum Warning Time (MT + CT)

Railroad Warning Time

Lights Flashing

Gates DownGates Descending


Track GreenMaximum RTT Yellow

Lights

Gates

Signal

Queue

Time

Warning

fixeddelay



0 10 20

Separation Time

Another case to analyze is where a preempt call isreceived when the controller is in the track clearancephase, depicting the "best case" scenario. The "bestcase" scenario can occur with any amount of railroadcrossing warning time variability. The significance of thiscase is the early termination of the track clearance phase;the traffic engineer should be aware of theimplications/results of this case occurring.

Simultaneous Preemption with Rail and Signal VariabilityCrossing Warning Time = Preemption Warning Time > 20 seconds

"Best Case" Scenario, Right-of-Way Transfer Time = 0

(variable) Minimum Warning Time (MT + CT)


Lights Flashing



Track Green Red / DwellYellow

Lights

Gates

Signal

Queue

Time

Warning

fixeddelay



RTT = 0 Seconds

Separation Time

0 10 20


Page 35

.

AdvancePreemption

Design Case

During simultaneous preemption design, the trafficengineer assumes that “only” 20 seconds of warning timeis available. The engineer then considers the “worstcase” scenario, where the controller has just started aphase conflicting with the track clearance phase, anddetermines the maximum right-of-way transfer time andthe necessary track clearance time. If the sum of themaximum right-of-way transfer time and the trackclearance time is more than 20 seconds, the engineerrequests advance preemption or additional CT. In somecases (and in accordance with jurisdiction policy, ifpresent), the engineer also has to provide time to clearpedestrians who have already entered the intersection. Insuch cases, the right-of-way-transfer time includes thepedestrian requirements. The traffic engineer requeststhe necessary advance preemption (for example 15seconds) from the railroad authorities and designs thesignal preemption plan based on total available warningtime (for example 20 + 15 = 35 seconds). The railroadauthorities will almost always provide the requestedadvance preemption time. A diagram of advancepreemption design is shown below.

Advance Preemption - Design CaseCrossing Warning Time = 20 seconds

Preemption Warning Time = 35 seconds"Worst Case" Scenario, Maximum Right-of-Way Transfer Time (RTT)

Track Green


APT Minimum Warning Time (MT + CT)

Lights Flashing


Lights StartTo Flash

Maximum RTT

Lights

Gates

Signal

Queue

Time

Warning

fixeddelay


Preempt ToController

Similar to Simultaneous Preemption,But the RTT Starts Earlier

0 10 20


Page 36

CrossingWarning Time

Variability(advance

preemption)

It is essential to remember that there is always variabilityin the actual warning times provided for each trainmovement. Reasons for the variation include traindeceleration or acceleration after the advance preemptcall is placed, computational variability in the railroad’sdetection equipment, and equipment lag time. Hence, thetrain will typically arrive at the crossing on time or earlierthan indicated in the design (due to variability and buffertime). But the train could occasionally arrive later due to arapidly decelerating train. Train deceleration and buffertime can result in early termination of RTT with respect tothe rail warning devices. In turn, this early RTTtermination can lead to early termination of the trackclearance green.

The figure below highlights several, but not all, of theincarnations of variability found in railroad warning times.The distribution to the left shows the variability in gatedown time with respect to train arrival at the crossing.The distribution to the right shows the variability inadvance preemption times sent from the railroad'swarning system to the traffic signal controller unit. Notethat these data are not meant to be representative of allcrossings; and field data collected at only one site.Further, train handling in the vicinity of the location whereresearchers collected the data has the impact ofproducing warning times that are more variable thanwould be found with only through train operations.

Warning Time Variability

P reem ption W arning T im e D istribution

0

5

10

15

20

25

30

0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68

W arning T im e (seconds)

Nu

mb

er

of

Tra

ins

95th Percentile= 57 Seconds

35

40

45

G uaranteed by Railroad:35 Seconds

20 s basic + 15 sadvance preem ption

AdvancePreem ption

Tim eM ean = 48Seconds

Tim e ofG ate D ow nM ean = 26Seconds

M inim um W arningTim e = 20 seconds


Page 37

Preemption andCrossing

Warning TimeVariability

(advancepreemption)

The figure below expands the advance preemption designcase diagram to include rail warning time variability.

Advance Preemption with Rail VariabilityCrossing Warning Time > 20 seconds

Preemption Warning Time > 35 seconds"Worst Case" Scenario, Maximum Right-of-Way Transfer Time

Variation in the advance preemption time and theminimum warning time can be unsafe when combinedwith variation in signal operations (i.e., entry intopreemption at any point in the cycle). A critical case insignal variation is when the preempt call is received whenthe signal controller is already in the phase that is used asthe track clearance phase. In this case, the RTT is non-existent (i.e., the track clearance green initiatesimmediately), and the track clearance phase canterminate before the railroad warning lights start flashing.In such a situation, drivers approaching the crossing afterthe termination of the phase (track clearing phase) maysee the railroad lights start flashing after already queuingup over the crossing. However, since the track clearancephase has already been served, the vehicles queued willnot be serviced again until the train completes its passagethrough the crossing and the preempt call is lifted. Thisset of circumstances has the potential to result in adangerous situation. Such a critical situation can occurwhen the preempt call is received during the track phaseand there is significant variation in the advancepreemption time and/or the minimum warning time. Thisscenario needs to be studied in detail and the designmodified to account for the track clearance phase

Time

Maximum RTT

��(variable)

��Separation Time

Gates Down

��Gates Descending

Yellow


Track Green

fixeddelay

��(variable)

Actual Advance Preemption Time

APT Minimum Warning Time (MT + CT)Railroad Warning Time

Lights Flashing

Red / Dwell


Lights

Gates

Signal

Queue

Warning



RTT Starts Earlier Than the Design Case

0 10 20


Page 38

terminating prematurely (i.e., before the railroad lightsstart flashing). The timings implemented in the controllermust be adjusted for to account for the “best casescenario.”

Advance Preemption with Rail and Signal VariabilityCrossing Warning Time > 20 seconds

Preemption Warning Time > 35 seconds"Best Case" Scenario, Right-of-Way Transfer Time = 0

The diagram above highlights deficiencies with thepractice of designing preemption only considering the”worst case" scenario. Even if warning times for thecrossing warning devices were minimal (i.e., only 20seconds of crossing warning time provided), the fact thatthe signal could already be in the track clearance phase atthe onset of preemption, combined with the trackclearance phase being a set duration, could lead to a"preempt trap." Such a situation can only be avoided ifboth the "best case" and "worst case" scenarios areexamined during preemption design.

The preceding presentation of simultaneous and advancepreemption, demonstrates that myriad combinations of railwarning time and traffic signal status at preemption arepossible. However, because the rail crossing's activewarning devices and the traffic signal controller of theadjacent intersection receive their notification of anarriving train at the same time under simultaneouspreemption, their actions are reasonably well coordinated.However, with advance preemption, the crossing warningsystem is explicitly designed to provide the traffic signalpreemption warning time before the crossing warningdevices begin operation.

Signal

Queue

��

��

(variable)

Actual Advance Preemption Time

APT

��(variable) Minimum Warning Time (MT + CT)


��Gates Descending

Yellow


Track Green

Preempt Trap

��Separation Time

Gates Down

Lights Flashing

Red / Dwell


Lights

Gates

Time

Warning

fixeddelay



RTT = 0 Seconds

0 10 20


Page 39

USE OFPRE-SIGNALS

One method that is becoming more commonly used toprevent vehicles from stopping on the tracks/crossing isthe use of pre-signals at a highway-rail intersections. Apre-signal consists of signal indications that are locatedon the near-side of the tracks either on a separate mastarm or as part of the mast arm for the railroad warningdevices. These displays are accompanied by a STOPHERE ON RED sign and/or a DO NOT STOP ONTRACKS SIGN.

The pre-signal indications are designed to turn red beforethe accompanying far-side indications to prevent peoplefrom queuing between the signalized intersections and thetracks. During signal preemption, only the far-side signalsare used to clear the tracks and the pre-signal remainsred throughout preemption. The photograph belowdemonstrates a pre-signal used in Coppell, Texas.

Guide for Traffic Signal Preemption near Railroad Grade Crossing Important Considerations

Page 41

PEDESTRIANS

As we have seen, preemption of traffic signals at highway-rail intersections is a very complicated process. Theprocess is further complicated since no two crossings areexactly the same. However, several situations arecommonly encountered that can have a significant impacton the time required to clear the tracks. The sectionbelow presents a few cautionary notes about some ofthese common considerations. Please be aware that thislist is not exhaustive and other situations may beencountered that require special consideration whendesigning the preemption sequence.

The issue of pedestrians at highway-rail intersections hasbeen touched on in earlier chapters but is repeated hereto emphasize how crucial it is to consider the effect ofpedestrian clearance times on the time required tocomplete preemption.

It is allowable (MUTCD) to shorten or omit pedestrianWALK and flashing DON’T WALK intervals for thepurpose of beginning the clear track interval earlier.However, keep in mind that by shortening these intervals,some pedestrians may be trapped in the intersectionfacing oncoming vehicles that are clearing thetracks/crossing. Shortening pedestrian times may not bea viable option at locations where pedestrian traffic is veryheavy (such as near a public plaza or college campus).

If the pedestrian clearance intervals are not shortened,the traffic signal design time must consider the entire timethat is necessary to clear pedestrians. When pedestrianpush buttons are the only mechanism calling thepedestrian phase, it is easy to forget to include thepedestrian clearance time in minimum time calculations,especially when pedestrian traffic is minimal.

��

Important Considerations Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 42

MULTIPLETRACKCROSSINGS

Multiple tracks at highway-rail intersections introduce twoproblems that must be considered when designing apreemption timing plan. The first problem is the additionalclearance distance that must be cleared during the cleartrack interval. The greater clearance distance increasesthe clear track green that is required and thus increasesthe total approach time that preemption requires. Thefigure below demonstrates how the minimum clearancetime is measured for various types of track configurations.

The second problem that multiple track crossings presentis the possibility that a second preemption call could besent to the controller immediately after the first preemptioninput is removed (and the exit from preemption sequenceis taking place). This situation occurs when a traintraveling on the second track approaches a crossing rightafter a train on the first track has left the crossing area.Some controllers will ignore this second preemption inputif it occurs during the return to normal operations, whichcould lead to skipping the clear track interval andpotentially trapping vehicles on the tracks/crossing. If thecontroller accepts the preemption input, short indicationson conflicting phases or an early entry into the clear trackphase could result. If the latter occurs, the red clearance


Page 43

MULTIPLECLEARANCEINTERVALS

that concludes the track clearance phase may occurbefore the warning gates have even fully lowered (thuspotentially trapping vehicles on the tracks). Contact yourcontroller manufacturer(s) to obtain complete informationon how your controller(s) will respond to a secondpreemption input at any and all times during thepreemption sequence. Test your equipment wheremultiple preemption inputs are a possibility. Also, if twotracks are present and APT is requested, make sure therailroad keeps the devices active if a second call (APT) isactive at the termination of the first call. This test isnecessary because the traffic signal controller does notrecognize the second call while in preemption.

When railroad tracks cross two or more approaches to asignalized intersection, multiple clear track intervals will berequired to ensure that that all vehicles are cleared fromthe tracks/crossings. The order or track clearancedepends on the direction that the train is traveling. In theexample intersection below, A Street would have to becleared first if a train was coming from the bottom-right ofthe figure. On the other hand, a train traveling in theopposite direction would force B Street to be cleared first.(Two approaches require setting up two preemption inputsin the controller that are activated based on the directionof train travel). Regardless of which street is cleared first,the need for multiple clear track intervals greatly extendsthe time needed for safe completion of the preemptionsequence. For example, if full pedestrian clearance isallowed during the right-of-way change interval, a totalapproach time of over 60 seconds may be required. Inthis situation, redesigning the traffic flow pattern or theintersection geometry may be more feasible thanproviding the necessary (lengthy) train detection circuit.

B Street

A S

tree

t

Requiredtrack clear


Page 44

INTERSECTIONDESIGNVEHICLE

TYPE OFTRAINS

When determining the queue length that vehicles mustclear and the time required to clear the tracks/crossing, itis necessary to have complete information about thelength and performance capabilities of the design vehicle.In most cases, a standard truck or bus should be used asa design vehicle. However, at crossings that are inindustrial areas, a larger truck with a more gradualacceleration rate should be used to account for the typesof vehicles that would commonly use the crossing.Similarly, if a crossing is near a school or on a knownschool bus route, the characteristics of the school busshould be used as the design vehicle when calculatingnecessary clearance time (if a bus is the largest designvehicle using the roadway and crossing).

Just as the type of vehicles using the crossing isimportant, the types of trains using the crossing can havean effect on the preemption as well. Preemptionsequences are easier to design when tracks carry onlyone type of train (such as freight or commuter). Underthis circumstance, the trains will be travelling atapproximately the same speed and thus arrive at thecrossing at a more consistent time after entering the trackcircuit detection zone. When faster passenger trainsshare a crossing with slower freight trains, differences inwarning time result. Preliminary engineering of thepreemption sequence should include knowledge of typesof trains using the crossing and train handling practices inthe vicinity of the crossing (e.g., nearby stops, depots,switching yards). Following implementation, the engineershould observe the preemption sequence during thepassage of all train types to ensure it is appropriate for allsituations.


Page 45

LIGHT-RAILTRAFFICSIGNALPREEMPTION

Constant warning time devices do help alleviate some ofthe problems mentioned above. However, if a commutertrain stop is located very close to a crossing, some trainsmight accelerate from the stop through the crossing, andshorter than expected warning times could result.

Urban light-rail transit trains often run on tracks thatparallel or run in the median of a major arterial roadway.LRT trains are lighter and travel slower than heavy-railfreight and commuter trains. For this reason, LRT trainscan be made to stop at a signalized intersection and donot always preempt the signal.

In some situations, signal priority is used instead ofpreemption for LRT. Priority routines will shorten orextend phases if possible to try to prevent the LRT fromhaving to stop. In other cases, a special cycle length isused to correspond with the train schedule to provide agreen band for the train as it crosses severalintersections.

In those situations that require preemption of a trafficsignal, the preemption routine should be designed as forany other type of train. It should be noted that CWTrailroad circuitry cannot be used on electrified tracks (asare found with LRT and some commuter and heavy railtransit rail systems. An example of a light-rail line in themedian of a roadway in Dallas, Texas, is shown in thephotograph below.


Page 46

FAILURE OFPREEMPTIONCIRCUIT

NOTE: U n der noc ircum s tances shou ld thep reem ption cap ab ilit ies o fthe con tro lle r be d isa b le dif the p re em ptio n c ircu itfa ils and the ra ilroa d trackis s till ac tive ly use d .

The last topic for discussion involves what occurs if thepreemption circuit that connects the traffic signal controllerto the railroad detection and processing equipment fails.If the circuit should fail for any reason, the preemptioninput will be opened (de-energized) and the preemptionroutine will begin. After completing the clear trackinterval, the signal will remain in the preemption holdinterval until the preemption circuit is repaired.

On some controllers, there is a timer that can be set tospecify the maximum time a signal can remain inpreemption before it assumes that the circuit has failed.Once the timer has expired, the signal will revert toflashing RED if it has been cycling in limited service orholding in green during preemption.

Once a problem has been identified, the traffic signalpersonnel should work with the railroad maintenancepersonnel to ensure that the problem is corrected asquickly as possible.

Guide for Traffic Signal Preemption near Railroad Grade Crossing Recommendations & Conclusion

Page 47

��

��

Safe operation of railroad preemption of traffic signalsrequires scrutiny of all aspects of the design andoperations of both railroad warning equipment and trafficsignal controllers. It is fundamental when designing suchsystems to realize that the design case establishes onlythe upper bound of warning time requirements; real-worldoperations will, in all practical cases, not require all of thewarning time that is provided. Both the "worst case" (i.e.,design scenario) and the "best case" (i.e., right-of-waytransfer time equals zero) situations must be fullyanalyzed in the context of crossing warning time andpreemption warning time variability.

In the absence of an analysis of variance in crossing andpreemption warning times, increase the track clearancetime in the controller by the duration of advancepreemption time to avoid displaying a "preempt trap" formotorists on the track clearance approach. In somecases, the extension of the track clearance phaseduration may mean that the track clearance green willextend beyond train arrival at the crossing. Analyze thegreen extension to determine an acceptable duration,realizing that the extra green time could result in inefficientoperations due to a delay in transitioning to thepreemption hold, or dwell state.

Some general recommendations for improving both thedesign and operation of railroad preemption of trafficsignals include:

• If using preemption without gates, install gates.• Request two preempt inputs to the signal controller for

each train arrival (first - early - preempt to initiatevehicle and pedestrian clearances, second - later -preempt to time the track clearance phase).

• Incorporate a “gate down” signal from the crossingwarning device into preemption sequence (extendtrack clearance phase until gates are down).

• To prevent re-queuing (i.e., "preempt trap") on thecrossing, install a “not-to-exceed” timer to force

RECOMMENDATIONS

Recommendations and Conclusions Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 48

CONCLUSION

activation of the crossing warning devices prior to thetime they would be activated by the railroad crossingpredictor, if necessary (used if train decelerates in thedetection circuit). This option is being recommendedby AREMA as a solution for the "preempt trap"problem.

Signal preemption practices have changed very littlethroughout the years; for instance, even in today'senvironment of microprocessor controller traffic signalcontrol equipment, the interconnect circuit with the gradecrossing warning equipment remains a single on/offconnection. However, complexities are introduced asvolumes of vehicles and trains increase, as real-worldoperational inconsistencies are considered, and as thesystems become integrated to provide relatively seamlessfunctions. The Fox River Grove school bus/train collisionin 1995 shed new light on the issue of preemption andhas sparked new efforts to improve the safety andefficiency of operations at highway-rail intersections.Though it is unfortunate that such a collision had to occurbefore the attention of the transportation engineeringcommunity focused on railroad and signal preemptionissues, valuable lessons have been learned about saferdesign and operations procedures.

It is important to have a thorough understanding of thecurrent state-of-the-practice in signal preemption that thisguide presented. While every highway-rail intersectionpresents unique challenges, understanding the processand being aware of the potential pitfalls can help theengineer design a safe preemption sequence for even themost complicated intersection. It was the purpose of thisguide to raise these pertinent issues.

Guide for Traffic Signal Preemption near Railroad Grade Crossing Sources

Page 49

The information in this handbook and workshop wasderived from the following sources:

Accidents That Should Not Happen: Executive Summary.Institute of Transportation Engineers. March 1996.[http://www.ite.org/blribrep.htm]

Accidents That Shouldn’t Happen. USDOT, Washington,D.C., 1996.

Campbell, R. Interconnection of Highway-Rail GradeCrossing Warning Systems and Highway Traffic Signals.Railroad Controls Limited, 1998.

Implementation Report of the USDOT Grade CrossingSafety Task Force. Office of Highway Safety, FederalHighway Safety Administration, USDOT, Washington,D.C., 1997.

Mansel, D.M., Waight, V.H. and Sharkey, J.T. SupervisedInterconnection Circuits at Highway-Rail Grade Crossings.Paper prepared for the 78th Annual Meeting of theTransportation Research Board, Washington, D.C.,January 1999. (Preprint)

Manual on Uniform Traffic Control Devices for Streets andHighways. Federal Highway Administration, UnitedStates Department of Transportation, 1988.

Manual on Uniform Traffic Control Devices for Streets andHighways. Federal Highway Administration, UnitedStates Department of Transportation, 2000 (draft -proposed text).[http://ohs.fhwa.dot.gov/devices/mutcd.html]

Marshall, P.S. and Berg, W.D. Design Guidelines forPreemption at Signalized Intersections. ITE Journal, Vol.67, February 1997, pp. 20-25.

��

http://www.ite.org/blribrep.htm

Sources Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 50

Marshall, P.S. and Berg, W.D. Evaluation of RailroadPreemption Capabilities of Traffic Signal Controllers.Transportation Research Record 1254, TRB, NationalResearch Council, Washington, D.C., 1990, pp. 44-49.

Oppenlander, J.C. and Oppenlander, J. StorageRequirements for Signalized Intersection Approaches.ITE Journal, Vol. 66, February 1996, pp. 27-30.

Preemption of Traffic Signals at or near Railroad GradeCrossings with Active Warning Devices: A RecommendedPractice. Traffic Engineering Council Committee TENC-4M-35, Institute of Transportation Engineers, Washington,D.C., February 1997.

Signal Manual: A Manual of Recommended Practices.Association of American Railroads, Volume 3, Section 1,1996. (The Signal Manual is now maintained by theAmerican Railroad Electrical Manufacturer’s Association -AREMA).

Traffic Control Systems (Standards Publication TS 1-1989). National Electrical Manufacturer’s Association,1989 (revised 1994).

Traffic Controller Assemblies (Standards Publication TS2-1992). National Electrical Manufacturers Association,1992.

Traffic Signal Operations Near Highway-Rail GradeCrossings. National Cooperative Highway ResearchProgram Synthesis 271, NCHRP, TransportationResearch Board, National Research Council, 1999.

Tustin, B.H., Richards, H., McGee, H., and Patterson, R.Railroad-Highway Grade Crossing Handbook, 2nd Edition.FHWA-TS-86-215. United States Department ofTransportation, Federal Highway Administration,September 1986.

Guide for Traffic Signal Preemption near Railroad Grade Crossing Appendix

Page 51

Vehicles Dynamics for Computing Track Clearance Phase Duration

Vehicles Dynamics when Accelerating from a Stopped Position

Time (s) to start moving for the Nth vehicleVehicle

(N)FastCar

AverageThru Car

AverageLeft T.

Car

SlowCar

AverageSU

AverageWB-15

SlowWB-15

AverageSchool

Bus1.0 2.0 2.2 2.0 2.5 2.5 4.0 4.6 2.72.0 3.0 3.4 2.9 3.9 3.9 7.0 8.2 4.53.0 4.0 4.5 3.9 5.4 5.4 10.0 11.7 6.24.0 5.0 5.7 4.9 6.8 6.9 12.9 15.3 8.05.0 6.0 6.9 5.8 8.3 8.3 15.9 18.9 9.76.0 7.0 8.1 6.8 9.8 9.8 18.9 22.5 11.57.0 8.0 9.3 7.8 11.2 11.2 21.9 26.1 13.28.0 9.0 10.5 8.7 12.7 12.7 24.9 29.6 15.09.0 10.0 11.6 9.7 14.1 14.2 27.9 33.2 16.710.0 11.0 12.8 10.7 15.6 15.6 30.8 36.8 18.5

Methodology from: Modeling Queued Driver Behavior At Signalized Junctionsby Jim BonnesonTransportation Research Record, 1992, (1365) p# 99-107

Numbers from: 1. Acceleration Characteristics of Starting Vehicles by Gary Long, University of Florida TRB 2000 Preprint Paper No. 00-09802. School Bus Acceleration and Sight Distance by J.L. Gattis, S.H. Nelson, and J.D. Tubbs Paper submission to ASCE

Assumptions: 1 seconds initial perception-reaction time (Bonneson's tau)The following saturation flows and vehicle queue spacing:

Sat. Flow Queue space(vphgpl) (ft/veh)

Fast Car 2600 25Average Thru Car 2200 25Average Left T. Car 2000 25Slow Car 1800 25Average SU 1500 36Average WB-15 900 61Slow WB-15 750 61Average School Bus 1500 46

��

Appendix Guide for Traffic Signal Preemption near Railroad Grade Crossing

Page 52


Distance traveled (ft) in time T from a stopped positionTimeT (s)

FastCar

AverageThru Car

AverageLeft T.

Car

SlowCar

AverageSU

AverageWB-15

SlowWB-15

AverageSchool

Bus0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.00.5 0.9 0.8 0.8 0.7 0.6 0.1 0.1 0.81.0 3.5 3.2 3.1 2.9 2.4 0.6 0.2 3.11.5 7.7 7.0 6.7 6.3 5.3 1.3 0.6 6.82.0 13.4 12.2 11.5 11.0 9.2 2.4 1.0 12.02.5 20.6 18.7 17.3 16.9 14.1 3.7 1.5 18.53.0 29.1 26.4 24.1 23.8 19.8 5.3 2.2 26.23.5 38.9 35.3 31.8 31.8 26.5 7.2 3.0 35.24.0 50.0 45.3 40.2 40.7 33.9 9.3 3.9 45.34.5 62.2 56.3 49.3 50.5 42.0 11.8 5.0 56.55.0 75.5 68.2 59.0 61.1 50.9 14.5 6.1 68.85.5 89.9 81.1 69.3 72.5 60.4 17.5 7.4 82.16.0 105.2 94.8 80.1 84.6 70.5 20.8 8.8 96.46.5 121.5 109.3 91.3 97.5 81.2 24.3 10.3 111.67.0 138.7 124.5 102.9 111.0 92.5 28.1 12.0 127.77.5 156.6 140.5 114.8 125.0 104.2 32.1 13.7 144.68.0 175.4 157.2 127.1 139.7 116.4 36.4 15.6 162.38.5 194.9 174.4 139.7 154.9 129.1 41.0 17.6 180.99.0 215.2 192.3 152.5 170.5 142.1 45.8 19.7 200.19.5 236.0 210.8 165.5 186.7 155.6 50.9 21.9 220.110.0 257.6 229.7 178.7 203.3 169.4 56.2 24.2 240.710.5 279.7 249.2 192.2 220.3 183.5 61.8 26.6 262.011.0 302.4 269.1 205.8 237.6 198.0 67.6 29.2 283.911.5 325.6 289.5 219.5 255.4 212.8 73.6 31.8 306.412.0 349.4 310.3 233.4 273.4 227.9 79.9 34.6 329.512.5 373.6 331.4 247.4 291.8 243.2 86.4 37.5 353.113.0 398.3 353.0 261.4 310.5 258.7 93.2 40.5 377.213.5 423.4 374.9 275.6 329.4 274.5 100.1 43.6 401.814.0 448.9 397.1 289.9 348.6 290.5 107.4 46.8 426.914.5 474.8 419.6 304.3 368.1 306.8 114.8 50.1 452.415.0 501.1 442.4 318.7 387.8 323.2 122.5 53.5 478.4

Methodology from: Modeling Queued Driver Behavior at Signalized Junctionsby Jim BonnesonTransportation Research Record, 1992, (1365) p # 99-107


Guide for Traffic Signal Preemption near Railroad Grade Crossing Appendix

Page 53


Speed (fps) at time T when accelerating from a stopped positionTimeT (s)

FastCar

AverageThru Car

AverageLeft T.

Car

SlowCar

AverageSU

AverageWB-15

SlowWB-15

AverageSchool

Bus0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.00.5 3.5 3.2 3.1 2.9 2.4 0.6 0.2 3.11.0 6.8 6.2 5.9 5.6 4.7 1.2 0.5 6.11.5 10.0 9.1 8.4 8.2 6.8 1.8 0.7 8.92.0 12.9 11.7 10.7 10.6 8.8 2.4 1.0 11.62.5 15.7 14.3 12.7 12.8 10.7 2.9 1.2 14.23.0 18.4 16.6 14.5 14.9 12.4 3.5 1.5 16.73.5 20.9 18.9 16.1 16.9 14.1 4.1 1.7 19.14.0 23.3 21.0 17.6 18.7 15.6 4.6 2.0 21.44.5 25.6 22.9 18.9 20.4 17.0 5.2 2.2 23.55.0 27.7 24.8 20.0 22.1 18.4 5.7 2.4 25.65.5 29.7 26.6 21.1 23.6 19.6 6.2 2.7 27.66.0 31.6 28.2 22.0 25.0 20.8 6.8 2.9 29.56.5 33.4 29.8 22.8 26.3 21.9 7.3 3.1 31.37.0 35.1 31.3 23.6 27.6 23.0 7.8 3.4 33.07.5 36.8 32.6 24.2 28.7 24.0 8.4 3.6 34.78.0 38.3 33.9 24.8 29.8 24.9 8.9 3.8 36.38.5 39.8 35.2 25.4 30.9 25.7 9.4 4.1 37.89.0 41.1 36.3 25.9 31.8 26.5 9.9 4.3 39.29.5 42.4 37.4 26.3 32.7 27.3 10.4 4.5 40.610.0 43.7 38.4 26.7 33.6 28.0 10.9 4.8 41.910.5 44.8 39.4 27.0 34.4 28.6 11.4 5.0 43.211.0 45.9 40.3 27.3 35.1 29.3 11.8 5.2 44.411.5 47.0 41.2 27.6 35.8 29.8 12.3 5.4 45.612.0 48.0 42.0 27.9 36.5 30.4 12.8 5.7 46.712.5 48.9 42.7 28.1 37.1 30.9 13.3 5.9 47.713.0 49.8 43.4 28.3 37.6 31.4 13.7 6.1 48.713.5 50.6 44.1 28.5 38.2 31.8 14.2 6.3 49.714.0 51.4 44.7 28.6 38.7 32.2 14.7 6.5 50.614.5 52.2 45.3 28.8 39.1 32.6 15.1 6.7 51.515.0 52.9 45.9 28.9 39.6 33.0 15.6 7.0 52.3

Methodology from: Modeling Queued Driver Behavior at Signalized Junctionsby Jim BonnesonTransportation Research Record, 1992, (1365) p # 99-107


Guide for Traffic Signal Preemption near Railroad Grade ... · PDF fileGuide for Traffic...

Documents

Transcript of Guide for Traffic Signal Preemption near Railroad Grade ... · PDF fileGuide for Traffic...