The following recommendations allow you to provide reliable power and communication to the SmartSensor Advance.

Understanding SmartSensor 6-conductor cable length limits

The 6-conductor cable has conductors for one power and two communication channels, each with its own requirements and length limitations:

• DC power – These two conductors (red and black) are 20 AWG wires. Based on DC power limitations and the power drawn by the SmartSensor Advance, this means that power can travel up to 1400 ft. (426.7 m) along these conductors.
• RS-485 communication – These two twisted pairs (blue and striped blue/white; orange and striped orange/white) are 22 AWG wires. Based on RS-485 limitations, this means that communication can travel up to 1400 ft. (426.7 m) along these conductors.

Proper sensor functionality requires DC power and one communication channel. Having a second communication channel allows for sensor configuration without detection data interruption.

This document discusses power and communication requirements for the SmartSensor Advance in more detail and provides recommendations on how to achieve sensor functionality in various applications based on voltage, needed cable length, and desired number of communication channels.

DC power and the SmartSensor Advance

The operating voltage for the SmartSensor Advance is 10–28 VDC. The recommended power supply voltage is 12–24 VDC, with a 5% voltage tolerance.

Due to Advance power consumption, 24 VDC can travel a maximum of 1400 ft. (426.7 m) along the 6-conductor cable. However, 12 VDC can only travel up to 200 ft. (61 m) along the 6-conductor cable. To extend 12 VDC to 400 ft. (121.9 m), sacrifice RS-485 conductors in the cable and combine them with the power conductors when terminating the cable into a Click device.

Extending 12 VDC up to 400 ft. (121.9 m)

Since the sensor only requires one RS-485 communication line to function, the second pair of conductors can be sacrificed to extend DC power up to 400 ft. (121.9 m) at 12 VDC.

Follow the steps below to correctly terminate the 6-conductor cable into a Click device and achieve a maximum cable length of 400 ft. (121.9 m) at 12 VDC:

1. Terminate the orange conductor (normally RS-485-) into the same terminal as the red (+DC) conductor.
2. Terminate the striped orange conductor (normally RS-485+) into the same terminal as the black (-DC/GND) conductor.

Communication and the SmartSensor Advance

As mentioned above, while only one communication channel is needed for sensor functionality, a second communication channel allows you to connect to the sensor for configuration without interrupting the data being sent from the sensor.

RS-485

The 6-conductor cable contains two twisted pairs for RS-485 communication. Cable runs over 1400 ft. (426.7 m) will begin to lose RS-485 capability.

Recommendations

The following tables give recommendations for various applications based on power, cable length, and number of communication channels. For more information or support, contact your Wavetronix representative.

SmartSensor 6-conductor cable

The table below shows applications based on power voltage, and the conductors needed to achieve maximum cable length.

Using an alternative cable

Below are Wavetronix recommendations for alternative communication cables. If you need an alternative power cable, any 2-conductor copper wire at the proper gauge will achieve the desired result.

Note. For cable runs up to 1400 ft. (426.7 m), we recommend you use a Wavetronix cable; if you choose to use an alternative cable, it must meet or exceed Wavetronix cable specifications. For cable runs longer than 1400 ft. (426.7 m), ensure the alternative cable used meets specifications for the power and communication standards being used. Failure to do so could cause devices to function improperly.

Communication cables

• Belden 3105A – One twisted pair with 22 AWG conductors used for one RS-485 channel.
• Belden 3107A – Two twisted pairs with 22 AWG conductors used for two RS-485 channels.
• Alpha 6453 – One twisted pair with 22 AWG conductors used for one RS-485 channel.
• Alpha 6455 – Two twisted pairs with 22 AWG conductors used for two RS-485 channels.

Power cables

The table below shows the voltage and wire gauges needed to provide DC power up to 2000 ft. (609.6 m).

Choosing a baud rate for wired communication

To achieve reliable wired communication, the selected baud rate must be compatible with the length of the cable run. The table below shows the cable length recommendations for wired communication.

*This is possible with an alternative cable.

‍

Ramp metering is a tool used by traffic and ITS engineers to manage the flow of vehicles entering controlledaccess highways or freeways. Controlled-access freeways generally consist of high-occupancy roadways, which in turn result in low or variable speeds and contribute to potential merge friction from vehicles entering the freeway. The principle behind ramp metering is to evenly space the merging vehicles by metering them, reducing the likelihood of congestion and accidents on the mainline.

The Wavetronix line of reliable, non-intrusive SmartSensors detect all the parts of the ramp metering system: the mainline flow, presence at the stop bar, and traffic queue on the ramps. There is always a SmartSensor for your application, whether it is system-wide traffic response or fixed-time operation. Although ramp configurations vary, Wavetronix will help you design a ramp metering solution that fits the needs of your application.

Ramp Stop-bar Detection

For stop-bar detection across the country, Wavetronix created the SmartSensor Matrix: a true presence radar vehicle detector with 16 antennas that can detect stopped vehicles in a 90 degree arc that expands out 140 ft. from the sensor.

In a multilane ramp metering configuration, there are two zones for each lane of travel. The first zone is configured before the stop bar. It is triggered when a stopped or slowed vehicle is waiting to enter the freeway. This zone can be as large as needed and must be long enough to work effectively as a presence detector when there are multiple cars in the queue.

The second zone is not always required, but is useful for systems that use a “passage zone” to ensure the vehicle has left the detection area. This zone signals back to the controller that the car has successfully cleared the area and the green light can be given to the next vehicle, as shown below. Wavetronix SmartSensors can have up to a 0.5-second delay in detecting a vehicle, so the passage zones can be small five-ft. zones. In some situations, it may be easier to make one zone that covers the whole roadway and tie it into the same channels on the controller.

Ramp Queue Detection

For ramp queue detection, the type of sensor you will want to use depends on the type of ramp: straight, clover-leaf, or long.

Straight-approach Ramps

The SmartSensor Advance detects vehicles as they approach the intersection and protects vehicles in the dilemma zone. This detection can happen at multiple zones along the approach, resembling a series of loops along a 600–900 ft. area of detection. There are several ways to detect the level of arriving traffic flow. One way is to monitor queue levels by configuring the SmartSensor Advance to detect the speeds when queue spillback reaches setback locations of interest. This method allows you to continuously activate different channels based on estimated queue length.

A 2009 ramp queue detection report conducted for Mn/DOT by SRF Consulting used four channels to estimate queue length in this way. The study reported that the average error in queue length was 9.2 ft (2.8 m) and the absolute average was 36.1 ft (11 m). These “excellent results” were achieved as the queue length on the ramp fluctuated from 0 to over 250 ft. and back six times during the almost two-hour period from 3:15 to 5:00 p.m.

Note. This feature can also be used in busy off-ramp scenarios where an extension on green at the end of the ramp may be needed to prevent backup onto the controlled-access roadways.

Clover Leaf Approaches or Long Ramps

If detection is needed beyond 500 ft. or the on-ramp is clover-leaf shaped, we suggest using the SmartSensor HD for detection. The SmartSensor HD can detect vehicles traveling perpendicular to the sensor up to 250 ft. away.

In several systems, the occupancy of a particular part of the roadway is used to adjust the signal timing at the stop bar. Sometimes this occupancy is detected by loops or micro loops. The SmartSensor HD can be used to emulate this occupancy detection over a particular space. Using the HD’s loop emulation and corresponding rack cards, the HD will output a call when the space in front of the sensor is occupied. This is especially useful for replacing loops with HDs.

‍

SmartSensor Advance improves the efficiency of an intersection by detecting queues and allowing vehicles to move through the intersection until the queue dissipates. These eight steps will help you understand how to use SmartSensor Advance to reduce queues at your intersection.

1. Understand advance detection

Advance intersection detection is important because it reduces the number of abrupt stops and rear-end and right-angle collisions. SmartSensor Advance is a long-range radar traffic detector that continuously monitors the progression of moving vehicles as they approach a signalized intersection. It calculates a vehicle’s estimated time of arrival at the stop bar based on that vehicle’s speed and range from the sensor. SmartSensor Advance’s accurate speed detection makes it uniquely suited for queue reduction because it can let the traffic controller know as soon as traffic has returned to free-flowing conditions.  

2. Select the mounting location

SmartSensor Advance can be mounted at any of the four locations shown below.  

3.  Mount and align the sensor

Once you have selected the location, simply mount the sensor to the pole. You will then need to align the sensor to the roadway. Use the sensor’s three axes of rotation to point it to the target area. SmartSensor Advance emits an elliptical footprint; the sensor needs to be aimed so that the footprint covers the entire approach you want to detect.

4. Connect the sensor to the cabinet

SmartSensor Advance will need to be wired to supporting devices that provide surge protection, power, and communication. The sensor typically communicates with an intersection controller via contactclosure cards in an input file rack. These devices will usually be housed in a traffic cabinet at the intersection.

5. Connect to SSMA

Once the sensor is installed, you can connect to it via the SmartSensor Manager Advance software.  

6. Configure SSMA

Once connected to SSMA, you will be able to see the vehicles that are being detected right away, but you will want to make sure the sensor is configured properly so you don’t miss any detections. To do this, use the Automatic Radar Configuration feature, and then fine tune the sensor’s sensitivity with the Manual Radar Adjustment tool.

7. Set up channels

The channel outputs are activated once the criteria for the channel are met, and these outputs are what the sensor actually sends out to the traffic controller. With the default Simple channel, all you have to do is set the range of the zone, then determine if the zone is activated by a user-defined speed, ETA or both.

Queue Reduction channel

The detection range of this zone should be about 100 to 150 feet from the stop bar. The speed setting means that vehicles in that zone that are going, in this case, 30 mph and under will activate the channel outputs, telling the traffic controller that there is a queue that needs to be cleared.

Dilemma Zone channel

The speed and ETA values should be such that they are activated when traffic is flowing freely, in this case, 30 mph and over. When a vehicle is in the zone and meets the specified criteria, the channel outputs will be triggered, telling the traffic controller that there is a vehicle in the dilemma zone and the green light needs to be extended for that vehicle.  

8. Verify channels

Once the channels have been set up, you can make sure they are working as you intended by using the Verify Channels-Alerts-Zones screen. On this screen you will see the speed, range and ETA values for each approaching vehicle, and you will also be able to see when a vehicle activates one of the channels you defined. Once you have verified that the channel settings are correct, your work is done. SmartSensor Advance will immediately start detecting traffic and reducing queues, improving the safety and efficiency of your intersection.

The goal of a traffic engineer is to keep traffic running smoothly, safely and efficiently for the driving public, and often this involves making changes to specific intersections and to the overall road system. The trouble is that it can be difficult to observe and quantify just how these changes have affected traffic flow. A method to measure the effects of those changes can be an invaluable tool—but how do you do that?

The solution? A performance measurement system, which will gather data from many different detectors and collate them into useful and concrete reports, graphs and tables. A system like this takes the guesswork out of evaluating the traffic on your roads and lets you fine-tune changes until you achieve your desired results. Read on to learn how to create such a system using the Wavetronix SmartSensor Matrix and SmartSensor Advance and other tools.

1. Install the SmartSensor Matrix and SmartSensor Advance at your intersections

These sensors are what will gather the data to send back to the TOC. (The backend software system you use will determine exactly how many total detectors and intersections you could tie into the system.)

The detectors you use need to be able to gather high-resolution data: counts in each lane, speeds of approaching vehicles, and the like. What makes the Matrix and Advance excellent choices for this application is first, the high-quality and reliable data, and second, that these sensors can gather the needed information while also performing their primary functions as intersection detectors.

So install your sensors, configure your detection channels, and then sit back while each device both delivers the necessary performance measure data and performs dilemma zone protection, stop bar detection, or any of the other useful applications these sensors can be used for.

2. Install traffic cabinet and components

At one corner of the intersection, put a traffic cabinet. These cabinets needs to have two devices inside:  

• Cabinet interface equipment, such as the Click 650, Click 600, or Click preassembled backplate. These provide the link between the sensors and the controller, and all of them provide the necessary power conversion, surge protection, and communication capabilities. The advantage of using the Click 650 is that its SDLC connector lets you plug right into the controller, negating the need for contact closure cards or racks.
• A high-resolution controller that takes a sample size every tenth of a second. A controller like this will make sure your system is getting the best-quality data possible.

3. Connect cabinets to the TOC

Each traffic cabinet in the system should be connected back to the TOC; fiber is a popular choice, but your options are open for this step.

At the TOC you’ll need a server for the performance measures system. This server, running backend software that you’ve created, borrowed from another DOT, or obtained from a third party, will poll all the connected sensors and gather their detection data. It then goes through all this data and provides statistical analyses.

4. Install the backend software

The backend software will then create useful graphs, tables and reports to deliver to you, the end user. This resources will allow you to see, and quantify when necessary, how well intersections are performing and whether changes you’ve made have affected those intersections.

You can see if your decisions have benefited the driving public, and you can find ways to further improve your roads.

This document provides a basic introduction to the use of Continuous Tracking Advance Detectors (CTADs) for advance detection at signalized intersections. This alternative type of detector has been used on hundreds of high-speed approaches to extend active through-movement phases to serve queue traffic; promote platoon progression; identify gaps in traffic; and provide safe phase termination.

SmartSensor Advance is the first long-range CTAD adeptly suited for through-movements because it can use multiple criteria (e.g. range, speed, stop-bar ETA, headway and qualified count) to determine if phase extension is desirable. To serve slowermoving traffic flow, time headway in advance of or near the stop bar can be estimated in a fashion similar to traditional presence monitoring with point detectors. For safe and efficient phase termination of high-speed traffic, ETA headway can be monitored in an improved way that dynamically adapts with the actual distribution of speeds on the roadway. Furthermore, speed measurements are used to quickly inhibit the queue output when traffic reaches free-flow in order to help create gap out opportunities and prevent high-speed max out.

In one short-term study (Study C – Impacts of Advance Loop Detection and the SmartSensor Advance on Driver Behavior at High-speed Signalized Intersection, David S. Hurwitz, ITE 2008), a single CTAD was reported to have approximately the same safety benefit (in terms of red-light running) as other innovative advance detection systems that have been evaluated over a longer term.

For more information on Study A and B, as well as the overall costs and benefits of advance detection systems on high-speed signalized intersections, please refer to the Wavetronix brochure entitled “SmartSensor Advance – SafeArrival Technology.”

This document offers guidance on CTAD configuration and controller timing necessary for basic integration at isolated intersections. On mainline high-speed through movements, CTAD advance detection has been determined to be the most cost-effective system.

Not only do CTADs have operation benefits compared to traditional embedded designs, but they offer the clear advantage of being easier to install and maintain. Embedded designs require that each distributed element continue to operate flawlessly, otherwise the phase will gap out prematurely. The FHWA publication Signal Timing on a Shoestring (FHWA-HOP-07-006) comments on this prevalent problem:

“One of the most common signal timing complaints is that the phase time is too short. This is frequently a result of a detector malfunction. The initial response, then, is to confirm that the hardware is operational and the timing parameters are operating as planned.”

With embedded designs, detectors are logically grouped together, making it expensive to replace a malfunctioning detector. It can also be challenging to confirm which detector is the culprit.

The SafeArrival Technology brochure, the SmartSensor Advance User Guide, this document, and the traffic controller manual documentation can all be used for reference so that your Wavetronix-certified representative can help you convert the concepts outlined for basic timing applications to more advanced scenarios, including volume-density, dynamic maximum green and actuated-coordinated operations.

Traffic Signal Timing

The list below shows some key terms and concepts that will be helpful to know before reading this document:  

The FHWA website, www.signaltiming.com, provides traffic professionals with a concise Traffic Signal Timing Manual (FHWAHOP-08-024) that reviews the objectives of signalized intersection traffic design in the context of long-practiced policies and standardized technologies. This document presents the material assuming an understanding of the traffic signal design concepts, basic signal timing procedures, and controller parameters as explained throughout the Traffic Signal Timing Manual (TSTM, especially sections 3.3–6.3). Please refer to the TSTM for explanations and definitions of key traffic signal engineering terms and concepts.

While the FHWA does not endorse specific products or manufacturers, it does encourage proactive management of signalized intersection traffic control in order to increase the safety and efficiency of operations. The TSTM stresses that a proactive review of the interaction between signal timing and detection decisions is important to achieve optimal performance. It states:

“The quality of intersection operation is particularly dependent on the relationship between the detection layout and the signal controller settings. For optimum performance, the detector layout and signal settings should be ‘tuned’ to the geometry of the intersection, its traffic volume, and the approach speed. The tuning process consists of finding a balance between detector location (relative to the stop line), detector length, passage time, and minimum green time for the prevailing conditions. However, there is no strong consensus in the industry with regard to what is the ‘best balance.’”

Furthermore, the purposes of signalized intersection detection specifically are stated in the TSTM as follows:

“The objective of detection is to detect vehicle presence and identify gaps in vehicle presence that are sufficiently long to warrant terminating the phase. There are many objectives of detection design that can be characterized with the following statements:

• Identify vehicle presence on a phase.
• Extend the phase to serve queued traffic and that which is progressed from upstream traffic signals.
• Identify gaps in traffic where the phase may be ended and extend the green.
• Provide a safe phase termination for high-speed movements by minimizing the chance of a driver being in the in decision zone [also known as a dilemma zone] at the onset of the yellow indication.”

The TSTM uses the inductive loop detector as the example in all of its applications and refers the reader to the Traffic Detector Handbook, 3rd edition (2006) to understand the strengths and weaknesses of alternative traffic sensing technologies. In particular the TSTM acknowledges one weakness of embedded detectors when it states:

“A designer may select a[n] inductive loop detector placement location based on a design manual or agency policy for the design travel speed. What the designer may not realize is once the loop detection is embedded in the pavement, it is difficult to relocate. If the operator or traffic engineer wants to change the posted speed . . . alternate loop locations may be necessary. This would force the costly relocation of the designed loops.”

The purpose of this document is to explain how proactive use of a CTAD-enabled system can achieve detection objectives 2–4 on high-speed approaches (>35 mph) more effectively, reliably, and at a lower cost than embedded detection.

Flow Capacity and Driver Behavior

TSTM objectives 2 and 3 are based primarily on traffic flow capacity and progression concepts that examine signalized traffic flow to determine if the active phases should be ended or extended. Typically, intersection capacity usage is maximized by queuing vehicles on one street while discharging the queue of vehicles on the cross street. Then, when the flow of vehicles on the active movements drop significantly below capacity, the traffic controller signals the lights to change the direction of flow. After the red clearance interval, and a couple seconds of initial startup time, the queued traffic on the newly activated movements discharges near capacity.

On the other hand, objective 4 is used to examine the exact same stream of vehicles based on principles from driver behavior analysis. This analysis will determine if high speeds prevail and if headways are conducive to safe green extension or termination. To prevent motorists from being lost between the cracks of signal timing, traffic engineers have used their statistical understanding of driver behavior at the onset of yellow to provide a low likelihood that the last-to-go driver will run a red light or incur a right-angle collision. At the same time, they are able to provide a safe headway buffer that helps minimize the chance that the first-to-stop vehicle will experience a rear-end collision.

Note. Before proceeding with a detailed discussion of the decision dilemma zone, it is important to clarify that a decision dilemma zone is not a physical dilemma zone (referred to as a Type I Dilemma Zone in the TSTM). Physical dilemma zones can be defined as zones in which law-abiding motorists cannot physically stop and cannot physically clear the intersection without running a red light.

To prevent a physical dilemma from occurring, the duration of the yellow interval should be long enough to provide motorists that cannot physically stop (based on reaction time, deceleration rate, grade, etc.) the time they need to proceed through the intersection. High-speed signalized approaches often need to have more than three seconds of yellow time in order to accommodate the longer stopping distances of high-speed vehicles. Consequently, engineering guidance often suggests yellow intervals from three to six seconds, depending on approach speed. By selecting an appropriate yellow time, physical dilemma zones can be avoided.

Decision Dilemma Zone

The decision dilemma zone (referred to as a Type II dilemma zone in the TSTM) is affected by several factors, including what is known as the option zone. Motorists with at least one safe alternative are not in a physical dilemma zone. If their only safe option is to stop, then they usually do. If their only safe option is to go, then they usually do. But often, the yellow timing will accommodate both the option to stop and the option to go. In this case, drivers are in what is termed the option zone. The option zone varies from one intersection to the next, and depends on the speed of the vehicle. Relatively speaking, the option zone can be large for slower-speed vehicles on a high-speed approach.

Driver behavior and knowledge of signal timing is fuzzy logic. Human perception of closing rate to the stop bar, the variance in yellow times from one intersection to the next, driver awareness, and a number of other factors can lead drivers to believe they are in the option zone, when really they only have one safe choice. In addition, a driver’s options are also constrained by other motorists. For example, even defensive drivers sometimes unintentionally run a red light, especially if a commercial vehicle or aggressive motorist is tailgating them. Defensive drivers may risk running a red light to avoid a rear-end collision—the most common type of traffic collision at a signalized intersection.

So even when the physical dilemma zone has been properly eliminated, a decision dilemma zone still exists. The dilemma is not that the driver does not know which decision he or she will make (although this may be the case, too); rather, the decision dilemma ultimately results from two factors:

• Motorists (trailing, leading or conflicting movement drivers) and traffic officials cannot reliably predict the decisions of drivers within this zone.
• Motorists within this zone cannot reliably predict whether they will clear the intersection if they decide to proceed.

Although not all intersections are the same, multiple driver behavior studies have reported that the size of the decision dilemma zone is approximately 2.5 seconds in length. These studies report that when the light changes to yellow, 85% of drivers will attempt to clear the intersection if they are within 2.5 seconds of the intersection stop bar, and 85% of drivers will try to stop if they are more than five seconds away.

Therefore, motorists with an estimated time-of-arrival to the stop bar roughly between 2.5 to 5.0 seconds are considered to be in the decision dilemma zone. As could be expected, the size of the decision dilemma zone (2.5 to 5.0 seconds) is about the same size as the variation in yellow times from one intersection to the next (3 to 6 seconds). Driver behavior has unconsciously adapted to the variance in yellow times, the existence of option zones, and/or the kinetic principles used to determine signal timing.

Maximum Allowable Headway

When green extension is used to achieve objectives 2–4, there is an opportunity to leverage synergies that exist between traffic flow capacity theory and driver behavior studies since the size of a safe following distance is roughly equal to the size of the decision dilemma zone. However, detection objectives 2–4 will not always be fully harmonized. This disjoint is shown on occasions when safe gap-out opportunities are not properly synchronized with phase termination, when gap-out opportunities are missed, or when the window of opportunity for green extension expires without any gap-out opportunities.

The traditional method of finding gap out opportunities focuses on estimating time headway using presence detection and controller passage time. With traditional use of presence detectors at discrete points, maximum allowable headway (MAH) in a single lane is a combined result of the detection zone length, vehicle length, vehicle speed and passage time. On multilane approaches, a gap of the desired MAH may occur in a single lane, but may be overlooked by the controller because of a requirement that all lanes on an approach gap out simultaneously. Simultaneous gap out criteria is often preferred in order to have dilemma-free termination. For example, the TSTM recommends the use of simultaneous gap out logic for concurrent through-movement phases when advance detection is used.

With the use of simultaneous logic on multilane roadways, it is imperative to keep the MAH reasonably small in order to avoid frequent max-out termination. By placing a point detector near the start of the decision dilemma zone, the MAH can be minimized. However, since the start of the dilemma zone is based primarily on stop-bar ETA rather than stop-bar distance, some traditional guidelines have allowed the MAH to be larger than necessary to accommodate for the fact that the embedded point detector cannot move based on the actual speed of the vehicle. This larger MAH can provide vehicles traveling at speeds lower than the assumed design with safe passage through the decision dilemma zone, but it does so at the risk of more frequent termination by max out.

For example, if a passage time of 3.2 seconds is used with a detector placed 405 ft. upstream of the stop bar to accommodate a 55 mph design speed (see figure above), the MAH detected on a 55-mph vehicle will be about 3.5 seconds. However, a 45-mph vehicle traveling over this same detector will have an MAH of about 3.6 seconds and will be provided safe passage to within about 165 ft. (or 2.5 seconds ETA) of the stop bar. To provide safe passage for vehicles at 40 mph, the passage time would need to be even larger if the detector was not moved closer to the stop bar.

Note. Some agency guidelines state that the objective of advance detection is to allow a vehicle to travel from the first advance detector all the way to the stop bar. This policy results in MAH values of 4.0 or more seconds. For safe passage through the decision dilemma zone, this large of an MAH is not necessary. Not only does this policy increase the likelihood of high-speed max out, it also decreases effective green time and increases clearance-lost time at the intersection. This policy may help minimize the risk of right-angle collisions during the clearance interval. When this methodology is used, the TSTM recommends the use of gap reduction in the traffic controller.

CTAD

CTAD is an alternative technology released subsequent to the latest edition of the Traffic Detector Handbook. A CTAD has the ability to continuously track the changing range, speed, ETA and qualified count of multiple vehicles as they approach an intersection stop bar. CTAD sensors can be programmed to signal a traffic controller to place a call to extend green only when specific criteria are met. By basing signal control on multiple criteria (see figure below), the effectiveness of signal control can adapt in real time to account for the dynamics in driver behavior that hange as the distribution of speeds on the roadway change.

The benefits of CTADs have been approximated in the past by using multiple advance detectors in a series of setback distances. The TSTM highlights a multiple-detector design with as many as three setback locations on the highest speed approaches. As with CTAD detection, the increased detection coverage of a multiple-detector design allows for more responsive operations. The TSTM states:

“This design is well-suited to approaches with a significant percentage of turning vehicles because it is likely these vehicles will gap out as the vehicle slows to make the turn at the intersection or upstream driveway.”

With the continuous coverage of CTAD detection, the speed and ETA of turning vehicles is actually tracked so that green extension can be determined based on the needs of each vehicle.

SmartSensor Advance (model SS200) is the most advanced CTAD sensor customized for traffic signal control applications. Classified as an LR-CTAD because of its long range of detection (up to 900 ft.), the sensor is uniquely suited for advance detection on through movements.

Technology Stack

The figure below shows the technology stack from the CTAD physical layer to the phase layer of the traffic controller. To provide real-time updates from the physical layer, the default settings achieve a communication latency to the controller inputs of 72.5 milliseconds on average. This latency is realized over wired communication distances up to 600 ft. and can be decreased by reprogramming the serial baud rate and output message push rate. Both power and dual port communications are brought to the sensor over a standard multi-conductor cable.

The CTAD physical layer is a 10.525 GHz implementation of FMCW Digital Wave Radar technology. FMCW Digital Wave Radar is a range-based, presence-sensing technology that performs consistently and accurately in harsh conditions including fog, snow, dust, rain and sun glare. The FMCW pulses are transmitted over 200 times a second and received using an elliptical antenna beam with a 70° elevation and 12° azimuth half-power bandwidth.

The tall elevation angle accommodates the long range of detection, while the narrow azimuth angle focuses the radar energy on the lanes of interest (see figure below). For approaches with more than one lane, two lanes can be detected 50 ft. from the sensor and three lanes can be detected about 100 ft. from the sensor depending on the selected alignment. Consult the SmartSensor Advance User Guide for instruction on placement, mounting, alignment and sensitivity configuration. Based on the configured sensitivity, trucks, cars, motorcycles, bicycles and pedestrians can be monitored.

Above the physical layer, the CTAD has several signal processing layers that provide for reliable tracking of vehicles even with roadside background clutter. A passage detection layer filters out stationary background objects (as well as stopped vehicles). A three-state (range, velocity and acceleration) tracking model monitors the propagation of up to 25 objects moving in sequence. A tracked detection filter helps eliminate unwanted monitoring of multipath detections or detections in the opposite direction so that the CTAD reports only objects moving in the selected traffic direction (either incoming or outgoing). A passing vehicle logic layer is also used to hold on to detections if vehicles pass each other or if they are momentarily occluded.

Above the signal processing layer are SafeArrival™ Technology logic layers. The zone layer processes the estimated range, speed, ETA and qualified count of detections to determine if each zone output is active. The alert and channel layers can provide additional conditioning of the CTAD’s eight outputs before their status is transmitted via serial and contact closure communications to the traffic controller. For a detailed understanding of all the flexibility in the channel, alert and zone layers, please refer to the SmartSensor Advance User Guide.

The outputs of the contact closure cards allow for simple integration with standard TS-1, TS-2, 170, and 2070 systems. Each sensor supports up to eight detection channels, but in most advance detection applications only a single two-channel card per sensor is necessary. In addition, the sensor protocol is an open standard for tighter integration with software programmable controllers.

SafeArrival Technology

The following flowchart shows the basic algorithm which is the core of SafeArrival Technology.

Based on the status of the zone outputs, the CTAD sets that state of its eight channel outputs. Each channel output is true (call active) while the selected criteria of range, speed, and ETA are satisfied for the specified number of vehicles. The output state of each channel is sent to the traffic controller at a programmable rate (default is about 10 Hz).

The zone output criteria are specified using upper and lower bounds. This specification flexibility allows for tight control of the criteria that warrant green extension in order to provide more gap out opportunities and superior prevention of high-speed max out.

ETA-based Extension

By specifying a minimum and maximum ETA, a CTAD can both tightly control MAH and locate the extents of safe passage based upon actual vehicle speeds (instead of an assumed design speed). This has the efficiency benefit of equalizing MAH for vehicles at all free-flow speeds. By keeping the MAH short and tailoring the extents of safe passage to individual vehicle speeds, dilemma zone protection can actually be increased without the overprotection that commonly causes point detector–based systems to become inefficient with low-to-moderate levels of capacity usage.

In addition, ETA of a vehicle to the stop bar is the best predictor of dilemma zone incursions. ETA is a better predictor than the position or speed of the vehicle alone, and with SmartSensor Advance, acceleration is also tracked in order to increase ETA accuracy. When all detected ETAs are outside the times that characterize the decision dilemma zone for a specific approach, then a safe gap in traffic has been detected.

The size of a safe gap can be defined using the sensor configuration software based on driver stopping probabilities. This allows for simple site specific customization. The nominal size of a safe gap is about 2.5 seconds, located 2.5 to 5 seconds upstream of the stop bar.

This relatively safe gap in traffic exists because multiple driver behavior studies have reported that the size of the decision dilemma zone is approximately 2.5 seconds in length. These studies report that when the light changes to yellow, 85% of drivers will attempt to clear the intersection if they are within 2.5 seconds of the intersection stop bar; and 85% of drivers will try to stop if they are more than five seconds away.

This level of certainty is also a strong deterrent to red-light running because less than 15% of drivers will attempt to clear when beyond 5 seconds from the stop bar. Even fewer will be likely to run the red light if the far boundary of the safe gap is increased to 5.5 or more seconds.

Advance Detector Channel

An advance detector channel is configured to achieve TSTM objectives 3 and 4. This channel uses its first zone to estimate capacity usage and provide efficient dilemma zone protection. It is often recommended to use a queue detector channel in combination with the advance detector channel. The queue detector channel is promptly inhibited after vehicle speeds are detected at free-flow levels.

The advance detector channel is configured by setting minimum and maximum bounds for vehicle ETA, speed and range. Here are some suggested settings:  

• ETA minimum = 3.0 seconds
• ETA maximum = 5.0 seconds
• Speed minimum = 35 mph
• Speed maximum = 100 mph
• Range minimum = 100 feet
• Range maximum = 500 feet  

If any one of the vehicles on the approach is detected to meet these criteria, then the zone output will be true (and the Qualified Count will be at least 1).

For an advanced detector channel, it is usually recommended to use the phase passage timer for output extension because it has the flexibility to be used with the controller gap reduction feature if necessary. A default minimum of 1.0 seconds of phase passage time is suggested.

When used with low-speed filtering, passage time will help verify that the changes in vehicle speeds with the advance detection zone are persistent. When used with ETA detection, passage time will help extend safe passage beyond the minimum ETA specified for the advance detector channel and will help interpolate in the event of momentary detection drop-outs.

Note. Extend timers include the following—the passage timer associated with each traffic controller phase, the extend timer associated with each traffic controller input, and the extend timer associated with each CTAD channel output. It is important to determine which extend timers will be used for each application in order to prevent overextension induced by programming extension in multiple places.

By using ETA-based extension in combination with an extend timer, the MAH becomes the combination of the ETA headway and the extension time if the CTAD assumes a vehicle length of 0 ft. (By default, the CTAD assumes a vehicle length of 20 ft., adding 0.25 seconds for a 55 mph vehicle.)

The following figure illustrates that with an ETA headway of 2.5 seconds and a passage time of 1.0 seconds, the resulting MAH is 3.5 seconds.

Since the minimum ETA is set at 2.5 seconds and the maximum at 5.5 seconds, safe passage will be provided by the CTAD from 330 to 165 ft. for a 45-mph vehicle. The passage timer will then count down while the vehicle travels another 66 ft. so that it is about 100 ft. from the stop bar. For a vehicle traveling at 55 mph, safe passage will be provided by the CTAD from 405 to 200 ft. The passage timer will then count down while the vehicle travels another 81 ft. so that it is about 120 ft. from the stop bar.

The above example serves as an illustration of what can be done to control the MAH and dilemma zone protection with a CTAD-enabled system. If the MAH needs to be smaller or larger at a particular intersection, or if dilemma zone protection warrants adjustment, customizations can be made using the configuration software.

For example, on trucking corridors, it may be more appropriate to set the ETA bounds from 6.5 down to 3.5 seconds and specify 1.0 second of passage time. This would result in a MAH of about 4 seconds and justify the use of gap reduction.

The SmartSensor Advance – SafeArrival Technology brochure illustrates how CTAD dilemma zone protection provides added priority for high-momentum vehicles like trucks and buses. In fact, if the intersections along the corridor are actuated, all standard controllers can be programmed to incrementally increase priority for large high-speed vehicles (when compared to smaller high-speed vehicles). This can be done using the gap-reduction feature. Near the beginning of green, a sizable amount of passage time can be used to prevent momentary occlusion of small vehicles from causing gap out. But then as the green time persists and no dilemma-free condition has been detected, the passage time can be decreased to favor termination with a small high-speed vehicle in the dilemma zone, as opposed to a large high-speed vehicle that may be occluding it.

Dynamic Dilemma Zone Protection

Since the advance detector channel uses ETA detection, it provides a dynamic form of dilemma zone protection that has both safety and efficiency benefits. As speeds increase, the need for high-speed protection moves further back on the approach. In addition, the size of a safe gap in traffic necessary to safely terminate a phase also grows. Conversely, when speeds decrease, the size of a safe gap shrinks, and its location on the roadway moves closer to the stop bar.

With SmartSensor Advance, high-speed protection can be actively adjusted as far away as 500 ft. from the sensor. This dynamic zone placement and sizing based on ETA provides additional safety when speeds reach their most dangerous levels, especially when compared to a single point detector located about 300 or 400 ft. upstream of the stop bar.

Dynamic dilemma zone protection based on ETA simplifies and validates design of advance detection. No longer do point detectors need to be relocated when the 85th percentile or posted speeds on an approach change.

Max-out Prevention

In order for an advance detection system to be most effective, it must minimize the number of cycles required to max out the traffic phase with high-speed vehicles in the dilemma zone. During moderate levels of congestion this can be very challenging for systems that use embedded presence detectors at discrete point locations. Inherently, embedded presence detectors do not have the ability to (1) dynamically change zone placement, and (2) detect vehicle speed.

With CTAD advance detector channels, calls for green extension can be constrained, using ETA, to situations where vehicles are over the areas of the roadway that need to be protected. If a leading vehicle is traveling fast and a following vehicle is traveling slowly, this area of protection can be very small (see SmartSensor Advance – SafeArrival Technology brochure).

Furthermore, when speeds drop to lower levels during moderate congestion, the use of speed-based detection will decrease the likelihood of terminating the phase by max out. With CTAD detection, green extension can be limited to vehicles traveling above a critical speed (e.g. 35 mph). This means that if speed drops low enough, green extension is no longer critical for safety, and the intersection can be controlled primarily based on the theories of capacity flow through an intersection rather than driver behavior.

Effective Green-time Recovery

One benefit of ETA-based advance detection is that it reliably predicts which motorists will go through the light even after it turns yellow. When a phase terminates by gap out with a vehicle 2.5 seconds away from the stop bar, this driver will have an 85% probability of clearing the intersection. In addition, any drivers ahead of this vehicle are also very likely to clear the intersection. Some traffic researchers claim that the light is effectively green for these drivers within 2.5 seconds of the stop bar.

Consider that if stop bar detection is used instead of advance detection, this extra-effective green time is not realized. In fact, with stop bar-only detection, it is actually expected that no vehicles will cross the stop bar in the last few seconds of green. So in comparison to stop-bar detection, a CTAD advance detection channel can regularly increase effective green time near the clearance interval transition from 2.5 to 5 seconds. And in comparison to other forms of advance detection, CTAD sensors do a better job of tracking arriving vehicles in order to help increase effective green time across a broad range of high speeds.

Queue Detector Channel

Often high-speed through movements are on minimum recall. Since these major movements are automatically called by the controller, there is no need for stop bar detectors in these lanes. However, in some cases stop bar detection is used to call minor high-speed through movements. On these minor movements the TSTM recommends programming the stop bar detectors as queue detectors so that they disconnect after detection objective 2 is satisfied.

If you already have stop bar detection on your high-speed approaches, you may not need to use the CTAD to achieve TSTM objective 2. However, configuring a queue detector channel with the CTAD can be very cost-effective for through movements on recall because stop bar detection is no longer necessary.

The queue detector channel can be configured using the first zone of the second channel of the CTAD. With the passage timer set at 1.0 seconds, this zone is configured to extend green with MAH of 2.6 seconds for vehicles traveling at 30 mph. The detection range of the zone can be 100 to 150 ft. from the stop bar. A vehicle traveling at 30 mph will activate this 50-ft. zone for approximately 1.6 seconds (with the default vehicle length set at 20 ft.). Slower vehicles will experience a larger MAH, which is useful during the startup period of queue dissipation.

The queue detector channel’s zone will not use ETA criteria or discriminate based on qualified count, but it will use speed in order to inhibit the detector once free flow is reached on the high-speed movement. To do this, the minimum speed bound will be set at 1 mph and the maximum speed bound will be set at 35 mph. Speed-based deactivation of the queue clearance channel is used instead of programming the associated controller detector input as a queue-detector type.

By placing the zone about 100 ft. back from the stop bar, you can get the benefits of setback detection. With setback detection, green extension is performed for vehicles that are still approaching the intersection, rather than vehicles that are already at the intersection. This setup helps recover some of the lost time between traffic phases, in the event that free-flow speeds are not reached on the high-speed approach.

When free-flow speeds are reached, the CTAD will stop signaling calls on the queue detector channel and start signaling calls on the advance detector channel. This speed-based deactivation and activation of channels is unique to a CTAD and is not something you get with embedded point detector systems. With embedded systems the queue detector channel will sometimes remain active and errantly force max out long after high-speeds are reached.

It is possible to decrease the size of the zone and increase the extension time if necessary on a particular approach. However, remember that once a queue detector channel is assigned to a phase, it inherits that base amount of extension time. So if the phase passage timer is set to 1.0 seconds and an additional 1.0 seconds is added to the controller input extend timer, then the total extension time experienced by queue detector channel will be 2.0 seconds (and the MAH would be even greater depending on the zone size).

Normally, it is not recommended to configure the controller so that CTAD’s queue detector channel will disconnect. Rather, the channel will be inhibited when speeds rise to free-flow level (e.g. 35 mph).

Minimum Green

The TSTM provides guidance on setting the minimum green time for different types of facilities. On a major arterial, 10 to 15 seconds of minimum green time is necessary to satisfy driver expectancy. The TSTM also indicates that 15 seconds is sufficient green time for the queue to clear when the nearest upstream detector is 126 to 150 ft. from the stop bar. Since the CTAD is a passage detector, it is recommended that the minimum green time be 15–20 seconds when used on a major arterial without stop-bar detectors. This will allow sufficient time for the vehicles that were stopped in the range of the queue clearance detector to begin moving.

It is also possible to use the CTAD’s advance detection channel to activate volume-density functions that help adjust the initial green. However, when configuring these functions, keep in mind that the CTAD’s advanced detection channel has not been configured for volume counting and will appear to under count. In essence, the channel will merge closely following vehicles because its output remains active as long as there is at least one vehicle that satisfies the specified criteria. To reduce under counting you may chose to reduce the ETA headway by increasing the minimum ETA bound. However, you will want to remember to increase the passage time if you want to maintain the same MAH.

If you are using the CTAD on a high-speed minor movement, it is likely that you will have stop-bar detectors to call the phase. If you do not, it is possible to mount the CTAD to detect vehicles at the stop bar. This will allow you to place the queue detector channel’s zone right at the stop bar so that the minimum green time can be reduced. In this case you may also be able to use a latched channel to call the phase (see the SmartSensor Advance User Guide for more information on latched channels).

For questions regarding the maximum green time and other signal timing questions, consult the TSTM.

Hardware Integration

Wavetronix supplies a SmartSensor Basic Preassembled Traffic Cabinet Backplate, Click contact closure modules, SmartSensor cable, and SmartSensor Pole-mount Surge Preassembled Cabinets for hassle-free integration of SmartSensor Advance into your signalized intersection. For more information on the integration process, consult the SmartSensor Advance Quick Start Guide, the SmartSensor Advance User Guide and the Wavetronix website.  

Device Specifications

For large vehicles like truck trailers, CTAD detection accuracy often exceeds 95% and the nominal detection accuracy is 90% for all vehicles within the first 400 ft. (see figure below). The detection accuracy drops off at further ranges depending upon the mounting height, vehicle type and lead-vehicle occlusion. Multiple detections of large vehicles like truck-trailer combinations are common and useful.

The nominal mean time of failure for the SmartSensor Advance is 10 years. It can be purchased with a 5-year warranty option. The CTAD is constructed with solid state electronics and has no batteries. For more information and up-to-date specifications related to CTAD accuracy, consult the SmartSensor Advance datasheet.

Last updated October, 2010

‍

1. Select Setup Channels-Alerts-Zones from the SmartSensor Manager Advance main menu.

2. In the Type drop-down menu, set channel 1 to be a Pulse channel. The pulse channel will allow you to record counts that can be used for the automated signal performance measures for the arrival of vehicles depending on the intersection phase it is on.

3. Create a 10-ft. zone from 350 ft. to 360 ft. along the approach.

4. From the main menu, go to Channels-Alerts-Zones > Setup Output Communication. In the Select trigger speed options: section, set Port to 1 and Channel to 1. The Tag field is used to identify which sensor or approach is sending the information.

The SmartSensor Advance has an option to output the trigger speeds of the vehicles that pass through the approach as a HEX string that can be used to gather the vehicle speeds. The example above uses the same channel that is providing counts to the controller to output the vehicle speeds as they pass the 10-ft. zone.

The SmartSensor Advance still outputs contact closure data through port 2 to the controller either via rack cards, the Click 600, or the Click 650. However, in addition to this, port 1 and channel 1 are used for the trigger speed option. Usually, this trigger speed information is sent over a serial-to-Ethernet terminal server in UDP mode, with another service listening to the UDP messages in the Traffic Operation Center.

5. Wire the Click 222 in order to send the trigger speeds to the TOC. The goal is to wire the port the trigger speeds are coming from to the terminal server that will eventually send it to the TOC (next step). Each Click 222 has the capability of hosting two sensors per device. In this example, you configure the sensor to push the trigger speeds HEX string over port 1 (control port). The control port in the Click 222 is composed of blue pairs for each respective sensor.

6. Connect to a terminal server. A terminal server with four serial ports can be used to send the trigger speeds to the TOC. After identifying the blue pair that corresponds to the respective sensor, wire from the control port to one of the serial ports in the terminal server. Make sure you look at the pinouts on the terminal server to see where each cable goes.

When wiring the Click 222, the RS-485 pair that is sending both the trigger speed packets and the contact closure data (port 2) will need to be wired to the terminal server device.

Setting the Bandwidth

Network bandwidth plays an important role in making sure the trigger speed information is received correctly. If you have low network bandwidth, you will probably want to separate contact closures from trigger speed packets to help avoid collisions and lost data. If you have high network bandwidth, you will be able to send contact closures and trigger speed packets simultaneously over the same port. The speed of the network allows the system to separate trigger speed packets and contact closure data. The settings in the figures below are recommended for networks using fiber or similar speeds and bandwidth capabilities.

Warning. When you are connecting to configure the sensor via a serial or TCP/IP connection over the control port, the trigger speeds are lost while the connection is active because port 1 (control port) is being occupied.

Adding Dilemma Zone Protection

1. While in Channel 2, click on the number 2 tab at the top of the screen.

2. Using the Type drop-down menu, set Channel 2 to be a Priority channel.

3. Set Level 1 for high-priority vehicles by creating a zone along the entire approach and setting the ETA at 2.5 — 7.0 seconds.

4. Set Level 2 for normal-sized vehicles by setting the ETA at 2.5 — 5.5 seconds.

5. Click on the Q tab and create a queue reduction zone from 100—165 ft., or whatever range meets your needs.

To connect to contact closure rack cards, the Click 600, or the Click 65x, please refer to the appropriate documentation by visiting www.wavetronix.com/support.

‍

The power plant is the collection of Click devices that provides power to the sensor and other Click devices. It provides surge protection and, if necessary, AC to DC power conversion (all Wavetronix devices run on DC power).

For installations supplied with AC power, the power plant includes the following:

• Click 210 circuit breaker and switch
• Click 230 AC surge module
• Click 201/202/204 AC to DC converter (in rarer cases, this could instead be the Click 203 unlimited power supply and battery)

For installations supplied with DC power, the power plant includes the following:

• Click 210 circuit breaker and switch
• Click 221 DC surge module

The most common way to obtain a power plant is to buy a preassembled backplate, of which the power plant will be part. In that case, the only setup that needs to be done is making sure that power is supplied to the backplate.

If you have purchased the individual components of the power plant, follow the steps in this document to assemble and install it.

If you have an installation with a traffic cabinet, the power plant will most likely be installed in there. If you’re just using a pole-mount box, no cabinet, that’s where the power plant should go.

Warning. An authorized electrical technician should install and operate these modules; there is a serious risk of electrical shock if the power source is handled unsafely.

AC Power Plant

Wiring in AC

1. Make sure power to AC mains is disconnected.
2. If you’re using a traffic cabinet, you will wire from its power source.
3. If you’re using a pole-mount box, push the AC power cable through one of the cable grip conduits on the bottom of the box (it’s easiest to use the bottom left). When you’re done, tighten the cable grip by twisting it until it’s tight.

Installing the Click 210

The Click 210 is a compact circuit breaker, which will interrupt an electric current if there is an overload. After a current interruption, reset the breaker by pushing the reset button (on the front of the device).

1. Mount the Click 210 onto the DIN rail.
2. Connect the line conductor (usually black) from the AC terminal block or cable to the screw terminal on the bottom of the module.
3. Connect another conductor (also black) to the screw terminal on the top of the device.

Note. For ease in troubleshooting, we recommend you follow the wire color scheme outlined here.

Installing the Click 230

1. Mount the Click 230 onto the DIN rail next to the Click 210.
2. Connect the black line conductor from the top of the Click 210 to terminal 5 on the IN side (bottom) of the Click 230. (The terminal numbers can be found on the side of the Click 230.)
3. Connect the neutral (usually white) wire from the AC terminal block or cable to the terminal marked 1 on the bottom of Click 230.
4. Connect the ground wire from the AC terminal block or cord to the terminal marked 3 on the Click 230.
5. Connect an outgoing and protected line wire (black) to the terminal marked 2 on the Click 230.
6. Connect an outgoing and protected neutral wire (white) to the terminal marked 6 on the Click 230.

Note. Terminal blocks 3 and 4 are directly bonded via the metal mounting foot of the base element to the DIN rail, so there’s no need for any additional grounding.

Installing the Click 201/202/204

The Click 201 provides 1 A (enough to power one SmartSensor HD). The Click 202 provides 2 A, and the 204 provides 4 A. Choose accordingly. (If you’re using the Click 203, see that chapter in the Click 100-400 Series User Guide.)

Follow the steps below to get power from the Click 201/202/204 to the T-bus, the power and comms bus that powers the rest of the installation:

1. Mount the Click 201/202/204 onto the DIN rail next to the Click 230.
2. Connect the line (black) wire from the Click 230 into the L screw terminal on the top of the Click 201/202/204.
3. Connect the neutral (white) wire from the Click 230 to the N screw terminal on the top of the device.
4. Connect a +DC conductor (red) to the + screw terminal on the bottom of the device.
5. Connect a -DC conductor (black) to either of the – screw terminals on the bottom of the device.
   
   Note. Don’t wire into the screw terminal marked DC OK; it provides only 20 mA and should be used only for monitoring the power supply.
   
6. Snap as many T-bus connectors as are desired onto the DIN rail and connect them together.
7. Connect a 5-screw terminal block to the end of the T-bus.
8. Connect +DC (red) from the Click 201/202/204 to the top screw terminal on the 5-screw terminal block.
9. Connect –DC (black) to the second screw terminal.

DC Power Plant

Wiring in DC

1. If you’re using a traffic cabinet, wire from its power source.
2. If you’re using a pole-mount box, feed the DC power cable through the cable grip conduit on the bottom left of the box. Tighten the grip down when you’re done.

Installing the Click 210

The Click 210 is a compact circuit breaker, which will interrupt an electric current if there is an overload. After a current interruption, reset the breaker by pushing the reset button (on the front of the device).

1. Mount the Click 210 onto the DIN rail.
2. Connect the line conductor (usually black) from the AC terminal block or cable to the screw terminal on the bottom of the module.
3. Connect another conductor (also black) to the screw terminal on the top of the device.

Note. For ease in troubleshooting, we recommend you follow the wire color scheme outlined here.

Installing the Click 221

1. Snap as many T-bus connectors as are desired onto the DIN rail and connect them together.
2. Mount the Click 221 on the first (leftmost) T-bus connector. This will connect power to the rest of the installation.
3. Connect the line (black) wire from the Click 210 into the +DC screw terminal on the bottom of the Click 221.
4. Connect the neutral (white) wire from the DC terminal block or cable to the -DC screw terminal on the bottom of the device.
5. Connect the ground conductor (green) from the DC terminal block or cable to one of the GND screw terminals on the bottom of the device.

Intersections with limited visibility, high speeds (55 mph and greater), temporary or newly installed intersections, or grade issues often need an advanced warning system (AWS). The AWS information prepares motorists for unexpected traffic conditions, and is therefore especially effective during the first several encounters.

Over time, however, drivers are prone to tune out this information because of its static nature. Another option employed in a number of state departments of transportation is adding intelligence to the AWS. This intelligence is used to provide the motorists with more information about the intersection they are approaching, and make the signal dynamic in order to increase its effectiveness over time.

This document describes how to increase the effectiveness of the AWS by combining it with advance detection. Specifically, the benefits of using Long Range–Continuous Tracking Advance Detection (LR-CTAD) technology in a combined AWS/AD system will be reviewed. These benefits include:

• LR-CTAD can be mounted on the AWS pole to reduce the cost of structures and trenching.
• LR-CTAD can detect vehicle speed and ETA at distances up to 500 ft. away in order to improve indirect protection of the decision dilemma zone.

This document also stresses the importance of a short lead flash time (the time between onset of flashers and onset of yellow) to prevent drivers from trying to race the system. Please refer to knowledge base article Basic Traffic Signal Timing and Advance Detection for detailed information regarding the decision dilemma zone and LR-CTAD technology.

Advance Warning Systems

To minimize the hazards of unprotected green termination, some agencies advocate the use of advance warning systems (AWS). Advance warning systems signal upstream drivers that a phase change interval is imminent or that the movement is now inactive. Alerted motorists are expected to prepare to stop. Agencies using AWS sometimes limit the posted approach speeds within their jurisdiction to about 45 mph in order to further reduce the hazard of the change period. However, AWS are not universally accepted. Many traffic officials point out that some drivers abuse the warning information by trying to race the system.

To discourage motorists from abusing advance warning information, advance warning systems with advance detection (AWS/AD) have been designed. With a combined AWS/AD system, traffic flow characteristics at an upstream location can be used to predict a time when the traffic stream within the decision conflict zone will be low-risk or risk-free. The onset of yellow is then scheduled to occur when the risk will be minimal.

Sometime before the onset of yellow, the advance warning system will activate its signals. The combined system is intended to send a clear message to motorists: if they see the active signal, they should expect to stop on red; whereas if they do not see an active signal, they should expect to clear the intersection. These systems can operate with a relatively short lead flash time in order to reduce the likelihood of motorists racing the system to beat the light.

Agencies that use any or all of these specialized systems sometimes have a warrant procedure that determines what types of intersection approaches need what type of system. The warrant guidelines are based on variables like intersection type, traffic mix, grade issues and design speed. Isolated high-speed intersections may warrant a system type other than high speed intersections within a coordinated corridor. Similarly, rural intersections with a high percentage of truck traffic and high speeds may warrant a different type of system than isolated intersections within an urban environment.

Even when these warrant guidelines are in place, the proficiency of the signal technician or signal engineer plays a critical role in the success of the system. For example, signal technicians often select the window of opportunity in which the system operates by selecting the minimum and maximum green times.

Clear Message

If the AWS is designed properly it will send a clear message to the driver. In an AWS/AD system the flashers provide advance warning of the onset of yellow. Motorists that see the flashers activate should be sent the message that they need to prepare to stop; they should be discouraged from trying to race the system and beat the light. Alternately, motorists that do not see the flashers come on should be sent the message that they may proceed through the intersection.

This type of clear message can be achieved with gap management techniques that tightly synchronize the timing of the AWS flashers and the yellow light. This technique creates a safer gap by maximizing the likelihood of the gap occurring over the onset of a yellow light. Motorists trailing the gap should see the flashers come on, then, at about 5.5 seconds from stop bar, they should see the yellow light come on (even if they try to accelerate). Motorists leading the gap should not see the flashers activate and should be closer than 2.5 seconds from the stop bar at the onset of yellow, making it safely through the intersection.

In the example shown above, the onset of flashers leads the onset of yellow by 3.5 seconds, and the AWS is looking for a safe gap in traffic from 5.5 to 9.0 seconds upstream of the stop bar. This represents a gap size of 3.5 seconds. The vehicle leading the gap (Vehicle A) is about 5.5 seconds from the stop bar at the onset of the flashers and 2.0 seconds from the stop bar at the onset of yellow. The vehicle trailing the gap (Vehicle B) is about 9.0 seconds from the stop bar at the onset of the flashers and 5.5 seconds from the stop bar at the onset of yellow. In this example, Vehicle A would make it safely through the intersection, while Vehicle B would be prompted—first by the flashers, then by the yellow light—to stop at the stop bar.

The synchronization between the AWS flashers and the signal light is essential because ultimately drivers give more obedience to the signal light. This system safely manages the gap, whereas systems that do not correctly manage the gap actually trap drivers in the decision dilemma zone. Over time, drivers lose confidence in this type of system. If the lead flash time is too large, drivers are prone to abuse the AWS system and try to beat the light.

Enhanced Visibility

In order to see the flashers come on, the AWS sign should be within a driver’s cone of vision for all lanes of the approach. Signs should be positioned far enough away from the stop bar that motorists have adequate time to respond after seeing the flashing lights.

Traditionally, the AWS sign is a diamond sign with flashing beacons. If this traditional sign is used, it can be more effective to put a sign on each side of the approach.

In some cases, it has proven effective to use an overhead blank-out sign. The benefit of an overhead blank-out sign is not only visibility, but also a measure of distinction. Drivers are less likely to confuse the blank-out sign with the type of AWS system where the flashers are always active.

Minimize Impact of Max Out

The AWS system is similar to other green extension systems in that it does not override the maximum green time programmed into the traffic controller. In other words, AWS systems are subject to max out. However, in the event of max out, AWS systems will still provide drivers upstream of the flashers with advance warning of the clearance interval. Since a gap was not found, these drivers may be in the decision dilemma zone at the onset of yellow, but they have been given advance warning and should be more prepared to make a safe decision. This type of advance warning is often crucial to drivers of commercial vehicles in order to avoid load spills, skidding and collisions.

Cabinet Configuration

In order to minimize the impact of high-speed max out, the system designer should increase gap-out opportunities by limiting the number of cases for which green extension is warranted. There are several ways to do this, including the use of speed thresholds. In addition, the designer should evaluate the splits and cycle lengths at the intersection. Are time-of-day, dynamic maximum green or volume-density options warranted?

The traffic cabinet is configured to drive the AWS flasher load. The flasher load is active when the phase is inactive. For example, if the AWS system extends green for phase 2, then the flashers will be active when phase 2 is inactive. This is accomplished using a control panel that inverts the phase 2 output to drive a flashing beacon circuit.

The signal light loads are driven using a trailing overlap. Overlaps are defined as customized outputs driven by one or more included phases. Typically, the included phases are sequential so that the green output “overlaps” the clearance interval between two included phases. However, with a trailing overlap, the overlap output is simply extended beyond the length of the phase output. The trailing overlap then also has its own yellow and red clearance times that then need to be programmed. In essence, a trailing overlap adds an unnamed phase that keeps the signal indication green for a fixed amount of time after the included phase output goes yellow. In this case, the fixed amount of time (green time of the trailing overlap) is recommended to be about 3.5 seconds.

The detector output channel from the SmartSensor is wired in like a loop amplifier rack card. The controller input extension time (or as an alternative phase passage time) can then be used to determine the amount of green extension once the sensor output goes inactive.

Sensor Programming and Timing

SmartSensor Advance and the traffic controller work together to detect a 3.5-second gap. A 3-second phase passage time (or as an alternative, an input extension time) is programmed into the controller to ensure that the gap is at least as 3 seconds long. The 3-second passage time is added to the duration of each vehicle call in order to detect the 3.5-second gap. Vehicles qualify to send in a call based on range, speed and ETA criteria. The call will last only as long as the qualification criteria are met. In this case, the calls have been designed to be approximately 0.5 seconds in duration so that the size of the gap will be about 3.5 seconds. The flexibility of SmartSensor Advance’s zone placement and zone filtering was used to design green extension for a range of approach speeds, not just a single design speed. Two zones were used: one for slower speed vehicles and another for higher speed vehicles.

Zone 1 was used to extend green for vehicles going from 1 mph to 45 mph. While the current posted speed limit is 40 mph, a large percentage of vehicles were going faster than this speed at the installation site. (An advisory speed of 35 mph is also posted.) It is anticipated that the 85th percentile speed is 45 mph or higher. The location of zone 1 was selected to be from 575 ft. to 610 ft. from the stop bar. Vehicles tracked by the radar through this 35-ft zone will extend green for about 0.5 seconds if traveling at 45 mph. One suggested adjustment to the current design is to reduce the number of vehicles that will extend green (increase gap-out opportunities) by increasing the lower speed threshold from 1 mph to 25 mph or faster.

Zone 2 was used to extend green for vehicles going 45 mph and faster. The location of zone 2 was selected to be from 600 ft. to 1000 ft. from the stop bar. Vehicles tracked by the radar through this 400-ft. zone will extend green for about 0.5 seconds. The exact distances for which extension will occur depend upon the vehicle speed because ETA filtering from 8.7 to 9.2 seconds is employed. ETA criteria help synchronize the lead flash time so that high-speed vehicles will not be in the decision dilemma zone at the onset of yellow. For example, a vehicle traveling at 55 mph will send a call to the controller while it is traveling from 742 ft. to within 702 ft. of the stop bar. Likewise, a vehicle traveling at 70 mph will send a call to the controller while it is traveling from 946 ft. to within 894 ft. of the stop bar. This type of dynamic detection overcomes some of the limitations inherent with traditional fixed-point detection.

One suggested adjustment to the current design is to reduce the number of vehicles that will extend green (increase gap-out opportunities) by bringing the far edge of the zone in from 1000 ft. to 950 ft. Vehicles traveling 70 mph or faster will still trigger the call with this adjustment (unless the higher speed filter is reduced from 100 mph), but the duration of the call may be less than 0.5 seconds since 70 mph is double the advisory speed at the installation site of 35 mph.

One last consideration for sensor programming and timing is the placement of the sign. With the sign positioned 500 ft. from the stop bar, 45-mph vehicles trailing the gap will be just about 1.7 seconds upstream of the sign when the flashers turn on. High-speed vehicles that are trailing the gap will be even further upstream at the onset of the flashers. This should provide good visibility of the flashing beacons and adequate response time.

‍

Traffic engineers use storage queues as a tool to help manage the variability in vehicle flow and optimize system throughput. For example, at signalized intersections, accumulation of cross-street demand during red is used to help maximize the throughput during the subsequent green.  

Additionally, in order to help coordinate arterial progression in tightly packed platoons, queuing at both inbound approaches on the extremities of a corridor is regulated. Or on freeway on-ramps, queuing is metered to prevent deterioration of flow along the freeway mainline.

SmartSensor Advance can detect vehicles up to 600 ft. (182.88 m) upstream or downstream of its mounting location, and the SmartSensor Advance Extended Range can detect semis and other large vehicles at a range of 900 ft. (274.32 m). The long-range detection capability makes it uniquely cost-effective for many queue detection applications. SmartSensor Advance is classified as a continuous tracking advance detector (CTAD), which means that it continuously tracks the speed, position and estimated time of arrival (ETA) of approaching vehicles.  

Queue Formation

At signalized locations, vehicle arrival rates fluctuate throughout the day and often from one signal cycle to the next. Because of the unpredictable variability, vehicle sensors are needed to maximize throughput and manage traffic on demand.

One way to detect the level of arriving traffic flow at an intersection is to count the number of vehicles. SmartSensor Advance can provide a good estimation of the number of vehicles that arrive during the red interval.  

Another way to monitor queue levels is to program SmartSensor Advance to detect the drop in speeds when queue spillback has reached setback locations of interest. This method allows the ability to continuously activate different channels based on the estimated queue length.  

A 2009 Ramp Queue Detection report conducted for Mn/DOT by SRF Consulting used four channels to estimate queue length in this way. Queue length was reported at points of interest 100, 200, 250, and 275 ft. (30.5, 61, 76.2, and 83.8 m) back from the flasher stop line. The study reported that the average error in queue length estimation was 9.2 ft. (2.8 m) and the absolute average error was 36.1 ft. (11 m). These “excellent results” were achieved as the queue length on the ramp fluctuated from 0 to over 250 ft. (0 to 76.2 m) and back six times during the almost two-hour period from 3:15 to 5:00 p.m.  

Queue Dissipation

Motor vehicles form first-in-first-out queues. When these queues dissipate, vehicle speeds first rise near the front of the queue. In some applications, like ramp metering, only one vehicle exits the queue every three or more seconds and vehicle speeds never reach free-flow levels.  

In other applications, like signalized arterial traffic flow, a large portion of the standing queue will reach free-flow speed during the green time. The table below presents the amount of time required for vehicles at incremental queue positions to reach the stop line on green.

For example, if a vehicle is stopped 500 ft (152.4 m) from the stop line at the start-of-green, it will be about 43 seconds before that vehicle reaches the stop line.  

Queue Channel Configuration

How a SmartSensor Advance queue management channel is configured will depend on both the application and the prevailing traffic control philosophy. SmartSensor Advance has a large toolbox of channel, alert, and zone controls available to accommodate a wide variety of traffic control methodologies including the following:  

• Off/on-ramp management
• Intersection management
• Gap detection for safe and efficient queue reduction
• Queue length estimation  
• Queue reduction
• Queue calling
• Vehicle counting to adapt signal timing

In the sections below we will explore how to configure the SmartSensor Advance for each of these methods.

Off-ramp Management

Spillback from an off-ramp onto a freeway presents serious dangers due to the high-speed differential between the exit lanes and the main line. Because of this safety hazard, at critical times it is important to give priority to the off-ramp approach to the signalized intersection.

For this application, consider the length estimation method of queue management. If the queue is estimated to reach a specific point of interest (e.g. 400 ft./121.9 m from stop bar) then the intersection traffic controller can prioritize demand from the ramp when a rack card activates its contact closure channel.  

On-ramp Management

If mainline flow is near capacity, on-ramp queuing can be used to prevent the freeway from entering breakdown. However, in some locations, care must be made to make sure that the ramp queues do not spill back into the nearby intersections and cause arterial gridlock.

For this application, consider the queue length estimation method to address on-ramp metering problems. If the queue length extends beyond a point of interest, then the signal cycle frequency can be adjusted accordingly. If the queue becomes too long, consider turning off the meter and releasing the queue onto the freeway.

Intersection Management

There are probably as many signal management methodologies as there are intersections. In some respects, that is a good thing because it allows operations to be customized for each intersection. However, it also makes us realize that a one-size-fits-all approach is not always the best policy. So while some intersections may best be controlled using queue length estimation, with others it is better to use vehicle counting to adapt signal timing. And in some cases, it may be beneficial to use a hybrid method. In the following sections we will explain how to program SmartSensor Advance for each control methodology.  

For intersection management, the default Wavetronix recommendation is to use gap detection for safe and efficient queue reduction. This methodology can be used to manage intersection highway queues at isolated intersections, coordinated intersections, high-speed intersections, and at intersections where the standing queue extends beyond the 600-ft. (182.88 m) reach of SmartSensor Advance.  

Queue reduction is a gap detection–based method of queue management that can be coupled with speed-based channel deactivation. When used in combination with dilemma zone protection, queue reduction provides superior management of high-speed signalized intersections. (See the knowledge base article Basic Traffic Signal Timing and Advance Detection for more information on dilemma zone issues and ETA-based gap detection.)

On coordinated arterials, using queue reduction on the mainline helps optimize split times on a cycle-by-cycle basis based on real-time demand. This is done by using a common controller feature sometimes referred to as “actuated coordinated” that allows the coordinated split to be extended by up to about ten seconds if warranted.  

The conditions that warrant extension can be based on efficiency and safety. For example, an extension can be granted if a platoon is progressing slower than anticipated or if a platoon is longer than expected. On a high-speed arterial, platoons may also exhibit decision dilemma zone hazards which warrant green extension for a fraction of a second or more.  

These possibilities underscore the advantage of using SmartSensor Advance on a coordinated mainline; demand on the cross-street and conflicting movements will be served more quickly if the mainline traffic flow does not warrant continued green extension.  

Basic Queue Length Estimation

The basic method of queue length estimation uses SmartSensor Advance’s ability to track moving vehicles as they stop and start in the queue. Length estimation is reported by activating as many as eight contact closure channels. When a contact closure channel is active, the queue has reached the associated point of interest on the roadway.  

The above example reports queue activity at four setback distances from a stop bar: 100, 200, 300, and 400 ft. The channels labeled “Q100” and “Q200” are red to indicate that the channel is active because queue length is currently at least 200 ft. (61 m) from the stop bar. On the roadway view, a detection is shown at a range of 245 ft. from the stop bar with a speed of 2 mph; this indicates that the queue continues to grow beyond 200 ft. from the stop bar. No vehicles are shown in front of this vehicle because SmartSensor Advance is a passage detector, which means it tracks moving vehicles, but filters out stopped vehicles.

In order for the “Q100” through “Q400” channels to continuously activate a contact closure output, even when vehicles are stopped they are configured to latch and release based on vehicle tracking. They latch when a vehicle is tracked to a stopping point near the selected point of interest on the roadway. Similarly, the latched channels release when a vehicle is tracked through the selected area of the roadway at a relatively high speed.

To program a latched channel, two alerts are used: one is an ON alert, which specifies the conditions that cause the channel to activate, the other is an OFF alert, which specifies the conditions that cause the channel to deactivate. For basic queue estimation, the ON alert uses a low-speed activation threshold and the off alert uses a high-speed deactivation threshold.

The figure below shows a screenshot of the queue at a point later in time than that in the figure above. The “Q300” contact closure channel still remains active because speeds in the vicinity have not yet surpassed the deactivation threshold. However, the “Q100” and “Q200” have deactivated because the vehicles at these ranges have begun to creep forward and the associated speeds are now above the selected thresholds.

Enhanced Queue Length Estimation

In addition to activation and deactivation speed thresholds, SmartSensor Advance provides a suite of channel, alert, and zone controls to enhance queue length estimation (see SmartSensor Advance User Guide for more information):  

• Channel Extend Timer – The extend timer can be used to smooth channel outputs by requiring that the channel stay on for a number of seconds each time it is activated. In the figure below, a three-second extension time is programmed.

• Channel Delay Timer – This timer can be used to require that specified ON-alert conditions persist for a selected duration of time before the channel activates. In the figure above, a 0.2 second delay is programmed. This setup time can be used to prevent premature activation due to momentary low-speed false detections.
• Channel Max Timer – This timer can deactivate a latched channel after a specified duration of time (assuming that the ON-alert conditions no longer exist) and can prevent the channel from sending a perpetual call in the event of a detection error.
• ON alerts and OFF alerts– These can be programmed to activate or deactivate based on criteria other than a single speed near a point on the roadway. For example, instead of activating a channel as soon as a speed less than 5 mph is detected between 400 and 450 ft., it may also be advantageous to specify that no fast vehicles are nearby. This additional logic can screen out false alerts when there is a mix of speeds on the roadway. In order to add the requirement that there are also no fast cars nearby, the on alert would be programmed with a second zone, AND logic and an inversion of zone 2 outputs. The second zone would be programmed to activate its output when nearby vehicles are traveling at relatively high speeds. The inversion of this output would then create the logical condition that vehicles in the vicinity must not be at high speeds. Finally, the AND logic would require that a slow speed vehicle must be detected at the same time that nearby high-speed vehicles are not detected.  

Queue Reduction

Queue reduction channels do not use queue length estimation. Instead, these channels monitor the range and speed of vehicles in a dissipating queue in order to detect gaps in traffic for efficient phase termination. On high-speed arterials, we recommend using queue reduction channels with advance detector channels to make sure detected gaps are both safe and efficient.  

A queue reduction channel should be used with a minimum green time that will ensure vehicle movement at the location of the associated zone. The default location for a queue reduction channel is from 100 to 150 ft. (30.5 to 45.7 m) from the stop bar (see the figure below); a 15-second minimum green time is usually sufficient.  

In some cases, you may also wish to use a queue reduction channel to hold green until a desired flow speed is reached. In the figure below, the Q Reduce channel will deactivate if a gap in traffic is detected or if the flow speeds are greater than 35 mph. This feature of queue reduction based queue management not only detects queues, but also provides additional control in managing their dissipation.

Speed-based deactivation is a critical tool when more than one detection channel is programmed to extend green. For example, when a queue reduction channels is used in combination with an advance detection channel, speed-deactivation of the queue reduction channel will ensure that the advance detection can terminate the phase as soon as the first safe gap in traffic is found.  

Queue Calling

If a phase or movement at a signalized intersection is on minimum recall, then detection is not necessary to call the phase. But if you would like to use SmartSensor Advance to call a phase, queue length estimation can be used.  

For example, on minor side streets it may not be desirable to call a phase as soon as a single vehicle arrives at the stop line; it may be better to delay this call in order to minimize interruption of the mainline. However, if the queue on the minor side street quickly builds up to a considerable length, then you may want to call the side street with little or no delay.  

Note. With queue calling channels, the delay should be programmed in the traffic controller (not using the SmartSensor Advance).

Queue Counting

At some intersections it may be beneficial to count the volume of queuing traffic and modify signal timing accordingly. There are a couple of ways to do this:

Added Initial Method

One simple way to modify timing is to increment the amount of initial green based on the number of vehicles arriving on red. The added initial method is an aspect of volume density control and works when the queue of traffic is contained within the 600-ft. (182.88-m) coverage area of the SmartSensor Advance. For more extensive queuing, additional methods must be used instead of, or in combination with, the added initial method.  

Volume Counts Method

A more comprehensive way to modify timing is to change to a different signal plan based on the volume of traffic. SmartSensor Advance volume counts are not produced by directly counting vehicles on a lane-by-lane basis as with SmartSensor HD, therefore SmartSensor Advance counts tend to be less accurate. Count errors can occur when vehicles in adjacent lanes merge, or when truck-trailer combinations are counted multiple times. However, these rough counts are accurate enough to be useful for making adjustments to signal timing.

In the figure below, the Q Count channel is configured by simply setting up a trip-line zone starting at 415 ft. from the sensor. When each tracked detection crosses through this area it will be signaled to the traffic controller as a contact closure. The traffic controller then counts these contact closures to detect incoming volume.

For proper counting via the traffic controller, SmartSensor Advance and the rack card must be configured appropriately.

This figure above shows the settings for communicating to a Click 172/174 rack card. The minimum duration of each contact closure is 130 ms and all standard traffic controllers should be capable of counting these pulses. When receiving data from SmartSensor Advance, the Click 172/174 racks cards should always be configured in Actuation mode.

‍

The SmartSensor cable used to be the standard cable for SmartSensor HD, Advance, and 105 installations; although this cable has been discontinued, certain installations may still use it. This cable is sometimes referred to as the 9-conductor cable, to differentiate it from the newer 8-conductor and 6-conductor cables. This cable is composed of three groups of wires, each containing color-coded wires and a drain wire surrounded by a shield.

This diagram shows the SmartSensor cable’s 26-pin socket assignment (seen from the solder cup side of the connector).

The codes listed in the diagram are to be used to solder wires into the back of the plug where the letters represent the individual solder cups.

This figure shows the SmartSensor cable wire connections into a Click 200 surge protector.

‍

The connector end of the SmartSensor 6-conductor cable mates to the 8-pin connector on the SmartSensor Matrix and SmartSensor Advance. The cable has seven wires. The sensor itself also contains internal wires that connect to the protective earth lug.

This figure shows the 6-conductor cable wire connections into a Click 222 surge protector.

‍