902.5 Phasing and Timing the Signal
From Engineering Policy Guide
902.5.1 Discussion
The full value of any signal installation is realized only when it is operated in a manner consistent with the requirements of traffic. The use of unduly long cycle lengths, improper phasing and the sequence of those phases can result in disrespect for and poor observance of signal indications and in greater crash potential. Observation of an efficient signalized intersection during full traffic demand is one having vehicles moving through it at all times.
| Multiple Left-Turn Phasing Strategies, Assessment |
| Report 2004 |
| Summary 2004 |
| See also: Innovation Library |
An efficiently phased and timed signal can lead to a significant increase in the capacity and/or safety of an intersection. Costly geometric revisions to improve an intersection may be postponed or eliminated with a thorough review of the signal operation.
Although the methods presented in this article cover most of the features available to the engineer to optimize the operation of a signal, there is no substitute to careful observation of an intersection. Operating characteristics that present themselves only at the intersection cannot be calculated or planned for on paper. Observations are to be made to determine the impact of any changes, and final revisions are to be based on those observations. Any method presented here is to be used only as a starting point, and not a final solution.
A signalized intersection information sheet is available to determine the setup of an intersection.
902.5.2 Signal Phasing
The phasing of a signal determines the order that movements are serviced.
A study of traffic movements at the intersection is made to determine permitted and controlled movements. From this, the number and sequence of traffic phases is determined, which in turn determines the interval or color sequence and types of signal indications to be used. In general, the most efficient operation is obtained with the fewest possible phases; however, each signal installation is designed to provide safe and efficient control of conflicting traffic movements. An exclusive left turn phase is to be considered when the number of left-turning vehicles is 100 vehicles or more during the peak hour. If the signal is within a progressive signal system or is interconnected with another signalized intersection, a time-space diagram is to be prepared to determine signal phasing. Signal timing programs are available for use with micro-computers to analyze phasing for both interconnect systems and isolated intersections.
Signal phases are identified by directional movement as follows:
Figure:
In general, phases 1, 2, 5 and 6 are used for the major street, and phases 3, 4, 7 and 8 are used for the minor street. Phases 2, 4, 6 and 8 are normally used to designate through traffic movements. Combined phases are shown with a + symbol (i.e. 1+ 6, 2+ 6).
The following articles provide guidelines for selecting phasing. A number of examples are shown in Signal Phasing and Layout Examples.
902.5.2.1 Sequences
902.5.2.1.1 Fully Actuated Controllers
The typical phase arrangement at most intersections is called the "8-phase quad" operation, with the eight phases grouped into two sets of movements, or "rings". NEMA designates the assignments as follows:
A form for fully-actuated controller sequencing of an intersection is available.
As shown, mainline leading left indications are displayed first. Phases on opposite sides of the "barrier" cannot operate together (such as phases 2 and 4). Phases on the same side of the barrier in one ring can run concurrently with phases in the other ring (such as phase 3 on at the same time with phase 8). Only one phase per ring can be operating at the same time. (In ring 1, phase 2 cannot be on if phase 1 is on).
Phase assignment is to be kept uniform in accordance with this ring structure whenever possible. Mainline left turns is to be assigned to phases 1 and 5, and mainline through movements to phases 2 and 6. Side street left turns is to be assigned to phases 3 and 7, with through movements assigned phases 4 and 8. Other phase numbering schemes may be used, but consistency is to be maintained throughout the district.
902.5.2.1.2 Semi-Actuated Controllers
There is no "semi-actuated" controller made for use, but any fully actuated controller can be easily configured to operate "semi-actuated", or with detection in most lanes. Usually, the mainline through lanes on a high-volume approach intersecting a low-volume road or the mainline through lanes on a coordinated roadway can do without loops. These movements have constant calls programmed to substitute for detection. Ring structure is identical to fully actuated control.
902.5.2.1.3 Pre-timed Controllers
At an intersection operated by a pre-timed controller (Figure 6.1), the sequence is controlled by a specific pattern of signal changes. Each change of this sequence is called an interval. The intervals are programmed for every change in the signal faces, and specify what color each vehicle movement will see for a set amount of time. A simple T-intersection with two directions would be set up in a pre-timed controller like this:
Table 902.5.2.1.3
| INTERVAL | EB & WB | NB | TIME (sec) |
|---|---|---|---|
| 1 | G | R | 25 |
| 2 | Y | R | 4 |
| 3 | R | R | 0.5 |
| 4 | R | G | 16 |
| 5 | R | Y | 4 |
| 6 | R | R | 0.5 |
A form for pre-timed controller sequencing of an intersection is available.
The addition of other directions and/or protected left turn movements would require additional intervals. The sequence of the intervals can be re-arranged to provide the most efficient use of green time; however, conflicting directions cannot be programmed to operate at the same time. In a pre-timed controller, the first interval is to always be the common mainline green time, as the previous example shows.
902.5.2.1.4 Pre-timed – Actuated
Most pre-timed controllers allow for a certain number of directions to have vehicle detection. This allows for certain intervals to be skipped or other intervals to be serviced depending on vehicle detection. Using the example in 902.5.2.1.3 with detection on the side street, the interval sequence would be:
Table 902.5.2.1.4
| INTERVAL | EB & WB | NB | TIME (sec) |
|---|---|---|---|
| 1 | G | R | 25 |
| 2 | Y | R | 4 |
| 3 | R | R | 0.5 |
| 4 | R | G | 15 |
| 5 | R | Y | 4 |
| 6 | R | R | 0.5 |
| 7 | G | R | 24.0 |
If the controller detects no vehicles on the side street during interval 1, it will proceed to interval 7 and continue the mainline green time. After interval 7 has timed out, interval 1 will be in effect, and the sequence continues. If during interval 1 there is a vehicle present, intervals 2 through 6 will be timed out, and then back to interval 1. Interval 7 will be skipped. This sequence can be better illustrated by drawing up interval paths:
These paths become more complex with the addition of detectors in other directions, but with proper planning, it can closely imitate a fully actuated controller.
902.5.2.2 Left Turn Phasing
There is a growing tendency by the driving public to violate traffic signal indications. One common violation involves improper left turns made from exclusive turn lanes governed by three section signal heads. In order to help reduce these violations, guidelines are available to aid in determining the proper left turn phasing for signalized intersections.
| Left Turn Phasing Warrants |
Left turn indications at signalized intersections are designed so they are neither overly restrictive nor inconsistent from the driver's point of view. Protected-only left turns are extremely limiting, therefore they are to only be used when tight control is absolutely necessary for a specific approach at an intersection. The Left Turn Phasing Warrants are available in an interactive spreadsheet for safety warrants and capacity warrants to determine the amount of protection to be given a left turn movement. These warrants are based upon accepted safety and capacity values for signalized intersections.
| Multiple Left-Turn Phasing Strategies, Assessment |
| Report 2004 |
| Summary 2004 |
| See also: Innovation Library |
When factors such as sight distance, speed of opposing vehicles, etc. make permissive turns undesirable, the permissive left turn option is to be removed. Safety warrants are to be checked first; if an approach requires protected-only phasing for safety reasons, it is unnecessary to check the capacity warrants.
Once safety considerations are satisfied, Capacity Warrants will need to be analyzed. Capacity Warrants are divided into three parts: Permissive-Only left turns, Protected/Permissive left turns, and Protected-Only left turns. This criteria is to be used when designing or upgrading a signal installation.
In order to provide the proper phasing at an intersection, it will be necessary to check Capacity Warrants for several hours for each approach. For example, if only the peak hour is checked, the phasing will most likely be too restrictive for the rest of the day. It is recommended that the peak periods plus a sample of off peak hours be checked before choosing the phasing.
When traffic volumes at an intersection are approaching the thresholds listed in the capacity warrants variable left turn phasing may be used by time of day. Variable left turn phasing allows for the selection of either protected only, protected/permissive, or permissive only left turn phasing. This can be used to provide appropriate phasing for varying volumes throughout the day. The protected left turn phase can be omitted by time of day, and flashing yellow arrow operation allows for removal of the permissive left turn in addition to removal of the protected left turn phase. It can be used only on approaches with “positive” signal lane control, in that each approach lane has its own signal indication. Refer to 902.11.28 for more information on flashing yellow arrow indications. Otherwise, the most appropriate left turn phasing should be chosen given the results of the Capacity Warrants.
When the flashing yellow arrow indication is used to provide variable phasing, each hour during a typical day should be evaulated to determine proper phasing throughout the day. The Variable Left Turn Worksheet can be used to evaluate each hour during the day. During initial installation the flashing yellow arrow indication can allow the selection of more restictive phasing initially and then change to a less restrictive mode if appropriate.
Note that some overlap may occur when analyzing the volumes at each approach (i.e., the data for one hour may satisfy parts of the criteria for both permissive-only and protected/permissive left turns). Therefore, it will be necessary to check at least two of the three parts of the criteria. When an overlap does occur, previous experience and/or evaluation studies at the location is to indicate whether the situation is better served by the more or less restrictive phasing that is determined using the criteria.
The left turn phasing guidelines, below, give safety and capacity considerations for selecting left turn phasing. The interactive spreadsheet allows for the user to directly enter criteria and see suggested thresholds based on these formulas. These guidelines are to be used when reviewing design plans and when modifying the phasing of an existing installation.
GUIDELINES FOR VARIABLE LEFT TURN PHASING
This is a guide for the selection of variable left turn phasing hour-by-hour. Guidelines based on safety and capacity are provided. Other issues may be considered in the selection of left turn phasing.
Definition of Terms
The following terms are used in these guidelines :
VLT = The left turn volume per hour per approach.
(VLT)pp = The number of vehicles attempting to make permissive left turns during the permissive part of a protected/permissive left turn per hour per approach.
VO = The opposing volume per hour per approach per lane (excluding free right turn volume and volume serviced by a separate right turn phase).
cp = The cycle length (in seconds) when those volumes occur using permissive-only phasing1.
cpp = The cycle length (in seconds) when those volumes occur using protected/permissive phasing1.
gp = The green time (in seconds) common to both VLT and VO during that cycle using permissive-only phasing1.
gpp = The green time (in seconds) common to both (VLTLT)pp and VO during that cycle using protected/permissive phasing1.
TP = The time allocated to the protected left turn movement using protected/permissive phasing1.
1NOTE: These green times are used in the calculations regardless of the existing phasing. For phasing configurations not currently used it will be necessary to develop realistic timing for that phasing configuration. A signal timing computer program can be helpful in developing this timing.
Adjustment of Left Turn Volumes
This evaluation considers the number of vehicles attempting to make permissive left turns during the permissive part of a protected/permissive left turn. Therefore, the effects of protected left turns should be eliminated. This can be handled using the following method:
- (VLT)pp = VLT - VP,
where the variable VP is the number of left turn vehicles served by the protected left turn indication. If this formula yields a negative number, use 0 for (VLT)pp. Assuming that vehicles enter the intersection at a rate of 2 seconds/vehicle, the volume using the protected movement in a one-hour period is:
VP = 
Safety Criteria
Protected Only Left Turns
NOTE: Protected-Only left turns should be provided full-time when any one of the following criteria are satisfied:
A. Number of Opposing Through Lanes > 3
B. Sight Distance:
- < 125 ft. for 20 mph
- < 150 ft. for 25 mph
- < 200 ft. for 30 mph
- < 250 ft. for 35 mph
- < 325 ft. for 40 mph
- < 400 ft. for 45 mph
- < 475 ft. for 50 mph
- < 550 ft. for 55 mph
C. Number of Correctable Accidents By Upgrading to Protected Only Phasing > 5 over 12 months
- NOTE: The 5 correctable accidents should involve the SAME Left Turn approach. Only those approaches satisfying that criteria should be upgraded.
D. Number of Observed Traffic Conflicts > 48 Conflicts / 11 Hour Day
- NOTE: Conflicts occur when motorists on the OPPOSITE APPROACH must respond to the actions of motorists making the subject left-turn movement. Therefore, conflicts should be measured by observing the intersection from the opposite approach. Only those approaches satisfying the criteria should be upgraded.
E. Speed (prevailing)
- > 50 mph AND > 2 opposing thru lanes
- = 45 mph AND a study indicates that the number of gaps is insufficient to turn safely
F. Number of lanes for left turns on the approach > 2
G. Unusual intersection geometrics that make permissive left turns difficult.
Protected/Permissive Left Turns
- NOTE: Protected/Permissive left turns should be provided when the following criteria is satisfied.
A. Number of Observed Traffic Conflicts > 29 Conflicts / 11 Hour Day
- NOTE: The number of conflicts are those occurring on the OPPOSITE APPROACH that are caused by the subject left-turn movement. Only those approaches satisfying the criteria should be upgraded.
Capacity Criteria
Permissive-Only Left Turns
- NOTE: Permissive-Only left turns may be provided when one of the criteria in (A.) is satisfied in conjunction with (B.).
A. VLT < 100 Vehicles per Hour
- VLT < 2 Vehicles per Cycle 1
- VO < 100 Vehicles per Hour
- 1NOTE: This criteria is only valid if observations at the intersection show that drivers tend to make left turns during the clearance interval on a regular basis. These field checks should be made during the hour(s) in which either the highest left turn volume or the highest opposing volume occurs.
B. VLT + VO < 600 x (gp/cp)
Protected/Permissive Left Turns
- NOTE: Protected/Permissive left turns should at least be provided when one of the criteria in (A.) is satisfied in conjunction with one of the criteria in (B.).
A. VLT > 100 Vehicles per Hour AND VO > 100 Vehicles per Hour
- VLT > 2 Vehicles per Cycle1 AND VO > 100 Vehicles per Hour
- VLT + VO > 600 x (gp/cp)
1NOTE: This criteria is only valid if observations at the intersection show that drivers tend to make left turns during the clearance interval on a regular basis. These field checks should be made during the hour(s) in which either the highest left turn volume or the highest opposing volume occurs.
B. (VLT)pp + VO < 1200 x (gpp/cpp)
- (VLT)pp x VO < 50,000
Protected-Only Left Turns
- NOTE: Protected-Only left turns should be provided when any one of the following criteria are satisfied.
A. (VLT)pp + VO > 1200 x (gpp/cpp) for 3 or more hours if considering permanent phasing change
B. (VLT)pp x VO> 50,000 for 3 or more hours if considering permanent phasing change
Protected left turn movements should be provided with an adequate turn bay or a separate turning lane, depending upon the volumes using the intersection and the existing intersection geometry. Shared lanes (LT+TH or LT+TH+RT) are undesirable for this purpose. Protected-Only left turns are not used with shared lanes unless split phase operation is used.
902.5.2.2.1 Simultaneous Left Turns
The phasing sequence which permits signal-controlled opposing left turn movements to move simultaneously, preceding the through movement, is the most efficient, provided auxiliary left turn lanes are available, left turn volumes are nearly equal, opposing traffic volumes are sufficient to justify simultaneous movements, and the intersection geometric configuration will permit the movements.
902.5.2.2.2 Off-Ramps
The phasing for signal controlled interchange exit ramps entering the crossroad is designed to prevent backup onto the freeway through lanes. A method of signal interconnection is to be provided between the ramp terminals.
902.5.2.2.3 Dual Left Turns
The modification of an existing intersection to provide dual left turns is to be carefully reviewed. Initially, left turn capacity can be increased with a dual left turn movement. However, without a new left turn lane built for this movement, the new left turn lane will most likely be shared with the corresponding through or right turn movement, and decrease those lanes' capacity by requiring split phasing. Also, the dual left movement is to be protected-only.
902.5.2.2.4 Leading and Lagging Left Turns
A leading left turn is a left turn that precedes or is accompanied by the first through movement in a direction. A lagging left turn is a left turn that follows the last through movement or is on at the end of the green time for a through movement.
At locations where a left turn lane is needed but cannot be provided, some relief is achieved by the use of a leading or lagging green period for the direction of traffic with the heavy left turn. When auxiliary left turn lanes are provided, it may be advantageous to lead and lag the left turn movements when volumes are unequal. A time-space study is used to verify feasibility if the signal is interconnected.
There is a danger in the indiscriminate use of "lead-lag" left turn phasing used with protected-permissive phasing. In this situation, the yielding left motorist who arrived after the leading left indication may try to clear the intersection during the yellow change interval, not knowing the opposite approach is continuing to display a green through indication while waiting for the lagging left indication. This is called a trap condition and is to be avoided if possible. This condition also exists where the left turn is permissive only and the opposite left is lagging. If either configuration is used, close monitoring is important. For existing trap conditions where a problem is apparent, possible solutions are to change the left turn phasing to lead-lead or lag-lag, use a flashing yellow arrow indication, “Dallas” left turn phasing or make the leading left turn protected-only.
The flashing yellow arrow display for left and/or right turns consists of a three or four-section head that uses a flashing yellow arrow indication instead of a circular green indication for permissive left-turn movements. It can be used only on approaches with “positive” signal lane control, in that each approach lane has it’s own signal indication. It’s capable of being operated in any of the various modes of left-turn operation by time of day, and it’s easily programmed to avoid the “yellow trap” that’s associated with some permissive turns at the end of the circular green display. The FHWA has issued an Interim Approval for the use of a flashing yellow arrow indication as an optional alternative to a circular green for permissive left turn movements and provisions of the Interim Approval are being incorporated into part 4 of the MUTCD. Refer to 902.11.28 for more information on flashing yellow arrow indications.
The flashing yellow arrow display for left turns is preferred but “Dallas” left turn phasing is also an options that uses a visibly-limited exclusive left turn signal face to display a circular green signal indication for the “trapped” left turn movement. The MUTCD allows this phasing only when using a protected/permissive signal face (5-section). See MUTCD Section 4D.06 for details on how to apply this phasing scheme.
Leading and lagging left turn phasing is typically used to improve coordination on mainline routes where modifying the left turn phasing will provide a significant improvement in coordination. Lead-lag protected-permissive phasing is not normally used on uncoordinated approaches. Lead-lag phasing can be helpful where a short left turn bay exceeds its capacity. The lagging left can prevent the turn bay overflow from blocking through traffic.
902.5.2.3 Overlaps and Right Turn Phasing
An "overlap" provides a green or flashing yellow arrow indication for a traffic movement during the green intervals of two or more phases. The overlap green indication may also be allowed during the change period between two or more phases if these phases are consecutive. An overlap can be integrated into an actuated controller to supplement the flow of traffic. A simple application of an overlap is shown below:
For this case, a through movement is labeled "OLA" (OverLap A). OLA indication can be green while either phase 1 or phase 2 is green. While phase 1 is timing out the change period in transition to phase 2, the OLA indication remains green since OLA follows concurrently. OLA indication is red when phase 3 is green or yellow.
Overlaps are to not be used when normal NEMA dual ring structure can be utilized. In the previous example, the overlap is to be assigned to phase 6. A more practical application is for signalizing a right turn movement as shown below:
In this case, the right turn phase is labeled "OLA" (Overlap A). OLA will display a green right while either phase 1 or phase 7 is green and will display a ball green when phase 2 and phase 6 are green. It receives no additional time, since its time comes from phases 1 and 7.
902.5.2.4 Split Phasing
Split phasing is servicing a street one approach at a time. Because split phasing is a very inefficient use of green time, other alternatives, including geometric improvements, is to be considered. This type of phasing greatly impedes co-ordination if used on a main line, and decreases the efficiency of the whole intersection by increasing the amount of time needed to serve both approaches separately. Reasoning to support this phasing is traffic from the two approaches would otherwise occupy the same space at the same time. Common examples for use are a dual left turn lane on one approach which is shared with a through movement, or safety concerns. For intersections with existing split phasing, care is to be taken when modifying phasing to include permissive left turns.
902.5.2.5 Alternate Sequences
When needed, the normal NEMA ring sequence can be altered to fit operating conditions, with restrictions as detailed in 902.5.2.1.1. The most common application is to provide lead-lag left turns. Assume this phasing assignment for the following intersection:
In order to allow the northbound phase 1 to become a lagging left turn, the sequence of phase 1 and phase 2 must be programmed to switch. Under this alternate sequence, the ring structure would be:
This ring sequence starts with southbound left and through, and ends on the left side of the barrier with northbound left and through. Sequencing on the right side of the barrier is unchanged. More than one alternate sequence can be programmed.
902.5.2.6 Pedestrian
Every effort is to be made to display the WALK indications with a green phase or interval, depending on the type of controller. This "phase-associated" pedestrian operation lessens the overall delay to drivers. Using a fully actuated intersection as shown:
The pedestrian WALK and flashing DON'T WALK indications for northbound-southbound on the east side of the intersection would be displayed only during phase 6 green and yellow time. Likewise, the indications for eastbound-westbound on the south side would be displayed only during phase 4 green and yellow times.
If this were a pre-timed, or pre-timed-actuated intersection, the WALK and flashing DON'T WALK displays would be active during the corresponding through intervals.
Under no circumstance will a pedestrian WALK or flashing DON'T WALK indication be active during a phase or interval which leads vehicles into the crosswalk. Using the previous example, the pedestrian indications for northbound-southbound cannot be active during phase 5, or any phase for the east-west direction of travel. Pedestrian indications are allowed in conjunction with the phase 6 right turn and/or the phase 5 southbound yielding left turn, if allowed, since the pedestrian movement has legal right-of-way over the northbound right turn or southbound yielding left turn.
Under extreme circumstances, an exclusive movement may be needed for the pedestrian indications. This is most commonly used for pedestrian crossings at school signals, as described in 902.4.3, School Signal Operations.
902.5.2.7 Diamond Interchanges
Due to the close spacing of both ends of a standard or compressed diamond interchange it is extremely important to offer coordination between both ends of the interchange. This may be accomplished with pre-timed or actuated control.
This discussion primarily relates to standard diamond interchanges, however similar considerations can also be made for half diamond and other diamond interchange variations. If a signalized roadway (i.e. outer roadway) is very close to a ramp intersection, a different configuration may be required.
Typically, pre-timed control uses two controllers: one for each end. Fully actuated control typically uses one controller to control both ends. In some situations, it may be possible to use two actuated controllers to control each end. This can be accomplished by running a pre-timed operation during the critical peak times and fully or semi-actuated operation during the off-peak times. This option is to be evaluated carefully, as coordination during fully or semi-actuated operation may not be optimal.
If the interchange is in a coordinated system, pre-timed operation may be used, however, actuated operation can also be effective in coordination. The merits of each setup are to be evaluated for the best operation at each location. The following are recommended criteria for selecting the best setup. A cost comparison may also be helpful in deciding which setup to use.
Fully Actuated Control With One Controller:
Overall interchange operates below capacity.
No more than one or two mainline or ramp left turn movements require critical coordination.
There is sufficient left turn storage between the ramps.
Actuated Control With Two Controllers (With Pre-timed Operation During Peak Times):
Overall interchange operates below capacity during off-peak.
None of the movements require critical coordination during off peak.
There is sufficient spacing and left turn storage between ramps for random (non-coordinated) operation during off-peak.
Pre-timed Control With Two Controllers:
Overall interchange operates near or at capacity.
Most or all of the mainline and ramp left turn movements require critical coordination.
There is not sufficient left turn storage between the ramps.
Diamond Interchange Examples provides examples of phasing configurations for diamond interchanges.
902.5.3 Settings
Once the proper phasing has been determined for an intersection, the proper timings for the signal indications must be developed in order to function efficiently.
902.5.3.1 Green Interval
902.5.3.1.1 Minimum Green
The minimum green ensures the green indication is displayed long enough after a red indication to clear the stopped traffic. Green times that are too short can lead to frequent and needless stops.
The following minimum green times are recommended: Mainline Through Movement: 10 seconds Side street Through Movement: 7 seconds Protected Left Turn: 7 seconds
In some cases, the minimums may be set higher. For heavily traveled mainline throughs, a higher minimum may be desired to reduce the chances of the controller quickly cycling off the mainline green to other phases.
In some cases, the minor approaches or movements may be set to six seconds minimum. Minimum green is to not be set below six seconds for green indications.
902.5.3.1.2 Maximum Green
Maximum green times is to be set on an actuated controller as low as possible, but high enough to adequately handle most of the vehicle demands. Maximum greens set too low result in less flexibility in the phase's timings based on detector activity, since there is very little time between the minimum and maximum for the fluctuation in traffic. Maximum greens set too high can result in unnecessary delays during periods of detector failures, and increase the delay for other approaches.
The following maximum green times are recommended:
Mainline Through Movement: 40 to 70 seconds
Left Turn Movements: 15 to 50 seconds
Side street Through Movements: 20 to 50 seconds
Observation is the final factor in deciding the proper setting for maximum green. In low volume and/or low speed situations, lower maximums may be advantageous. Some approaches may need more than the usual times at different times of the day. Several newer controllers allow for different maximum settings to be enacted through the time clock. This is useful if heavy demand on a certain phase can be accurately predicted and set to a time of day. This is also useful with semi-actuated control where the mainline timing is controlled by a maximum recall since mainline demand typically changes by time of day.
Although the concept of a cycle length is usually reserved for pre-timed control and coordinated actuated control, it is to also be applied to isolated, actuated control. The temptation to set all maximums at an isolated intersection extremely high is to be avoided. Maximum settings too high result in longer delay for other approaches, and defeat the flexibility of actuated control by creating needless backups. See 902.7, Coordination, for more discussion on cycle lengths for all types of control.
902.5.3.2 Clearance and Change Intervals (Change Period)
The change interval (yellow indication) and clearance interval (all indications displaying red, if used) are required to prepare the intersection for the transfer of right of way. These intervals permit vehicles that are either within the intersection or so close to it that they cannot comfortably stop to clear the intersection, and to permit those vehicles that can come to a comfortable stop to do so. The total time of the yellow change interval and the red clearance interval (if used) is the change period.
According to the MUTCD, the yellow change intervals are to have a minimum of three and maximum of six seconds. The yellow time should never be less than the sum of the first two terms in the formula below. The last term is the red clearance interval and should not exceed six seconds. If a phase change period longer than the selected yellow change interval is needed or it exceeds six seconds, then the additional time is provided by an all-red interval.
The addition of an all-red clearance interval is to not be automatically provided after every movement. The all-red time has become nearly automatic at intersections and with this has come increased driver expectancy of an all-red. Over-use may lead to drivers incorrectly assuming a change will be timed out with all-red and failing to clear the intersection in time. The use of an all-red is to be reserved for phases where either a phase change period longer than the selected yellow change interval is needed or where observations show a need. A common need is for unusually wide intersections. Proper application of the following formula is to eliminate most cases of needless all-red intervals.
The duration of change and clearance intervals, as well as the appropriateness of red clearance intervals, is a topic with no clear consensus. The following formula is developed based on a kinematic model of stopping behavior to determine the duration of the yellow and red indications, and is in common use throughout the country.
- Change Period:
CP = nondilemma change period (yellow plus all red), seconds
t = perception-reaction time, recommended as 1.0 sec.
V = approach speed, ft/sec
g = percent grade (positive for upgrade, negative for downgrade)
a = deceleration rate, recommended values as follows:
- 10 ft/sec2 - low speed approaches, i.e. CBD
- 12.5 ft/sec2 - typical arterial approaches
- 15 ft/sec2 - high speed approaches
W = width of intersection, ft.
L = length of vehicle, recommended as 20 ft.
NOTE: CP greater than seven seconds not recommended.
A spot-speed study on an approach to an intersection will produce a range (or distribution) of speeds. Typically, the 85th percentile speed or the prevailing speed limit has been used to determine the yellow change interval. It is important, however, to also consider slower traffic going through the intersection at the 15th percentile speed. Low speeds and wide intersections or large left turn radii are a combination that may require a longer change period (yellow plus all-red). It may be necessary, therefore, to calculate the equation using both the 85th and 15th percentile speeds and to employ the longer of the two calculations.
Using this equation for approaches with steep downgrades yields such long intervals that they appear unreasonable to drivers as well as the engineer. The remedy is not to ignore the physics of the situation when an unusually long phase change period results from a steep grade or from high approach speeds. The remedy may come from other devices such as warning signs or other countermeasures.
902.5.3.3 Recalled Phases
At isolated actuated intersections, it is common for one roadway to serve considerably more traffic than the others. In these cases, the use of the recall feature for those approaches carrying the heavier traffic will increase the potential for drivers on the major roadway to receive a green signal as they approach the intersection, thus minimizing the number of stops. There are three main recall options:
1. Min(imum) Recall. This will place a continuous request for service for minimum green on the selected phases. If the minimum green setting is below the maximum green setting, then the selected phases' green can extend to maximum time with vehicle detection before the next phase is serviced, or this phase can stay green if no other movements have vehicle calls.
2. Max(imum) Recall. This will place a continuous request for service of maximum green on the selected phases. Green indications are displayed for the selected phase up to the programmed maximum, regardless of vehicle calls. Once the maximum time has expired, the next phase will be served if there is a detected call. It is commonly used for movements without detectors in semi-actuated, isolated intersections.
3. Soft Recall. This is similar to Minimum Recall, except the phase set for Soft Recall is served if no other detected calls are present on any other phases. All phases must be actuated, including the Soft Recall phase. Typically used on main street throughs so the controller will not serve the main throughs if other calls exist and there are no actuated calls on the main throughs.
902.5.3.4 Detector Settings
With actuated control and properly timed detectors, the green time can be distributed to the needed movement and taken away when demand is gone. In order to keep this rotation of phases moving along without dwelling on a movement with little demand, the settings must be programmed to match the type, size and location of the detectors.
902.5.3.4.1 Stop Bar Detectors
The most common placement of detection senses the vehicle at the stop bar. This location is generally used for minor approaches, mainline turning lanes, and low-speed mainline through lanes. When this location is used, the type of detection is set for "presence". Presence detection allows for a call to be placed to the controller whenever a vehicle is in the zone of detection, and the call is not removed until the vehicle leaves the detected zone.
The setting to control the time to elapse without a vehicle call before changing phases is the "passage" or "gap" time. This timer will reset if a vehicle is detected before the max green has timed out and extend the green indication. If the timer reaches zero before another call is placed, the controller "gaps out" and serves the next phase.
Settings for the passage time vary on how much of a gap to allow between successive vehicles before the following vehicle loses the right of way. If the phase is a turning phase or a minor through phase, the setting would be low, less than two seconds, in order for major movements to be served quicker. If the phase is for the major through movement, longer passage times are needed in order to favor the larger volumes. The passage time is to not be set too high, generally not above four seconds. This will result in long green times because the right of way will be retained by traffic that is too far apart.
902.5.3.4.2 Advance Detectors and Gap Reduction
Advance detectors are used for through movements on the major roadway to create a detection zone well back of the stop bar, especially if the 85th percentile speeds on the major roadway are 45 mph or more. These detectors are typically placed a distance of five seconds behind the stop bar at the 85th percentile speed. Because of the long spacing from the stop bar, sole use of the standard passage time would result in extended green times with large gaps. Gap reduction control provides a means of decreasing the passage time as the speed of the flow increases. The needed settings for gap reduction control are:
1. Passage Time. When used in gap reduction control, passage time becomes the maximum gap time allowed. Its purpose is to provide sufficient time so a vehicle moving at the prevailing speed can travel from the detector to the stop bar.
2. Minimum Gap. This is the smallest gap time allowed, and is the lowest time allowed after gap reduction has occurred.
3. Time Before Reduction. This setting is the time to postpone the start of gap reduction from passage time to minimum gap time. This time is usually set up to allow any queue to accelerate up to a free flow speed.
4. Time To Reduce. After "Time Before Reduction" has reached zero, this timer begins. This is the time allowed for gap reduction to go from its passage time to the minimum gap time. Reduction is linear.
5. Added Initial Per Actuation. This setting allows the controller to increase the minimum green time to account for the vehicles stored between the detector and the stop bar at the beginning of the green interval. Without this setting, the minimum green would have to be set high enough to insure all vehicles could clear the intersection. This time is set based on the number of seconds added for each detection while the phase is not green, usually 0.5 to one second added to the programmed minimum per actuation.
The benefit of these settings is best realized when stop bar detection is not used on the approach or is de-activated during the approach green interval. Stop bar detectors are not needed on mainline approaches with advance detection unless there is a driveway or street between the advance detectors and the stop bar.
The following diagrams show the relationship between the gap reduction settings and the concept of added initial time:
Advance detection may also be advantageous on lower speed mainline approaches in lieu of stop bar detection. Where a minimum recall is used for mainline phases, presence detection (stop bar detection) is not required. In this case, the advance detectors would typically be installed 75 to 150 ft. behind the stop bar. This setup will make more efficient use of the passage timer and can reduce detector installation cost. The minimum green will need to be sufficient to clear the storage between the advance detectors and the stop bar. The gap reduction and added initial settings are typically not needed in this case.
902.5.3.4.3 Detection on High-Speed Approaches
With signals installed on high-speed roadways (85th percentile speed of 45 mph or more), a single back detector may not be able to be placed in the proper location to keep vehicles from the "dilemma zone" conflict. The "dilemma zone" is the area approaching the intersection where drivers first see a yellow indication, but is too far away to proceed through the intersection, and too close to stop comfortably. If a back detector is placed too close to the intersection, it may not detect fast vehicles in time to control them with gap timing. If placed too far back with high gap times, the mainline will be needlessly favored with long green times.
A solution is the placement of two pulse detectors per lane per approach spaced far enough apart to take a high-speed vehicle through the intersection. One detector is placed equal to the distance a vehicle takes to travel at the 85th percentile speed in eight seconds back from the stop bar, and the second detector is placed in a similar fashion five seconds back from the stop bar. Minimum gap is set to three seconds using gap reduction control.
This allows for a vehicle approaching the intersection at the 85th percentile speed to hit the first detector and extend their call for at least three seconds once gap reduction is finished. At the 85th percentile speed, they will hit the next detector five seconds away from the stop bar, and extend their call another three seconds. After this second extension, the vehicle is two seconds away from the stop bar. This is close enough to allow it to clear during the yellow and red intervals.
Any vehicles moving faster than the 85th percentile speed will hit the five second detector before the gap times out, and vehicles traveling slower will gap-out before reaching the five second detector. But this allows for a comfortable stopping distance at speeds below the 85th percentile speed. The "dilemma zone" is avoided with this strategy, except when the max green timer reaches zero. The green will be terminated regardless of the location of vehicles.
Table 902.5.3.4.3 is a reference for detector placement.
Table 902.5.3.4.3 Vehicular Distance Traveled
| Speed | Time, in seconds | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| mph | ft./sec | 1 | 5 | 8 | 10 | 15 | 20 | 25 | 30 | 35 | 40 | 45 | 50 | 55 | 60 |
| Distance Traveled in Feet | |||||||||||||||
| 1 | 1.5 | 1.5 | 7.3 | 11.7 | 15 | 22 | 29 | 37 | 44 | 51 | 59 | 66 | 73 | 81 | 88 |
| 2 | 2.9 | 2.9 | 15 | 23 | 29 | 44 | 59 | 73 | 88 | 103 | 117 | 132 | 147 | 161 | 176 |
| 3 | 4.4 | 4.4 | 22 | 35 | 44 | 66 | 88 | 110 | 132 | 154 | 176 | 198 | 220 | 242 | 264 |
| 4 | 5.9 | 5.9 | 29 | 47 | 59 | 88 | 117 | 147 | 176 | 205 | 235 | 264 | 293 | 323 | 352 |
| 5 | 7.3 | 7.3 | 37 | 59 | 73 | 110 | 147 | 183 | 220 | 257 | 293 | 330 | 367 | 403 | 440 |
| 10 | 14.7 | 15 | 73 | 117 | 147 | 220 | 293 | 367 | 440 | 513 | 587 | 660 | 733 | 807 | 880 |
| 15 | 22.0 | 22 | 110 | 176 | 220 | 330 | 440 | 550 | 660 | 770 | 880 | 990 | 1100 | 1210 | 1320 |
| 20 | 29.3 | 29 | 147 | 235 | 293 | 440 | 587 | 733 | 880 | 1027 | 1173 | 1320 | 1467 | 1613 | 1760 |
| 25 | 36.7 | 37 | 183 | 293 | 367 | 550 | 733 | 917 | 1100 | 1283 | 1467 | 1650 | 1833 | 2017 | 2200 |
| 30 | 44.0 | 44 | 220 | 352 | 440 | 660 | 880 | 1100 | 1320 | 1540 | 1760 | 1980 | 2200 | 2420 | 2640 |
| 35 | 51.3 | 51 | 257 | 411 | 513 | 770 | 1027 | 1283 | 1540 | 1797 | 2053 | 2310 | 2567 | 2823 | 3080 |
| 40 | 58.7 | 59 | 293 | 469 | 587 | 880 | 1173 | 1467 | 1760 | 2053 | 2347 | 2640 | 2933 | 3227 | 3520 |
| 45 | 66.0 | 66 | 330 | 528 | 660 | 990 | 1320 | 1650 | 1980 | 2310 | 2640 | 2970 | 3300 | 3630 | 3960 |
| 50 | 73.3 | 73 | 367 | 587 | 733 | 1100 | 1467 | 1833 | 2200 | 2567 | 2933 | 3300 | 3667 | 4033 | 4400 |
| 55 | 80.7 | 81 | 403 | 645 | 807 | 1210 | 1613 | 2017 | 2420 | 2823 | 3227 | 3630 | 4033 | 4437 | 4840 |
| 60 | 88.0 | 88 | 440 | 704 | 880 | 1320 | 1760 | 2200 | 2640 | 3080 | 3520 | 3960 | 4400 | 4840 | 5280 |
| 65 | 95.3 | 95 | 477 | 763 | 953 | 1430 | 1907 | 2383 | 2860 | 3337 | 3813 | 4290 | 4767 | 5243 | 5720 |
| 70 | 102.7 | 103 | 513 | 821 | 1027 | 1540 | 2053 | 2567 | 3080 | 3593 | 4107 | 4620 | 5133 | 5647 | 6160 |
| 75 | 110.0 | 110 | 550 | 880 | 1100 | 1650 | 2200 | 2750 | 3300 | 3850 | 4400 | 4950 | 5500 | 6050 | 6600 |
| 80 | 117.3 | 117 | 587 | 939 | 1173 | 1760 | 2347 | 2933 | 3520 | 4107 | 4693 | 5280 | 5867 | 6453 | 7040 |
| 85 | 124.7 | 125 | 623 | 997 | 1247 | 1870 | 2493 | 3117 | 3740 | 4363 | 4987 | 5610 | 6233 | 6857 | 7480 |
| 90 | 132.0 | 132 | 660 | 1056 | 1320 | 1980 | 2640 | 3300 | 3960 | 4620 | 5280 | 5940 | 6600 | 7260 | 7920 |
902.5.3.4.4 Delay and Extend Detector Settings
(A) Delay Settings. When a vehicle travels into the detection zone, the detector amplifier immediately receives the call. In some cases the call may not be needed immediately. A common situation is a dedicated right turn lane where free right turns are possible and stopping the opposing direction is usually not needed. A delay is programmed to keep the call from registering in the controller until a certain amount of time has passed. This time may be programmed in some detector amplifiers, or in the controller. After the programmed time has passed, the call is recognized by the controller.
Care must be taken as to where the delay time is programmed. If the delay time is set up in the detector amplifier, then every call going through that detection zone will be delayed. This will cause quick gap-outs if the delay time is near the gap time and no other normal detection is set up for that movement. If delay time is programmed in the controller, then the delay time is for all detectors in a movement, but delay is usually turned off when that movement is green. This will not allow for an immediate call in a lane where detection delay is needed when facing a red indication.
(B) Extend Settings. In other cases, the detector call may be needed longer than the time the vehicle is within the detection zone. An extension time can be programmed into either the amplifier or controller to hold the call for a certain period of time. Once the vehicle leaves the detection zone, the extend timer begins to countdown, and holds the call in until reaching zero. Common applications are where the detection zone is back from the stop bar, and the call is needed until the vehicle passes the stop bar. Back loops on pulse setting for dilemma zone prevention are another application. Extension time allows the call to stay on while the vehicle clears the intersection.
As with delay settings, care is to be taken as to what movements need extension timing. Extension timings set in an amplifier will extend every call regardless of that movements indications, while extension settings in the controller will usually extend only when that movement is green. However, all calls for that movement will be extended.
902.5.3.4.5 Locking and Non-Locking Detector Setting
When a signal is red for an actuated movement with no recall option, the vehicle detection is registered in the controller whenever a vehicle enters the detection zone. When the vehicle is allowed to leave the intersection before getting a green indication, usually on a right turn on red, it may not be necessary to call that movement if all vehicles have left the detection zone. The detector input for that movement can be set to "non-locking" in order to keep the call from stopping opposing directions. The movement will be served with green if a vehicle remains in the detection zone while set to non-locking. If the movement is set for "locking", then a call remains for that movement until it is served with a green indication, regardless of the presence of vehicles after the initial call.
Commonly, non-locking is used for dedicated right turn lanes, and protected-permitted left turn lanes. Locking is usually for through lanes and protected left turn lanes. Other situations, such as odd detection zone locations, may require a different locking technique.
902.5.3.4.6 Guidelines for Location of Closed Loop System Detectors
Closed Loop System Detectors: The objective of system detectors is to gather data that the system master uses to make decisions on timing plan and offset patterns (see 902.7, Coordination, for more information). The data from these detectors can also be used as a monitoring tool for the system. The primary difference between system detectors and standard detectors is that system detectors do not have direct control over signal phase times. The master uses data from the system detectors to make system wide decisions based on parameters set by the user.
Types of System Detectors:
(A) Volume and Occupancy (V + O) Detectors: These detectors are used to measure the amount of traffic at a point in the system. Several sets of V + O detectors can be used together to select system timing plans. They can also be used to select offsets by comparing the traffic flow in opposing directions.
(B) Queue Detectors: Queue detectors are used to identify congestion on a particular movement or approach and select an appropriate timing plan.
(C) Speed Trap Detectors: Speed trap detectors are pairs of detectors used to monitor traffic speeds. The speed data from these detectors is typically not used to select timing plans however can be a measure of system performance.
(D) Intersection Detectors: Intersection detectors can be used simultaneously as system detectors and as standard phase detectors. These detectors are typically used to select timing plans, however they can also be used as V + O or queue detectors as described above. These detectors allow the system to respond to changes in side street demand. During times of low side street demand, the system can maximize the green band for mainline. During times of high side street demand the system can prevent cycle failures and congestion on the side street. Stop bar detectors (i.e. 6 x 30 loops) require special dual output detector amplifiers to operate this way.
Determining System Detector Locations: Successful operation of a traffic responsive closed loop system is dependent on well placed system detectors. The following offers guidelines for placement of system detectors.
Planning the System: It is advantageous to have a drawing or sketch showing the entire system in the form of plan sheets, aerial photos or a good sketch. On this you will want to show the signalized intersections in the system and the spacing between them. It is helpful to show on the layout the traffic volumes on mainline and ramp approaches entering the system, significant side road approaches and significant turning movements. Intermediate mainline volumes can also be helpful, especially for larger systems.
For systems with more than a few intersections, the system can be divided into zones. The dividing line between zones would be interchanges or major intersections that cause a significant change in traffic on mainline. Zones typically have at least 3 or 4 signalized intersections.
Identify any potential queuing problems. This may be closely spaced intersections or high volume turning movements that exceed their storage space. These can be determined by field observation, traffic modeling or using queuing formulas from 233, At-Grade Intersections.
Volume and Occupancy (V + O) Detectors: At least two sets of V + O detectors are needed, one for each mainline direction. If the system is divided into zones, at least two sets of V + O detectors are to be provided for each zone.
These detectors are placed where they can monitor free flow conditions. They are typically installed just downstream from an intersection. There is to be adequate spacing between intersections so that traffic will not queue over the system detectors. Avoid placing system detectors in front of driveways or unsignalized side roads.
Advanced detectors can also be used as V + O detectors. These may be dilemma zone detectors, ramp detectors or advanced detection on major side roads. See Figures 1 and 2 for examples. Advanced V + O detection can be useful for small systems with closely spaced intersections or for side road approaches that contribute a large percentage of system traffic. Advanced ramp system detectors are recommended where the ramp accounts for a significant amount of traffic entering the system.
Queue Detectors: Queue detectors are located on approaches or auxiliary lanes where a potential queuing problem exists. Queue detectors are to be installed in the affected lanes in advance of the maximum allowable queue to allow the system time to respond to the queue. For instance, queue detectors in a 400 ft. turn lane could be installed 300 ft. back from the stop bar.
For long turn lanes or approaches where long queues are expected, two queue detectors or two sets of queue detectors can help the system be more responsive.
Speed Trap Detectors: Speed trap detectors are two detectors in the same lane spaced a certain distance apart. The system manufacturer specifies the distance. For plans development, it is recommended that the exact distance not be specified, but referenced to manufacturers recommendations. A distance of 10 ft. from head to head of the detectors can be used for drawings.
Speed trap detectors are usually installed at the same location as V + O Detectors. One of the V + O detectors can serve as a speed trap detector. One speed trap pair is usually adequate for each set of V + O detectors. For multi-lane approaches, the inside or middle lane is best for monitoring speeds, especially where there are mid-block driveways or unsignalized intersections.
Intersection Detectors: Intersection detectors at intersections with the highest side street approach volumes are typically set as system detectors. The detectors would be installed as standard intersection detectors. For long detectors (i.e. 6 x 30 detectors), dual output detector amplifiers are used to provide this operation. One output provides presence detection and the second output provides point detection for the system. For point detectors (i.e. 6 x 6 detectors) the controllers can typically address these as both a system detector and a local detector.
902.5.3.5 Detector Call Switching
Most modern actuated controllers allow for detector calls to be transferred to phases other than the ones assigned. This option is very useful on approaches with protected-permitted left turn phasing and detection on a left turn lane and the opposing through lanes.
Take a standard 4-way intersection with phase 1 a northbound left turn, phase 6 the northbound through, phase 2 southbound through, and standard ring structure:
Phase 1 is a protected-permitted left turn. After phase 1 gaps or maxes out, phase 2 goes green along with phase 6 until they both max or gap-out. Without detector switching, a vehicle waiting to make a yielding northbound left turn would not be detected and would be susceptible to gap-outs caused by phase 6 detectors, even though time is left in the max timer for a yielding left turn. When detector switching is programmed for phase 1 calls to be transferred to phase 6, the vehicle waiting on the phase 1 detector is then placing a call on phase 6 once phase 1 goes yellow, and continues to call phase 6 until leaving the detector or reaching phase 6's max time. This allows extension of the ball green for the yielding left turn. Once phase 6 goes yellow, the phase 1 detection returns to phase 1 and allows the protected left turn to come up next cycle.
Advantages to this setting are the reduction of the "pinballing" or quick changes of phases late at night with sporadic traffic, and the reduction of yielding left-turn conflicts by extending their green time. This setting can also help prevent a yellow trap in the absence of side street calls. Care is to be taken when using this setting at a coordinated intersection. Detector switching will have little effect when used with coordinated phases, since mainline green time is usually unaffected by detection. It can be used on non-coordinated phases to prevent early returns to the mainline direction by extending the yielding green time.
902.5.3.6 Phase Re-Service
Phase re-service, or conditional re-service, allows for the standard phase sequence to reverse and display green for non-conflicting directions that have already been served. The most common application is re-serving protected left turns when the opposing throughs gap-out.
Take a standard actuated 4-way intersection with phase 1 a northbound LT, phase 6 the northbound through, phase 2 southbound through, and standard ring structure:
In this case, the phase 1 indication is a protected-only left turn. After phase 1 is served, phase 2 begins its green time along with phase 6. Re-service of the odd-numbered phase is allowed under these conditions:
1. The even phase in the same ring (phase 2 in this example) has gapped out and is resting in green.
2. There is a call across the ring barrier to another phase (a side street call for this example).
3. The even phase in the opposite ring is still extending and there is enough time left in its max timer. This time must be equal to or greater than the re-serviced phase's minimum green time plus opposing through phase yellow and all-red time.
During the re-service period, gap control is timed by phase 6 detectors and not phase 1 detectors. Therefore, if the northbound through gaps out with demand still present on the northbound left turn, both directions will terminate together.
Advantages to this setting are in reducing the delay for re-serviced left turns. Again, care is to be taken when using this setting at a coordinated intersection. Re-service will not be possible when used with coordinated phases, and available time to re-service side street phases will be almost non-existent. This setting works best at isolated intersections.
902.5.3.7 Pedestrian Intervals
If pedestrian indications are used to provide WALK and DON'T WALK messages, they must conform to the MUTCD. Careful balance must be decided between the time required for the pedestrian to cross and the time required for vehicles. A pedestrian crossing a large arterial with a minor side street approach will likely need more than the given time for side street vehicles to cross. This results in increased delay for the major approach. It is preferred the pedestrian indication be push-button actuated. See 902.3.4.4, Pedestrian Control, for more information on pedestrian control.
902.5.3.7.1 Pedestrian Indication Considerations
In some circumstances, the engineer can find it difficult to time for pedestrians. On one side is the duty to consider the time needed to allow pedestrians of all travel speeds to cross wide roadways. On the other side is the responsibility to operate busy arterials to their peak capacity by minimizing stops and delay for the vehicles. These two goals are usually in conflict if every side street green must be timed long enough to accommodate pedestrians. In situations where pre-timed controllers are used, consideration must be given to determine whether the pedestrian indications will result in an overall negative impact. If the pedestrian usage at these intersections is minimal, then other methods other than pedestrian indications are to be explored. Options include converting to an actuated controller or providing indications at a nearby intersection that can better accommodate the time required and equipment necessary for a proper crossing.
Confusion is common among pedestrians as to the meaning of the indications. They tend to associate the WALK indications with a "green" meaning, and the flashing DON'T WALK with a yellow indication. This results in a pedestrian turning back once the WALK time expires and not using the flashing DON'T WALK to cross. Every effort is to be made to educate the public to the meanings of these indications instead of lengthening the WALK time, since the flashing DON'T WALK time is the critical time. Sign R10-3b provides a detailed explanation of the indications and is to be used at all new crossings and other crossings where confusion is prevalent.
902.5.3.7.2 Walk and Pedestrian Clearance Intervals
The WALK interval is to be at least four to seven seconds, with four seconds adequate when fewer than 10 pedestrians per cycle are expected. The four-second minimum allows a pedestrian time to react to the change in signals and begin travel.
Flashing DON’T WALK or flashing DON’T WALK plus Vehicle Change Period indicates pedestrian clearance time. The flashing DON'T WALK display is always used whenever pedestrian indications are provided. Pedestrian clearance setting is to be equal to the time required for the pedestrian to leave the curb and travel to the far side of the farthest travel lane on the other side of the intersection. Typically, a rate of 4 ft. per second is used to calculate this time, but research from the MUTCD Traffic Control Devices Handbook (TCDH) verifies that one-third of all pedestrians cross streets at a rate slower than 4 fps and 15 percent walk at or below 3.5 fps. The timing for clearance intervals is to conform to areas that serve segments of the population with slower walking speeds by taking the slower rate into account. Engineering studies and careful judgment is to be used to obtain the proper clearance times for each situation.
There are two methods available to time out the pedestrian clearance before opposing vehicles receive a green indication. Either method may be used, but the primary objective in selecting a method is to be consistency between adjacent intersections. Confusion may result if the two methods are each used along an arterial operated by MoDOT or other agencies.
(A) Method 1: End of Yellow. The flashing DON'T WALK indication is terminated at the end of the yellow(1) indication for the associated vehicle phase.
(1) Pedestrian clearance may also include all-red vehicle clearance, if used.
(B) Method 2: End of Green: The flashing DON'T WALK indication is terminated at the end of the green indication for the associated vehicle phase. The flashing DON'T WALK is displayed only during the green portion of the pedestrian clearance time. The pedestrian clearance time is the total of the flashing DON'T WALK and yellow(1) time.
(1) Pedestrian clearance may also include all-red vehicle clearance, if used.
Either method may be used at actuated or pre-timed intersections. However, for actuated approaches where maximum green for force off time is not dictated by the ped timing, Method 2 is preferred. It is possible some actuated controllers may only allow Method 2.
902.5.4 Preemption
It is often necessary to interrupt the normal operation of a traffic signal or a group of traffic signals to facilitate the clearance of traffic that might be backed up onto an active railroad track or to facilitate the movement of emergency vehicles. This is accomplished by the activation of an external input into the controller, which causes the controller(s) to run a special pre-determined routine in order to best clear the intersection for the situation. After the special routine has completed, the signal returns to normal operation.
902.5.4.1 Railroad Preemption
When a train approaches a crossing connected by a preemption output to a signalized intersection, a controller begins a routine to accomplish the following in this order:
1. Cut off traffic flow to the track crossing.
2. Clear out vehicles between traffic signal and track crossing.
3. While preemption input is active, keep traffic from approaching the track crossing while permitting other movements headed away from the track to continue to cycle.
4. After input has dropped, the movements stopped because of the track crossing is to be resumed to remove queued traffic as much as possible before returning to normal operation of the signal.
The following 8-phase intersection with a railroad crossing is used as an example:
Upon preemption input, the signal will terminate any movement other than southbound through and left (phases 1 & 6). This may be done with a special minimum green that is less than the normal programmed times. The yellow change interval, and any red clearance interval that follows, shall not be shortened or omitted.
If or after these movements have terminated with a red indication, the controller will time out the southbound through and left (phases 1 & 6) for a pre-determined time in order to clear out vehicles which may be left on the tracks. The amount of time given to this movement is critical, and must be sufficient for the slowest of vehicles to get away from the tracks.
After the southbound direction has cleared, and while the preemption input is running, it is allowable for the westbound through and left (phases 3 & 8), eastbound through (phase 4), and northbound protected left (phase 5) to cycle. The protected eastbound left movement cannot be brought up (phase 7). If phase 7 is a protected/permissive left turn, a red indication must be displayed for the movement. This will require special wiring and cabinet configuration to separate the ball indications in the phase 7 head. This special cycle allows for the side street to move along since their movements are not in conflict with the tracks.
After the preemption input has dropped off, northbound and southbound through directions (phases 2 & 6) is to be favored with a green indication for a pre-determined time before resuming normal signal operation.
If the distance between the tracks and the intersection is extremely close, measures to prevent any movements from approaching the tracks is to be explored. If a pre-signal is installed at an interconnected highway-rail grade crossing near a signalized intersection, a STOP HERE ON RED (R10-6) sign shall be installed near the pre-signal or at the stop line if used. If there is a nearby signalized intersection with insufficient clear storage distance for a design vehicle, or the highway-rail grade crossing does not have gates, a NO TURN ON RED (R10-11) sign shall be installed for the approach that crosses the railroad track.
Different controllers will call these special movements with different names. There is no consistent standard for these settings. Care is to be taken to thoroughly review a controller manual to make sure the programmer has positively identified which setting will accomplish the best preemption routine. Thorough testing, both on the shop bench and at the intersection, is a must before leaving the controller to operate the intersection. Testing is to also be done as part of the yearly preventative maintenance schedule.
902.5.4.2 Emergency Vehicle Preemption
Emergency vehicle entrance to a road can be accomplished with a signal at the driveway of the emergency institution, and a special routine at any nearby traffic signals. The signal at the emergency driveway rests in a yellow flash for the main direction, and cycles to solid yellow, then red when the preemption input is activated.
If there are signals near the emergency driveway connected to the preemption input, the signals begin a special routine to facilitate the rapid exit for the emergency vehicle. The routine accomplishes the following in this order:
1. Cut-off movements not critical to the path of the emergency vehicle.
2. Dwell in green for directions critical to the path of the emergency vehicle to allow for all emergency vehicles to clear the intersection.
3. Movements stopped to allow passage of the emergency vehicles is to be started up after preemption routine ends to remove queued traffic as much as possible before resuming normal operation of the signal.
The following 8-phase intersection with a nearby emergency vehicle entrance is used as an example:
Upon preemption input, the emergency driveway signal goes from flashing yellow to solid yellow to solid red. The indications for the driveway will then go ball green. At the controller north of the driveway, the controller begins its preemption routine. The first item to determine is whether to delay the start of the routine. If the emergency vehicle driveway is too far away from the signalized intersection, the routine may expire before the emergency vehicles have cleared the intersection. This delay time must be determined in order to achieve proper operation.
When the preemption routine begins, the signal will terminate any movement other than northbound through and left. This may be done with a special minimum green that is less than the normal programmed times. The yellow change interval, and any red clearance interval that follows, shall not be shortened or omitted.
After other movements have terminated with a red indication, the controller will dwell in northbound left and through green for a time that allows for northbound vehicles between the emergency driveway and the signal to clear out before the arrival of emergency vehicles. It will remain in green for these directions to allow for emergency vehicles to clear the intersection.
After the northbound dwell has timed out, the heavy directions that were stopped is to be brought up with a green indication for a pre-determined time before resuming normal signal operation.
As with Railroad Preemption, different controllers will call these special movements with different names. There is no consistent standard for these settings. Care is to be taken to thoroughly review a controller manual to make sure the programmer has positively identified which setting will accomplish the best preemption routine. Thorough testing, both on the shop bench and at the intersection, is a must before leaving the controller to operate the intersection. Regular testing is to also be done as part of the yearly preventative maintenance schedule.
902.5.5 All-Direction Flashing Operation
There are two reasons for a signal to be in flashing operation for all directions: to reduce the level of control when traffic volume is low and to provide the safest method of control when a signal is inoperative. The benefits of all-direction flashing operation include:
Reducing stops and needless delay to mainline traffic.
Reduce delay to side street traffic when mainline traffic is light.
Less fuel consumption due to fewer stops and delays.
Electrical energy costs reduced by 50-65%.
Programmed all-direction flash operations at non-factory or non-school signals are to be done once a day. It is not advisable to switch back and forth between all-direction flashing and cycling operation during the day, although in special instances this may be appropriate.
902.5.5.1 Yellow/Red vs. Red/Red Flash
Generally, yellow/red flash is used for intersections where traffic volumes during flashing operation greatly favor one direction. Yellow indications are flashed for the major direction and red flashed for the minor direction.
Red/red all-direction flashing operation is used for intersections where volumes during flashing operation on all approaches are approximately equal. Complex intersections (5-leg, large widths, multiple dual left turn approaches) will likely require red/red flash, regardless of volumes. When a system of interconnected signals is flashed, care is to be taken to be consistent in flash color for the major direction so as not to violate driver expectancy. A red/red flashing signal in the middle of a system of yellow/red flashing signals may lead to violations of the flashing red along the mainline due to the expectancy of yellow flashing indications.
902.5.5.2 Light Traffic Volumes
There is no magic number that can be derived for deciding when to implement all-direction flashing operation and when to cycle normally for light volumes. It is acceptable to use 50% of the warranted volumes to decide when to begin and end the all-direction flashing operation time.
Several other factors are to be used with good engineering judgment when making the decision to implement all-direction flashing operation. Among these is crash history and sight distance. If correctable crashes were a significant warrant for installation, then all-direction flashing operation may not be acceptable at any time. If sight distance is restricted for an approach below acceptable AASHTO standards, then no flashing operation is to be considered. If no all-direction flashing operation will be used, then at a minimum semi-actuated operation, and preferably fully actuated operation, is to be used during times of light traffic volumes.
902.5.5.3 Signal Malfunctions and "Soft" Flash
Fail-safe operation for a malfunctioning signals leads to all-direction flashing operation when tripped by a conflict monitor. Generally, this means red/red for fully actuated intersections and yellow/red for semi-actuated or pre-timed intersections. Cabinet wiring determines all-direction flashing operation when malfunctioning. Many modern controllers allow for programmed flash colors to differ from malfunctioning flash colors. This "soft" flash can be set in the controller to allow for yellow/red all-direction flash during light traffic demands. This replaces cabinet wiring, which will in the event of a malfunction go to red/red operation, regardless of controller flashing operation.
Flexibility of this type is useful at intersections where yellow/red all-direction flashing operations for malfunctions during the peak traffic periods will cause excessive delays for the side street, or red/red operation at night would needlessly delay traffic. Where this option is not available, careful consideration is to be given to covering a worst-case scenario of programmed all-direction flash vs. malfunction all-direction flash, and cabinet wiring is to be set up to match the preferred situation.
902.5.5.4 Fully Actuated All-Direction Flash
Consideration is to be given to all-direction flashing isolated fully actuated signals yellow/red where high-speed approaches are involved. It has been common to leave a fully actuated signal operating free at all times. However, if sight distance is adequate and mainline volumes are light, then yellow/red all-direction flash may reduce the chances for crashes since mainline traffic will not have to stop for a single vehicle on an opposing approach.
Fully actuated signals within a coordinated system are to be analyzed for possible free operation when other signals within the system are on all-direction flash. It may be more beneficial to all-direction flash the fully actuated signal yellow/red with the other signals in order to be consistent with the rest of the system.
When malfunctioning loops are present at a fully actuated intersection, and normal programming calls for free operation, it may be desirable to go to all-direction flashing operation if the loop cannot be replaced soon. Recall operation for the phases with the bad loops may not provide the best form of control, since the recalled phases will be serviced regardless of vehicle presence.
902.5.5.5 Flashing Indications
In accordance with the MUTCD, all signal faces on an approach shall flash the same color, except that separate signal faces for turning lanes may be flashed a different color. When a signal face consists of all-arrow indications other than red, or when a signal face contains no circular indication of the color that is to be flashed, the appropriate arrow indication shall be flashed.
For dual or protected-only left turns programmed to flash, the indications are to flash yellow arrow when the purpose of the phasing was high volume. When the phasing is in place for safety reasons, the indications for the left are to flash red.
When a signal face includes both circular and arrow indications of the color that is to be flashed, only the circular indication of that color shall be flashed. A common application of this is a five-section head with a yellow arrow and circular yellow. The circular yellow is to be flashed and the yellow arrow remain dark. For a signal head with a flashing yellow arrow indication, the flashing yellow arrow is to be dark.
Whether the signal flashes by cabinet wiring or controller programming, care is to be taken to not flash all indications at the same time. This causes a severe load fluctuation that reduces the life of the electrical equipment and can lead to malfunctions when the signal is in operation.