Traffic Signal Timing

With all this talk about yellow lights and whatnot, I thought it would be a good time to review the how’s and why’s of Traffic signal timing, and then on to how it affects cyclists because this is a distinct problem. On the surface it’s simple, green, red, yellow but like everything else there’s more to the story.

The law on red lights

In brief, the law says a driver must not enter the intersection when the light is red.

As usual, the first stop is a review of the Arizona Revised Statutes. Arizona law does not follow the UVC with regard to where a red violation occurs. Here are the relevant Arizona statutes:

  • §28-645 Traffic Control Signal Legend: “3. Red indication…traffic facing a steady red signal alone shall stop before entering the intersection“
  • §28-601 “8. ‘Intersection’ means the area embraced within the prolongation or connection of the lateral curb lines” — this is called the “enforcement line”

So, the enforcement line is well forward of what many people perceive as the start of the intersection.

Curiosly (well, curious to me anyway) is that unlike at stop sign intersections, drivers are under no obligation to stop before entering the crosswalk at a signalized intersection.

Oh, and by the way; irrespective of signal timing, drivers must not enter an intersection even if the light is green, until the intersection has cleared of any/all traffic from a previous cycle: “Vehicular traffic facing a green…  shall yield the right-of-way to other vehicles and to pedestrians…”. [clarification/update: I should not have said “not enter”, that’s too strong; what i meant was drivers with a green must yield to anyone still in the intersection left over from the last cycle]

Tucson problems: as usual, there are special problems in Tuscon. Sigh. Apparently they use trailing green arrows. So, i guess this precipitates a need for a yellow arrow; this apparently leads to a lot of consternation with photo-enforced intersection with arrows; see e.g. this news item. This coupled with some oddly-shaped intersections causes much grief, e.g. River and Oracle.

Photo enforcement and the new law

Now, a sidenote; It’s sad, but true that something like traffic signal timing becomes a political football, particularly with respect to the duration of the yellow phase.

This tidbit from Phoenix will keep using traffic camera, The Arizona Republic, May 14, 2010: “(Phoenix police officer Bill) Fisher and others at the Phoenix Police Department said the city’s yellow-light timing is set at four seconds or longer, one second longer than the federal standard of three seconds, to provide motorists with additional time to stop on red” — is incorrect; there are definitely 3-second yellows in Phoenix. See, for example 44th and Chandler Blvd, or 46th and Ray Road.

In any event, our legislators (who consistently maintain that we already have enough laws) enacted a law that mandates that the engineering standard be modified to include a provision that the minimum yellow cycle be 3 seconds. Having legislation dictate engineering standards seems like a bad idea — but there you have it. The engineering standard is (already) specified by §28-641 is the MUTCD with Arizona supplements, and is maintained by ADOT.

A new provision of the law adds that any photo-red enforcement must meet engineering standard for yellow duration at that signal. So, if you get an officer-issued ticket at an out-of-spec signal, I guess you are out of luck.

A provision that would have provided a mandatory 1-second “grace period” to phot0-red was dropped.

[news item about new law] [another news item about new law]

HB2338 (select 49th Legislature, 2nd Regular session), here is the enacted law, chapter 213.

I don’t know of any intersection that doesn’t already meet the requirements of the new law — therefore I would tend to believe that the new law was more a statement by anti-photo-enforcement partisans than about improving safety.

The Engineering Story

Here is the section of the ADOT manual with calculations; section 621 Signal Phase Change Intervals.

There is a calculated minimum green time; but that’s based on how many cars are expected to be waiting to go. In effect, the minimum is practically zero.

The main attraction is calculation of yellow. Note that the formula takes into account the approach speed of traffic, and that figures for deceleration and reaction-time are taken to be conservative constants.

The design of the formula gives legally operating (e.g. not speeding, not so impaired their reaction time is longer than normal) time to stop during the yellow interval, thus avoiding a violation of 28-655.

I would point out, that the “approach speed” is a huge bone of contention with certain groups. The speed variable is supposed to be the 85th percentile speed, however “The posted speed limit may be assumed to be the approach speed when an engineering study or 85th percentile speed data is not available.”

Calculation of the all-red phase is discussed below.

Impact of timing on cyclists

Cyclists will often complain that, at the minimums, there isn’t enough time to cross before the light changes to red; or worse still, before the conflicting traffic gets a green.

To deal with this, the yellow phase should not be lengthened, “Excessively long yellow vehicle change intervals may encourage driver disrespect and unsafe operating practices”. It also doesn’t fix anything, cyclists will just enter the intersection later.

Rather the only guaranteed-safe thing to do is to lengthen the all-red phase.

The calculation of the all-red phase depends on the width of the intersection (plus 20 feet as a safety margin), and the posted speed limit. So for crossing an typical wide-ish arterial road (80 feet), at 40mph posted it comes out to be, say, 2 seconds. However for a cyclist traveling at, say, 15mph, it would almost 4 seconds without any margin of safety added.

[more goes here]

5 thoughts on “Traffic Signal Timing”

  1. Lots of anti-cyclist sentiment from a few traffic engineers.

    http://www.iterisprojects.com/octec/news/OCTEC%20Oct%202009%20(final).pdf

    Looming Debate on AB 1581
    Panelists gather at OCTEC [Orange County Traffic Engineers Council]
    meeting to discuss the text of AB 1581, what is wrong with the bill’s
    wording, and how to fix it.

    The ramifications of Assembly Bill 1581 (AB 1581) will be felt across
    California. The bill and the directives involved are subjects of this
    month’s [October 22] OCTEC meeting. According to Ron Keith from OCTA,
    the goal “is to educate the traffic engineering community to a directive
    from Caltrans regarding bicycle detection and timing mandated by AB 1581.”

    He continues: “The directive gives guidance as a SHALL to install 6′ x
    6′ limit line hot loops in all lanes to detect a
    non ferrous bicycle with a person weighing minimum 90 lbs. and 4 feet in
    height. The directive gives a timing scale depending on the width of the
    intersection and the yellow and all red time. The directive makes the
    minimum initial for a phase equivalent to clear a bicycle whether one is
    present or not unless there is a way to discriminate a bicyclist from a
    motor vehicle. This will result in severe increases in Lost Time and
    cause unnecessary delays and increases in emissions and GHG.

    “The objective is to get the directive rescinded and/or amend the bill.
    We also want request that a legal investigation and report from Caltrans
    or the State be made to determine the meaning and the intent of the
    legislature when in the bill it says ‘… detection and related timings’.
    We do not feel the intent was to time signals adversely for the motoring
    public, undo what we have spent millions of dollars in equipment and
    retiming efforts all over the state to reduce delay and GHG, and finally
    with size and weight of the person in the directive, we are now timing
    signals for adolescents or children who do not have the mental capacity
    or experience to make competent decisions when riding a bicycle.”

Comments are closed.