[Top: John S. Allen's Home Page]
[Up: bicycling articles]

[contact John S. Allen by e-mail]

Site logo - Track bike (1.3 KB GIF)

Traffic Signal Actuators: Am I Paranoid?

1993, 1997, 2003, John S. Allen

Running red lights is a losing game for bicyclists -- not only because it is hazardous but even more so because of the anger it arouses against all bicyclists.

However, the problem of bicyclists' running red lights has an additional and even more troubling dimension. If I told you that bicyclists are being psychologically programmed by evil, unseen forces to run red lights, would you think that I am paranoid? Well, I'm not. No, really, not. Let me explain.

Until about 1970, most traffic lights were actuated by timers, and a few by treadle devices in the road. These were visible and obvious, and they worked for bicycles. But more and more traffic lights are now actuated by loops of electrical wire buried in the pavement. The wire loops operate as metal detector antennas, sensing the presence of vehicles overhead. You must have seen the narrow lines in square patterns cut into the pavement ahead of intersections. That's where the wires are buried.

The photo above is of the intersection of Housatonic Street, in Lenox, Massachusetts, with the busy, four-lane Route 7 bypass. You can see the actuator loop cuts in the pavement. The group of touring bicyclists has been waiting for a long time for a traffic light which would not change. A car has just driven up to the intersection. Its brake lights and left turn signal are on as it waits for the light to change.

The photo shows two problems:

  1. The actuator clearly did not work for the bicyclists. They would have ridden over the actuator loop in the foreground to get to where they are. That loop should have triggered the signal.

  2. The bicyclists don't understand the actuator. If they did, they would have positioned their bicycles along the edge of the actuator loops nearest the intersection to send the strongest possible signal.

In this photo, taken a few seconds later, the car has tripped the actuator, and is making its left turn. The bicyclists and a second car also are crossing the intersection.

Do you still think that I am paranoid now? Should bicyclists have to wait for a car to show up before they get a green traffic signal? Or should they risk running the red light when a car doesn't happen to show up to trip the signal?

Not only do these actuator loops work very unreliably for bicyclists, most bicyclists do not even know what they are. Often, the loops are buried under a new layer of pavement and invisible. If you don't understand the actuator loops, all you know is that you have been waiting for the traffic light since Tuesday and it hasn't turned green. If this happens at a substantial percentage of traffic lights, a reasonable human being will draw the logical conclusion: why bother to wait at all?

The traffic law of all 50 states grants bicyclists the rights and duties of vehicle drivers. It follows directly that the traffic control system must make it possible for bicyclists to obey the law, or else a state is acting in contempt of its own legal authority. Yet the Institute of Traffic Engineers specified the defective actuators, and the state highway departments adopted them, with the full knowledge that bicycles are legal roadway traffic. What was the underlying attitude here? "They all run the lights anyway"? "Maybe if we make it hard enough for them they'll go away"? "Who cares"? Whatever it was, it does not amount to responsible, professional public policy.

The real tragedy is that bicyclists are being injured and killed, and public outrage against bicyclists is being created because of this policy, while the police, the grieving survivors and the public all unknowingly pin the blame on the bicyclists who ran the red lights. Cries go up that we must restrict bicycling because it is too dangerous and bicyclists are too lawless. Motorists treat us with contempt because they perceive us as scofflaws.

So, what should we do?

First: we can try to make the signals work. If the pavement cuts are visible, you can often trip the signal by riding along one side of the loop with your wheels directly over the wire (but if your bicycle has nonmetallic wheels and a nonmetallic frame, good luck, Charlie). If a car is waiting or is about to come up behind you, it will trip the signal. If you're alone and trying to get across a busy street, try laying your bicycle flat on the pavement with the wheels on the wire at one side.

Often, more than one entrance to an intersection is programmed to get a green light at the same time, and sometimes certain entrances (usually, the ones with heavier traffic) will automatically get a green light; so it is worth the trouble to learn the actumrk.gif (1590 bytes)signal sequence, at least for intersections you ride through frequently. Then you will know when you actually have to run the red light. You may also use the pedestrian push-button -- if there is one and if you can put up with the delay and inconvenience. You can take some ironic satisfaction that a pedestrian phase that would otherwise be unnecessary delays the cars in the cross street too.

Putting an end to the actuator problem is a job for bicyclists' organizations: put your local and state authorities on notice that bicycle-insensitive traffic signal actuators are defective and illegal. Be friendly and cooperative, but firm. You are not asking for anything which is either difficult or expensive to implement. A simple solution is to turn up the sensitivity at the signal control box until the actuator will detect a bicycle traveling along the loop wire, and to mark the location wire with a standard symbol (image, right). A technically better solution is also to lay the wire in a different pattern called a "quadrupole" or "California D-type" loop, which is much more sensitive directly overhead while avoiding false triggering from adjacent lanes. These solutions are widely used in enlightened West Coast communities.


Quadrupole loop, Port Townsend, Washington, USA (1989)
senses a bicycle positioned over its center wires. (This loop is
less than optimal, because its great length reduces its
sensitivity -- nonetheless it works.)

8707N05R14Quadrupole closeupa.jpg (39310 bytes)


I thank my friend Sheldon Brown for an inspired idea for an exercise in civil obedience. It's similar in principle to a work-to-rules job action.You gather a group of bicyclists to wait for a traffic light to change, while cars back up behind you. When a police officer appears to wave you through the defective light, wait at the next intersection. Or parade across on the pedestrian signal, which causes almost as much delay and looks even sillier.

For those of us with carbon-fiber composite bikes: metal cranks or rims should be enough to trigger the better actuators. But the long-term answer might include an industry-standardized loop of thin wire built into the bicycle's wheels or frame, or an active electronic device. If the traffic engineers had taken a responsible approach toward bicyclists, they would have foreseen the problem and worked with the bicycle industry to develop a standard solution. After all, animals and wooden wagons also are legal traffic on most roads and they're nonmetallic too -- and before too many years, many motor vehicles will have nonmetallic composite bodies and ceramic engines. We'll have a lot of traffic light problems then!

An additional , though depressing approach: make it known to your close friends and relatives that if you are killed or incapacitated while running a red light, your survivors are to contact a lawyer and consider suing the responsible parties. The Institute of Traffic Engineers has been on notice about the actuator problem since 1982, when it was sued over the death of a cyclist who was forced to run a red light. The legal status of bicycle traffic and other traffic with small metal mass has never been in question. As the saying goes, eternal vigilance is the price of liberty: there's nothing like a good, stiff lawsuit to bring irresponsible bureaucrats and Neanderthal engineers to their senses.

And a final caution: all in all, you will find that the actuator problem will cause you serious delays at only a minority of intersections, and usually late at night when there is little other traffic to trip the signals. Do not use the problem as an excuse for lawlessness. That only makes bicyclists' situation worse. Rather, understand that the actuator problem, because it involves outright, inexcusable illegality, is a legal crowbar which can work to bring the traffic engineers to respect bicyclists, just as the struggle for Constitutional voting rights forced the nation to respect African-Americans in the 1960's.

Be optimistic, but work hard. If the lobby for people with disabilities could enforce the installation of sidewalk ramps at every corner throughout this nation, we should be able to get the actuator specification changed.

Worker with diamond saw,
cutting pavement slot for actuator loop

diamswsm.jpg (15050 bytes)

Update, January 2005 -- video detection.

Comments by Eli Mowbray of the City of Santa Cruz, California Public Works Department:

I came across your article on bicycle detection while researching a reply to a local newspaper (I work as the Engineering Associate in the Traffic Engineering Division for the City of Santa Cruz Public Works Department.)

Thought you might be interested in our success using Video Detection at signalized intersections. These detectors have no problem finding bicycles thus they provide an excellent solution to the bike detection issue. They use cameras and pixel analysis to infer the presence of vehicles or bikes graphically (when enough pixels change quickly enough.) They work great at night so long as the bicycle rider has a light (as required by California law.)

What may be most crucial: Video Detectors are a "good sell" to the traditional road building types (aka - people who seem to be against bicyclists and only for cars.)

  1. Inductive loops are saw-cut and permanently installed in the road; this is a maintenance headache for the road and the traffic signal. (It lessens the useful life of both as water inevitably seeps through.)

  2. Loops also need to be replaced whenever the road is reconstructed (or even overpaved); this increases the cost to improve roads thus limiting the amount of new actuator control that is placed.

  3. Traffic engineers like their flexibility as it helps lower costs to shift lanes around: cameras can simply be reprogrammed for a new detection area and we avoid the expense of loop replacement.)

Video detection or camera type systems can therefore save money in the long run even though they cost more than loops at first.

About the Massachusetts specification for bicycle detection

Links to other articles on this topic...


[Top: John S. Allen's Home Page]
[Up: bicycling articles]

[contact John S. Allen by e-mail]

Contents 1993, 1997 John S. Allen

Last revised 19 August 2010