Saturday, November 5, 2016

Tripping the Light Fantastic

Many cyclists have probably encountered this scenario: You come to a stoplight and wait for the signal to turn green. You wait. And wait. And wait. And . . . nothing. The light refuses to change. If this happens, chances are, you have encountered a demand-actuated vehicle detector - and the sensors, which are primarily designed to detect cars, are not sensitive enough to detect your bicycle. There are a few myths about these detectors out there, and lots of different ideas regarding what cyclist should do when they are unable to trip the lights.

First off - what exactly are these things, and how do they work?

Look for those cuts in the pavement.
Obviously, these demand-actuated devices are supposed to detect a vehicle that approaches the intersection, and send a signal to the traffic light to change. No vehicle, no change. Though some people assume that the detectors are tripped by vehicle weight, and others think it has to do with magnetism, neither is exactly right. The most common type of vehicle detectors today work by using an inductive loop which is basically a large metal detector set into the pavement. The inductive loop creates a faint electromagnetic field, and a vehicle entering the loop creates a small disturbance in that field and should trip the signal. Obviously, cars are much larger, and have a lot more metal on them so they are nearly always detected, but the detectors should react to any kind of metal, including aluminum, even though aluminum is not magnetic. These sensors are supposed to be able to detect bicycles, but depending on the sensitivity, both bicycles and motorcycles can occasionally have difficulty tripping the lights.

Searching the internet, you can find people who swear by attaching magnets to their pedals or their bottom brackets to help them be detected. Based on what I've read from engineers who design and work with the systems, I'd have to conclude that this is a persistent myth, or possibly a cyclist would have to be carrying a lot of magnets (as in, several pounds worth) on their bike to actually make a difference. In any case, we retrogrouches are at an advantage in that there should be enough metal in a steel-framed bike with aluminum-rimmed wheels to trip many vehicle detectors. That can't be said of carbon fiber bikes with carbon wheels and other components. There's a very informative article about the technical aspects of inductive loop detectors by Steven Goodridge to be found on the HumanTransport site.

There are several different layouts for inductive loops: Dipole, which is the simplest - a single rectangular loop; Quadrupole, which doubles the loop on itself in a figure-8 (it appears as a double rectangle in the pavement) which increases sensitivity; and the Diagonal Quadrupole, which has several loops of a smaller size and is the most sensitive. From what I understand, a smaller loop will generally be more sensitive to bicycles than a larger loop.

Assuming that the vehicle detector is sensitive enough, how do you ensure that a bike will trip it?

There are a few different layouts for inductive loop sensors.
These graphics show the "sweet spots" in the various layouts.
Apparently the diagonal quadrupole design is the most sensitive
for bicycles. (Graphic from
A major factor in tripping the light is knowing where the inductive loop is located, and stopping in the optimal position to activate it. Often, you can see where the pavement has been cut to install the loop -- it may look like a large rectangle cut into the pavement right at the front part of the intersection, just behind the white line that marks where vehicles are supposed to stop. On a bicycle, it isn't enough to stop inside the rectangle, or near it. You really need to be stopped right on the loop, preferably with your bike in-line with the cuts in the pavement.

It is possible that the loop can be paved over, and the cuts would not be visible. In some communities, the road dept. may actually mark with paint the optimal stopping locations for cyclists to trip the lights, but I have no doubt that is pretty rare. If your community uses vehicle detectors that are invisible because they've been paved over, I'd suggest getting fellow cyclists to complain and petition to have such markings put into place, as ultimately it becomes a safety issue. In the meantime, I've read that it may help to get off the bike and lay it down on its side approximately where you might expect the loop to be located, thereby increasing the chance that you might cross over it and trip the signal.

In my experience, simply understanding how these things work, and knowing about the "sweet spots" makes a big difference in being detected. If you are indeed able to trip the lights, you should really know within a minute or two. If after a reasonable time you can determine that you are unable to get a green light, it's time to go to plan B. Unfortunately, plan B can be a bit of a legal gray area.

Although only a handful of states actually have bicycle-specific laws regarding what to do when the vehicle detectors fail to work, most states (if not all) have language in their traffic laws regarding malfunctioning traffic signals. What are car drivers expected to do when they get to an intersection and the traffic lights aren't functioning? Typically, the expectation is that they treat the malfunctioning light like a stop sign. Come to a complete stop, yield to traffic with the right of way, and then proceed when it is reasonably safe to do so. There is a strong legal argument to be made that a vehicle detector that is unable to sense a bicycle should be treated as malfunctioning, and that the affected cyclist is justified in proceeding with caution after reasonably determining that the sensor is indeed not working, and determining that it is safe to proceed.

I say that this is a bit of a gray area because not everyone agrees. For example, in my researching this issue, I found that in Florida, officials with the state's Highway Safety bureau specifically advise against cyclists treating such a traffic light as a stop sign. Instead, they recommend that cyclists make a right turn, then later make a u-turn to get back to where they intend to go - or find an alternate route (FloridaCyclingLaw). I have no idea how many other states make such a recommendation, but if I lived in Florida, I'd be teaming up with fellow cyclists and lawyers to challenge that interpretation. My point is that I truly believe that such an interpretation of the law would not stand up to a legal challenge. First of all, the laws in all 50 states define bicycles as vehicles and hold them accountable to the same laws as any other vehicle drivers. That being the case, then it is the responsibility of the state to apply its own laws equally to cyclists. If the state or local road authorities install traffic control devices that fail to recognize all vehicles, or knowingly place an undue burden on a segment of vehicle users, then the state is violating its own laws.

In specific regard to the advice that cyclists should make a right turn, then later make a u-turn, this in and of itself may be against the law in some situations. Even in states that allow right turn on red, there are always some intersections where it is prohibited. Likewise, there are many places where a u-turn is not legal. Also, many states (including Florida, I believe) have laws specifically prohibiting vehicles from performing the right-turn/u-turn maneuver because it is sometimes used by drivers to avoid waiting for the traffic lights! Whether or not the maneuver is legal, it certainly opens a cyclist up to even more risks than simply treating the malfunctioning signal as a stop sign and proceeding through the intersection accordingly with due caution. There is no way to legally argue that the intent of the law is to put an unfair burden on certain vehicle users, and to put vulnerable road users in even greater risk. As to the advice to "find an alternative route," again it places an unfair and undue burden on cyclists, and may be completely impractical in some cases.

Okay then, what does a person do when they encounter this situation? I am of the mindset that says any traffic device that doesn't recognize a bicycle should be considered malfunctioning, and treated as stop sign. Do the same thing you would do in a car if you came to a traffic light that was not working. Stop, yield, proceed with caution when it's safe to go. That is the advice given in John Forester's Effective Cycling, as well as by John Allen and other bicycle safety experts and advocates. As I've already pointed out, there is a pretty sound legal argument for this, whether or not a state has bicycle-specific language on the subject. If you're still uncertain (or live in Florida) keep your eyes open for police. I mean, really, what are the odds that one will suddenly pop up just as you're going through the intersection?

So then, let's say you are the incredibly unlucky person who gets stuck at a light that won't change, you've done what you can do to make sure the vehicle detector is not "reading" your bike (good luck to you carbon bike users!), and there are no cars in sight that can trip the light for you, so you decide to proceed cautiously through the red light, and BAM! A cop shows up out of nowhere to write you a ticket. If that happens, you calmly and respectfully explain the situation to the officer. It's worth noting that many police officers are also motorcyclists (at least in my experience), and it's possible they may have encountered the same situation themselves. Unless the officer is a completely irrational bike-hating A-hole, it's hard to imagine they'd still write a ticket. And if they did, I'd absolutely take that ticket to court and fight it.

If there is an intersection that you know of that consistently does not detect bicycles, it's really important to note the location and notify the proper authority. It might take some effort to determine exactly whom to call, whether it is the city road/transportation department, or the county, or whatever the jurisdiction. Until you reach the right office, I wouldn't even mention that I have a bicycle-specific issue -- if they think you are a driver reporting a malfunctioning traffic signal, they might give you less run-around. Once you reach the right person, be calm, rational, and non-argumentative. One really shouldn't have to argue too hard, as it really is a safety issue if a light doesn't function properly for cyclists - it essentially forces cyclists to engage in potentially risky behavior, and the road/traffic authorities have the legal obligation to ensure the safety of all road users. Still, it's possible that the wheels may get greased a little faster if you can encourage others to report the malfunctioning signal as well. Get your local bicycling clubs involved in reporting such intersections likewise.

The thing is, if one of these inductive loop detectors is adjusted properly, it should be able to detect a bicycle. And if it can't, it is apparently possible for the road dept. to make adjustments to the sensitivity. If they cannot adjust the sensitivity to detect a bicycle without getting interference from traffic in adjoining lanes, then they may need to install the loop in a different layout, because as described earlier, there are some designs that work better than others. As previously mentioned, if the loop is completely hidden, then the "sweet spots" can be marked with paint, which costs next to nothing but makes a big difference.

I suppose it would be easy for some to argue that this is why cyclists shouldn't bother following the law, or use it as justification for blowing off lights. It should be clear that I would not make that argument. As I've explained, knowing how to trip the devices makes a big difference, and in my own experience, the devices are getting better at detecting bicycles. Also, unless your state has some specific laws regarding bicycles and traffic lights (such as the "Idaho Stop") I definitely don't recommend simply running the lights as a first response. Don't do it until you've determined that you can't get the lights to work - then proceed cautiously.


  1. I live in South Florida, and most "light loops" don't trip for me. So, I follow the "two wrongs don't make a right, but three rights make a left" velosophy.

    1. Yeah - you guys really need to petition on that. Find a cycling congressman to help you.

    2. The law here in Virginia is that we have to wait two cycles of the light or 2 minutes, whichever comes first. Then we can legally run a red light.

    3. That seems pretty reasonable. If you routinely encounter a light that doesn't detect you, it's probably still worthwhile to report it to see if they can make it more sensitive.

    4. I hop off the bike and become a pedestrian. I have had cops watch me do it and they don't bat an eyelash.

    5. Though not necessarily convenient - that certainly is legal. So, no, police really shouldn't have an issue.

  2. Some of the larger intersections here in New York have four-way red signals. They have "walk" signals for pedestrians that flash for twenty seconds before the traffic signal turns green. That gives pedestrians time to cross before motor vehicles make turns onto the street the pedestrian is crossing.

    According to the law here, that signal applies only to pedestrians. Cyclists have to follow the traffic signal--which, of course, can be dangerous, especially if the cyclist is crossing from a bike lane. So, now there is a proposal to let cyclists follow the pedestrian signal. I hope it passes.

  3. Surely if you are in a left turn lane, and determine that the light is not going to change, Florida's "advice" of making a right turn from the left lane is just as illegal as making a left turn against the light.

    I'm in Illinois where we have the "wait two minutes then treat as a stop sign" rule. I go through one traffic light each morning that will not change for me. I should probably report it. I am not going to wait two minutes every time when I've determined already that it will never change, so I go when safe unless there is a car there waiting with me to trigger it.

    1. Ok, I reported it. Their response:

      "We will attempt to adjust the sensitivity on the sensor, but it may still have a hard time detecting a bicycle. I will forward your concern to IDOT, who has jurisdiction at this intersection."

    2. It would be interesting to see what they can do. It still might help if you can convince others to report them too.

    3. There aren't that many transportation cyclists in this town, unfortunately; and fewer who would wait for the light.

  4. How do pedestrians get across? If you get off your bike and walk across, you are pedestrian. Seems simpler than a lot of right turns/U turns/jockeying... ?