When Signal Detectors Don’t Work

Safety Corner logo


by John Allen



We had a crash on a recent CRW ride. The rider was signaling a left-hand turn while on a downhill and hit a pothole with only one hand on the handlebars. We might venture that he went down because his weight shifted forward when he hit the pothole and his one hand pushed one side of his handlebars forward, throwing him off balance when the wheel turned. Lesson learned—keep both hands on the bars on an uneven road, or anytime that you need two hands for comfort or stability. Under these circumstances, use verbal signals rather than hand signals to indicate your intentions.

Most people on CRW rides know how to turn left (just as motorists do) – by first merging to the left-turn position on the roadway. Everyone did that in this video I shot during July’s Climb to the Clouds ride: https://vimeo.com/73215541. The rest of this article comments on the Climb to the Clouds video, so please have a look.

As indicated in the video, the left turns on red shown are by no means entirely the bicyclists’ fault. The left-turn signal is triggered by a metal detector buried in the roadway, which in this case did not detect bicycles. The bicyclists who ran the red did wait -- but the signal did not change.

On the other hand, if the bicyclists had known to line up carefully over the pavement cuts that indicate the loca-tions of the detector wires, the signal probably would have changed.

Where to wait to trigger wire–loop actuators. Image kindly provided by Dan Gutierrez, www.DualChase.com. Usually a“type Q” (quadrupole, 3–line) loop is indicated by a pavement marker. If no cuts are visible and there is no pavement marking indicating where to wait, try waiting ¼ and ¾ of the way across the lane.

A motorist was waiting behind the bicyclists, but too far back to trigger the signal actuator. The motorist appeared to be staying out of the way of the group of waiting bicyclists -- and waited through an entire extra light cycle. Perhaps the motorist was trying to be considerate. On the other hand, the motorist could have resolved the situation without anyone’s running the light, by pulling forward over the detector – but he or she probably didn’t know that. After the first group of bicyclists left, I motioned to the motorist to pull forward, and the light did change, allowing the remaining bicyclists to proceed on a green signal.

Marking indicating to wait
at the middle of the lane in newer,
quadrupole installations.

The most effective time for a teachable moment about signal detectors isn’t while waiting in a group of bicyclists who are arriving at random. My failed attempt in the video demonstrates that well enough. Even more so for a motorist, who is out of earshot inside a vehicle. So, what can you do to help?

1. Alert ride leaders to the problem spots on our major rides, so ride leaders can mention problem spots during pre-ride talks.
2. Spread the knowledge of how to trigger signal actuators during discussions with fellow cyclists
3. Develop a culture of orderly assembly when queuing at intersections. This includes not only avoiding dumb moves like the close passes of a moving motor vehicle shown in the video, but also forming neat double lines over the wires of the metal detector loops. An orderly group also might encourage motorists to feel more comfortable in pulling forward to help trip signal actuators. A final point about neatness. Note the motorcyclists in the video. Both bicyclists and motorcyclists have problems with public image, but the motorcyclists in the video were doing a fine job of dispelling their image problem. Even if a non-functional detector forces a left turn on red, orderly appearance conveys a more positive message than the ragged group of bicyclists in the video. Your thoughts? Please email me at jsallen@alum.mit.edu.

Please send corrections, additions, comments and praise to

© 1997- CRW, Inc. All rights reserved. Revised: