A tale of a malfunctioning sensor

2011-12-16 by . 2 comments

Post to Twitter

Most urban cyclists have probably encountered the problem of triggering traffic signals once or twice. Most of the time, sensors at traffic lights detect bicycles right away, and are often helpful in reducing wait-time. I know of a few intersections where the sensors are finely tuned so that the light turns green before you even get there, if nobody is coming the other way. What happens though when one of these sensors is out of whack?

I go to school in Corvallis, OR, which currently holds the record for the highest percentage of bicycle commuters nationwide. There's a particular intersection here where the under-pavement loop-type sensor simply wasn't working for me. Some friends had mentioned having problems at the same intersection as well. The sensor is for a left-turn arrow in a turn-only lane, so it probably doesn't get huge amounts of bike traffic, but I use the turn lane there, as do others. You can see the intersection in the streetview image below:

As you can see, there's a pavement indicator for where bikes should stop to trigger the signal. There is also one for the bike lane on the right. It's difficult to see in this image, but there are pavement cuts which indicate a buried sensor loop, which works by magnetic induction, not a visual cue. When I arrived at the intersection, I stopped just past the marker, and waited for the light to turn. It never did. Fortunately, the intersection has one of the new "flashing yellow" sequences, so I was able to turn left when there was a break in traffic, although I did have to wait two light cycles before I was able to make it across.

So, what to do when a sensor seems to be malfunctioning? In hind sight, I should have tried again before seeking help, but having heard similar reports from other cyclists, it sure sounded like there was a problem. Fortunately, it's easy to contact the city public works and get problems checked out. I sent an email on a Sunday afternoon, and on Monday I got a response back from them:

Thank you for the description of the concern you have at location.  We went today to test the southbound left turn loop.  This movement of the intersection uses a quadrapole loop for vehicle detection.  In layman's terms, a loop is essentially an antenna in the pavement, tuned to a specific frequency, to detect metal objects.  When a metal vehicle passes over the loop the frequency of the loop is changed and acknowledged in the traffic signal controller.  A quadrapole loop looks like two long narrow rectangles, placed side by side, with a common line in the middle.  We have a bike marker placed on the pavement, in the center of the quadrapole, at the very front, behind the stop bar.  The bike's crankset should be placed directly over this marker.  We use a Specialized MTB from Public Works to perform our bicycle detection.  We feel this type of bike represent the majority of commuter cyclist in Corvallis.  We did not find any issue with the southbound left turn loop at location.  Every time we installed or removed the bike from the detection, it was acknowledged in-kind at the traffic signal controller.  If you have any questions or concerns please contact me via email or phone.

Wow! In less than 24 hours, a work crew was sent out to check the sensor, and verified that it's properly working. I was slightly embarrassed to hear that the loop is functioning just fine and that I had not been triggering it properly, but that feeling was vastly overshadowed by how impressed I was with the city's quick and helpful response. This is the same city where a police officer was dispatched to cut loose my friend's bike when her lock got jammed (after proving ownership of course), and where the University just finished installing about 500 new bike racks across campus. No wonder so many people bike commute here!

I have yet to return to this intersection since filing the report, since I don't bike that direction often. The next time I do, I'll be sure to try the public works official's tips for positioning the bike to properly trip the sensor. It's odd that this problem occurred, since I've never had trouble with any other intersection. It's possible that the light timing is slightly different and if I had waited another cycle, I would have gotten an arrow. It's also possible that I was just not paying close enough attention to where I was positioning my cycle - if I had originally triggered the sensor, but then moved off it slightly, it might have thought that I went through the intersection during the flashing yellow cycle.

Regardless of the true cause of the light mishap, which I now know won't happen again, the lesson learned is this: if something looks amiss, don't hesitate to contact your municipality's public works department. They're here to help you out, and if my experience so far is any indication, they'll be more than happy to help.

Filed under Commuting Traffic

2 Comments

Subscribe to comments with RSS.

  • Neil Fein says:

    You’re very lucky to live in an area where the public works responds to issues like this. In New Jersey, complaints about sensors not responding to bicycles are met with either disinterest or “that’s not out signal, talk to [other municipal entity].”

  • freiheit says:

    It seems like part of the problem is the pavement indicator design. Instead of a single dash, I’m used to seeing two dashes with a bike logo in between, which intuitively leads most people to place a wheel on each dash.

  • Comments have been closed for this post