Right off the top of my head, on all but the most high-end bikes, all the spokes are ferrous metal anyway, so I'm surprised that a loop of thin wire adds anything to the detectivity. Unless it's just something about the configuration of a loop that does it. But if they say it works, hey. Interesting.
FAO Nick and other Bike/Engineering Geeks
I found this post in a listserv that I belong to.
Just wondering what people think about the baling wire concept from an implementation and/or maintenance point of view, and whether anyone has any idea how close this ferrous metal would have to be to the sensor in order to be functional. Could the same thing be accomplished by stringing/wrapping a larger amount of ferrous wire elsewhere, as long as it was low on the bike? (I am wondering on behalf of both my bicycle and my scooter....)
Original Message
All,
We are having an issue with bicycles not being detected by traffic light
sensors. The sensors are a "puck" about 4 inches in diameter imbedded in
the pavement and it detects ferrous metals. It is not the traditional
"loop" detector. I know that the issue of traffic light sensors not
responding to bicycles is an old one, but I thought it had been sorted out
and that there were readily available systems that would detect a bicycle.
Our conversations with the engineers responsible for this system tell us
that they will try increasing the sensitivity though they doubt that a
system designed to detect cars will pick up the amount of steel in a modern
(CF or aluminum) bicycle. At their current sensitivity setting they do not
appear to even detect a steel bike (on trial). Their suggested alternative
is an infrared detection system that will cost $10K per crossing ($5K for
each sensor unit).
Do any of you have suggestions on dealing with this issue? We have 3 such
crossings in our city bike route network and it doesn't seem like the best
use for $30K. We do have a state and local Complete Streets program but
these intersections were installed before those requirements were in place,
and besides there is a pedestrian pushbutton available (cyclists would have
to dismount and go up on the sidewalk to reach the button).
Any thoughts/suggestions will be appreciated.
Response 1:
Hi Folks
When Caltrans did their testing of various loop detector configurations it was learned that by placing a single strand of bailing wire (ferrous metal) under the tube protector band (is there a technical term for the big rubber band that you place over the spoke nipples to protect the inner tube from being punctured ??) that most loop detectors could be activated.
Such wire placement add only a few grams to the tires/wheel's overall weight and doesn't interfere with handling.
Just a thought to pass along to some of your bike folks who may have non-ferrous bikes.
Response 2:
Bailing wire, in interesting idea. The cut ends would need to be taped to keep from cutting thru the strip and tube.
(is there a technical term for the big rubber band that you place over the spoke nipples to protect the inner tube from being punctured ??)
That would be the rim strip.
----
Thoughts? Comments?
I really don't like having to dismount to operate a button. Usually I'm trying to make a left, and would either have to drag the bike over to the pole, or leave the bike in the middle of the street, in order to do that. Cars making rights on red are a major danger. It's just a lot easier to "3112" the light (i.e., go when safe, never mind the light), or wait for a car to show up behind, and hope it gets close enough to trigger the light. Think Perrymont at Perry/US19 in McCandless, a light I have to "3112" every day.
Will a small, powerful (neodymium) magnet trip the puck sensor?
If not, maybe a small tuned coil would trip it...but you'd have to know more about how the device works.
http://www.wikihow.com/Trigger-Green-Traffic-Lights
http://www.amazon.com/High-Power-Motorcycle-Green-Trigger/dp/B0038B4BW6
hmm... not posting for me...
http://www.wikihow.com/Trigger-Green-Traffic-Lights
http://www.amazon.com/Traffic-Light-Changer-Specifically-Motorcycle/dp/B001DC7L42
@Orionz - Thanks! That first lick was great. I can't wait to try it out!
At Motorcycle Safety class they recommend revving you engine. Not very helpful on a scooter with a) an automatic transmission, and b) handbrakes only.
I agree with Stu. Its funny to me that when making a left from Washington Blvd. into the POLICE station parking lot I have to do this frequently. Whatever sensor is there does not pick up bikes wanting to make that left.
@edmonds http://www.fhwa.dot.gov/publications/research/operations/its/06108/02.cfm
http://en.wikipedia.org/wiki/Eddy_current
http://en.wikipedia.org/wiki/Magnetic_field_sensors
"Their suggested alternative
is an infrared detection system that will cost $10K per crossing ($5K for
each sensor unit)."
Whoever made that suggestion must be up for salesman of the year. I find it somewhat unbelievable that the same type of sensor that turns on the light when I walk into the bathroom at work would cost $10,000 (US) to use at an intersection and that it's the only viable alternative. Granted, this is only my knee-jerk reaction and I haven't bothered to read anything on how traffic light sensors work, but it still seems highly unlikely that $10K per intersection is a reasonable price to pay.
Also, I don't know of any cyclists that would feel okay about putting pointy metal inside their tires. There must be a better system out there.
Back home, they installed a button that cyclists could reach without dismounting on either end of a two-lane tunnel that activates warning lights so that drivers know there is a cyclist in the tunnel. Why can't they install similar switches in this case?
Edit: I'm not sure how to get the streetview link to embed properly, but Google "Stringer's Ridge Tunnel, Chattanooga, TN" and check it out for yourself.
My 2 cents:
There's a super old infrared system above the inbound left turning lane on the south 10th street bridge. Ask County DPW about it?
Are we considering systems installed in NEW pavement or in existing pavement?
Best I can think of is a piezoelectric pressure sensor installed in the middle of the travel lane - away from prevailing motor vehicle tire paths - before the stop bar where a cyclist (pedal or motor) can sit on. (EXAMPLE) Also, it'd work best if there was a separate, smaller signal which notifies the cyclist of their detection by the sensor.
Also, here's a BLOG post by Bike Delaware showing how awesome their DOT is to cyclists...
Mikhail's first link says that early inductive-loop detection devices, back in the 1970s, couldn't detect bicycles, but designs since then can. They were "developed to meet the demands of the European market, where bicycles must be detected."
It sounds like somebody invented a new "puck" design that returns to the problems of the 70s. The best solution would be to avoid the defective design, and use a properly set up induction system. But perhaps that would cost even more than the infrared approach if it means digging up the street to correct the screwup.
Having everybody modify their bikes to compensate for inept roadway construction seems less than ideal. Imagine if a city asked motorists to do that.
I've had no problem triggering the sensor lights with my steel Trek 520, so long as I line up with the pavement the right way.
Some of the sensors here in Illinois (I'm there for school) are triggered by weight and even motorcycles won't trigger them.
Illinois passed a law this year allowing motorcycles and bicycles to run redlights when necessary. http://www.ilga.gov/legislation/billstatus.asp?DocNum=2860&GAID=11&GA=97&DocTypeID=HB&LegID=60182&SessionID=84
My little aluminum bike has definitely never triggered the one on the Schenley Drive Extension thingy coming from the Cathedral. Cars never want to pull up far enough behind me, either, so it always turns into a really hilarious game of charades when I try to mime "This light has a sensor that I can't trip, so you need to move your car forward, please"
+1 pearmask. I love that game of charades with the drivers behind me.....
Still curious on the ferrous wire thing though. I'd certainly never put one inside my wheel, as proposed, but if it was enough to trip a certain type of sensor, might consider wrapping some element of my frame, or rear brackets, with enough to trip a difficult sensor.
I used to have the same problem at the intersection of rt 40 and Northern Pike, I'd be trying to turn from south bound to east, and cars would politely give me enough room that the light wouldn't trip.
Usually having to wait a cycle while I'm waving them forward gets it through the windshield what needs to happen.