Traffic Loop Sensor Trigger For Bikes

In article
,
Jay Beattie wrote:

On Nov 11, 6:48*pm, Mike Rocket J Squirrel
wrote:
On 11/11/2008 6:34 PM Patrick Lamb wrote:





On Tue, 11 Nov 2008 16:52:11 -0800, Mike Rocket J Squirrel
wrote:

On 11/11/2008 1:56 PM Arthur Shapiro wrote:

Would the presence of a strong magnet affect these sensors? *I don't
appreciate the behavioral distinction between magnetic flux and
metallic mass
in the situation in question.

I'm not anxious to add extra riding weight, but these fancy new rare
earth
magnets can pack a huge effect in a couple of ounces.

Art
Nope -- AC induction loop doesn't care about magnetic fields. All the
loop
would "see" is the metal of the magnet, the field would be invisible.

Is there a single common frequency that all these things use? *I've
toyed with the idea of building a light weight circuit with a small
antenna to disrupt the flux, but I don't care enough to build much
smarts into the system.

Dunno. If you're smart enough to build a circuit to counter the signal in
the loop, you're smart enough to hook a small loop to a battery-powered
oscilloscope and plunk it atop the sensor loop to see what it picks up.

I have a friend who says that if the light does not turn green, you
can treat it as broken and proceed through the intersection when
safe. I have not been able to find any support for this statement in
the UVC. -- Jay Beattie.

The basic principle is this: a light system that does not properly
detect traffic is de facto broken. The UVC does have a standard for
non-functioning traffic lights: treat as a four-way stop.

Bikes are traffic, so if the loop can't detect you, it's not working as
expected. Note that the key difference between this broken intersection
and most others is that the cross traffic doesn't know it's broken.

Nonetheless, traffic laws are not meant to be a death sentence. You
aren't expected to wait at the light until dehydration ensues.

That said, let's be realistic: there are busy intersections where, laws
be damned, the only rational choice may be to give up and activate the
pedestrian signal.

--
Ryan Cousineau http://www.wiredcola.com/
"In other newsgroups, they killfile trolls."
"In rec.bicycles.racing, we coach them."

On Nov 11, 11:23*pm, Jay Beattie wrote:

I have a friend who says that if the light does not turn green, you
can treat it as broken and proceed through the intersection when
safe. *I have not been able to find any support for this statement in
the UVC. *

Hmm. I suppose an alternative strategy would be to stand there and
wait for a traffic light repair crew. Or wait for starvation,
whichever comes first.

Anybody have a report of someone ticketed for treating a non-
functioning signal as broken, then proceeding when safe?

- Frank Krygowski

On Nov 11, 11:25*pm, peter wrote:
On Nov 11, 6:34*pm, Patrick Lamb wrote:



On Tue, 11 Nov 2008 16:52:11 -0800, Mike Rocket J Squirrel

wrote:
On 11/11/2008 1:56 PM Arthur Shapiro wrote:

Would the presence of a strong magnet affect these sensors? *I don't
appreciate the behavioral distinction between magnetic flux and metallic mass
in the situation in question.

I'm not anxious to add extra riding weight, but these fancy new rare earth
magnets can pack a huge effect in a couple of ounces.

Art

Nope -- AC induction loop doesn't care about magnetic fields. All the loop
would "see" is the metal of the magnet, the field would be invisible.

Is there a single common frequency that all these things use?

Don't think so. *They're generally in the 20 - 40 kHz region but AFAIK
it's not standardized.
BTW, the advantage of iron for E-M induction is primarily at very low
frequencies. *At typical loop detector frequencies non-magnetic metals
work just as well. *So aluminum bikes and rims are equally effective
as steel ones. *The problem is just insufficient quantity unless the
detector sensitivity is turned up.

Do you know the frequency that industrial inductive proximity sensors
use? I ask because those are definitely less sensitive to aluminum
than to steel.

(Unless the aluminum is thin foil, oddly enough.)

- Frank Krygowski

On Nov 12, 8:08*am, Frank Krygowski wrote:
On Nov 11, 11:23*pm, Jay Beattie wrote:



I have a friend who says that if the light does not turn green, you
can treat it as broken and proceed through the intersection when
safe. *I have not been able to find any support for this statement in
the UVC. *

Hmm. *I suppose an alternative strategy would be to stand there and
wait for a traffic light repair crew. *Or wait for starvation,
whichever comes first.

Anybody have a report of someone ticketed for treating a non-
functioning signal as broken, then proceeding when safe?


And for Ryan, could you give me a section number for the UVC or the
langauge -- I am trying to find a cross-reference under the Oregon
version of the Code. So far, I have only found a case where someone
who went through a "stuck" red light and got hit sued the city. The
court noted that the city should assume that a driver at a "stuck" red
light will go through it eventually. It cited to the general statute
about following traffic control devices but did not cite any section
about how people are supposed to treat malfunctioning devices. In
fact, our version of the UVC makes it clear that it is an infraction
to go against a light unless ordered to do so by a police officer.
The section where you would expect to see an exception for
malfunctioning devices says nothing about them. -- Jay Beattie.

On Nov 12, 11:08*am, Frank Krygowski wrote:
On Nov 11, 11:23*pm, Jay Beattie wrote:



I have a friend who says that if the light does not turn green, you
can treat it as broken and proceed through the intersection when
safe. *I have not been able to find any support for this statement in
the UVC. *

Hmm. *I suppose an alternative strategy would be to stand there and
wait for a traffic light repair crew. *Or wait for starvation,
whichever comes first.

Anybody have a report of someone ticketed for treating a non-
functioning signal as broken, then proceeding when safe?

- Frank Krygowski

No ticket, but I've been stopped and warned, and told I should use the
crosswalk and walk across in the future. I didn't feel like arguing
at the time, so I let it go and went on my way. The intersection in
question had 2 lanes coming from my left, in a 35MPH zone that just
came down from 50, and 3 lanes from my right, also in a 35.
Personally, crossing that intersection I'll take the extra speed of
the bike every time. Had I been on the shoulder and crossed as a
pedestrian, even using the bike, he probably wouldn't have bothered
me.

In article
,
" wrote:

On Nov 10, 3:01*pm, " wrote:
On Nov 10, 4:44*pm, Michael Press wrote:
" wrote:

It need not be steel, only a conductive body that will have
eddy currents induced in it by the inductive loop in the pavement.

I take it that aluminum is far less conductive than steel then? *I
know from experience of several of these sensors that my older mostly
steel Ninja would trip that the newer mostly aluminum R6 wouldn't.

http://www.howstuffworks.com/question234.htm

Aluminum is a very good conductor. The issue here
is the inductance. Placing the metal object near the
wire loop changes the inductance of the loop. The
control circuit is monitoring the inductance by sending
an AC current through the loop and effectively measuring
the time constant. The metal bike acts like the
ferrous core that one puts inside a loop of wire to make
a useful inductor as an electrical component.
Aluminum also increases the inductance, but
not as effectively as iron because it has a much
lower magnetic permeability than iron.

http://en.wikipedia.org/wiki/Inductor
http://en.wikipedia.org/wiki/Magnetic_permeability

Traffic loop sensors run a 20000 Hz or more. A conducting
body near the loop makes the combined system a different
circuit. The varying magnetic field of the loop induces
currents in a nearby conductor that reduce the inductance
of the combined circuit, raising the resonant frequency
of the combined circuit. Note however that a ferromagnetic
effect will _increase_ the inductance of the loop,
increasing the resonant frequency and evading detection
by the signal detection system. The ferromagnetic effect
of the steel in a car is overwelmed by the large silhouette
of the conducting body, so the overall effect is reduction
of the overall induction.

--
Michael Press

In article ,
Mike Rocket J Squirrel wrote:

From the "More Stuff on a Bicycle Is Better," Dept.

http://gizmodo.com/5079430/traffic-l...e-green-lights

Al bicycle rims lend themselves well to detection by
inductive loop detectors because they provide an excellent
conductive loop and are located close to the ground where
the loop wires are. Here is what to do. Find a section
of the sensor loop that is parallel to your direction of
travel an put your wheels directly above that section
with the bicycle straight up. This maximizes the induction
in the rims.

--
Michael Press

On Nov 12, 4:30*pm, Michael Press wrote:
In article ,
*Mike Rocket J Squirrel wrote:

*From the "More Stuff on a Bicycle Is Better," Dept.

http://gizmodo.com/5079430/traffic-l...for-bikes-gets...

Al bicycle rims lend themselves well to detection by
inductive loop detectors because they provide an excellent
conductive loop and are located close to the ground where
the loop wires are. Here is what to do. Find a section
of the sensor loop that is parallel to your direction of
travel an put your wheels directly above that section
with the bicycle straight up. This maximizes the induction
in the rims.


That definitely helps, but it's not 100% reliable. If the sensitivity
is set too low, it won't trip the light.

In that case, if the coil is rectangular, tipping the bike down toward
the inside of the rectangle (tires placed as before) sometimes does
the trick.

Sometimes they put a figure 8 coil down to help cyclists. It shows up
as a double rectangle, i.e. three parallel lines in the direction of
travel, boxed in by perpendicular lines fore and aft. IME, these work
well only if the bike is on the center parallel line. Aligning over
the outside parallel lines doesn't work. This is because the field is
tightly contained between the two rectangles.

I know this is all on a website somewhere. I'll google later, if
nobody else has.

- Frank Krygowski

In article
,
Frank Krygowski wrote:

On Nov 12, 4:30*pm, Michael Press wrote:
In article ,
*Mike Rocket J Squirrel wrote:

*From the "More Stuff on a Bicycle Is Better," Dept.

http://gizmodo.com/5079430/traffic-l...for-bikes-gets...

Al bicycle rims lend themselves well to detection by
inductive loop detectors because they provide an excellent
conductive loop and are located close to the ground where
the loop wires are. Here is what to do. Find a section
of the sensor loop that is parallel to your direction of
travel an put your wheels directly above that section
with the bicycle straight up. This maximizes the induction
in the rims.


That definitely helps, but it's not 100% reliable. If the sensitivity
is set too low, it won't trip the light.

In that case, if the coil is rectangular, tipping the bike down toward
the inside of the rectangle (tires placed as before) sometimes does
the trick.

Sometimes they put a figure 8 coil down to help cyclists. It shows up
as a double rectangle, i.e. three parallel lines in the direction of
travel, boxed in by perpendicular lines fore and aft. IME, these work
well only if the bike is on the center parallel line. Aligning over
the outside parallel lines doesn't work. This is because the field is
tightly contained between the two rectangles.

The figure 8 is a quadrupole loop,
an improvement over the dipole loop.

I know this is all on a website somewhere. I'll google later, if
nobody else has.

http://www.humantransport.org/bicycledriving/library/signals/green.htm

--
Michael Press

Patrick Lamb wrote:

Is there a single common frequency that all these things use? I've
toyed with the idea of building a light weight circuit with a small
antenna to disrupt the flux, but I don't care enough to build much
smarts into the system.

Pat


I was told by someone with the highway dept that there is not a single
frequency.

All the loops at any one intersection are from 20Khz to 60Khz, because
if any two shared the same frequency, they would suffer from RF
cross-talk inside the control box where the wiring converges. So each
one is adjusted different and randomly, and there are no guidelines
except for the street dept to change a frequency any time that a nearby
business reports a problem near a frequency that the traffic loops were
set to.

The actual frequency adjusters are unmarked or only rough-marked, so the
crews don't normally know exactly what frequencies are in use--they only
need to know to set each loop at any intersection different from any
others.

You could (I would guess) build a "sweeping" oscillator, that would go
from the lower limit through to the upper limit, at the press of a
button. That wouldn't require active detection. I don't know how much
time the sensor has to detect before it triggers the light, however....
~