Camera Systems at Stoplight Intersections to catch the red light runners



[email protected] wrote:

> You seem to believe a bicycle should be in the automobile path at a
> traffic light. I find that unnecessarily intrusive, considering that
> when the light turns green, cars behind the bicyclist will progress at
> whatever rate the bicycle chooses. Putting a post with push button at
> the roadside for bicyclists (and possibly pedestrians) to activate is
> not an undue expense.


Most of my issues with non-responsive traffic signals are when I'm
trying to turn left and the left arrow doesn't trigger. Putting a pole
in the middle of the road where I'd be able to reach it doesn't seem
like a very safe thing to do whereas making the loop detectors
sensitive enough to detect my bike wheel is safe, convenient, and
frequently requires no additional expenditure beyond proper training of
the personnel who adjust the sensors.

> By thew way, there are simple ring detector loops and single and
> multiple figure-8 loops. The figure-8 loop couples closely from one
> half of the loop into the other and can detect smaller conductors such
> as a bicycle wheel because it is closely coupled.


True, provided that the sensitivity is properly adjusted.
Unfortunately many are not adjusted that way until enough cyclists
complain and someone from the transportation department has to come out
and readjust the sensor.
 
[email protected] wrote:
> Bob Prohaska writes:
>
>
>> Another trick that sometimes helps is to lay the bike down in the
>> sensor loop.

>
> I think you are falling prey to the same thinking that traffic
> engineers and many bicyclists use. Bicyclists are not inside motor
> vehicles and when stopping for a traffic light put a foot down.


If you're suggesting I should complain to the TE's when this happens
you're right. Laying the bike down is a last resort when I'm stuck.
Trouble is, I never remember to complain during business hours :cool:.

> Placing a manual button at that place is simpler, less ambiguous, less
> expensive and more convenient for the user. The need for a detector
> loop for bicyclists is rare, occurring mainly when a bicycle lane lies
> between other lanes.


I'm not convinced it's a technical problem to make traffic sensors
that can pick up bicycles. It might require a little more tuning
of detector thresholds, but that's a one-time thing. Now, if you're
talking about a graphite bike with composite rims, that's a harder
problem, but it's not yet common and there's still plenty of metal.

> Locally we have a large eight lane intersection with a pedestrian and
> bicycle crossing side by side. Bicycles have slot in the traffic
> island with a paint marked detector loop with a post on which, instead
> of a push-button, a 4x6" sign saying "Bicycles Wait Here" which few
> riders do. Instead they go to the pedestrian walk and press the
> pedestrian button on the light pole. This triggers the light to stay
> green for 40 seconds instead of 20 seconds for a bicycle. Meanwhile,
> everyone waits for the light to change at an empty intersection that a
> bicyclist crossed.
>

That behavior suggests the pickup does not sense the bicycles in the
way the riders expect; either it does not work well or the delay is
too long.

> At least today's manual buttons can be slapped with the palm of the
> hand, unlike the old ones that had a little 3/8" brass button shielded
> by a raised rim.

Big buttons certainly help, but they have limitations:

Going straight through an intersection I'd rather not pass
through space used by right-turning traffic, if I can avoid it.
Buttons are no obvious help in situations with dedicated, protected
left turn lanes unless there's a center island.

bob prohaska
 
Bob Prohaska writes:

>>> Another trick that sometimes helps is to lay the bike down in the
>>> sensor loop.


>> I think you are falling prey to the same thinking that traffic
>> engineers and many bicyclists use. Bicyclists are not inside motor
>> vehicles and when stopping for a traffic light put a foot down.


> If you're suggesting I should complain to the TE's when this happens
> you're right. Laying the bike down is a last resort when I'm stuck.
> Trouble is, I never remember to complain during business hours :cool:.


>> Placing a manual button at that place is simpler, less ambiguous,
>> less expensive and more convenient for the user. The need for a
>> detector loop for bicyclists is rare, occurring mainly when a
>> bicycle lane lies between other lanes.


> I'm not convinced it's a technical problem to make traffic sensors
> that can pick up bicycles. It might require a little more tuning of
> detector thresholds, but that's a one-time thing. Now, if you're
> talking about a graphite bike with composite rims, that's a harder
> problem, but it's not yet common and there's still plenty of metal.


If it isn't a figure-8 detector loop, then it won't work for bicycles
reliably.

>> Locally we have a large eight lane intersection with a pedestrian
>> and bicycle crossing side by side. Bicycles have slot in the
>> traffic island with a paint marked detector loop with a post on
>> which, instead of a push-button, a 4x6" sign saying "Bicycles Wait
>> Here" which few riders do. Instead they go to the pedestrian walk
>> and press the pedestrian button on the light pole. This triggers
>> the light to stay green for 40 seconds instead of 20 seconds for a
>> bicycle. Meanwhile, everyone waits for the light to change at an
>> empty intersection that a bicyclist crossed.


> That behavior suggests the pickup does not sense the bicycles in the
> way the riders expect; either it does not work well or the delay is
> too long.


Not so. The detector works perfectly and I use it... as do others to
whom I have explained the different phase times. One of the problems
is that the loop is not at the forward edge of the stop area because a
pedestrian path crosses it and most riders want to be up front to make
the start when the light turns green. Besides, as someone already
mentioned, they don't care whether cars have to wait at an empty
intersection.

>> At least today's manual buttons can be slapped with the palm of the
>> hand, unlike the old ones that had a little 3/8" brass button
>> shielded by a raised rim.


> Big buttons certainly help, but they have limitations:


> Going straight through an intersection I'd rather not pass through
> space used by right-turning traffic, if I can avoid it. Buttons are
> no obvious help in situations with dedicated, protected left turn
> lanes unless there's a center island.


You don't have that option anyway in many intersections. For
instance, the one near my house allows left turns from both lanes as
well as straight ahead and right turns from the right lane. If I
don't get there before cars stops on the stop line I prefer to cross
illegally between phases, there being no other way except to dismount
and take pedestrian right-of-way.

Jobst Brandt
 
In article <[email protected]>,
[email protected] writes:

> Locally we have a large eight lane intersection with a pedestrian and
> bicycle crossing side by side. Bicycles have slot in the traffic
> island with a paint marked detector loop with a post on which, instead
> of a push-button, a 4x6" sign saying "Bicycles Wait Here" which few
> riders do. Instead they go to the pedestrian walk and press the
> pedestrian button on the light pole. This triggers the light to stay
> green for 40 seconds instead of 20 seconds for a bicycle. Meanwhile,
> everyone waits for the light to change at an empty intersection that a
> bicyclist crossed.


I won't push the button if I'm the only one crossing, unless
I've had to wait for more than 2 cycles of a timed traffic light.
While I'm waiting I'll look for anyone approaching who can use
the same traffic light. If there's anyone coming, I'll wait for
them and push the button when they're close enough. If there's
nobody coming, I can usually get through a break in the cross traffic
without having to push the button and needlessly delaying people.

A lot of riders will push the button as soon as they get to it,
without checking to see (or caring) if anyone else is approaching
who will "just miss" the light when they get to the intersection.

> At least today's manual buttons can be slapped with the palm of the
> hand, unlike the old ones that had a little 3/8" brass button shielded
> by a raised rim.


We've got some newfangled buttons here (in Vancouver) with
microswitches and a chirp tone & a little red light to give
feedback that it's really been pushed. We also have the
palm-slapper ones, and the old little brass button ones that some
people feel they have to keep depressed until the light changes.

One nice thing about the palm-slap buttons is that you can press
them noisily, so as to let anyone else crossing know that you
really did press the button.


cheers,
Tom

--
-- Nothing is safe from me.
Above address is just a spam midden.
I'm really at: tkeats [curlicue] vcn [point] bc [point] ca
 
On Fri, 03 Mar 2006 20:00:27 +0000, jobst.brandt wrote:

> Patrick Lamb writes:


>> I think you are falling prey to over generalization. I'm a
>> bicyclist, and I don't know of a bicycle lane within 100 miles. I
>> ride in a traffic lane, and when I put a foot down, there's no pole
>> I can reach on which a switch could be mounted. Adjusting the
>> sensitivity of the traffic sensor for a bicycle is simpler, less
>> expensive, and more convenient for the user (me).


Yes, if it works.

What Jobst is referring to are dedicated switches for cyclists, if not on
an existing pole then on their very own. These are common in CA on
high bicycle traffic routes.

> You seem to believe a bicycle should be in the automobile path at a
> traffic light. I find that unnecessarily intrusive, considering that
> when the light turns green, cars behind the bicyclist will progress at
> whatever rate the bicycle chooses.


So what. You can move over if you want after you get rolling, without
delaying anyone more than a split second. I know drivers in CA can be very
impatient and treat a traffic light as a drag race signal. It's easy to
accept this as normal when you live there, but it's not. It's up to them
to pass safely, even if they have to wait to do so. This is the
convention, the law, and the assumption traffic engineers make when
designing roads and traffic control devices.

> Putting a post with push button at
> the roadside for bicyclists (and possibly pedestrians) to activate is
> not an undue expense.


> As for bicycle lanes, I made no assumptions abut anyone else's road
> markings. The discussion was about detector loops for bicycles and I
> believe that most of these are a wasted expense and should generally be
> replaced by push buttons.


I would guess the loops are much cheaper, especially since
they're already being used for motor traffic. So why not just tweak them
so they work for bicycles, instead of spending jillions of dollars to put
buttons at ever intersection?

I talk to VDOT engineers frequently in my advocacy work. If you're
interested I can probably find the answer to this, in cost terms.
Unfortunately I've lost track of my contacts at Caltrans, which would be
more relevent to your area.

Matt O.
 
[email protected] (Tom Keats) wrote in
news:[email protected]:

> One nice thing about the palm-slap buttons is that you can press
> them noisily, so as to let anyone else crossing know that you
> really did press the button.



Tom it is common knowledge that you have to press them 100 times before
the activate, try it.
 
On Fri, 03 Mar 2006 10:08:42 -0500, Matt O'Toole
<[email protected]> wrote:

>On Thu, 02 Mar 2006 18:43:21 +0000, jobst.brandt wrote:
>
>> I think you are falling prey to the same thinking that traffic engineers
>> and many bicyclists use. Bicyclists are not inside motor vehicles and
>> when stopping for a traffic light put a foot down. Placing a manual
>> button at that place is simpler, less ambiguous, less expensive and more
>> convenient for the user. The need for a detector loop for bicyclists is
>> rare, occurring mainly when a bicycle lane lies between other lanes.

>
>This is fine where bikes wait at the right edge of the road. In fact such
>buttons are common around southern CA. But what about left turn lanes?
> They're everywhere, and they should have functioning detector loops
> for cyclists.


And don't forget intersections with right turn lanes. IME, these are
worse than getting off the road to hit a button, because almost none
of the right-turning drivers think of a pedestrian or cyclist wanting
to get across their lane.

Pat

Email address works as is.
 
On 04 Mar 2006 05:33:41 GMT, [email protected] wrote:

>Bob Prohaska writes:
>> Going straight through an intersection I'd rather not pass through
>> space used by right-turning traffic, if I can avoid it. Buttons are
>> no obvious help in situations with dedicated, protected left turn
>> lanes unless there's a center island.

>
>You don't have that option anyway in many intersections. For
>instance, the one near my house allows left turns from both lanes as
>well as straight ahead and right turns from the right lane. If I
>don't get there before cars stops on the stop line I prefer to cross
>illegally between phases, there being no other way except to dismount
>and take pedestrian right-of-way.


I must be missing something. It's entirely reasonable for you to
climb the sidewalk at an intersection to hit a button, but you don't
want to cross at the crosswalk??

Pat

Email address works as is.
 
In article <[email protected]>,
Mike Latondresse <mikelat@no_spam_shaw.ca> writes:
> [email protected] (Tom Keats) wrote in
> news:[email protected]:
>
>> One nice thing about the palm-slap buttons is that you can press
>> them noisily, so as to let anyone else crossing know that you
>> really did press the button.

>
>
> Tom it is common knowledge that you have to press them 100 times before
> the activate, try it.


heh :) Sometimes, 101 times.

Y'know those people who lean on the button until the light changes?
I figure they must have those toilets that don't completely flush
unless you hold the lever down through the whole process.

Maybe some day I'll continue inventorying all the non-bike-route
cyclists' push buttons around town. Some of those include:
E 19th/20th/Findlay/Commercial-Victoria Diversion, Kingsway & Glen,
22nd & Victoria, 23rd & Fraser, 19th & Fraser, 19th & Cambie
(20th or 18th would be better), Prince Edward & King Edward,
Broadway & Pine, and 18th & Main. They'd be nice to have at
41st & Prince Edward.

The one at 37th & Fraser activates instantly. The one at
37th & Main is a dice-toss, and activates only when it
damn well feels like it -- often so you end up stopping
one solitary car after a whole stampede of 'em has gone by.


cheers,
Tom

--
-- Nothing is safe from me.
Above address is just a spam midden.
I'm really at: tkeats [curlicue] vcn [point] bc [point] ca
 
[email protected] wrote:
<snip>
> Instead they go to the pedestrian walk and press the
> pedestrian button on the light pole. This triggers the light to stay
> green for 40 seconds instead of 20 seconds for a bicycle. Meanwhile,
> everyone waits for the light to change at an empty intersection that a
> bicyclist crossed.


Pedestrian buttons work differently in different places. (I don't know
the exact URL, but you could go to www.straightdope.com and search their
archive for it.) Some do nothing at all; some give you a few extra
seconds of green; some actually change the light sooner. Case in point:
there's a light I have to go through every morning that is set at to stay
green the other way for just under two minutes. I think that's
ridiculously long, but the highway department won't change it. I have
tried to trigger the detector, but can't do it, perhaps because my bike
has an aluminum frame. Fortunately, it is simple for me to ride onto the
sidewalk and push the button, which immediately gets me five seconds of
green light. That's the same the detector gives, enough for a single car,
but not nearly adequate for the second car--or for a pedestrian. I wrote
the mayor about this light. He is a former head honcho of both the local
PD and the state highway patrol, and he agrees that pushing the button is
my best alternative at that intersection.
Incidentally, I seem to vaguely remember someone somewhere around here
mentioning a magnet that you can attach to your frame that will trigger
detector loops. Anybody remember any specifics?


Bill


------------------------------------------------
| Blind faith in your leaders--or in anything-- |
| will get you killed. |
| --Bruce Springsteen |
------------------------------------------------
 
Bill who? writes:

>> Instead they go to the pedestrian walk and press the pedestrian
>> button on the light pole. This triggers the light to stay green
>> for 40 seconds instead of 20 seconds for a bicycle. Meanwhile,
>> everyone waits for the light to change at an empty intersection
>> that a bicyclist crossed.


> Pedestrian buttons work differently in different places. (I don't know
> the exact URL, but you could go to www.straightdope.com and search their
> archive for it.) Some do nothing at all; some give you a few extra
> seconds of green; some actually change the light sooner. Case in point:
> there's a light I have to go through every morning that is set at to stay
> green the other way for just under two minutes. I think that's
> ridiculously long, but the highway department won't change it. I have
> tried to trigger the detector, but can't do it, perhaps because my bike
> has an aluminum frame.


As I said, at this crossing over an 8-lane expanse, pedestrians are
allotted twice the time that bicycles are and it's not a guess either.
My point is that many bicycle detector loops are a waste of money and
that this one in particular, not only was a waste but an annoyance to
the bulk of traffic at the intersection concerned. A bicyclist button
would resolve most of the problem.

> Fortunately, it is simple for me to ride onto the sidewalk and push
> the button, which immediately gets me five seconds of green light.


I doubt that it "immediately" turns the light green for you.

> That's the same the detector gives, enough for a single car, but not
> nearly adequate for the second car--or for a pedestrian.


So you found a dumb signal circuit. Most are capable of sensing
whether there are more cars behind the first and stretch the duration
of the green light up to a prescribed maximum.

> I wrote the mayor about this light. He is a former head honcho of
> both the local PD and the state highway patrol, and he agrees that
> pushing the button is my best alternative at that intersection.


Hey! "Are you telling ME, the boss how to do my job." is the usual
response, if not in words, then in action.

> Incidentally, I seem to vaguely remember someone somewhere around
> here mentioning a magnet that you can attach to your frame that will
> trigger detector loops. Anybody remember any specifics?


That's BS, You would need a large field to make a significant signal
in an inductive detector.

Jobst Brandt
 
[email protected] wrote:
<SNIP>

> > Fortunately, it is simple for me to ride onto the sidewalk and push
> > the button, which immediately gets me five seconds of green light.


> I doubt that it "immediately" turns the light green for you.


I push the button, and the light turns yellow as fast as a light turns
on when I flip the switch. Three seconds later it's red. I've done it
hundreds of times. That's immediate enough for me.

> > That's the same the detector gives, enough for a single car, but not
> > nearly adequate for the second car--or for a pedestrian.


> So you found a dumb signal circuit. Most are capable of sensing
> whether there are more cars behind the first and stretch the duration
> of the green light up to a prescribed maximum.


Not this one, at least at this time of day. My point was that there
are a lot of different kinds of pedestrian buttons, inductor loops, and
intersections.

> > I wrote the mayor about this light. He is a former head honcho of
> > both the local PD and the state highway patrol, and he agrees that
> > pushing the button is my best alternative at that intersection.


> Hey! "Are you telling ME, the boss how to do my job." is the usual
> response, if not in words, then in action.


He's actually a pretty nice guy. He did change some other things I
complained about: a bike-eater storm sewer grate in a bike lane, and a
pedestrian crossing button in another place near a hospital that didn't
give wheelchair patients enough time to cross a busy road.

> > Incidentally, I seem to vaguely remember someone somewhere around
> > here mentioning a magnet that you can attach to your frame that will
> > trigger detector loops. Anybody remember any specifics?


> That's BS, You would need a large field to make a significant signal
> in an inductive detector.


Long time since I took electronics, but I do seem to remember that
magnets behave differently than masses of unmagnetized metal. Any magnet
that moves near a conductor will cause a current to flow in the conductor.
I have no idea whether road sensors are sensitive enough for a
bike-mounted magnet to trigger them, and I'm kind of skeptical, but I do
remember reading about such a claim here. I was hoping someone would have
actual experience one way or the other with trying one. Someone who
wouldn't just say that the idea was BS, but who would have actually
tried the magnet.

Incidentally, in case anyone is interested in it, here's the Straight
Dope's article on pedestrian buttons:

http://www.straightdope.com/classics/a3_153.html


Bill, or if you prefer, you may call me Emperor William the Ineffable



__o | Every person takes the limits of their own field
_`\(,_ | of vision for the limits of the world.
(_)/ (_) | - Arthur Schopenhauer
 
Bill who?

>>> Fortunately, it is simple for me to ride onto the sidewalk and
>>> push the button, which immediately gets me five seconds of green
>>> light.


>> I doubt that it "immediately" turns the light green for you.


> I push the button, and the light turns yellow as fast as a light
> turns on when I flip the switch. Three seconds later it's red.
> I've done it hundreds of times. That's immediate enough for me.


Either you are operating at a time when there is no cross traffic or
you have a miserable signal controller.

>>> That's the same the detector gives, enough for a single car, but
>>> not nearly adequate for the second car--or for a pedestrian.


>> So you found a dumb signal circuit. Most are capable of sensing
>> whether there are more cars behind the first and stretch the
>> duration of the green light up to a prescribed maximum.


> Not this one, at least at this time of day. My point was that there
> are a lot of different kinds of pedestrian buttons, inductor loops,
> and intersections.


As I said, what time of day? I've got crossings like that all over
this area when it's in the wee hours.

>>> I wrote the mayor about this light. He is a former head honcho of
>>> both the local PD and the state highway patrol, and he agrees that
>>> pushing the button is my best alternative at that intersection.


>> Hey! "Are you telling ME, the boss how to do my job." is the usual
>> response, if not in words, then in action.


> He's actually a pretty nice guy. He did change some other things I
> complained about: a bike-eater storm sewer grate in a bike lane, and
> a pedestrian crossing button in another place near a hospital that
> didn't give wheelchair patients enough time to cross a busy road.


Both of those could have caused him grief with the law once he was
made aware of them (as you did). It's not the kindness of his heart
but the law of which he is aware.

>>> Incidentally, I seem to vaguely remember someone somewhere around
>>> here mentioning a magnet that you can attach to your frame that will
>>> trigger detector loops. Anybody remember any specifics?


>> That's BS, You would need a large field to make a significant
>> signal in an inductive detector.


> Long time since I took electronics, but I do seem to remember that
> magnets behave differently than masses of unmagnetized metal. Any
> magnet that moves near a conductor will cause a current to flow in
> the conductor. I have no idea whether road sensors are sensitive
> enough for a bike-mounted magnet to trigger them, and I'm kind of
> skeptical, but I do remember reading about such a claim here. I was
> hoping someone would have actual experience one way or the other
> with trying one. Someone who wouldn't just say that the idea was
> BS, but who would have actually tried the magnet.


Urban legend. The magnetic field is essentially a point source,
strong as it may be and falls off to effectively zero at the distances
involved in detector loops. The loop can only see a dPHI/dt and that
ain't much for a hand held device. "near to a conductor" is the
defining operator. You would need to slide the magnet on the ground.

Jobst Brandt