OT: red light jumping



C

Colin Blackburn

Guest
In the many threads on this I have probably more than once mentioned the
lights near my office as a prime example of a set of lights where
motorists routinely go through red lights. Well, this morning as I
approached the lights, walking, they were just changing. As the red
light came on the lead car came to a halt so quickly it skidded several
yards over the stop line. The sound was quite dramatic. Interestingly
there was a police car parked on the opposite corner attending a broken
down car.

Colin
 
Colin Blackburn wrote:
> As the red light came on the lead car came to a halt so quickly it
> skidded several yards over the stop line. The sound was quite
> dramatic. Interestingly there was a police car parked on the
> opposite corner attending a broken down car.


The police of course were probably much more attracted by the
"emergency" stop than they would have been by routine red light
running. The cager's poor decision making and lack of observation are
more worrying to me than his simple disregard for the lights.
--
Dave...
 
dkahn400 wrote:
>
> The police of course were probably much more attracted by the
> "emergency" stop than they would have been by routine red light
> running. The cager's poor decision making and lack of observation are
> more worrying to me than his simple disregard for the lights.


There is a problem that the amber phase of many lights is shorter than
the reaction and stopping time so that depending on where you are when
the light changes to amber and how fast you were going at the time, this
can be the consequence if you try to stop

Tony
 
Tony Raven wrote:
> dkahn400 wrote:
>
>>
>> The police of course were probably much more attracted by the
>> "emergency" stop than they would have been by routine red light
>> running. The cager's poor decision making and lack of observation are
>> more worrying to me than his simple disregard for the lights.

>
>
> There is a problem that the amber phase of many lights is shorter than
> the reaction and stopping time so that depending on where you are when
> the light changes to amber and how fast you were going at the time, this
> can be the consequence if you try to stop


Are you really suggesting that somebody driving at a legal speed towards
a green traffic light can find that the light changes to amber *and
red*, before the vehicle passes the stop line, so quickly that the
vehicle cannot be brought to a controlled stop before the stop line?

I'd be interested to know of some examples.

--
Joe * If I cannot be free I'll be cheap
 
JLB wrote:
>
> Are you really suggesting that somebody driving at a legal speed towards
> a green traffic light can find that the light changes to amber *and
> red*, before the vehicle passes the stop line, so quickly that the
> vehicle cannot be brought to a controlled stop before the stop line?
>
> I'd be interested to know of some examples.
>


I am indeed. Its known as the dilemma zone - you are too close to stop
at the line but too far away to cross the line before the light goes red
An example of TRL research is:

> UG131: REAPPRAISAL OF STOPPING AMBER
>
> TRL reviewed the operation of traffic signals on high-speed roads, where the 3-second duration of the amber signal has caused concern. Some drivers may find it difficult to know whether or when to brake sharply and stop, or to risk being recorded by cameras for crossing traffic lights at red. Simulator trials at TRL by a sample of 40 motorists recorded and compared reactions of drivers to the use of an advanced warning of signals changing from green to amber, with the normal arrangement where there is the standard sequence. The simulations included a variety of situations whereby the presence of other vehicles would be influencing the driver's decisions.
>
> The research conclusion was that i) advance warning of a change to amber should not be used, ii) the current value of 3 seconds stopping amber should be retained for traffic signals, even on high speed roads, the all red period being adjusted to provide a safe change between opposing flows and iii) on red light cameras a minimum activation time, following the amber, should be commensurate with the safe operation of the signals.
>
> Consequently, no change to the policy of 3-seconds amber is recommended. The implications for the setting of red-light cameras on high-speed roads are being considered.


For these purposes "high speed" means over 35mph and the clear
recommendation is 3 seconds is too short for stopping reliably within
the amber so accept that cars will cross at red and set the trigger
delay on the red light cameras so as to not catch them and run a longer
dual red phase so they won't hit crossing traffic.


Tony
 
Tony Raven wrote:
> JLB wrote:
>
>>
>> Are you really suggesting that somebody driving at a legal speed
>> towards a green traffic light can find that the light changes to amber
>> *and red*, before the vehicle passes the stop line, so quickly that
>> the vehicle cannot be brought to a controlled stop before the stop line?
>>
>> I'd be interested to know of some examples.
>>

>
> I am indeed. Its known as the dilemma zone - you are too close to stop
> at the line but too far away to cross the line before the light goes red
> An example of TRL research is:
>
>> UG131: REAPPRAISAL OF STOPPING AMBER
>>
>> TRL reviewed the operation of traffic signals on high-speed roads,
>> where the 3-second duration of the amber signal has caused concern.
>> Some drivers may find it difficult to know whether or when to brake
>> sharply and stop, or to risk being recorded by cameras for crossing
>> traffic lights at red. Simulator trials at TRL by a sample of 40
>> motorists recorded and compared reactions of drivers to the use of an
>> advanced warning of signals changing from green to amber, with the
>> normal arrangement where there is the standard sequence. The
>> simulations included a variety of situations whereby the presence of
>> other vehicles would be influencing the driver's decisions.
>>
>> The research conclusion was that i) advance warning of a change to
>> amber should not be used, ii) the current value of 3 seconds stopping
>> amber should be retained for traffic signals, even on high speed
>> roads, the all red period being adjusted to provide a safe change
>> between opposing flows and iii) on red light cameras a minimum
>> activation time, following the amber, should be commensurate with the
>> safe operation of the signals.
>>
>> Consequently, no change to the policy of 3-seconds amber is
>> recommended. The implications for the setting of red-light cameras on
>> high-speed roads are being considered.

>
>
> For these purposes "high speed" means over 35mph and the clear
> recommendation is 3 seconds is too short for stopping reliably within
> the amber so accept that cars will cross at red and set the trigger
> delay on the red light cameras so as to not catch them and run a longer
> dual red phase so they won't hit crossing traffic.


Thanks! I do find that interesting. To paraphrase the suggested policy,
going through a red light on such junctions will not bring about
enforcement action until after the red has been on for some period of
time. It looks as though people will not be told what that period of
time is. This is not exactly ideal from several points of view, though I
can see why some other options might be thought even worse.

The best answer would be a blanket national speed limit of 30 mph.
Except for cyclists.
--
Joe * If I cannot be free I'll be cheap
 
JLB wrote:
>
> Thanks! I do find that interesting. To paraphrase the suggested policy,
> going through a red light on such junctions will not bring about
> enforcement action until after the red has been on for some period of
> time. It looks as though people will not be told what that period of
> time is. This is not exactly ideal from several points of view, though I
> can see why some other options might be thought even worse.
>
> The best answer would be a blanket national speed limit of 30 mph.
> Except for cyclists.


Even that wouldn't work. A three second delay works if you are staring
at the light waiting for it to change. If you have just checked your
mirrors or you are keeping an eye on a pedestrian waiting to cross the
road, you can easily add another second before the light changing will
register. The TRL problem is they did it in lab conditions not in real
life busy road environment lots of other things to be dealing with
environment. In those environment delay times before people start
braking can be a couple of seconds or more rather than the normal
3-600ms reaction time.

Tony
 
Tony Raven wrote:
> JLB wrote:
>
>>
>> Thanks! I do find that interesting. To paraphrase the suggested
>> policy, going through a red light on such junctions will not bring
>> about enforcement action until after the red has been on for some
>> period of time. It looks as though people will not be told what that
>> period of time is. This is not exactly ideal from several points of
>> view, though I can see why some other options might be thought even
>> worse.
>>
>> The best answer would be a blanket national speed limit of 30 mph.
>> Except for cyclists.

>
>
> Even that wouldn't work.


I did not say, nor did I think, that it would work all the time for
everyone, solve all problems or produce peace, happiness and eternal
bliss on earth. I was only suggesting that it would be the best of the
practical options available. It would work for most people most of the
time (except for those who want to exceed 30 mph, who might not be too
happy, but, hey, stuff them).

A three second delay works if you are staring
> at the light waiting for it to change. If you have just checked your
> mirrors or you are keeping an eye on a pedestrian waiting to cross the
> road, you can easily add another second before the light changing will
> register. The TRL problem is they did it in lab conditions not in real
> life busy road environment lots of other things to be dealing with
> environment. In those environment delay times before people start
> braking can be a couple of seconds or more rather than the normal
> 3-600ms reaction time.


At 30 mph a driver who is distracted for two seconds has travelled about
27 metres (apologies for the inconsistent units) without paying
attention to relevant matters. I don't suppose anyone has credible
figures for how frequently drivers do this. However, any problems this
causes at 30 mph will be proportionately worse at higher speeds; and the
minimum braking distance increases more than proportionately.


--
Joe * If I cannot be free I'll be cheap
 
Tony Raven wrote:
>>There is a problem that the amber phase of many lights is shorter than
>>the reaction and stopping time so that depending on where you are when
>>the light changes to amber and how fast you were going at the time, this
>>can be the consequence if you try to stop


JLB wrote:
> Are you really suggesting that somebody driving at a legal speed towards
> a green traffic light can find that the light changes to amber *and
> red*, before the vehicle passes the stop line, so quickly that the
> vehicle cannot be brought to a controlled stop before the stop line?
>
> I'd be interested to know of some examples.


I've had this happen on my bike. It was a couple of years ago on
Warmley High Street. I would have been riding at about 20-25mph, when a
ped pressed the button at a pelican crossing. The light changed
instantly to amber. I was still far enough from the crossing that I
*could* have stopped, if I'd slammed everything on, but I decided it was
safer not to try (AFAIR the road was wet, which would have influenced my
decision). Then the light changed to red before I actually reached the
crossing - I had no choice but to carry on through the red light.

Most embarrassing.

--
Danny Colyer (the UK company has been laughed out of my reply address)
<URL:http://www.speedy5.freeserve.co.uk/danny/>
"He who dares not offend cannot be honest." - Thomas Paine
 
Tony Raven wrote:
>>There is a problem that the amber phase of many lights is shorter than
>>the reaction and stopping time so that depending on where you are when
>>the light changes to amber and how fast you were going at the time, this
>>can be the consequence if you try to stop


JLB wrote:
> Are you really suggesting that somebody driving at a legal speed towards
> a green traffic light can find that the light changes to amber *and
> red*, before the vehicle passes the stop line, so quickly that the
> vehicle cannot be brought to a controlled stop before the stop line?
>
> I'd be interested to know of some examples.


I've had this happen on my bike. It was a couple of years ago on
Warmley High Street. I would have been riding at about 20-25mph, when a
ped pressed the button at a pelican crossing. The light changed
instantly to amber. I was still far enough from the crossing that I
*could* have stopped, if I'd slammed everything on, but I decided it was
safer not to try (AFAIR the road was wet, which would have influenced my
decision). Then the light changed to red before I actually reached the
crossing - I had no choice but to carry on through the red light.

Most embarrassing.

--
Danny Colyer (the UK company has been laughed out of my reply address)
<URL:http://www.speedy5.freeserve.co.uk/danny/>
"He who dares not offend cannot be honest." - Thomas Paine
 
JLB wrote:
>
> At 30 mph a driver who is distracted for two seconds has travelled about
> 27 metres (apologies for the inconsistent units) without paying
> attention to relevant matters. I don't suppose anyone has credible
> figures for how frequently drivers do this.


Well a good driver should be regularly checking his mirrors and that
takes 0.5 - 1s per time and I don't think anyone would suggest a
competent driver should not do so. Meanwhile they have travelled 14-27m
without looking forward except for peripheral vision.

Tony
 
MadDog62 wrote:
> Maybe I am missing something here but the Police don´t give a **** about
> red lights and I NEVER expect anyone to obey red lights. I am a cyclist
> and usually better off and safer ignoring the red lights sometimes. I
> behave though. I just don´t hang up my life on stats and lab tests.


Possibly you are missing something here, since the thread began with a
description of a car going through a red light despite doing its best to
stop; that led to a discussion of the consequences of the standard
three-second period of amber, and how it has been shown that for
junctions where drivers approach at speeds in excess of 35 mph many
drivers will not be able to halt in time for the change to red, leading
to a recommendation that at such junctions the other lights stay red for
long enough to let the stragglers through before the next group of
drivers enter the junction. Hence it was pretty much off topic, because
it was about vehicle traffic and had little to say about cyclists or
cycling.

Just a small suggestion: on Usenet, including urc, it's customary to
place the new post below whatever is being quoted, rather than
top-posting. It helps readers to get the context before reading the new
contribution.

--
Joe * If I cannot be free I'll be cheap
 
On Mon, 13 Dec 2004 16:09:31 +0000,
JLB <[email protected]> wrote:
>
> Are you really suggesting that somebody driving at a legal speed towards
> a green traffic light can find that the light changes to amber *and
> red*, before the vehicle passes the stop line, so quickly that the
> vehicle cannot be brought to a controlled stop before the stop line?
>
> I'd be interested to know of some examples.
>

I think the amber phase of most lights is of the order of 3 seconds.

Allow a respectable 0.5g braking and you get a speed a little over
30mph that you can stop from.

But at that same 30mph you will do 132 feet if you don't brake at all
so, in theory, at 30mph you shouldn't have a problem. More than 100 feet
away and you can stop easily, less than 100 feet and you can cross the
junction easily. (I'm ignoring thinking times here)


According to the highway code it takes 4.8 seconds to stop from 70mph.

And in three seconds you will travel 300 feet which is just about the
highway code stopping distance.

But you have to make a split second decision to emergency stop or carry
on. Further away isn't usually a problem - you will stop easily, it's
when it starts getting a bit closer. Say you are 200 feet away when the
lights change - if you don't brake it won't be a problem. If you do
brake you won't be able to stop before the line but there will be time
for the lights to go to red before you get to the line.

Add on the 0.7 seconds of reaction time (you are expecting the lights to
change so going for the lower limits of reaction is reasonable) and at
70mph distances from 306 to 315 feet are lose-lose - you can't stop in
time but you can't get across before the lights turn to red.

There are a few dual carriageways with lights - it's scary when you
start to brake from 70mph, realize you won't stop, back on the throttle
at about 50mph and then see the light go to red a long time before you
actually get to them.

Now on any fast road when approaching traffic lights I always have a
"point of no return" in my mind after which I ignore the lights
(other than being ready for someone coming the other way to jump their
lights but that can happen regardless of what my lights are saying)

Tim.


--
God said, "div D = rho, div B = 0, curl E = - @B/@t, curl H = J + @D/@t,"
and there was light.

http://tjw.hn.org/ http://www.locofungus.btinternet.co.uk/
 
Tim Woodall wrote:
>
> I think the amber phase of most lights is of the order of 3 seconds.
>
> Allow a respectable 0.5g braking and you get a speed a little over
> 30mph that you can stop from.
>


You've forgotten to allow for reaction times before you start braking.
You need at least 1.25s for that.
http://www.visualexpert.com/Resources/reactiontime.html


Tony
 
On Mon, 13 Dec 2004 22:34:48 +0000, Tony Raven <[email protected]>
wrote:

>a good driver should be regularly checking his mirrors and that
>takes 0.5 - 1s per time and I don't think anyone would suggest a
>competent driver should not do so. Meanwhile they have travelled 14-27m
>without looking forward except for peripheral vision.


Surely the inevitable result here is that a child will run out and be
mown down? Or is that only when they look at their speedo?

Incidentally, it is good practice to slow down and cover the brake
when approaching light-controlled junctions. Or so my driving
instructor told me.

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

88% of helmet statistics are made up, 65% of them at CHS, Puget Sound
 
Provence1971 wrote:
> The police officer did nothing of course :mad:
>

To be fair he was actually dealing with another incident at the time.

Colin
 
Just zis Guy, you know? wrote:
> On Mon, 13 Dec 2004 22:34:48 +0000, Tony Raven <[email protected]>
> wrote:
>
>
>>a good driver should be regularly checking his mirrors and that
>>takes 0.5 - 1s per time and I don't think anyone would suggest a
>>competent driver should not do so. Meanwhile they have travelled 14-27m
>>without looking forward except for peripheral vision.

>
>
> Surely the inevitable result here is that a child will run out and be
> mown down? Or is that only when they look at their speedo?
>
> Incidentally, it is good practice to slow down and cover the brake
> when approaching light-controlled junctions. Or so my driving
> instructor told me.
>
> Guy


Indeed. There's a set of lights on a dual carriageway near me. The
limit as you approach the lights (they are only applicable to one
direction) is 70. The lights themselves change fast, and seem to be set
up such that the both-red period can't be more than a second, if that.
I suspect in actual fact the side-junction is leery of pulling out till
they can see the cars on the dual have actually stopped, in practice, so
quite possible (as with other light-controlled junctions around here)
that there is no all-red period to them.

Anyway, the upshot is that there are those who will not slow approaching
the lights and will continue through on red, because they can't stop in
the time it takes to react/brake when they go amber (not very long spent
on amber, either, it has to be said). This isn't a problem, because if
they *genuinely* are too close, then they'll be through before the side
road moves off - because it's well known. Others will slow approaching
it, or are slowed by those in front filtering off to the left at the
junction - the majority of those will stop in time, due to having
already lifted off the throttle, if not already covering the brake.

Having driven through this junction many many *many* times, I've had
ample opportunity to try different tactics. What I've perceived whilst
doing this is that either all my cars stop on a sixpence (unlikely) or
it's a case that people don't know how fast their cars *could* stop if
they really wanted them to (and if you've been using your mirror, you
know if you'll end up with a following-too-closer in your boot if you
throw the anchor out). And I don't class the stops I do as emergency
ones - I could brake a fair bit sharper if it really was a 'stop or kill
something' situation. Many of my decisions to continue through are due
to it being a close call - being able to stop in time but being followed
too close. Often I'll stop, and a car or two in the lane next to me
will go flying through. If I approach at 70, though, without having
lifted off and the speed dropping, then I know I don't have enough time
to stop even though I know the instant the lights change I'll see them
(if you're coming up this fast to lights, you have already looked in
your mirrors and are looking at the lights/side junction!).

Not all 70mph trafficlight junctions have this characteristic of
approaching at 70 - in this instance it's the only one on the length of
the dual carriageway, the lanes merge to 1 shortly after as you climb up
a hill, and who knows what's about to turn out of the side junction and
then crawl up said hill at snails pace.

That said, there's a lot of lazyness around, and I think it's lazyness
and not knowing how fast you can stop your car that contributes to the
majority of red-light running on this junction, and I'm willing to bet
on others, too. Red light running is so common these days that I'm wary
of moving off till I can see the other traffic is slowing enough that
they'll stop, and not just wavering in indecision. This makes the
actual throughput of junctions even lower than normal, of course, and
increases congestion ;-)

--


Velvet
 
On Mon, 13 Dec 2004 22:34:48 +0000, Tony Raven <[email protected]>
wrote:

>JLB wrote:
>>
>> At 30 mph a driver who is distracted for two seconds has travelled about
>> 27 metres (apologies for the inconsistent units) without paying
>> attention to relevant matters. I don't suppose anyone has credible
>> figures for how frequently drivers do this.

>
>Well a good driver should be regularly checking his mirrors and that
>takes 0.5 - 1s per time and I don't think anyone would suggest a
>competent driver should not do so.


AND also ANTICIPATE that the lights may change to red, and be prepared
to stop in time in that case.
 
On Tue, 14 Dec 2004 07:44:43 +0000, Tony Raven <[email protected]>
wrote:

>Tim Woodall wrote:
>>
>> I think the amber phase of most lights is of the order of 3 seconds.
>>
>> Allow a respectable 0.5g braking and you get a speed a little over
>> 30mph that you can stop from.
>>

>
>You've forgotten to allow for reaction times before you start braking.
>You need at least 1.25s for that.
>http://www.visualexpert.com/Resources/reactiontime.html
>
>


OR 8.3 seconds if you have had 6 pints, 4 whiskies and a joint :-|

Believe me they are out there !!
 
Just zis Guy, you know? wrote:
>
> Incidentally, it is good practice to slow down and cover the brake
> when approaching light-controlled junctions. Or so my driving
> instructor told me.
>
> Guy


The slowing down will help but covering the brake only saves about 200ms.

Tony