Traffic lights and cyclists

Discussion about street lighting, road signs, traffic signals - and all other street furniture - goes here.

Moderator: Site Management Team

Herned
Member
Posts: 1363
Joined: Thu Jul 20, 2017 09:15

Traffic lights and cyclists

Post by Herned »

During lockdown I have been out cycling quite a bit... one thing I have noticed is that many traffic lights do not detect cyclists, so if no cars want that direction the green arrow for the direction I want is skipped. This has much more of an issue when traffic has been quieter. Obviously I can just go when safe, but I would prefer to be able to obey the signals and not give any comfort to people who believe cyclists are a menace and don't obey the laws

Is it that detection loops just don't work for cyclists, or they are set up incorrectly? Is this a common issue?
User avatar
FosseWay
Assistant Site Manager
Posts: 19621
Joined: Thu Feb 23, 2006 22:26
Location: Gothenburg, Sweden

Re: Traffic lights and cyclists

Post by FosseWay »

It's a common issue IME.

The simple answer is part time signals. If there's enough traffic to warrant having signals in operation, then you probably wouldn't notice on a bike that you weren't triggering the circuit, as there would be motor vehicles ahead of and/or behind you that do trigger it. If you're finding that you're sitting at an untriggered red light for long enough to consider going anyway, that suggests traffic is very light. Equally, notwithstanding the law which is very clear on the subject of red lights, if you in fact can go against red without putting yourself or others at risk, that also suggests that the traffic levels don't warrant lights.

I've never understood the British objection to having part time lights at normal (i.e. non-roundabout) intersections, with a flashing orange light at all entry points.

If cycle traffic disproportionately comes from one entry point, or if one entry point is disproportionately lightly used by motor vehicles compared to the others, you can also default to green on that arm. There'd be plenty of motor vehicles on the other arms to trigger the signal for those arms, but once they'd had green for a given time, it would default back to green for the side road.

Or councils can stop being incompetent and install stuff that works. Supposedly we have pedestrian crossings that react dynamically to whether there is actually someone on a crossing and to how fast they are clearing it. If technology can do that, it ought to be able to detect the same person seated on 10-20 kg of metal.
Did you know there's more to SABRE than just the Forums?
Add your roads knowledge to the SABRE Wiki today!
Have you browsed SABRE Maps recently? Try getting involved!
User avatar
Chris Bertram
Member
Posts: 15744
Joined: Tue Nov 13, 2001 12:30
Location: Birmingham, England

Re: Traffic lights and cyclists

Post by Chris Bertram »

“The quality of any advice anybody has to offer has to be judged against the quality of life they actually lead.” - Douglas Adams.

Did you know there's more to SABRE than just the Forums?
Add your roads knowledge to the SABRE Wiki today!
Have you browsed SABRE Maps recently? Try getting involved!
User avatar
jervi
Member
Posts: 1596
Joined: Thu Aug 17, 2017 16:29
Location: West Sussex

Re: Traffic lights and cyclists

Post by jervi »

Like you I have been out cycling a lot since the start of lockdown. I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle. If the loops or IR cameras fail to detect you, you can report it to the highway authority and they will fix it (keep pestering them if they don't).

Sometimes signals will go through a phase where no traffic was detected. This is usually only done for busy arms at busy junctions incase there is a failure in the loops. For example when the A14 Huntingdon Bypass first opened, the junction with the now A1307 and A141 was giving the closed slip road a green light, despite there being no vehicles, I believe that a traffic light engineer changed the phases in the controller shortly after due to the queues.

When I get stuck at a red light and it hasn't detected me and there is little traffic around to trigger my phase, I either proceed through the red light with caution (naughty me), or dismount and walk across the junction. It is somewhat important to try and go across all the loops on the approach too, if you are too close to the kerb then it may not detect you.
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Traffic lights and cyclists

Post by traffic-light-man »

It's a fairly typical problem that some loops are not sensitive enough to detect cycles, which is also a problem that particularly persists with non-metallic cycles for obvious reasons. An 'easy' fix to both those problems would perhaps be an above ground detector, which has a decent chance of detecting a pedestrian approaching at pace, so will usually not have an issue with cycles.

I was at a site with radar detectors in the road surface (in the same style as magnetometer) the other day, which react particularly well to cycles as they're expecting to 'see' an object pass above them rather than dealing with metallic objects and magnetic fields. That particular site is also fitted with above ground radar detectors and the cycle-only arm has an additional a PBU for the cycles too.
jervi wrote: Thu Apr 22, 2021 15:58Sometimes signals will go through a phase where no traffic was detected. This is usually only done for busy arms at busy junctions incase there is a failure in the loops. For example when the A14 Huntingdon Bypass first opened, the junction with the now A1307 and A141 was giving the closed slip road a green light, despite there being no vehicles, I believe that a traffic light engineer changed the phases in the controller shortly after due to the queues.
There's a number of reasons this could happen, but I think the chances are that it's either being forced through the cycle by something plan-based such as Fixed Time, SCOOT or CLF operation, or that the detectors have 'broken'. I say 'broken', because they're usually set to fail active, and will do so after a specified time. If you have a road closure in place, depending on what time has been specified, there's a likelihood the lack of traffic will cause the detectors to fail as they won't have detected anything for so long, and therefore failing active will constantly input a demand and extend the green time to its maximum time.
Simon
Al__S
Member
Posts: 491
Joined: Sat Mar 04, 2017 10:56

Re: Traffic lights and cyclists

Post by Al__S »

most induction loops can be adjusted such that they will reliably detect almost all bikes. I know this, as for all their faults Cambridgeshire County Council highways department do actually manage to do so.
Herned
Member
Posts: 1363
Joined: Thu Jul 20, 2017 09:15

Re: Traffic lights and cyclists

Post by Herned »

jervi wrote: Thu Apr 22, 2021 15:58 Like you I have been out cycling a lot since the start of lockdown. I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle. If the loops or IR cameras fail to detect you, you can report it to the highway authority and they will fix it (keep pestering them if they don't).
I hadn't actually thought to do that, I will do that and see if anything happens
Herned
Member
Posts: 1363
Joined: Thu Jul 20, 2017 09:15

Re: Traffic lights and cyclists

Post by Herned »

jervi wrote: Thu Apr 22, 2021 15:58 Like you I have been out cycling a lot since the start of lockdown. I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle. If the loops or IR cameras fail to detect you, you can report it to the highway authority and they will fix it (keep pestering them if they don't).
I hadn't actually thought to do that, I will do that and see if anything happens. There is a set just down the road from my house which I have tried repeatedly taking care to roll over the loop, including going backwards and forwards over it to get it to do something, to no avail
Last edited by Herned on Thu Apr 22, 2021 21:43, edited 1 time in total.
WHBM
Member
Posts: 9707
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Traffic lights and cyclists

Post by WHBM »

TfL decided that to suit cyclists they would move to fixed stages 24x7. Which is so wasteful at quiet times. I suspect it was the cheapest thing to do to the controller.

Have always thought that a variation on the pedestrian push button unit alongside the cycle stopping point would be the most effective.
User avatar
jervi
Member
Posts: 1596
Joined: Thu Aug 17, 2017 16:29
Location: West Sussex

Re: Traffic lights and cyclists

Post by jervi »

The best solution to the problem is having some sort of indicator that shows wether the controller is aware of waiting traffic on that arm. If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected. Some junctions in Brighton and Hove have push buttons, however these are current exclusive to filtered turns (which have their own phase) where cycles would not go over any loops when using the cycle tracks/lanes to the side (they move over the to the right turn filter lane within the bike box)
cb a1
Member
Posts: 5361
Joined: Wed Jun 30, 2004 07:30

Re: Traffic lights and cyclists

Post by cb a1 »

jervi wrote: Thu Apr 22, 2021 15:58I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle.
To correct you if you are wrong would involve knowing every law which I doubt anyone does. When it comes to proving a negative, the burden of proof generally sits with the person wanting to prove the positive.

Thus, IMO, the onus is on you to prove there is a law that traffic signals must be able to detect any bicycle.

I'm certainly not an expert in traffic signals and the law around them, but in all my years working in the industry, this is the first time I've ever seen this suggestion. I'm certainly not claiming there isn't and I would genuinely be fascinated to see such the reference and wording of such a law if it exists.
Education makes the wise slightly wiser, but it makes the fool vastly more dangerous. N. Taleb
We tend to demand impossible standards of proof from our opponents but accept any old rubbish to support our beliefs.
The human paradox that is common sense
The Backfire Effect
User avatar
solocle
Member
Posts: 806
Joined: Thu Jun 02, 2016 18:27

Re: Traffic lights and cyclists

Post by solocle »

I had a similar issue last April. I'd taken to carefully jumping the traffic lights (almost invariably Somerset, so not particularly busy at the best of times) on my road bike. Once or twice met something coming the other way on light controlled single lane stretches and never had an issue pulling over to let them pass... in fact I'd say there's more conflict when the light timings don't consider cyclists and both parties go through on green!

Anyway, I'm out shopping on my steel bike and I wait (after all, steel should be easy to pick up with a loop). And I wait. And I wait. I'm about to jump the light when a van finally arrives behind, which naturally triggers an immediate change. The kicker? We're both turning left, he decides it's a great idea to overtake me in the 100 yards between that light and a mini roundabout, now with oncoming traffic, and nearly runs me off the road.

As a consequence, I'm a bit more quick to jump to the conclusion that a light isn't going to detect me and proceed. Sometimes perhaps a little too quick... but based on previous experience with the lights along here, I believed that they didn't turn green for cyclists. Shouldn't be an issue, because I'm on the main road, so they default green... until the traffic lights look like they're being actively vindictive and turning red as I approach.
69741A87-3DA2-4647-93C6-7333A13DA034.jpeg
In fact, the only set that didn't do that was the one set that had a car coming the other way.

But it turns out that at least the first set of lights does that at quiet times, to everyone. It actually turns red as cars approach!

Probably some scheme to encourage compliance with the 40 limit, but all it achieved in my case was me continuing at 20, now illegally.
cb a1 wrote: Fri Apr 23, 2021 11:01
jervi wrote: Thu Apr 22, 2021 15:58I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle.
To correct you if you are wrong would involve knowing every law which I doubt anyone does. When it comes to proving a negative, the burden of proof generally sits with the person wanting to prove the positive.

Thus, IMO, the onus is on you to prove there is a law that traffic signals must be able to detect any bicycle.

I'm certainly not an expert in traffic signals and the law around them, but in all my years working in the industry, this is the first time I've ever seen this suggestion. I'm certainly not claiming there isn't and I would genuinely be fascinated to see such the reference and wording of such a law if it exists.
Rule 176, while not being law, does say "If the traffic lights are not working, treat the situation as you would an unmarked junction and proceed with great care." It's not an unreasonable interpretation that not detecting a cyclist counts as "not working".
Last edited by solocle on Fri Apr 23, 2021 11:47, edited 4 times in total.
linuxrocks
Member
Posts: 135
Joined: Tue Jul 19, 2016 15:31

Re: Traffic lights and cyclists

Post by linuxrocks »

jervi wrote: Fri Apr 23, 2021 08:08 If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected. Some junctions in Brighton and Hove have push buttons...
That is what they did in Adelaide where I'm from. The post with the button on was handy too for holding yourself up without having to get out of the toe clips.
AndyB
SABRE Developer
Posts: 11057
Joined: Tue Dec 26, 2006 21:58
Location: Belfast N Ireland
Contact:

Re: Traffic lights and cyclists

Post by AndyB »

I have found that, all other things being equal, you have a better chance of activating the loop if you cycle along the actual loop rather than through it, ie along the tar track where they laid the detection cable.
WHBM
Member
Posts: 9707
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Traffic lights and cyclists

Post by WHBM »

The silly thing is that the old pneumatic detector strips from our forefathers' generation detected bicycles going over them fine. I wonder how magnetic inductive loops ever got accepted if they could not replace all the prior functionality.
User avatar
traffic-light-man
Member
Posts: 4728
Joined: Fri Sep 01, 2006 18:45
Location: Liverpool, UK

Re: Traffic lights and cyclists

Post by traffic-light-man »

jervi wrote: Fri Apr 23, 2021 08:08 The best solution to the problem is having some sort of indicator that shows wether the controller is aware of waiting traffic on that arm. If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected.
I'm championing this idea initially for dedicated cycle facilities, but it'll probably fall on deaf ears, unfortunately. My suggestion is a white aspect with a 'WAIT' mask in it for both LLCS and standard signals, in both stand-alone and incorporated-in-signal options to give some flexibility, providing DfT would be happy to approve those. Basically, the same idea as the Dutch-style 'WACHT' and the 'Bitte warten'/'WARTE' aspects found in some German-speaking countries.

At the junction I mentioned in my last post, exactly as you describe, the demand confirm ring (WAIT, of old) illuminates when either of the radars detect cyclists, but I think it'd be nice to give a clear response to approaching cyclists either way.
solocle wrote: Fri Apr 23, 2021 11:04 Probably some scheme to encourage compliance with the 40 limit, but all it achieved in my case was me continuing at 20, now illegally.
Would it be possible that they're 'rest-in-red' (as you've suggested elsewhere in your post), and it's just that the approaching vehicle has just missed the window in which the controller has decided to move to the all-red revert, meaning it's then got to clear that all-red before moving back to that stage?

There's one on the A580 that used to do that. If you just missed the window and it moved to revertive all-red, you'd just be coming to a complete stop from 50mph by the time it dropped back to green. I think it's since been put on to MOVA, and even approaching at 50mph, you're probably only down to around 30mph by the time it drops back to green. That all depends on when you start slowing mind you, I tend to ease off fairly early when it's quiet to give signals time to change stages before I have to come to a complete stop.
Simon
User avatar
jervi
Member
Posts: 1596
Joined: Thu Aug 17, 2017 16:29
Location: West Sussex

Re: Traffic lights and cyclists

Post by jervi »

cb a1 wrote: Fri Apr 23, 2021 11:01
jervi wrote: Thu Apr 22, 2021 15:58I have come across a few signals that fail to detect me, however I believe by law (correct me if I am wrong) all lanes that allow cycle traffic must be able to detect any bicycle.
To correct you if you are wrong would involve knowing every law which I doubt anyone does. When it comes to proving a negative, the burden of proof generally sits with the person wanting to prove the positive.

Thus, IMO, the onus is on you to prove there is a law that traffic signals must be able to detect any bicycle.

I'm certainly not an expert in traffic signals and the law around them, but in all my years working in the industry, this is the first time I've ever seen this suggestion. I'm certainly not claiming there isn't and I would genuinely be fascinated to see such the reference and wording of such a law if it exists.
I heard it from someone else (not sure if it was on here or elsewhere). It could be written in guidance/standards somewhere which is defacto law.

traffic-light-man wrote: Fri Apr 23, 2021 14:26
jervi wrote: Fri Apr 23, 2021 08:08 The best solution to the problem is having some sort of indicator that shows wether the controller is aware of waiting traffic on that arm. If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected.
I'm championing this idea initially for dedicated cycle facilities, but it'll probably fall on deaf ears, unfortunately. My suggestion is a white aspect with a 'WAIT' mask in it for both LLCS and standard signals, in both stand-alone and incorporated-in-signal options to give some flexibility, providing DfT would be happy to approve those. Basically, the same idea as the Dutch-style 'WACHT' and the 'Bitte warten'/'WARTE' aspects found in some German-speaking countries.
That sounds like a good idea to me!
User avatar
solocle
Member
Posts: 806
Joined: Thu Jun 02, 2016 18:27

Re: Traffic lights and cyclists

Post by solocle »

traffic-light-man wrote: Fri Apr 23, 2021 14:26
jervi wrote: Fri Apr 23, 2021 08:08 The best solution to the problem is having some sort of indicator that shows wether the controller is aware of waiting traffic on that arm. If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected.
I'm championing this idea initially for dedicated cycle facilities, but it'll probably fall on deaf ears, unfortunately. My suggestion is a white aspect with a 'WAIT' mask in it for both LLCS and standard signals, in both stand-alone and incorporated-in-signal options to give some flexibility, providing DfT would be happy to approve those. Basically, the same idea as the Dutch-style 'WACHT' and the 'Bitte warten'/'WARTE' aspects found in some German-speaking countries.

At the junction I mentioned in my last post, exactly as you describe, the demand confirm ring (WAIT, of old) illuminates when either of the radars detect cyclists, but I think it'd be nice to give a clear response to approaching cyclists either way.
solocle wrote: Fri Apr 23, 2021 11:04 Probably some scheme to encourage compliance with the 40 limit, but all it achieved in my case was me continuing at 20, now illegally.
Would it be possible that they're 'rest-in-red' (as you've suggested elsewhere in your post), and it's just that the approaching vehicle has just missed the window in which the controller has decided to move to the all-red revert, meaning it's then got to clear that all-red before moving back to that stage?

There's one on the A580 that used to do that. If you just missed the window and it moved to revertive all-red, you'd just be coming to a complete stop from 50mph by the time it dropped back to green. I think it's since been put on to MOVA, and even approaching at 50mph, you're probably only down to around 30mph by the time it drops back to green. That all depends on when you start slowing mind you, I tend to ease off fairly early when it's quiet to give signals time to change stages before I have to come to a complete stop.
No, because observation as a pedestrian showed it staying green indefinitely, only turning red when a vehicle approached! When I approached that light on the bike, it had been green for quite some distance, then turned red as I approached, and I was the only vehicle within range.
swissferry
Member
Posts: 319
Joined: Tue Aug 22, 2017 20:42

Re: Traffic lights and cyclists

Post by swissferry »

jervi wrote: Fri Apr 23, 2021 08:08 The best solution to the problem is having some sort of indicator that shows wether the controller is aware of waiting traffic on that arm. If having a push button similar to pedestrian ones then the red circle around the button could light up once someone is detected. Some junctions in Brighton and Hove have push buttons, however these are current exclusive to filtered turns (which have their own phase) where cycles would not go over any loops when using the cycle tracks/lanes to the side (they move over the to the right turn filter lane within the bike box)
There used to be a push button for cyclists at Pulpit Rock on the A82 before the road was widened. Saw a car passenger use it once when the lights appeared to be stuck on red and the lights changed to green.
Al__S
Member
Posts: 491
Joined: Sat Mar 04, 2017 10:56

Re: Traffic lights and cyclists

Post by Al__S »

A lot of Dutch junctions have, on the cycleways, induction detectors with an indication that you've been detected and a button to press if you haven't
Post Reply