Puffin crossings - a blunder?

[Chris5156]

To be fair a lot of the waiting times are not down to the Puffin or DfT but are a set-up issue and could and can occur with pelicans.

And yet I have never encountered a pelican that does not change immediately (unless it has recently given a ped cycle), nor a puffin or toucan which does not make you wait until there is a gap in the traffic big enough to use without the lights.

The puffin crossing on the A212 Westow Street in Crystal Palace will change the instant you push the button, unless it has only just gone back to the red man. Even then, if it has only just completed a cycle, it will only wait about twenty seconds before stopping traffic again.

I use this crossing a lot and in general my experience is that pedestrians will clear the crossing and about five seconds later traffic is moving.

[UTCPaul]

Then I've seen others where wait times at night when there are 5 cars at the junction are near 90 seconds. It must be a set-up and maintenance issue.

Almost certainly. West Lothian's lights have always worked well from a demand responsive point of view so it certainly can be done right.

IIRC WLC outsource the maintenance to Siemens, so I suspect there may be some pretty strict SLAs in place.

Speaking as a pedestrians I'd be happy to see the back of puffins. Whoever though that designing a pedestrian crossing that makes pedestrians wait until there is a gap in the traffic that they would be able to cross on anyway was a bit of a numpty frankly.

But that is EXACTLY how they are supposed to work! They are a facility to help people, particularly the less able, to cross the road. They are not a right of way for pedestrians to immediately leap out - they're called "Zebra Crossings".

I am assuming you are not an 80 year old arthritic with sight problems, and that the gap is plenty for you to cross in, otherwise you wouldn't be raising this.

The fairly typical setup for a standalone crossing is that there is a maximum timer for vehicles. Generally 20 or 30 seconds, but can sometimes be longer. Depending upon the mode of operation, the timer starts when someone presses the button, or the timer starts when traffic reverts to green ("Pre timed max"). In areas of higher traffic flows, the former is usually applied.

So you walk up, press the button, and wait. There are sensors on the crossing to detect approaching vehicles. If the timer expires before the traffic clears, the crossing changes (a "force change"). However if there is a gap in the traffic before the timer reaches its limit, the crossing will "gap change".

Pre Timed Max is generally used where vehicle flow is lower or sometimes if there is very high ped flow. In this case the max timer starts when traffic gets a green. If you arrive before the max timer expires and press the button, then you simply wait until the end of the timer before they change. If the max timer has apready expired (say at night when ped flow is low and nobody has pressed the button for a minute or 2) then the lights change immediately.

PTM also has safety implications. Car drivers are used to seeing people waiting to cross, and don't expect the lights to change as soon as a ped arrives and presses the button. Peds get used to the lights changing as soon as the button is pressed, especially if they usually cross at a time when it changes straight away, and can just press it and step out without waiting for the green man. Now whilst I am a firm believer in natural Selection, I am also averse to Coroner's enquiries, consequently on our very few PTM sites there is a delay between pressing the button and the demand registering and the wait lamp coming on. What would be ideal would be a PTM timer followed by a 5 second maxset, but this isn't available on the old kit we have.

Now the third type of setup is where they are co-ordinated with a junction. If a crossing is at a junction, then srictly speaking it isn't treated as a ped crossing, and is a stage in the cycle at which a ped movement is permitted. The ped crosnig can be physically separate from the junction, but still may only run at a point in time when the system knows thre are no cars approaching, otherwise drivers set off from a green, around the corner, into a red crossing they arn't expecting, or worse still are queueing through a crossing point at a red light, the lights go green and traffic sets off, and cars waiting in the queue at the crossing opint just set off and blindly follow the vehicle in front, irrespective of what colour the crossing is.

At a junction crossing point the green man can come on for a few seconds, or until the lights are about to change. The disadvantage of the second mode is that you need to allow ped clearance time before giving traffic a green, so get into all sorts of complicated phase delays and dummy phases if you want to avoid dead time.

[Benny]

Instant changing to red is a bad idea. What if the roads are icy? I'm not going to risk skidding out of control and wrecking my car just so a pedestrian can cross the road in front of me (especially when very often there isn't a line of traffic behind me).

[PeterA5145]

Instant changing to red is a bad idea. What if the roads are icy? I'm not going to risk skidding out of control and wrecking my car just so a pedestrian can cross the road in front of me (especially when very often there isn't a line of traffic behind me).

Another problem if the lights are set up to change to red immediately the button is pressed, even if there are cars coming, is that they can be abused by mischievous children (or students) who then proceed to run off

[cb a1]

Instant changing to red is a bad idea. What if the roads are icy? I'm not going to risk skidding out of control and wrecking my car just so a pedestrian can cross the road in front of me (especially when very often there isn't a line of traffic behind me).

Don't worry, the three second amber phase is still mandatory. It's not an 'instant' change to red. It just means that when the button is pushed, the lights go to amber, then red.

[Gareth]

And yet I have never encountered a pelican that does not change immediately (unless it has recently given a ped cycle), nor a puffin or toucan which does not make you wait until there is a gap in the traffic big enough to use without the lights.

I struggle to believe that they can somehow manage to set pelicans up right but not puffins and toucans.

I've encountered many a pelican with a stupidly long wait time. That said, plenty of the 70s/80s vintage pelicans seemed to change immediately. It's probably more due to the age of the installation than the type of crossing, which just happen to somewhat align. I agree that some have a way too long waiting time, though the situation locally seems to have actually improved somewhat over the years. If it takes so long that it's no quicker than if the crossing wasn't there, then people won't use it; particularly if one needs to walk up the road to it. I've been crossing the road with many a person where I've suggested walking to the nearest proper crossing only to be told in a matter-of-fact tone that it would be "no quicker".

Another thing to clear up here is that a pelican is the one with the flashing amber/green man phase. If it doesn't have that, it isn't a pelican, regardless of whether the pedestrian signals are farside or not.

Getting back on topic: puffins and nearside indicators have been around since 1996. The one in Liverpool's Queen Square bus station was one of the first in the country. People found it beyond confusing and the positioning of the poles and nearside indicators have been reconfigured at least two times to my knowledge. They're not new anymore. We've had them for 18 years, so it isn't a case of getting used to them. The only thing that seems to have changed in people's behaviour is that instead of being all confused when both the pedestrian and vehicle signals show red for an extended period of time (before more often than not deciding to cross anyway), people in these parts are now just out of the habit of looking at them at all; no doubt due to the awkwardness of the nearside signals but also to the perceived lack of reliable information they give out. The only time it remotely works is in a more suburban area where only one or two people are crossing. The lights change, the pedestrian starts crossing (mainly because they see that the cars have stopped and hear the beeping start rather than looking at the nearside display). The pedestrian finishes and the lights change back quickly because no one is on the crossing. More commonly in urban areas, city centres especially, the lights change, pedestrians again move due to seeing the cars stop (most can't see the nearside indicators anyway as they're obscured by other people). The green man changes to red man but more people just keep coming onto the crossing. Eventually, the clearance period maxes out and the lights change to green with people still crossing. The motorists then rev their engines and slowly move closer to try and establish precedence. It's pants.

[jw24]

A new edition of TSRGD is currently under consultation and one of the proposals is to cease installation of pelicans and suggest instead installing a London 'pedex' type crossing.

Pelican crossings
6.6 This revision also provided an opportunity to look at the
various crossing types and see if changes needed to be made.
Pelican crossings have been in use since the 1970s and while
still useful, cannot provide the benefits available with more
modern crossings such as puffins. Many authorities are now
choosing to install puffin crossings as their default crossing
type.
6.7 Puffin crossings give more benefits to pedestrians than
pelicans by using detectors to monitor the crossing and give
people extra time to cross if needed. This is especially useful
to more vulnerable pedestrians, such as older people, and
people with mobility issues. Research has shown that these
crossings are considerably safer than pelican crossings.
6.8 Authorities that want to retain the farside signals but provide
the benefits of puffin crossings can also use what is known in
London as a 'pedex' crossing. These crossings use the familiar
farside signals of a pelican, but do not have the flashing green
man or flashing amber. They can be used with similar
detectors to puffins, and the new countdown signals (included
in the new TSRGD) developed to show how much time is left
to cross the road during the blackout period.
6.9 The number of pelican crossings has been declining steadily
as puffin crossings increase in numbers. With this, and the
development of countdown and pedex crossings, we are
proposing that pelican crossings are no longer prescribed.
6.10 This will not mean that pelican crossings will need to be
removed from roads. Local authorities will not be required to
remove or replace any crossing and existing pelican crossings
can stay in place until the equipment naturally reaches the end
of its life. In most cases, this is about 15-20 years.

Taken from: https://www.gov.uk/government/uploads/s ... -final.pdf

Further reading inc. draft of TSRGD 2015 sched; https://www.gov.uk/government/consultat ... tions-2015

[Gareth]

It actually feels weird writing this but that's actually a good effort by the DfT.

Firstly, pelicans are archaic and so I think they've had their day. New pelicans should've been discontinued 15 years ago already.

Secondly, it explicitly mentions that farside installations can still be used and with puffin detector equipment. It also mentions the countdown signal but I can't see these working with countdown signals though. I think a "pedex" and a farside puffin should be considered distinct. I think this is the way to go.

The great news is that's it seems to offer assurance that authorities may retain farside signals if they prefer. I feel the reason many authorities migrated to nearsides was not that they thought they were so good but that the DfT was going to remove type-approval for farsides. Also, not being allowed to use puffin equipment with farsides (which never made sense) also loaded the dice.

Looks like common sense may eventually prevail.

Just an additional wishlist...

- A signalised clearance interval for farsides that isn't blackout. Blackout to be banned regardless.

- A new standard for audio signals to be more in line with most other developed countries. It should be able to be used at junctions and staggered mid-block crossings and give feedback to the pedestrian at all times. I'd like something similar to in the Netherlands where they have a quiet slow/fast click. It's clear but quiet enough to minimise confusion about which particular crosswalk it applies to and also not to annoy people nearby.

- black arrow stencils for red and amber aspects on turn signals.

- a new design for mast arms that are more similar to continental designs and doesn't look like a motorway gantry.

- eye level repeaters to be allowed though recommended to only be used where a farside secondary is not practical for whatever reason. Also to be used for cycle signals, though these should contain a black bike image on the red and amber and then a green bicycle.

[traffic-light-man]

- a new design for mast arms that are more similar to continental designs and doesn't look like a motorway gantry.

Something like these TfL Standard ones?

I'd also like to see some continuity added to Toucans - perhaps the introduction of a red cycle for farside installations. It's alway irritated me that nearside Toucans have a red cycle aspect, but the farside ones don't.

[OLD GIT]

Now ,meanwhile in the Midlands, we don't seem to get many birds crossing the roads. Penguins- no, Puffins, bit too inland for them and as for Toucans, perhaps a bit too cold. I remember in one part of nearby Leicester a sign warning of Ducks crossing. And in one stretch outside Coventry ,i remember two signs warning of Elderly crossing and cows on road. So perhaps we have somewhere to find Alf Garnet's "Silly old moo"

[ianwallis]

It actually feels weird writing this but that's actually a good effort by the DfT.

I second that Gareth. It's been too long that DfT have avoided giving a strong and positive message but that's what we need whether you agree with it or not. We've got such a confusing mix of crossing types now and by still allowing pelican crossings to be installed we're not getting anywhere nearer that idea of using the flashing amber as a fault condition (which was one of the driving forces behind the puffin design).

[Jam35]

It actually feels weird writing this but that's actually a good effort by the DfT.

I second that Gareth. It's been too long that DfT have avoided giving a strong and positive message but that's what we need whether you agree with it or not. We've got such a confusing mix of crossing types now and by still allowing pelican crossings to be installed we're not getting anywhere nearer that idea of using the flashing amber as a fault condition (which was one of the driving forces behind the puffin design).

That is one of the worst design flaws of puffins. The flashing amber is a sensible indication that allows people to proceed with care and encourages them to pay attention to the road. If anything, it should replace the silly red-amber stage on other traffic lights.

If a fault indication is required, they could always use all three colours on at once.

[M4Mark]

That is one of the worst design flaws of puffins.
The flashing amber is a sensible indication that allows people to proceed with care and encourages them to pay attention to the road. If anything, it should replace the silly red-amber stage on other traffic lights.
If a fault indication is required, they could always use all three colours on at once.

I've read some strange ideas about traffic lights on forums over the years but having all three colours showing to indicate a fault takes the gold medal.

The red with amber 'starting amber period' is one of the best things about our lights, we get far more efficient junctions and less wasted green time compared to countries that go straight from red to green.

[Gareth]

- a new design for mast arms that are more similar to continental designs and doesn't look like a motorway gantry.

Something like these TfL Standard ones?

I'd also like to see some continuity added to Toucans - perhaps the introduction of a red cycle for farside installations. It's alway irritated me that nearside Toucans have a red cycle aspect, but the farside ones don't.

Meh, they're okay I guess but I'm not a huge fan. I don't know what it is about the UK and brackets, rather than the sexier sloping curve that mast arms and lamp standards usually have in continental Europe and elsewhere. Still, those examples are many times better than the motorway gantry-style ones, especially when they're used to support tiny helios signals, like those ones on Queens Drive. Seriously, those look so ridiculous.

[Gareth]

It actually feels weird writing this but that's actually a good effort by the DfT.

I second that Gareth. It's been too long that DfT have avoided giving a strong and positive message but that's what we need whether you agree with it or not. We've got such a confusing mix of crossing types now and by still allowing pelican crossings to be installed we're not getting anywhere nearer that idea of using the flashing amber as a fault condition (which was one of the driving forces behind the puffin design).

Yes, pelicans need to go. Vehicle and pedestrian signals should behave the same, whether they're at junctions or mid-block. For vehicle signals, that should be the standard red-red&amber-green-amber and for pedestrian signals, it should be red man- green man and then some sort of clearance period (though preferably not blackout). I suggest a flashing red man. This would clear up any confusion with pelicans and would give a more negative signal than flashing green man. At crossings with on crossing detectors, the flashing red man would be the variable stage, whilst on normal crossings, it would be a fixed time clearance and could be used with countdown timers, where the blackout is currently used in London. Blackout definitely needs to be shown the door along with the pelican though.

I agree with traffic-light-man about red cycle signals on toucans. They should be used consistently, whether farside or nearside.

[Patrick Harper]

I must admit to avoiding the use of 'controlled' ped crossings, even on busier roads I feel unsafe trusting technology rather than my own intuition.

[jouef]

It seems to me that crossings are an exercise in ergonomics - the design of equipment that fits the human body and its cognitive abilities.

Farside indicators make ergonomic sense. They comply with a putative convention for all road users including pedestrians of signals being in line of sight along the expected path, so they are easy to detect (a similar convention for all road users is red commands 'stop' and green permits 'go'). Farsides are not as easily obscured by people, umbrellas or vandalism as nearsides, though nearsides could be freely installed as repeaters whenever helpful.

Farsides provide on-crossing feedback to pedestrians - this is psychologically necessary when their locus of control is partly external. Comparing light-controlled crossings to zebras: a pedestrian on a zebra has an internal locus of control. They invite and make their own decision to cross and their presence on the stripes is the 'stop' signal to the traffic. They can obtain confirmatory feedback while crossing by observing the traffic and predicting that its expected behaviour will continue to be to stop while they are crossing. On light-controlled crossings, locus of control is partly external. The pedestrian can press the button but the stop signal is the red on the carriageway which is external to the pedestrian in that they do not directly control it, know how it is controlled or predict its behaviour. Some pedestrians may have been given instructions or know how the equipment works but its decisions are still programmed in and not varied by the pedestrian so locus of control remains external. To continue to predict while crossing that the expected behaviour of waiting or approaching traffic will be to stop, the pedestrian requires feedback from the carriageway red. A farside green signal provides this confirmatory feedback, which is what people miss when only nearsides are present. Behaviour of equipment is mechanical while behaviour of humans is emotional; this is why humans require confirmatory feedback even if the equipment is programmed to accommodate their time taken to cross and they are informed of this.

The lack of confirmatory feedback with nearside-only is a design flaw because people lack the informstion on which to confirm or amend their behaviour. This would not matter if there was no need to take responsibility for their own safety. However that responsibility is essential, and to temporarily surrender it to the equipment during the crossing phase, even if safe to do so, creates cognitive dissonance which people find uncomfortable. People need consistency and the equipment should present feedback even when not mechanically necessary - for humans during crossing it is emotionally necessary.

The best design requires the least explanation. I advocate education of general conventions, not instruction on the use of specific equipment. Thus crossings could be designed with whatever detection and automation is available as long as they present the right signalling and feedback to road users. These conventions might be:
(a) Your signal is in your path in front of you (whoever you are),
(b) Stop at red, proceed with caution at green (whoever you are),
(c) Check that other traffic is behaving as you expect, and
(d) Take responsibility for your own and other's safety.

The addition of countdowns, ambers or flashing would be possible to any signals already obeying the red/green and in-your-path conventions.

[Fenlander]

A question about near sides, (I'm thinking of a junction just up the road from me) what about where there's an island between the opposing lanes - one half might be green but the other red, there's a pair of near sides on the island itself, how do you know which is for the direction you're facing?

[michael769]

If it's not staggered then it's a single crossing, so both sides will be the same.

[jw24]

It actually feels weird writing this but that's actually a good effort by the DfT.

I second that Gareth. It's been too long that DfT have avoided giving a strong and positive message but that's what we need whether you agree with it or not. We've got such a confusing mix of crossing types now and by still allowing pelican crossings to be installed we're not getting anywhere nearer that idea of using the flashing amber as a fault condition (which was one of the driving forces behind the puffin design).

That is one of the worst design flaws of puffins. The flashing amber is a sensible indication that allows people to proceed with care and encourages them to pay attention to the road. If anything, it should replace the silly red-amber stage on other traffic lights.

If a fault indication is required, they could always use all three colours on at once.

And what about if you have failed lamps / cables / controller components? There is chance of only greens lighting on an approach causing a conflict. Unlikely I know, but still possible.