Monday, 15 August 2022

Summer CYCLOPS Safaris: Part 1 - Manchester

At long last, I have manged to get out to go and experience the CYCLOPS junctions which are starting to be rolled out across the UK.

I have covered the Cycle Optimised Protected Signals junction designs before with a comparison with the Dutch approach and I have seen plenty of photographs and videos of the design. However, there's nothing like seeing infrastructure in the flesh and so in this post, I will have a look at the first and third examples of the arrangement which were built in Manchester.

A crossroads junction with grey footways, red cycleways and then a black carriageway. There is a pedestrian island at each corner of the junction with a mini-zebra crossing over the cycle track to get to the corner islands. The cycle crossings of the side roads are outside of the pedestrian crossings.


First, let's have a little recap. The sketch above is a general arrangement of the CYCLOPS junction. The red shows a series of cycle tracks and in this case, they are all with flow. This means that cycle traffic circulates around the junction in a clockwise fashion. The light grey is footway and island areas and the dark grey is the carriageway. You'll notice the mini-zebra crossings over the cycle track which in this case are placed on small speed humps and the dark red is tactile paving. 

The 'L' shaped tactile paving essentially shows the start and end point of a complete crossing of the junction and personally, I prefer this to leaving the cycle track crossings uncontrolled. The little speed humps are not about slowing cycle traffic, they are more around giving a level crossing point. Some visually impaired users do prefer having dropped kerbs as the slopes help show they are entering a crossing area.

In terms of guidance, Local Transport Note 1/20 "Cycle Infrastructure Design" uses a catch-all term of Circulating Cycle Stage Junction in S10.6.21 for signalised junctions where walking and cycling run together orbitally around an entire junction of which CYCLOPS is a style. It is possible to use two-way cycle tracks with the design, but I'll cover that another time.

Probably down to timing of publication, but LTN1/20 doesn't include a CYCLOPS junction, and the section on circulating stage junctions is very brief for what, in my view, is going to become a very important design tool in the coming years - as an aside, LTN1/20 doesn't go into enough detail on design generally, but that's another post. Here's a more detailed technical document explaining the concept.

A CYCLOPS junction with a cyclist crossing one of the side roads on a green cycle track.

The first CYCLOPS junction was delivered in 2020 on the Manchester to Chorlton Cycleway at the junction of the A5067 Chorlton Road and Royce Road (above). In the photograph above you can see one of four pedestrian islands on the left (one at each corner) with the pedestrian crossing in red. On the right of that, there is a the one-way cycle crossing (green) running in parallel.

A CYCLOPS junction with a pedestrian island shown prominently with various islands, kerbs, coloured surfacing and road markings.

Above is a slightly different view of the junction, this time showing more of the corner island and the traffic signals. Transport for Greater Manchester has elected to use a pair of low level cycle signals - personally I like to see a full-sized signal plus a low level sign with the former being more easily seen from a distance back. There is a push button, but that's a back-up because there is detection and it's too close to the stop line for my liking (coming at it from a non-standard cycle user's perspective. It is nice to see the button light up to show you are detected.

Another view of a CYCLOPS pedestrian island.

For pedestrians, there is no priority onto/ off from the island and I personally would have preferred a mini-zebra crossing. The island is quite complicated as it is really three small islands forming a larger space which sits at carriageway level and the further island (in the middle of the three in the photograph above) helps to keep left turning drivers out wide. 

The islands provide a vertical upstand to help cane-users to navigate, but they create a problem for wheelchair and mobility scooter users who cannot get close to the push button. A better approach would have been to place the posts at the same level as the walking surface. 

Because the access to the island is not in line with the signalised crossing of the road, people have to turn through 45° to move between the crossing points. This means that it is less likely that visually impaired people will mistake the crossings as single movement as can be the concern with in-line crossings. This is something DfT worries about in LTN1/20 (10.6.22) where mini-zebras are used with signals over the road, although I think this is overstated to some extent. However, it address some issues, so I am not against it.

Two cyclists crossing a side road. One on the CYCLOPS cycle track and one on the road.

The other design decision was to provide a gap in the approaching cycle track protection to allow cycle traffic to move out of the cycle track into general traffic, complete with an advanced stop line (ASL). This means that some cyclists may elect to move into traffic if the signals are showing in favour if that movement. The photograph above shows someone who has chosen to do this. 

This was the first CYCLOPS and so there was concern that people cycling wouldn't want to wait for a green signal, so the access to the ASL was an adjustment to counter the problem. The real problem of course is how much time we give to general traffic verus how nimble and efficient cycling is given that the CYCLOPS junction is motoring infrastructure. The signals engineers have been tweaking the detection and so it isn't a huge issue and in theory, the all round green for walking and cycling could be run twice per cycle.

225 metres south of the first CYCLOPS, we have the UK's third, which is part of the same scheme, this time at the junction of Chortlon Road with the A5067 Stretford Road

A corner island with a diagnal crossing point surfaced in red.

In many ways, it's the same layout as the first, but an immediate difference is the addition of a diagonal pedestrian crossing between the southeast and northeast islands (above). The junction is skewed which means one of the diagonals is quite short and because walking and cycling run under a single green, the diagonal is easy to accommodate. In fact, had the junction been designed in the Dutch way with the cycles on the inside, then the diagonal wouldn't have been possible.

A wider view of the crossing and orbital cycle tracks.

The walking layout on the diagonal islands is a little more complicated with two crossings of the cycle tack each, but as things are offset as with the first scheme, it all works very intuitively. There are no ASLs on this scheme, but there are gaps in the kerb protection if one feels the need to move out. When I cycled the two junctions I didn't have to wait too long and as you'll see in the video below, a complete U-turn is easily possible within the green which is a very important feature to allow people to change direction if they didn't want to cross the road to turn right at a proceeding junction.

The CYCLOPS junction is interesting because on the face of it, we get fully protected intersections which feels very familiar to anyone who has used such in other places, especially northern europe. However, at the same time, they are very much British and have a look and feel that will be familiar to politicians and designers (including signals engineers) as well as the public and that is why we'll see them all of the UK in the coming months and years.

Overall, I think the cycle tracks are a touch narrow, there are perhaps too many fiddly kerbs and islands, and the colour scheme is a bit garish, but the junctions work well and most importantly, they give a great sense of protected and action protection from traffic.

I'll leave you with a video of a bit of cycling in the general area, including these two new junctions.


Sunday, 31 July 2022

Hare-Brained

I do enjoy discovering highway engineering relics which turn out to be clever little interventions that we have forgotten to do and in this post, we head to Peterborough to look at such a relic.

On the one hand, it's great to see old layouts that make walking, wheeling and cycling easier because it shows that we've always been able to do this stuff. On the other hand, it's also sad, because rather than rolling them out everywhere, it reminds us that the UK chose the wrong path to mass car use which is so difficult to face today.

To the north of the city of Peterborough, the A47 and A15 meet with both roads amusingly called parkways. Just south of this junction we have Harebell Close, part of the larger Dogthorpe part of the city. The suburbs of Peterborough being a product of it becoming a New Town in 1967 with lots of low density development surrounding the historic city centre. 

In common with many New Towns, there is also a network of walking and cycling routes which are separated from the road network, but like the others, it's low density development and where driving was made far easier. Over the years, the quality of the separate networks has been allowed to degrade and it hasn't really been expanded. The 2011 cycling commute mode share was 5.7%, compared to the national 2.7% and so at least being fairly compact, flat and with some cycling infrastructure, it gives the place something to build upon.

Anyway, Harebell Close has an interesting junction where a cycle track meets the road (below). It is a junction because cycles are vehicles and the cycle track is meeting a carriageway.

A view of a street where the road is narrowed at the place a cycle track crosses from each side. there is a footway on both sides with bushes behind those. The paving changes type and colour at the junction.

I'll go through the details shortly, but the the location came to my attention through tweets from the local MP, Paul Bristow (Con, Peterborough). Who had apparently been campaigning for barriers to be placed across the cycle track.

Tweet from Paul Bristow MP sent on 27/7/22 at 6.04PM "A year ago I went to see the residents of Harebell Close in #Peterborough about putting in some kissing gates where the cycle path meets the road 👍  This is about safety. Officers agreed with me and promised action. We have e messed around for year. This is not good enough." The tweet has a photo of the MP standing by the cycle track which runs off into the distance.

As you would expect from Twitter, the MP received a schooling in access issues and the Equality Act 2010, but even when noting comments from the Peterborough Cycle Forum, he doubled down and wrote to the council's chief executive about the issue.

Text from the MP's letter of 29th July 2022. "PAUL BRISTOW MP, PETERBOROUGH  HOUSE OF COMMONS  LONDON SW 1A OAA  Matt Gladstone Chief Executive Officer Peterborough City Council  Sent via email  Dear Mr Gladstone, Re: Harebell Close  29 July 2022  Around a year ago I went and met with residents at Harebell Close regarding the cycle path crossing near the entrance to the close. Residents expressed concern that cyclists can not easily see that a road crossing is coming up and as a result cross the road at speed which creates risks for all road users - pedestrians, cyclists and motorists alike. There is also a children's play area near by which means many families are using the road in close proximity to the cycle crossing. As a result of my visit last year officers agreed that it was a dangerous situation and some sort of gate or barrier would be installed so that cyclists were aware they were approaching a road crossing. I have been contacted again by residents who are disappointed that no action has been taken for a year. I understand that under Equality Act 2010, Section 20, any physical feature installed should not put a disabled person at a substantial disadvantage. I would not support a barrier that would do this and I have been reminded of the law several times in a 'spirited' way by members of the Peterborough Cycle Forum on social media. However, this is an issue of safety and the Act must not be a reason to do nothing. Working with local residents, I am sure a solution can be found that suit cyclists, pedestrians, motorists and disabled people. Furthermore, I understand that there is confusion about whether responsibly for the maintenance of the busies and foliage along the path is the responsibility of Longhurst or the City Council. When overgrown these buses also affect accessibility and viability and it would be if this could also be resolved.  I would be very grateful if you could look into this matter and ensure something is done as soon as possible for the safety of all road/ path users. I am determined to hold the council to their promise to put something in place that will improve the safety of this junction.  Yours sincerely,  Paul Bristow MP Member of Parliament for Peterborough "

This is a fascinating insight into day to day local political campaigning because "road issues" is a great subject to generate a bit of noise given how many politicians only listen to noisy adult drivers with everyone else ignored. 

Like traffic signs and pothole repairs, a campaign for a couple of barriers is also a good tactic because it's pretty cheap, it's something that generates a good photo of Things Getting Done and it plays well to those who are listened to. Within the local authority, a CEO is able to push just a request down the food chain as inevitably, there will be a manager somewhere happy to instruct the work to shut the politician up - it's easier to acquiesce because it takes so little time and effort, compared with actually resisting such nonsense with data, facts and logic.

I don't think I need to explain why barriers across a cycle track is a bad idea, but you can read this post from 2016 for more, even where there might be an issue of compromise (which isn't the case here). Meanwhile, lets look at the details of Harebell Close and try to get under the skin of what the issue really is.

A view of a street where the road is narrowed at the place a cycle track crosses from each side. there is a footway on both sides with bushes behind those. The paving changes type and colour at the junction.

If we go back to the first image (above to save scrolling), there are several points of detail which can be gleaned. The general layout has a 5.5m wide carriageway with a couple of 1.8m wide footways; a layout which can be found all over the UK for estates of this age. 

At the crossing, the carriageway is narrowed to about 3.2m which means drivers have to give way to each other. The narrowing is also asymmetric with the nearside (on each approach) being narrowed far more than on the offside. For drivers, this gives two messages. First it is very obviously that they should slow down, but more subtlety, that oncoming traffic has priority. There are no signs to explain this because the layout is self-explanatory.

The same layout as before, but zoomed in.

The asymmetric narrowing also physically helps to slow drivers because they have to steer to the right fairly abruptly. The area that is narrowed is also paved in a completely different material (above) with bollards to give some visual verticality to the horizontal deflection. The lighting columns on the nearside to each approach make sure the junction is well lit and so further provide verticality. 

The very interesting thing about the asymmetric narrowing is creates a bit of a zig-zag through the junction that you only notice on a closer inspection. Across the junction, kerbs flanking the carriageway continue through and while disappearing into the background, they do add to a suggestion that general traffic has some priority.

We can also note that the junction is actually on a raised hump which provides some vertical deflection for drivers as well as a flatter place for cycle traffic to cross. For the cycle track, there are give way markings before the junction which indicates priority for general traffic and the cycle track ends short of the carriageway on each side to provide tactile paving, perhaps recognising that people may well be walking along the cycle track which is signed for cycles, but over which people have a right to walk.

A view of the tactile paving as described in the main text.

The tactile paving is also interesting. At the edge of the carriageway, there is red crossing blister paving which is actually reserved for zebra crossings and signalised crossings. This should ideally buff, or at least a non-red contrast. 

The grey paving behind is actually a UK rarity - a guidance path surface. This is used to guide visually impaired people along a route when the traditional cues, such as a property line or kerb edge, are not available. This has been deliberately provided here because the arrangement seems to guide people to the crossing point. It's an odd arrangement.

So there we have it. A layout which is a junction, does some clever things with vertical and horizontal deflection and with a change of materials, but which still prioritises the road. Throw in a warning signs for a cycle route crossing ahead on each side of the junction aimed at drivers, we have an interesting layout which generally seems to mitigate the risk of a collision. In terms of data, there haven't been any actually injury collisions here for 20 years and while we should be careful with applying casualty data (as it's only a small part of an overall consideration), the junction seems pretty safe to me.

Let's go back to Bristow's letter and look at the detail.

"Residents expressed concern that cyclists can not easily see that a road crossing is coming up and as a result cross the road at speed which creates risks for all road users - pedestrians, cyclists and motorists alike." 

This is really a projection. The concern is coming from adults who drive and despite living here and knowing the layout intimately, still have concerns. There are just 50 homes beyond this junction and so this is genuinely a low traffic environment and so it hints that driving residents don't really want to slow down and check the junction. Even if someone cycling didn't slow down, they are still not going to be faster than someone driving through, even with the traffic calming. What the residents actually want is for the state to mitigate any liability they might be carrying into the junction under the veil of "safety".

Throwing in "all road users" is absolutely classic politics. The whole "all road users" trope is used by those trying to suggest that everyone has come to the game on a level playing field with the same power. This is absolutely not the case and argue as they might, physics is the final arbiter. He throws in some more Legitimate Concerns.

"There is also a children's play area near by which means many families are using the road in close proximity to the cycle crossing"

There is a little playground at the end of Harebell Close and because the street curves back on itself, it is directly served by the cycle track. I wonder if by "families using the road" he is worried about locals driving to the playground which would be odd. He goes on.

"I understand that under Equality Act 2010, Section 20, any physical feature installed should not put a disabled person at a substantial disadvantage. I would not support a barrier that would do this and I have been reminded of the law several times in a 'spirited' way by members of the Peterborough Cycle Forum on social media. However, this is an issue of safety and the Act must not be a reason to do nothing. Working with local residents, I am sure a solution can be found that suit cyclists, pedestrians, motorists and disabled people."

I'll take this as some progress that he understand that barriers can be discriminatory, but his passive aggression against the local cycle forum shows just how embedded the thinking is that a barrier is a solution per se. Then we're back at a solution that suits everyone - the point is, we are dealing with physics and the second best solution is reducing the energy of the situation which has to be aimed at drivers and it is they who should be made responsible. The best solution is removing general traffic, but that's not an option.

Let's also consider the users a little more. Drivers are trained, licenced and insured. They are adults and have a better developed understanding of their environment and the risks associated with it. Then let's say we have a couple of 10 year old children using the cycle track to get to the playground. They have a less developed idea of risk, speed perception and they certainly haven't been trained in the rules of the road, or at least not to the extent that a driver has. 

I would contend that users of the cycle track might, on balance, be less understanding of the role of the give way markings than drivers, but cyclists can also adjust their speed so they don't need to stop before crossing (which is a time penalty). To a driver approaching the junction at a speed excessive to the situation, a young cyclist going straight across might appear fast, but in reality this isn't the case. Throw in the general proposition that it isn't in cyclists' interest to be in a collision and although we have a good layout, it is still set up to favour and indeed to absolve the driver of responsibility and so in providing a solution to the "problem", we need to go further.

Let's swap the propositions around. Here we have a cycle track which is useful for utility purposes and it is the (car) based route which has to cross it. Cyclists have an energy penalty for every stop and so maintaining momentum is important. We also have users who may not be trained in what traffic signs mean and who might not be the best at judging risk. The answer here is to change the junction to put the absolute responsibility for crossing the space on drivers which is completely appropriate for the conditions.


The image above has the cycle track made continuous across the junction with the kerbs along the carriageway edge removed in favour of kerbs running across a driver's line of travel. These are design cues which require drivers to slow right down and check before crossing. One thing that does need doing, and I am in agreement with Bristow on it, is that the bushes need cutting back to improve visibility between modes. This is not to absolve drivers of responsibility, but to ensure that when they are in the correct position, they can actually see to check before crossing. Some blocking of visibility is helpful as it stops people glancing early and running through.

This approach has been available for decades and it is featured in Local Transport Note 1/20 Designing for Cycle Traffic (see Figure 10.6) - a "cycle priority crossing". Whist I would argue that the design layout is far more important than administrative control through the use of traffic signs (including markings), the minimum requirement is to provide a pair of dotted give way lines (diagram 1003) and the additional of a painted give way triangle and upright give way signs is overkill in my view for this type of situation with low traffic flows. I would even look at omitting the give way lines based on a risk assessment.

A plan drawing of the junction as explained in the main text.

The sketch above shows what an updated version of this layout could look like. Rather than the gentle slopes up to the junction, I have used entrance kerbs to provide both the vertical deflection and the visual cue across the line of travel and I have added give way markings to be explicit. I have also assumed the bushes have been dealt with to provide better visibility at the junction. 

Really, this is the kind of layout which Bristow should be pushing for. One which makes allowances for the knowledge and energy of those cycling and which makes driver responsibility explicit. I am not coming at this from a premise that cycling should be prioritised at all times, but from the premise that this type of situation should very much prioritise cycling and through good design, we require the party with the capacity to do the most harm to have the responsibility for the safety of all road users.

Saturday, 4 June 2022

A Cheeky Monkey

Well Dear Reader, this is my 500th blog post and I can scarcely believe it. My first post was on 28th November 2012 and so this year I shall also be celebrating 10 years of writing this conduit from my brain.

This really *has* been a chronicle of my adventures in space and time in the strange universe of highways and transport both professionally and personally. I think the journey to starting this actually started back in January 2011 because I started cycling to work and this enabled me to see the streets from a lost perspective, having barely cycled since being a child. My conversion wasn't because of any green or health-related ideals, it was the fact that I had just moved offices from the suburban edge to the town centre and got fed up with sitting in traffic jams twice a day. In many ways, journey time predictability and convenience remains a key reason why I still cycle today, but I'm increasingly learning there is more to it than that.

From that point onwards, it started to dawn on me that our approach to urban transport in particular was utterly warped. I could see cycling was part of the solution, although I didn't really grasp just how important it was. I already had a professional interest in walking and so I saw cycling as an natural extension, which in some ways it is. When I started to write this blog, I was working for a London borough and after cycling for 18-months I started to become frustrated with some of my colleagues, certainly management and definitely politicians when they couldn't see the world as I saw it.



I had started to read other people's blogs at the time and one in particular got my brain thinking and that was "Crap Cycling & Walking in Waltham Forest." I was reading lots of others, but CC&WIWF got my brain going because of its focus on infrastructural solutions to the problems we had (and still have in many ways). I also need to give a shout out to the Cycling Embassy of Great Britain which really introduced me to how the Dutch design for cycling as well as showcasing the best of the UK. The organisation is currently resting for a variety of reasons and I don't know if it will get going again, but I am indebted to all involved, plus those who took time to lead rides and talk at the Embassy's AGMs - a much missed annual highlight for me.

In 2012, cycling for transport seemed a marginal idea in UK transport terms, at least it was in my own professional circles. Sure, we had places like Cambridge, but they had been dining out on local peculiarities such as being a university town with skint students needing a cheap way to get around. There really wasn't a coherent UK-wide policy approach, although in my London bubble things were starting to change. At the time, Johnson was freshly into his second term as London mayor and there was outcry that his first generation of so-called cycle superhighways had been killing people. These were just painted stripes and in many cases, not even actual cycle lanes.


Cycleway 8 in London. The wands are a Pandemic response. The paint was how Mayor Johnson thought in his first term.

In 2013, I looked at Transport for London's Cycling Safety Innovation Trials 
being undertaken by the TRL and later the same year I actually visited trials twice to see them for myself. First as a cycling participant and second on a professional institution visit. Change to the cycling design approach with regulatory change was on the move, although it took until 2020 for the Department for Trasport to update its design guidance (for England) with LTN1/20 - Cycling Infrastructure Design. While the devolved administrations have their own guidance, LTN1/20 probably just pushes ahead, although I'd say it needs updating already, but that's another post.

I think looking back at the approach to cycling infrastructure design, it has increasingly became a mainstream subject in my profession from a conceptual point of view, but it is still taking time to permeate best practice. That is personally beneficial as almost three years ago, I quit my local authority job for my third private sector role which majors on walking and cycling design, but hopefully I'll end up out of a job one day.


The test roundabout at TRL (long since removed). It's a great pity the "shark's teeth" give way markings didn't make into UK regulations.

Beyond the potted roundup of how we and I got here in 2022 from a cycling design point of view, I think the posts on this blog have broadly crystalised into three themes:
  • In-depth technical posts,
  • Infrastructure Safaris (a term coined by Sally Hinchcliffe), and
  • Rants - subjects which had got my goat and which needed a venting.
The in-depth technical posts have been very enjoyable to write because they have pushed me to undertake some research which in itself has been fascinating because it often showed that many "new" ideas weren't new and we were just rediscovering things lost to a generation or two of designing and managing our streets for cars. For example, my research into traffic orders supporting "Low Traffic Neighbourhoods" (LTNs) earlier this year definitely showed they were nothing new at all.

My infrastructure safaris have been very enjoyable, whether solo or with other people. During the Pandemic lockdowns, they became vital to my wellbeing, but more generally, they have been a great way of looking our how other designers have been working. As I have said, nothing is new and so borrowing (and hopefully improving on) other ideas is as old as time. I also think that as design approaches have matured, I've probably shifted more to safaris in recent because they are a great way to explain design concepts. 



Ranting has been there in the background, although I've probably had it knocked out of me a little bit over the last couple of years because it takes too much energy! One of the subjects which rose to some controversy has been the deployment of LTNs and I am contantly irritated by the bad faith of people who use increasingly contorted arguments against them without admitting that they want the status quo. There's often a suggestion that main roads need treating too (they do), but that's more of a delay tactic than a genuinely held position because in my experience most who oppose LTNs also oppose main road changes. In the summer of 2020 I wrote about this and my post was commended in the Active Travel Media Awards in the blogs cateogory that year. Completely unexpected and very much appreciated.

I don't tend to keep an eye on the statistics generated by this blog (apart from posts like this) and so it's always interesting to see what is popular. The blog itself has passed 875,000 views (whatever that actually means), but more importantly, the most popular post with 10.7k views remains my 2013 "Kerb Your Enthusiasm" piece which was my original attempt at explaining how kerbs are used. I think this truly demonstrates that people are interested in how our roads and streets fit together in detail. I don't know what the least popular post was because Blogger doesn't seem to rank all of them, only the top 20 or so!


The very first deployment of entrance kerbs in the UK on the Coundon Cycleway in Coventry.

Speaking of kerbs, my ramblings have actually helped get the UK a new civil engineering element - the "Dutch Entrance Kerb" which is common outside of the UK (and of course in the Netherlands). It's a simple premise which uses a large ramped kerb to support the use of continuous treatments over side streets where walking (and cycling) space is prioritised over motor traffic accessing the side street.

My writing about the concept led to a meeting with a couple of Charcon's team about four years ago. As ever, ideas are great, but you need someone willing to to put their money where their mouth is and so the first potential customer for the kerb units was Salford City Council which wanted to use them in their Chapel Street scheme. Catriona Swanson was working for the council at the time and was instrumental in getting them included in the proposals and all of a sudden, the idea was a reality. Well actually, it wasn't quite there because the scheme ended up getting delayed. At the end of 2019 I was able to reveal the details of what was happening and at the start of 2021, I was able to talk about the first deployment of the units in Coventry on the Coundon Cycleway.

Although I have written quite a bit about cycling, I have also tried to cover walking, given it was my original point of interest. Walking is harder to talk about because the infrastructure already exists in most people's minds. In fact, given it's 2022, we should be talking about walking and wheeling if we are to be inclusive. In terms of infrastructural tools, there's probably only the side road zebra we need to add to our tool box. On the other hand, the use of infrastructure seeks to reduce the attractiveness of walking with staggered crossings and long wait times.


An experimental side road zebra crossing in Salford.

Back in 2014, I was reminded how important properly constructed dropped kerbs were as a new addition to our family got me back into practice pushing a buggy. This also was a good lesson in remembering that what we do to help one group in society, invariably makes life easier for many other people too. Maybe that's the key to walking and wheeling because it's the mode that requires the most personal effort and every twist, turn and long crossing can be tiring and works against the idea of a 15-minute city.

I think some of the problem with walking and wheeling being the original mode of transport with its (UK) infrastructure just being there is that there is very little research work going on to update practice. At the end of last year, I considered the PV² assessment framework which should have been consigned to the bin in 1995, but still lingers in highways departments today. It was actually a pretty shocking piece of research to undertake because it relies on the number of people crossing now - the fit and the brave. It was therefore nice to propose a different assessment framework, especially as the official one from 1995 which replaced PV² has also been withdrawn.


People don't behave like traffic models.

My writing hasn't always been about this blog, although did have the Silvertown Mole writing about the awful Silvertown Tunnel scheme in 2018, my only guest post so far. In March 2014, my first spin-off column appeared in Highways Magazine and my 84th article will be out in the next week or so. This has enabled me to reach a different audience and one which needs to change its approach in the face of the climate emergency. I started my micro-consultancy, City Infinity, in 2017 because I was frustrated with the direction of my day job (lots of annoying politics) and that has spawned three design guides and some interesting little projects. As it turned out, I didn't end up doing this as my day job because an exciting permanent role came up, but I still take small commissions and who knows what might happen in the future?



As a practical person in terms of learning and doing, I have found the more academic concepts in life challenging and often hard to understand. In the last couple of years I have take some more challenging training courses to try and get out of my comfort zone and it was an absolute revelation to complete the Urban Cycling Institute's Unravelling the Cycling City course in 2020. I think it was this piece of study which finally allowed me to conceptualise the fact that most of the "stuff" we have on our streets is motoring infrastructure. It's an area of thinking which extends to ideas of how people's behaviour responds to the environment you give them and for walking and cycling, rules created for driving often fail because of the way people behave when under their own power. The "how" of unravelling driving from urban areas is a subject which continues to fascinate me and I look forward to learning more.

Talking of unravelling driving, another little piece of my mobility jigsaw changed last August when our aging car died on us just before going on holiday. The hire car we arranged was another opportunity to consider changing technology and 10 months later, the idea of actually another car becomes more distant by the day.


My changing mobility choice.

Luckily, my Christiania tricycle has picked up the slack for local heavy lifting and the train for the odd long trip. There have been some trips we just don't make any more because they were only accessible by car, but to be honest, they're really not missed and there are other things to do. Life moves on and life changes.


My tricycle.

Let's go back to my original blog post where I asked "what do we really want". It was probably a pretty rhetoric question and it probably still is. Maybe it's because I have become immersed in my professional area of interest, but I think the UK in 2022 has become more aware of the importance of streets as places. Whether that has been a general shift in thinking, or whether highlighted by the Pandemic, we have (in some places) rediscovered our neighbourhoods. We have also realised that many places have become very inaccessible through gradual creep of motorisation.

One thing I haven't really written about is the climate emergency. This is mainly because I really don't understand enough to have any sort of authority on it (there are plenty who do), and partly because it's a tough thing to conceptualise. Even within the last few weeks, I heard Professor Glenn Lyons quote the economic anthropologist Jason Hickel who said "The Status Quo is a death march and we must abandon it". We are already beyond the point where we can pretend that business as usual is possible. Professor Lyons' words are going to haunt me going forward (with thanks to Glenn for the link). What I do know is our urban transport future and climate change cannot be solved by betting on a technological solution, it needs a huge reduction in car use and repurposing of streets to enable walking, wheeling and cycling as well as creating countless local responses to heating and flooding.



So what's next? Well, writing a weekly blog post has been hard work and at times it has been very time-hungry with site visits and research. I had been toying with retiring the blog when it hit 500 posts for some time, but it's not something I am ready to give up just yet. What I am going to do is to be less worried about posting weekly and I'll be having some time off. I have some family trips in the next few weeks into which I'll hopefully squeeze a few mini-safaris into and I've other places to visit and revisit on my long list. I've also a pile of books I want to get through, so local campaigning I need to help with and some other ideas at the early stage of thought.

Other than that, thanks for reading, thanks for the feedback and look out for my next post in a few weeks time which will most likely be another safari post. Oh, and where does the cheeky monkey come in? It's apparently Indian slang for the 500 rupee note which used to have a monkey printed on it and it became very familiar in London to mean 500. We're very good at nicking everyone else's ideas you see.

Saturday, 28 May 2022

Traffic Signal Pie: Innie Vs Outie

Over the last several months, I have been thinking about Circulating Cycle Stage Junctions and thought that it's high time I wrote about them.

Circulating Cycle Stage Junctions is the formal name given to signalised junctions which have all cycle movements running at once in an orbital fashion around the junction (usually with-flow and clockwise). They're covered in LTN 1/20 (for England) in Section 10.6.21. There's very little guidance on them because we haven't built many, although (reading between the lines), the Department for Transport seems to be a little nervous, especially as they don't like the idea of mini-zebra crossings over the cycle track. In past days, the DfT would have undertaken research on such innovation and published specific detailed guidance.


"Green at the same time (bottom); free right turn (top)"
Deventer, The Netherlands, 2015, now rebuilt.

The other thing to note is the junction will run pedestrian movements at the same time as cycles (it's not a total requirement as variation is possible), but pairs of diagonal cycle movements are frowned upon because apparently, people cycling can't negotiate the space with each other. I am still waiting for a signals engineer to point out where this is prohibited in law however. On this, the Dutch, use the "simultaneous green" which I picked up on in my 2015 visit to Deventer in The Netherlands (above).

Perhaps the closest UK example of a simultaneous green of this I have found is outside Hatfield Station (below) where there are toucan crossings on all arms of the junction with a pair of diagonal puffin crossings. In reality you shouldn't cycle the diagonals, but people can and do. The world doesn't end. Still, this post isn't strictly about simultaneous greens, it's about Circulating Cycle Stage Junctions of which simultaneous greens would be a subset if the UK "allowed" them.

The junction outside Hatfield Station with crossings marked in a large X in the middle all controlled by traffic lights.

The type of protected Circulating Cycle Stage Junctions that people might have in mind will be the Dutch approach which (in urban areas) places cycle tracks between general traffic and pedestrian space. The cycle tracks being "innies" as being on inside of pedestrians (well, I have to have a title for the post!).

A cartoon sketch of the junction outside Weesperplein underground station.

The photograph above is a wayfinding map at the Weesperplein underground station in Amsterdam, which is a handy diagram for the kind of layout I am talking about. Footway is grey, cycle tracks (all with-flow/ one-way) are pink and the carriageway is dark grey. In general, there is a buffer between the carriageway and the cycle tracks with the footways to the rear of the space. 

From a walking and cycling point of view, the desire lines through the junction are fairly straight and the turns for cycling relatively tight (but people will be slowing down to turn anyway. The design is very scalable and can be used in some pretty tight spaces (with tightness sometimes a factor of the number of motor traffic lanes being accommodated).

The approach to the junction from a cyclist's point of view with traffic to the left and a footway to the right. All controlled with traffic signals.

The photograph above is of one of the protected junctions on Middenweg in Amsterdam. This quite an interesting street to look at for various design features. The main junctions follow similar principles, but there are lots of little tweaks and adaptions for local circumstances. You can see some of the common features;
  • The cycle stop line is beyond the traffic stop line,
  • Cyclists at the stop line are protected by right turning drivers by an island.
  • There is a pedestrian refuge between the cycle track and carriageway.
In this particular location, general traffic, cycle traffic and walking get a green together, with right turning drivers held on red with their own signals, but even that varies a little bit on each arm. The Dutch tend to have some fairly complex signal arrangements compared to the UK. For example, walking and cycling may get a green to run ahead with right turning drivers getting a green a few seconds later (called a "leading green"). People walking and cycling ahead have priority over the turning traffic later into the stage, but it is a conflict that simply isn't permitted in the UK and as far as I know, it is a practice which is gradually being eliminated in the Netherlands anyway.

A corner island protecting cycle traffic in the junction from right turning drivers.

The photograph above is another part of this junction which again shows a protection island. The protection island helps to keep the turns tight and also works to try and provide stopping space for a turned driver about to cross the ahead cycle track. You can also just see on the top left  of the photo, the ends of a central islands jutting into the edge of the junction which also physically works to slow left turning drivers. These are preferable to have, but not always applied (often because of space).

Anyway, this isn't meant to be a post about the detail of Dutch signal design and for more information, have a look at this video by Not Just Bikes. So, can we have this arrangement in the UK? The simple answer is yes and in fact, the London Borough of Waltham Forest have built several of this type of junction under UK signal rules. Before I go on, I will mention that there are UK rules, but equally, I have been told by signals engineers that certain arrangements are not permitted (as the simultaneous green I mentioned earlier), yet they've never shown me an official list of things which are thus banned. Always worth challenging in my view.

A red two-way cycle track crossing one arm of a large signalised junction.

The photograph above is the junction of Lea Bridge Road with Orient Way and Argall Way. The photograph is actually of the anti-clockwise direction because with this scheme, Waltham Forest decided to make the orbital cycle track two way. I am not entirely sure why it was done this way, because in use, following the clockwise UK convention works fine for right turns which can easily be completed in the green stage. In fact, I managed to cross three of the arms to perform a U-turn on my bike in this video:


In fact having the two way operation technically means that people could cycle diagonally through the junction and "escape" on the other corner and I have seen the odd person it. Most follow the rules and cycle clockwise, although because the approach cycle tracks are with-flow. In fact, it's not really following any rules, it's just following the obvious layout (below). In terms of the traffic signal sequence, there are two traffic stages (north-south and east-west) and the third stage is the all round walking and cycling green.

The walking and cycling green stage and we can see people cycling across the junction arms on the cycle tracks.

This junction is a little untidy in places with awkward kerbs and lots of flush areas between cycle track and footway, but it does a very good job and in the context of lower pedestrian flows, it seems to work fine for walking and it certainly has larger pedestrian waiting areas than some Dutch junctions I have experienced (below).

Three people wait to cross the road, one is holding a toddler and they are stood on a narrow paved strip with the road in front of them and a cycle track right behind them.

A pedestrian waiting area in Amsterdam. About 300mm deep!

The sketch below gives an idealised and anglicised general arrangement of this type of junction. It's squeezed in and so the central islands are omitted in the example, but with 2 metre cycle tracks, 2 metre floating pedestrian waiting areas and internal radii of no sharper than 4 metres for cycle traffic, it gives a good idea of the space needed - a total of 32 metres on the narrowest diagonal points which is fairly tight, but things could be squeezed some more.

A sketch of a crossroads. On all arms there is a pavement, then a cycle track and then the road.
The Dutch are always tinkering with layouts and with my example, if we were starting to get cycle traffic congestion, then there's a couple of things we could do to increase capacity.

A zoomed in view of the sktech as described in the text below.

In the sketch above, I have "hollowed" out one of the protection islands to give a little more waiting capacity at the crossing of the junction's western arm. The crossing markings narrow on exit of the crossing which reflects the fact that people move at different speeds and will naturally merge ahead. This example could be due to a busy northbound flow blocking the westbound flow. The extra space created at the northbound stop line just allows people to wait in a row to free up space for westbound cycle traffic to pass behind. This is known as the "banana and chip cone" treatment and there is much more from Bicycle Dutch in his blog post on the concept.

That's the "innie", what about the "outie"? Well, in Greater Manchester, the design geeks had been cooking up a UK version of a Circulating Cycle Stage Junction a while. I first saw the idea presented by "Cyclops Dave" at the Cycling Embassy of Great Britain's AGM back in the summer of 2018, but it took two more years before anyone could see one in the asphalt and concrete. I am of course referring to the CYCLOPS junction which stands for CYCLe Optimised Protected Signals junction and there is a whole paper you can read about the concept.

The first example was built on the Chortlon Cycleway in Manchester at the junction of Chorlton Road with Royce Road and despite the several Dutch-style junctions Waltham Forest had already built, this design seemed to be one which captured the imagination. Maybe it was marketed better, but I think the first looked quite familiar to many people, including engineers and politicians in terms of how it was laid out.

The floating pedestrian island of a CYCLOPS junction. There is a signalised crossing of the road to the left and ahead with a cycle track crossing in the foreground.ead

As you can see in the photograph above, the CYCLOPS places the cycle track crossings outside of the pedestrian crossings and hence "outie". In effect, this means instead of the 8 smaller floating pedestrian waiting areas, the CYCLOPS has 4 larger waiting areas which double up as the protection islands for cycle traffic. The Chorlton CYCLOPS has the simple three-stage signals arrangement as with Lea Bridge Road. This first example has lots of fiddly sub-islands in the pedestrian area which form tapping features for long cane users.

A CYCLOPS sketch with large pedestrian islands with cycle tracks passing behind.

The sketch above is at the same scale as my Dutch example, but you can clearly see the change in the layout. At 37 metres across the diagonal, it's a little larger, but again, it could be made a tighter. One the of main disadvantages to the design is that the adaptability is more challenging because any reduction in the protection island is a reduction in the pedestrian waiting area. However, this design could be more accommodating where there are greater pedestrian flows given the waiting area capacity. There is also the potential for diagonal pedestrian crossings, but no later conversion to simultaneous greens if they were ever allowed.

Another view as described in the text below.

The other thing I like about the approach is that the turns for cycles is a little more gentle size for size with the Dutch layout. This is a bit more forgiving for users of non-standard and adapted cycles. The photograph above shows a little more of the easier turning space. It also shows an advanced stop line which was a feature of this first scheme because there were concerns that "more confident" cyclists would want to go back into traffic if a general green was available. If that's a problem, then in fact the junction should be rebalanced to walking and cycling.

The other thing which is being done within the CYCLOPS arrangements is the introduction of more complex junction layouts, although many seem to end up with staggered 2-stage crossings for pedestrians while cycle traffic gets a single stage. It might be a pragmatic way to get cycling infrastructure in (as cycles need shorter greens than pedestrians), but we should be prioritising walking and wheeling far more than we are.

The same CYCLOPS sketch with mini-zebra crossings over the cycle tracks.

One of the issues with the Dutch style approach is where we want to place mini-zebra crossings over the cycle tracks. As I mentioned above, the DfT doesn't like the idea of zebra crossings over cycle tracks being in line with signalised crossings of the carriageway and this is mentioned in 10.6.22 of LTN 1/20 citing potential confusion, especially for visually impaired users.

I think that they might be over-cautious here, but if we want to help people with visual impairments to navigate what is more complex than a basic signalised junction (where we force cycles to mix with traffic), then the offset nature of the CYCLOPS crossing of the cycle track helps, plus adding a mini-zebra crossing is less confusing than the in-line arrangement. I'm happy to go with that (sketch above).

Dozens of CYCLOPS junctions are planned in Greater Manchester, but the idea has already escaped the region and one has been built on Histon Road in Cambridge (the UK's fourth), although Google hasn't caught up yet, and I'm aware of quite a few across England in various stages of being thought about and designed. If you have time, it's worth watching the Ideas With Beers session on CYCLOPS junctions as there's even more about the clever detection for cycle traffic used on the first scheme which is being applied elsewhere.

As an exercise, I took the concepts and applied them to a junction in London which I had recently visited - the junction of Whitechapel High Street (the western end of Cycleway 2) and Mansell Street, the latter of which has just been linked for cycling to the south with Cycleway 3. Appendix B of LTN 1/20 contains the Junction Assessment Tool (JAT) which is used to asses the safety of major junction schemes for cycling. To go through the process would take a post in its own right and if I am honest, I am still practicing using the tool. However, based on my knowledge, I applied it to the junction and it scored 50%. 

The two way cycle lane meets a junction. The road is on the right and bolt down islands protect cycle traffic. There are various traffic signals ahead. A cyclist can be seen going ahead.

The Mansell Street approach to the junction with a cycle gate.

70% is considered to be what is needed to be operating in the Dutch-level of excellence and although 50% is a "fail" in this regard, it is significantly better that it was and indeed, it's really good considering just how low cost the recent interventions were. The JAT can also be used as a design tool to see what could be tweaked to improve the score, but again, that's a different post.

My two redesigns have removed the staggered pedestrian crossings on the northern and western arms, but otherwise the number of lanes are the same, as are the motor traffic movements. Both designs score 100% on JAT as every cycle movement is fully protected. First the Dutch (innie) design.

A Dutch version of the junction.

The key thing to mention is the southern arm (Mansell Street) isn't completely Dutch in the context I have been explaining, because it has a two-way cycle track on the eastern side; but a Circulating Cycle Stage Junction can neatly integrate this different arrangement easily, it's just there's no "free" left turn to the west (heading into The City).

The junction all fits in quite well, give or take, but the immediate issue is the width of the floating pedestrian islands. I have designed them at a minimum depth of 2 metres, but it's still tight and pedestrian flows can be high here. They could be made a little larger, but they still represent a tight spot in places. I have also added in-line mini-zebra crossings over the cycle tracks to annoy the DfT.

A CYCLOPS version of the junction

Then we have the CYCLOPS version which gives far larger pedestrian waiting areas. Cycle traffic is diverted off the desire a bit, but that could be improved - the gentler turns and "outie" nature is a directness penalty. I could probably push the Mansell Street cycle crossing a bit further north and for fun, I have added a double mini-zebra to the northeast corner for a crossability improvement.

This design probably fits better with the local context which can have some busy pedestrian flows and plenty of bus movements. It also marks the point at which the east - west corridor turns from a "road" into a "street" as you enter The City of London to the west of the junction. A city centre street should ideally be low enough in terms of motor traffic to integrate cycling and The City is really pushing this idea. Seems like a logical place to make the transition to me.

We have two ways of achieving the same thing, both having advantages and disadvantages. The key for me is using the space, context and junction layout to think around the best solution. The Dutch approach works perhaps because there are so many people cycling in The Netherlands, so walking space is less of a issue. At the Mansell Street junction, walking is key and I think the CYCLOPS is the better option. You pay your money and you take you choice and maybe we'll be swapping successful CYCLOPS junctions for Dutch layouts in 30 years!