Monday 12 August 2013

Ten Bus Stop Bypasses for Bicycles. Bikes, buses and bus passengers can be in harmony only when separated (floating bus stop)


This video shows ten of the bus stop bypasses for bikes in Assen. They're not especially good "cherry picked" examples but simply the ten nearest my home. Most are alongside normal roads, one is alongside a bus road. All are within a couple of km and they were videoed in half an hour early last Friday morning. They're entirely typical of normal cycling infrastructure quality in this area. Note that these are all designed to  be convenient for cyclists to use. They do not narrow or have raised sections to slow cyclists because cyclists are already slow compared with motor vehicles and we need to encourage efficient cycling not to slow cyclists down.

Large and small vehicles can never "share" equally. In order to encourage true mass cycling, where the entire population uses a bicycle for a proportion of their journeys, conditions for riding must be subjectively safe. Where bicycles mix with motor vehicles, this feeling of safety is reduced. Where bicycles mix with large motor vehicles it is reduced further.

Don't combine buses and bikes
Dutch bus-stop with no obvious bypass.
Bicycles don't travel on this road at all
but are behind the noise barrier. Watch
this video to see how it works.
Buses and bicycles should never be combined in one lane. This is not only because of the subjective safety issue, but also because the two modes move in fundamentally incompatible ways. Cyclists try to maintain a constant speed because this maximises efficiency (greater efficiency makes riding a bike being practical over a longer distance and reduces the time taken - further aided in the Netherlands by cycling routes being unravelled from motor routes and avoiding traffic lights). Buses, on the other hand, stop regularly to take up and let off passengers.

Where bus and bicycle infrastructure is combined on a long road it often leads to leapfrogging as buses repeatedly overtake bikes and cyclists are given a choice either to wait behind the bus, wasting time and making cycling less attractive, or overtaking the bus while it is stopped, which can be dangerous for the cyclist.

Decades old and unimproved, the least
good example of a bicycle bypass in
Assen still has a place for passengers
to stand before crossing the cycle-path.
New-build doesn't look like this.
Segregate bikes from buses and also from passengers
Bus passengers also clash with cyclists. If a cyclist tries to pass the bus on the "wrong" side between the stop and the bus then this puts cyclists directly into conflict with those who are boarding or alighting the bus. This is avoided in the Netherlands as shown in the video above, though you'll note that in the oldest example, pre-dating modern practice, cyclists are routed on the wrong side of the bus shelter which could cause conflict. No modern bus stop would be designed like this, with conflict built in, but note that even in this example there is somewhere to stand after leaving the bus and before crossing the cycle-path.

Why doesn't Britain copy the best examples ?
After making the video above and while writing this piece I discovered that new bus stops claiming to deal with the problem that the Dutch were had already tackled more than 30 years ago were being introduced in London. Unfortunately, instead of copying from the best tried and tested examples, attempts have been made to design something new. Time will tell whether these are good examples, but there would seem to be reasons to expect them not to be as successful as normal Dutch practice:

New design for London. This will cause clash between cyclists and bus passengers. Why didn't they copy from best practice in NL ?


Bus passengers clashing with
cyclists in Royal College St
(Thanks to @AlternativeDfT)
New bus-stop for Royal College Street, London. It was obvious from the design that conflict should be expected here between cyclists and bus passengers because bikes are being routed between the bus and the bus stop without even a small place for bus passengers to stand between the bus that they're entering or leaving and the bicycle path with through traffic.

Also note that at just two metres wide, the bicycle lanes shown here are narrower than any of the examples in Assen. The high kerbs and the planters between the cycle-lanes and road reduce the safely usable width of the facility.

Even the oldest example shown in my video from Assen is 2.3 m in width. That's on a relatively quiet residential access road and it is just one small weak link in a very dense grid of high quality cycling facilities within a small city. It deals with far fewer cyclists than could be the case in a larger city with fewer facilities.

(a few days later, The Alternative Department for Transport blog included an interesting blog post about how this bus-stop has worked out in practice)

Another compromised design for London. Sharp bends on the cycle-path which is not wide enough. Opened several months after this blog post was written and immediately caused problems due to bad design. Why doesn't London copy best practice from the Netherlands ?
A second example from London is the proposal for extension of Cycling Superhighway 2 in London. This looks as shown in the brightly coloured illustration above.

Dutch example from 1981. Not angular
and you could expect the cycle-path
to be of usable width.
From the picture it initially looked like this example could turn out better than Royal College Street as there is at least supposed to be somewhere for passengers to stand as they leave the bus. In that sense, it's similar to 30 year old Dutch designs. Similarity only goes skin deep, though, as the illustration shows an overly angular design and this could make it difficult to ride around easily, quickly and safely. There also appear to be dangerous high kerbs. It's not obvious why these kerbs are necessary at all, but they could at least be safe like these ones. Lastly, the plans for this cycle-path again suggested it would be just two metres wide.

It can be difficult to pass other cyclists safely within just two metres and if the route is well used this will be a problem. Given that "superhighways" in London are few and far between and that even with the low cycling modal share of London there is a huge population to draw cyclists from, this route probably will be well used.

The real bus stop bypass in London
under construction. "barely 1 m wide"
(Thanks to @AsEasyAsRiding)
Unfortunately, it's not actually been built even two metres wide. Mark Treasure tells me that it's "barely 1m wide at narrow point". A one metre wide cycle-path, with a post in it which narrows its effective width further, deep kerbs either side and passing a bus stop where bus passengers may or may not be aware of an approaching cyclist is very very far from best practice so far as bus stop bypasses are concerned.

The idea of a bypass is to make cycling convenient. It should not cause cyclists to slow down or place them in danger. It should not cause conflict between cyclists and bus passengers. It should also not cause frustrated cyclists to use the road in order to bypass the bypass.

(Hackney Cyclist wrote a very good blog post giving more details of the HS2 route extension)

"40 years behind" is a choice
London is "40 years behind" by choice not by accident. Even now, after supposedly having "gone Dutch", the city is still designing and building inferior infrastructure as seen in the two examples above.

This simply isn't good enough.

Why is London still not learning from the best examples ? Why is the city still trying to find its own novel ways to solve problems which were identified decades ago in the Netherlands and the solutions to which have since been refined to a very high standard ?

Instead of continuing to make costly mistakes, why not send planners to find out what proper cycling infrastructure looks like ?

Thanks to @EdinburghTom
Go North, find an even worse idea...
An amusing blog post from The People's Cycling Front of South Gloucestershire reminded me of another idea for passing buses which has appeared in Scotland recently. The rightly much criticized "Nice Way Code" suggests that cyclists should never pass buses on the left and on the right only "if you must".

Sadly, the "Nice Way Code" campaign has consumed a considerable fraction of the not very generous funding for cycling in Scotland. Instead of making it easy for bikes to pass buses safely and without conflict, they're using cycling money to tell cyclists not to pass buses.

This is just one of the many errors made by the "Nice Way Code", an organisation which sadly is backed by both CTC and Sustrans amongst other organisations who really should have known better.

Campaigners need to campaign effectively.
There is also excitement this morning about the British government having announced the "largest ever investment in cycling". This is apparently a figure of 77 million pounds to be spread amongst eight cities "in an effort to put Britain on a level footing with countries [...] such as [...] the Netherlands".

Unfortunately, this sum of money isn't nearly enough to achieve that aim. The Netherlands spends roughly €30 per person per year, every year in order to improve the cycling infrastructure and continue to expand the network. The total funding available over the next two years is about £148 million and this will allow investment in just eight cities to be at around £10 per head for just two years. This low figure of £10 per head is the figure which CTC actually asked for a few months ago and it's also the figure which will be debated in parliament on the 2nd of September.

Why is there so much complacency amongst campaigners ? Why such low ambitions ? Why do they offer support for inadequate policies and why do they not fight for what is really needed ? When you are already decades behind you cannot catch up by doing a third as much so why are campaigners putting their names to calls to do so little that the UK will inevitably remain in "dark ages" so far as cycling is concerned ? Low aspirations will not result in mass cycling and campaigners working with such aspirations are very much part of the reason why the UK is behind.


"Armadillos" in Assen. These have sometimes featured on
the study tour as an example of what not to do. They get
bumped by passing vehicles.
Update January 2014
The "armadillos" in Royal College Street are a failure. If we had been consulted we could have told TfL in advance that this would happen. It's not something you find often in the Netherlands because it doesn't really work. There are just a few old examples like that shown in the photo on the right. When copying from the Netherlands it's important to look for good examples.

This is not a good example, it's one of those things that you should not copy. Probably the oldest cycle-path in Assen. It works OK in this location because it is alongside a residential area (apartment blocks) and there are very few vehicles turning across the cycle-path. But in time this will be replaced (it was supposed to have happened in 2010).
Update March 2014
The "Armadillos" shaken loose over the last few months by large vehicles which have been accessing a site on this road have been re-fitted and broken examples replaced. Much tidier now.

You'll see from the photo that the cycle-path is generous enough that it doesn't feel cramped and also that cars pass at a reasonable distance. Nevertheless, this is outdated infrastructure. We were originally told there would be a proper cycle-path in this location by 2009.

Each Assen concrete block "Armadillo" is 1 metre long, 25 cm wide and 12 cm tall. When some were loose, I tried picking one up. It was heavier than I thought I could lift without damaging my back. These are not the insubstantial plastic used elsewhere. Nevertheless, this is not adequate.

Also see Hackney Cyclist for more and The Alternative Department for Transport for more about Royal College Street. This post was updated to include the words "floating bus stop" because this newly invented term has become commonly used to refer to bus stop bypasses.

13 comments:

The Ranty Highwayman said...

I have not ridden CS2X yet, but the photos make the cycle track look very narrow, despite there being so much space available. All makes me a very frustrated engineer!

Koen said...

I think you left out a "not" in the caption of your first photo. Not cherry-picked is your standard choice, I believe?

David Hembrow said...

Ranty: There's just no excuse for London to continue to aim so low and I'm sure it is extremely frustrating indeed for people like you who'd like to do a better job. The buck stops with the politicians but they seemingly won't learn.

Koen: Thanks for proof-reading ! It's fixed now.

Titus said...

You should take a look at the newly resurfaced Jan Fabriciusstraat near the recently opened Media Markt store. How they handeled the bus-stop there will proberbly surprise you.

David Hembrow said...

Titus: I'm struggling to find anything at all that I like about that newly opened shopping centre. It's really very poor indeed, an example of several things which should not be done.

I started a blog post a while back about it. It's really quite awful and that needs to be pointed out - hopefully changed.

I know the local Fietsersbond have already complained, and quite rightly so.

Jim Moore said...

If corridor width is a real design constraint then a short section of 2.0m wide lane past the bus shelter *only* is OK and I posit it would possibly be safer than continuing a lane (of at least 2.5m) that allows overtaking of two cyclists riding abreast, especially at bus stops within a CBD where conflicting travel paths between bus passengers and in a location (i.e. a CBD) where bike riders probably have to give way at zebra crossings between the bus stop landing and the footpath.

Just on minimum bike lane widths, with the increase in cargo bikes and the slightly extra width these require, then where these bikes tend to be in numbers (the CBD again) all the bike lanes should be designed to be at least a little bit wider than the minimums stated in the CROW manual.

Trick of the eye: on the CS2 bus bypass graphic I always see a twin-celled open pit just ahead of the bike rider, rather than the single arrow. Is it just me?

Lastly, thanks for another great post David, covering both tactical on-road details as well as strategic objectives including the big one: proper amounts of funding.

Clark Nikolai said...

This is something I had thought of when the separated bike lane on Dunsmuir street here in Vancouver went in. When waiting for the bus, you have to cross the bike lane to get to it.
I found this picture of it:
http://flic.kr/p/8PE7AZ

It works for the most part but I've seen a few minor conflicts and startled people walking across to the bus as someone cycled by.
It was initially a trial but now that it's been made permanent, it's time to add a curve to the path and move the bus shelter. (And tree too I guess.)

Robert said...

In Birmingham (UK), the third option has been implemented outside the recently completed Queen Elizabeth hospital. This was built on derelict land, so the designers had an empty canvas. Space was found for two lanes of traffic in addition to the bus bays, a large central island, and even a verge behind the bus stops, but the designer elected to take the cycle lane right through the bus shelters. Note that this is a PFI (private) project, so politicians cannot be blamed for it.

Manouchk said...

What a great article again! I'm here in Brazil in Vitória and we are trying to get our cycle lanes and if possible with improved quality.
It is very interesting to have this kind of detailed reports on your blog! We can share it here with our cycle lane planners.
Here we already have a cycle lane which does not necessarily make cyclists interact with bus stop. In few (new) cycle path, cycle lane design near bus stop looks very similar to good examples you showed but the bus stop is so crowded that spaces is missed for pedestrian and they tend to walk on the cycle path at rush hour!
We recently discussed exactly this problem with our planners on a new project of a cycle lane. I really should film that kind of negative interactions and analyze it in order to show good points and problems to our planners in order to make it clearer for them!
We have so many other cycling related problems here that we sometimes left behind this kind of problems as if they were details but in fact, even if our existing cycle infrastructures are scarce and yet often badly designed, is does not mean that we have to admit also new infrastructures with problems! One the problems that remains here is ramp inclination. Small ramp can be as steep as 20° for small ramp by law which is quite inconvenient.
We recently measured new cycle lane in construction we found that it was 21°...

Unknown said...

Mostly irrelvant video for most British cities where there is nowhere near as much space as in the sub urban dutch examples. You should think about real contraints before posting.

David Hembrow said...

Benny, you've missed the point. These examples are what we have. Obvious "bypasses" don't exist.

The small inner city streets where someone might have tried to build such a thing were cleared of cars and buses long ago, and that's what Britain should have done as well.

Dutch cities and Dutch suburbs have exactly the same constraints regarding space as do their British counterparts.

Multiparty Democracy Today said...

Do you think those armadillos, the plastic ones, if they were modified to something like concrete to be more durable and embedded into the asphalt, would be worth anything as a cycle lane? Cycle lanes are generally not desirable, I know that, but maybe as a bit of an aid to a 1.9 metre wide cycle lane and a 2.9 metre wide car lane and the same for the other side? As long as it is understood as a cycle lane, not a cycle path or as all that much protection and that it is not ideal.

David Hembrow said...

Restlesstablet: No. The concrete lumps shown at the bottom of the post above are concrete and they go flying when a truck hits them. Cyclists need to be separated from motor vehicles, not have flimsy posts, concrete lumps, fences or whatever. In the example shown, the motoring lanes are 2.9 m wide and the cycle-lane is 2.2 m wide but this is still not really good enough. It has survived so long in this specific location only because the problems caused are relatively minor here. i.e. it's another example of this. Don't copy from less than ideal examples.