Monday 6 July 2015

A "Pinch-Point" design which slows cars without "pinching" bikes

On-road cycle-lane approaching a pinch-point. A potentially dangerous situation for cyclists. Note how from this view the driving lanes appears to narrow at the pinch-point.
Pinch points are often installed on roads to slow motor vehicles and to provide crossing places for pedestrians. They are often dangerous for cyclists. Road lanes which suddenly narrowed to encourage drivers to slow down can force them to come too close to cyclists resulting in close and dangerous overtaking or even a collision. However pinch-points do not have to be built in this way. In this blog post I illustrate a "pinch point" which doesn't really pinch at all. While the pinch-point appears from a distance to restrict the space for cars, resulting in drivers slowing down at the junction, the driving lane actually widens through the pinch point. The cycle-lane has constant width.

Thorbeckelaan from ground level
Thorbeckelaan has a 50 km/h speed limit. These photos were both taken while riding along the road from west to east.
Before the pinch-point. The driving lane is 2.8 metres wide, the cycle-lane 2.1 m wide. Note reasonably wide parking bays and a 0.5 m wide drain. The width of the drain helps to reduce dooring incidents (explained here), as does this being relatively infrequently moving residential parking and not business parking.
At the pinch-point, the cycle-lane remains the same width while the driving lane increases to 3.5 m in width. This still causes drivers to slow down. The narrowing of the driving lane back to 2.8 metres is visible after the central reservation. At this point there is of course no space for car parking.

From the air
Aerial photos (from Google Maps and Ongelukken Kaart) provide another view of the same cycle and driving lanes. The photos above were taken facing east (i.e. view is from left to right on the photos below):

On-road cycle-lanes - not really a good idea
This blog post highlights a good example of how a pinch-point can be built which successfully reserves space for pedestrians crossing the road and also slows drivers, without squeezing cyclists. It's certainly an improvement over other pinch-points I've seen where cyclists are squeezed by drivers as they pass through pinch-points. However I don't mean to suggest that this road layout, with an on-road cycle-lane, is actually particularly good. Such infrastructure shouldn't be an aspiration.

On-road cycle-lanes are well known to cause a myriad of problems for cyclists worldwide, but context is always important. This arrangement works fairly well on this road because cyclists travelling west-east along this particular road are protected from conflict by several other factors which are specific to the character of this road:
  1. There are few destinations on this road so few clashes with drivers starting and stopping.
  2. Residential side streets do not provide through routes by car and therefore few cars turn into or out of the side roads.
  3. There are no bus-stops along this road.
  4. The cycle-lane is of a good width, providing wiggle room.
  5. The parking along the road is for residents, so those cars rarely move and the drain provides a gap where there are parked cars, reducing the risk of dooring.
Because of these factors, almost all motor traffic on this road travels all the way from one end to the other without stopping and there are few conflicts caused by drivers cutting across the cycle-lane. Note that both ends of this road have extremely safe junctions. At one end there is a very safe roundabout and the other has a very safe traffic light junction. Neither of those relatively large junctions have ever caused cyclist injury.

In other locations the problems would be greater. That is the reason why this layout is by no means a substitute for proper cycle-paths. Though it is overall quite well designed, this road (when riding west to east) is still one of the least pleasant along which to cycle within Assen. In particular, being overtaken by a bus or truck while riding side by side in the cycle-lane is not especially pleasant.

On the other side of the road there's a segregated cycle-path.
A video shows the popularity of the sports club on that side
The other side of the road
This road is unusual in that while there is an on-road lane for cyclists heading from west to east, cyclists heading in the east-west direction are provided with a kerb separated cycle-path on the other side of the road. The cycle-path not only feels far safer but due to being 2.5 metres wide and having a 0.5 metre buffer, it provides more space for cyclists further from motor vehicles.

On this side of the road, there is little or no discomfort due to being passed by a large vehicle.

Note that with a proper separate cycle-path, pinch-points have no effect whatsoever on cyclists so can be made quite narrow in order to reduce the speed of motor vehicles. However on this road the driving lane widens on both sides through the pinch-point.

Read about other examples of pinch points with cycle-paths and at village entrances.

This junction doesn't have a good safety record. Why ?
This road is a busy west-east route for both bicycles and motor vehicles. On its length there are three pinch points of a very similar design to that emphasized here. Two of these pinch points have a good safety record while this one has a relatively bad record by the standards of Assen. Why ? The reason is that this pinch point also happens to be a junction with a busy main cycle-route north-south which provides access from some suburban areas in the North to the city centre.

The red flag with "10" within it on the second aerial photograph tell us that this junction has seen ten incidents since 2007, four of which caused injuries to cyclists, one of which injured a moped rider and one of which caused the unfortunate death in 2009 of a 76 year old female moped rider, in collision with a large car.

A cropped version of the first photo in this blog post. Does
each participant in this scene know what the others are doing?
The most common recorded cause of crashes at this location is "no priority given". This suggests that people find it difficult to see each other well while maneuvering or find it difficult to judge whether they have time to complete a maneuvre. Turning across traffic from an on-road cycle-lane is always difficult because the cyclist's proximity to the driving lane results in having limited rear visibility. 180 degree head swiveling is required to see behind, which complicates control of the bicycle and makes correct decisions more difficult to make. The lack of gap between the cycle infrastructure and road also makes it more difficult for drivers to predict cyclist behaviour. Could you tell at a glance what everyone in the photo beside this paragraph was about to do ?

It has been known for many years that unsignalled crossings are far more dangerous than either well designed roundabouts or well designed traffic light junctions. Therefore it should not be surprising to us that the well designed junctions at both ends of this road (both of which are illustrated at the links in the previous sentence) have a far better safety record than this relatively minor junction half way along this road.

Due to the designed higher speed of this road (speed limit 50 km/h), the heavier traffic at this location and limited sight-lines compared with motor vehicle speed, this is not a suitable location for a cycle priority crossing like those which I looked at a few weeks ago.

How not to build a pinch-point
Pinch points which cause problems for cyclists are far more common than those which do not. Here are two examples which featured regularly on my commutes in Cambridge:

Don't copy this: Several narrow pinch points exist on the narrow but busy trunk road which passes through Harston in Cambridgeshire. Cyclists have no good alternative to using this road. Drivers habitually exceed the speed limit here, especially at the ends of the village where the speed limit drops sharply from 60 mph to 30 mph. I estimate the width of the road through this pinch point to be around 3 m. No additional space is provided for cycling. Note also dangerous steel railings which cyclists can be pushed against and that the paint pattern used in the middle of the road through almost the entire village also encourages drivers to give as little space as possible when overtaking cyclists.
Don't copy this: When the new development of Orchard Park was built in Cambridge, the re-design of roads near the development brought many new dangers for cyclists. Several crossings and other features were added to Kings Hedges Road and they were nearly all built with pinch points in widths known to encourage close passes by drivers. Orchard Park could have been built similarly to new Dutch developments. e.g. Kloosterveen
Both of the examples of pinch-points above increase danger to cyclists. They reserve no separate space for cyclists at all and they encourage bad overtaking behaviour by drivers, especially those who misjudge the speed of a cyclist and believe they have more time to overtake than in actually the case.

But it's not only in other countries that mistakes have been made. We don't have to look very far from home in the Netherlands to find bad examples from this country. For example, we also have these examples near the centre of Groningen and in a residential area in Assen:
Don't copy this: This very poor pinch-point in Groningen demonstrates nearly the opposite of what is shown in the good example above. The road lane decreases from 3.5 m to 2.8 m through the pinch point while the cycle-lane is an inadequate 1.2 m the whole time. This leaves a width which encourages close overtakes - e.g. by that bus. Just because something exists in the Netherlands, that doesn't mean it's a good example
Copy this only with caution: This is a residential street in Assen which has a relatively minor speeding and through traffic problem. These planters create a different type of pinch-point, potentially causing conflict between cyclists and drivers heading in both directions. Even when the same idea is used with a cycle bypass there is still potential for danger.
Just because something exists in the Netherlands, that doesn't mean it's a good example. Context is always important and ideas which may work well enough in one street in the Netherlands cannot necessarily be translated to a different place and work well there too. Older examples like this are perhaps more understandably sometimes of poor design, but there are also examples of less well thought through brand new infrastructure which work poorly.

Not enough space ?
One of many very narrow streets in
Assen which didn't have enough space
for a pinch-point or for motor vehicles
either. Assen moved the through
motor traffic
I predict that some peoples' response will be there is not enough space in a given location for a safe pinch point to be built. This may be true for some locations, but if it is true then I would contend that before considering building a dangerous design of pinch point you should think very carefully about whether any pinch point at all is appropriate in that location.

Other ways of removing conflict are possible. For instance, moving motor vehicles onto completely different routes. This has resulted in many minor streets and roads in the Netherlands working extremely well for cyclists and pedestrians without the need for pinch points or other obvious physical infrastructure.

Learn more by seeing the real thing
Book now
We offer the only independent study tours of Dutch cycling infrastructure with a native English speaking guide to minimise the chance of "lost in translation" errors. On our study tours we take care to demonstrate not only many good examples of good infrastructure but also explain why the less good examples should not be emulated.


Unknown said...

here we have possibly the most incompetently designed pinch points ever...

absolutely encourage motorists to go for it...

these are bad as well:

and I really hate this one on a 50 mph speed limit:

Shred use path on the right, but as usual it's difficult to get to for those going in the other direction.

Kevin Love said...


That first one is really bad. Why do I suspect that a lot of people ride on the pavement? Probably because I would.

seiklmeikl said...

Berlin is quite fond of cycle lanes in these years. At pinch points however, the solution is to simply let the cycle lane end some meters before. Its up to drivers and cyclists to negotiate. Guess who "wins" in these negotiations. The two by far most important causes of incidents in Berlin are: right turning vehicles and mistakes whilst moving into a priority lane.