Cycle Traffic and the Strategic Road Network – Cycleway design

Earlier this week I blogged about a new Highways England standard – Cycle Traffic and the Strategic Road Network, or IAN 195/16. This is a 68 page document with plenty of detail, and for convenience I’m breaking it up into chunks.

That first post looked at the basics of ‘Cycle Traffic’ covered by IAN 195/16 – this post will look at what the document has to say about how cycleways should be designed, in particular, and what form they should take, according to context.

One of the first things we encounter in this section is a table of desirable and absolute minimum widths, according to the expected flow at peak times, and the nature of the cycleway.

As explained in the previous post, ‘absolute minimums’ can only be used where there are existing physical constraints. But even these ‘absolute minimums’ are reasonable generous – a two-way cycleway with a peak hour flow of over 150 cycles per hour (two-way) has to be at least 3.5m wide, and it can be that narrow for only 100m at a time. This is roughly equivalent to sections of the new superhighways in London, so a good standard, even for an absolute minimum.

The 4m wide cycleway on Blackfriars Bridge

Explicitly, these values also do not include the additional width required ‘to maintain effective width’ – i.e. the usable width of a cycleway with kerbs or vertical features beside it.

From this table, a cycleway with vertical kerbs requires an additional 20cm of width, while one with a feature like a railing or a parapet requires an additional 50cm of width. All very sensible stuff.

We also have guidance on how to improve social safety, including lighting, making sure the route is overlooked by passing people and traffic, ensuring good sight lines, and low vegetation. This is also refreshing –

Sign posts and lighting columns shall not be placed within the width of a cycle track, and shall have a minimum clearance of 500mm between the edge of the cycle facility and any parts of the sign or lighting assembly that are less than 2.3m in height.

The word ‘shall’ here is a requirement, not a recommendation.

IAN 195/16 recommends (throughout) separating pedestrians and cycle traffic, and is explicit that the difference between the footways and cycleways should be clear, either with verge separation, or with height separation. Forgiving ‘splayed’ kerbs are recommended –

Using splayed kerbs along the edges increases the effective width of the cycle track and helps to prevent collisions by reducing the risk of pedals striking the kerb.

Table 2.3.2 in the document is too large to include here, but is a very good summary of the potential , respectively, of using one-way or two-way cycleways, and the appropriate contexts for their use. Again, it’s all sensible stuff – for instance –

If cycle users persist in using one-way tracks the wrong way, this suggests that the facility may need to be made two-way.

This kind of behaviour suggests an obvious desire line, with people cycling not wishing to cross a road to cycle a short distance in the ‘wrong’ direction. Similarly –

In situations where there are one-way cycle tracks on links approaching junctions, designers should provide two-way cycle tracks within the junction if they offer a safer more direct way to negotiate the junction.

We also have an important table setting out the minimum requirements for horizontally separating a cycleway from a road, according to the speed of motor traffic on that road.

This table means that ‘stepped’ cycleways (or ‘hybrid’ cycle tracks) are only appropriate on roads with 30mph speed limits, or less.

While I agree with the height stipulation for stepped tracks (50mm above the road) I do not agree with this –

Stepped tracks shall return to the carriageway and become initially mandatory lanes before changing to Diagram 1010 (reference TSRGD) [15]2 markings through junctions.

My view is that stepped tracks should continue across side road junctions, unchanged – that visual priority is lost if the track ‘returns to the carriageway’ and becomes a mere painted lane (the dashed ‘1010’ marking), and indeed any advantage of having the cycleway raised above the road (which would slow drivers) is lost too. Old Shoreham Road is used as a photographic reference in IAN 195/16, yet the stepped tracks at side roads here (generally) continue unchanged across side roads.

This ‘continuous’ kind of arrangement also ensures the cycleway remains level, rather than bumping up and down at each side road, as would be the case with the IAN 195/16 stipulation.

There follows the familiar guidance on cycle lanes, which includes some advice on how to apply ‘light segregation’ features. I mentioned in the previous post that I think 30mph limits with 5000 vehicles per day is too weak for mere painted lanes, so applying light segregation in that context would remedy that weakness. IAN 195/16 mentions wands, ‘low height separators’ (presumably things like armadillos), and short sections of raised kerb – but doesn’t give any view or advice on which is superior or appropriate, according to context. My own view is that armadillos are pretty much a waste of time; the raised sections of kerbs or wands are much more effective (and indeed they appear to be being used by TfL for the diversion of CS3) and it would be good to see them being given greater weight here.

In this section we also have advice on dealing with cycling alongside buses, and bus lanes. It’s not explicitly stated, but the impression given is that designing for cycling in bus lanes should be avoided as much as possible. Where ‘sharing’ has to take place, it must be at 30mph or under, and the bus lane must be ‘no narrower than 4.5m wide’, ruling out sharing in conventional bus lanes in standard lane widths. I think this is still weak, however – IAN 195/16 would (for instance) still allow cycling in 4.5m, 30mph bus lanes with up to 5,000 buses a day (two-way flow). That’s not an environment I can envisage my other half cycling in.

The bus stop bypass recommendations are good, however, with an explicit requirement that bus stop islands are used –

The bus stop shall be placed so that users on the bus do not directly step down onto a cycle track when leaving the bus.

That is to say – ruling out the ‘Copenhagen’ style of bus stop where bus passengers step straight onto the cycleway from the bus, familiar from Royal College Street in Camden.

The final noteworthy element in this section is a similar, explicit rejection of all kinds of ‘access control’ on cycleways, except for bollards.

In most cases, a single bollard (reference Figure 2.3.8) is sufficient to prevent motor traffic from entering routes for cycle traffic. The gap between posts and other physical constraints shall be no less than 1.5m so as to prevent access by cars while retaining access by cycles. Bollards shall be aligned in such a way that enables a cycle design vehicle to approach them in a straight alignment.

A frame and K Frame type barriers, often used to prevent motorcycle access, shall not be used on cycle routes because they cannot be negotiated by the cycle design vehicle.

… the ‘cycle design vehicle’ being an (abstract) vehicle of fixed dimensions that, if designed for, would allow access by all types of cycles, including hand cycles, cargo bikes, adapted cycles, tandems, and so on.

So, all in all, there is more excellent stuff here, although it has to be said it is a little weak in places, and contains at least one requirement I don’t agree with. The next section to be covered is junctions, which I will examine in the next post.