Surveying Sustrans’ Skillards

My day out in Bath as a Sustrans quality control technician…

With a to do list sagging under the strain of new documents landing on it, I did the sensible thing and allowed myself to be sidetracked by the latest Sustrans barrier design. Initially I was incredulous – hasn’t my (admittedly tiny) contribution to the sustainable transport charity’s barrier bureau been used to conjure up enough ways to disadvantage disabled bike users? Apparently not.

SkillardSpecAs far I can tell it doesn’t have a name like york-chicane or A-frame so I’ve christened it the skillard: (n) a collection of bollards in a skittles formation requiring additional skill to negotiate.

I imagine they’ll be scalable, so rather than the skillard5 seen here a skillard7 would be used on a wider path and, as I discovered, a skillard4 on a narrower path.

Then a chance encounter with Tyne Tom in Leeds pulled me up short. He put the new cycle-friendly design handbook it’s taken from into some context, and explained the issues around trying to limit speed on bike paths which is the primary purpose of this new barrier – it’s nothing to do with motorbikes – along with the problems encountered when trying to get a desired solution accepted by a local authority. Like the boy named Sue, I came away with a different point of view.

I put aside my experience of riding in the Netherlands where obstructions are rare, and conceded that a skillard is probably less restrictive than overlapping gates for trikes, handbikes, tandems etc. To reach a rational conclusion I decided to have a go at calculating if this new barrier design is compatible with the spec we’ve been developing for a Standard Inclusive Bike.


But the description of the skillard is a little vague: “2 rows of staggered bollards. 1.5m between bollards, 5m from junction”. To make sure I was evaluating the right thing I wanted to answer the question: is it a 1.5m clear gap between bollards or 1.5m between centres? Recalling a tweet in my favourites giving the location of one in the wild, I packed a tape measure and set off to find out.


On the way from Bath Spa station there are reminders of both, how barriers once installed appear to be set in concrete forever…


…and how sometimes, between barriers, you might for a moment think you were in a cycling paradise across the North Sea; the worst and the best of the NCN within just a few turns of a crank.


And on to the business at hand. The junction between Inverness Road and Bellotts Road forms a crossroads for people on bikes and on foot. This is the view looking south from the centre of the crossroads and is the same view shown in the tweet above. The good news: the holes for the second row of bollards have been filled in and the two outer bollards have been reset further out.


The bad news: it’s neither a 1.5m clear gap nor is it 1.5m between centres, it’s a narrower 1.2m clear gap give or take a couple of centimetres. The wooden posts are 143mm square. So what was it before they were moved? More of that in minute.

These are views looking east-ish from the skillard4 at the end of Inverness Road, and west-ish from the skillard5 adjacent to Bellotts Road. Every bollard in this location has a clear gap of about 1.2m to its neighbours.



I’d been tweeting these pictures while there and just when I thought I knew all I needed to know about Bath’s skillards I got a tip-off to head north down the hill to this…


It’s on what feels like a 1 in 4 slope, on a curve, way closer than 5m to a junction with no drop kerb, and to top it all a clear gap between bollards of no more than 1.05m. I’m guessing that was the spacing initially used at the top of the hill.

I struggled to wriggle through it – there must be an MTB technical difficulty rating applicable to this – and if someone on a trike attempted it I think they would tip over as they turned across the slope. A hand bike could well end up stuck with the rider helpless to reverse up the steep incline to make adjustments. Coming down the hill in a wheelchair with just your hands as brakes would make for an interesting experience. Quite the challenge.

So where did that leave my evaluation of the compatibility of the Standard Inclusive Bike and this new barrier design?


In a state of confusion. I didn’t find the answer to my original question and all I can say is that if it were a 1.5m clear gap between bollards it might well be sufficient, that if not it probably won’t be, and that based on the 2 different implementations found shouldn’t a tolerance of -0.5m (-33%) be introduced into the manual?

I don’t believe it was a wasted journey though. Looking at it from a user’s perspective, particularly a user of an inclusive bike, it demonstrates that Sustrans has a quality control problem. One that is so endemic I think it prevents the organisation from meeting the needs of its users and its charitable aims when it comes to access for all.

I can’t find a Sustrans Quality Manager or Director of Quality to contact. In fact, I couldn’t find any information about who’s responsible for what on the website. If I could I’d ask…

  • Has Sustrans verified which bikes are compatible with its design manual?
  • Has Sustrans validated its network against its design manual?
  • Has Sustrans recorded and published the noncompliances in its network?

If the answers to these questions were yes, anyone could go into a bike shop knowing whether the bike they were about to buy was ‘NCN compliant’ and exactly where they could and couldn’t use it – get quality right and equality is sure to follow. It should be that simple.

But it isn’t. As I understand it Sustrans has little or no control over its network; it’s a hostage to the preferences and prejudices of each and every landowner and authority it has to deal with. Ranty put his finger on the reason Sustrans has produced its own design manual and based on my experience of parts of the NCN I can’t help wondering if it’s worth the paper it isn’t written on.

A position within Sustrans Quality Control department would be many people’s idea of a dream job – riding the NCN all day, towing a trailer measuring the network’s parameters as they go – but what good will the data be until it can be acted upon?

Without a government endorsed national standard for cycle infrastructure that applies throughout the country, can Sustrans ever get away from ‘local conditions’ compromising its charitable aims? Will it ever achieve a consistent standard of quality across its network? Will it ever be inclusive?

Lots of questions. Here’s one more: Who can answer them?

31 thoughts on “Surveying Sustrans’ Skillards

  1. carmarthenbaywatcher

    “Cantstandup” sadly (fortunately – I’m not sure) it is not necessary to be a person with a disability to appreciate the insanity of some of Sustrans design failures. “A Frames” too narrow to get a bike through and chicanes where you have to carry a tandem through/over are common. All this from an organisation chosen to drive (pun intended) through the National Cycle Network. People who are, for whatever reason, unable to easily negotiate the Sustrans obstacle course are suffering a level of discrimination normally reserved to the private sector!

  2. davidhembrow

    Very interesting. I’m not at all surprised to hear that such barriers limit access. When I reviewed Sustrans’ handbook I specifically pointed out that their designss would cause problems for people with disabilities. This confirmation of the problem is helpful.

    In any case, why are Sustrans obsessed with reducing speed of a low speed mode anyway ? Is it because they build in conflict due to the way they design the infrastructure ? Why are such speed reduction measures never required in the Netherlands ? Is that perhaps because paths n that country are designed to facilitate cycling using a range of different bikes so that they are accessible to as large a proportion of the population as possible ?

    1. Kevin Hickman Post author

      I had something to say about the second picture in your post too – signage – but I wanted to keep the focus on access for now. This issue of accessibility has been an ongoing problem for as long as I can remember, and for you as well I guess. It’s an issue that has been taken seriously in other forms of transport but not cycling (perhaps there’s a pattern there!) and I wonder if the reason comes down to responsibility. It’s easy to determine who’s responsible for ensuring access onto a bus, a train, into a station, but not so easy in a bike network.

      Sustrans has a duty under the Equality Act and through the Public Sector Equality Duty, but is Sustrans itself responsible when it comes to ensuring access to and accessibility on its network? If the answer’s no, then that’ll be one of the reasons this problem has proved impossible to resolve to date and continues to dog new routes.

  3. Tom Bailey (@TyneTom)

    Hi Kevin,

    I’ll do some digging on who has been putting these in down in Bath and why the widths appear to be too narrow.

    It isn’t hard to track down our engineering department in Bristol and I’m sure they’d be happy to talk to you.

    The background to “slowing cyclists” is that is often identified as a requirement by the Safety Audit process that Local Authority engineers follow to meet their obligations under the Construction Design & Management regulations. The safety audit may well identify the approach speed of cycle traffic as a “risk” that the engineer feels obligated to mitigate. The default solution to this risk in a highway design toolkit is guardrail installed across the desire line.

    I’m only aware of Sustrans showing an interest in slowing cyclists in a couple of cases where speed has been an issue between cyclists and pedestrians. I’m not aware of whether this has led to us putting anything like this in ourselves.

    The skillards (good name) and putting 90 degree bends in the path are solutions offered to local engineers as an alternative to guardrail. It may well be that there have been problems in the past with poor implementation and that that including them in the design manual with 1.5m spacing makes it less likely for this to happen again.

    Sustrans are not in a position to demand that changes are made to the NCN. We can attach conditions to our funding (if we have any) which require adherence to a standard when new routes are built. We can try and persuade LAs to fix problem points but can’t really insist.

    The issue is cropping up right now at a junction on NCN7, local authority officers wanting to install barriers on a small section of path on their land. I ask for them not be installed, and are then confronted by very emotive talk of “children running out” and am asked to come up with an alternative. Hopefully can deal with this by suggesting some deflection in the path but if all else fails then I guess properly spaced skillards may be on the agenda. Not ideal but that is where we are at.

    If you look for a root cause in all this you find it in the UK approach to road safety.

    1. Kevin Hickman Post author

      Can’t disagree about the approach to road safety Tom.

      I looked up the Construction Design and Management regs and was surprised to find myself on the HSE site…
      It would be great if HSE did cover bike paths and streets, but aren’t these regulations only applicable during construction? If so, where does the requirement for the Safety Audit come from?

      If a national regulation or responsibility is the source of these decisions/requirements it’d be far more efficient to tackle that than dealing with every LA individually. I think picking one LA and holding it to account on its Public Sector Equality Duty would still be a worthwhile exercise though.

      However, I think the first thing Sustrans needs is a quantitative definition of accessibility. Who would be responsible for that?

      1. Tom Bailey (@TyneTom)

        CDM also covers use of the finished design (so car hits bike entering road from path) as well as safety during construction. For highway engineers this is codified as safety audit.

        to answer the query below, good modern safety audit should look at how safe the alternative if cycling infrastructure becomes too compromised to use. But doesn’t always work well.


    tom –

    sustrans can and should be demanding that the ncn is built to certain standards (particularly in terms of accessibility to all users – including less able, elderly, families..). if local authorities won’t comply, there is always the option to build elsewhere. there’s plenty that needs doing.

    in terms of safety – i think the key is to compare the cycle path to the most minor of minor roads – and look at the safety implementation there. the cycle path is always going to be safer, in terms of pedestrians, and so should need less traffic calming than the road.

  5. davidhembrow

    It was Kevin, the author of this post who proposed “the fifty bollard game” a couple of years ago. I’m quite sure “Skillards” aren’t what he had in mind because I still remember that discussion. The idea was to use a minimal number of bollards to facilitate cycling, not to use an array of them to get in the way of cyclists.

    Sustrans are not alone in doing this and it’s not only started now. The long gone Cycling England’s “skillard” design using no fewer than seven bollards as well as some similar three bollard designs made it into my “bad examples” list. I.e. showing how not to apply bollards.

    Sadly, Sustrans seems to be intent on putting its name on ever more bad examples of infrastructure, again and again picking out ideas which have been tried, failed, discussed, tried again and failed again for implementation on the NCN. It’s simply not good enough. You should be among for high standards, not passing off past failures as best practice for the future.

    1. Kevin Hickman Post author

      In an ideal world we wouldn’t need them but it’s true I’m a big fan of the well placed bollard to create filtered permeability.

      I came to the conclusion that a 1.5m clear gap should suffice for accessibility re inclusive bikes, which ties in with Ranty Highwayman’s assessment that 1.5m is the sweet spot for excluding motor vehicles.

      I’d clean forgotten about the skillard3s shown in your post David. Nothing’s new in cycling is it? :o)

      But, technically, that Cycling England one doesn’t look like it’s on an equilateral grid, although it’s hard to tell from the picture.

  6. Dave H (@BCCletts)

    The poor management of quality and resulting safety issues isn’t only linked to cycle facilities, or even cycle & pedestrian works. It also exists on the roads network generally – compare that to how the rail and air traffic networks are managed, and then deliver their Vision Zero results.

    1. Kevin Hickman Post author

      Comes down to responsibility again I guess. ‘Public’ highway not applicable to rail network or air space?

      Did you mention swept volumes for calculating clearances on bends at some point Dave? Is there an app for that?

      1. Dave H (@BCCletts)

        One of our church (the cycling one) actually wrote the early programming for swept areas of trucks and buses, might see what he comes up with for bikes and bollards.

        I did have some useful .vsd software but ‘someone’ took off with the PC that we used for graphics and I’ve been struggling with the lesser stuff I have on the other machines. It was good for doing swept area stuff.

        If you can get to use something like Visio, it is quite user friendly and I have a bike template.

        The whole ‘excuse’ lexicon applied to any matters attaching to how we move motor vehicles around and remain blissfully care-free for the safety of people in those vehicles and outside them on the roads and paths of this country is appalling.

        Kill someone with any bit of dangerous equipment = manslaughter – unless it happens to be a motor vehicle and then it is sanitised with the euphemistic causing death by … driving. Same applies to investigating crashes and sorting out remedies to prevent future crashes – RAIB AAIB MAIB and HSE all have independent and objective investigations with clear action points Section 39 has the roads authority investigating crashes where their roads may be the root cause of those crashes quo custodiet ipsos custodes!

        If the cycle route is adopted – and presumably on the list of roads it could justify some pressure for Section 39 reporting. If the site is private – Ie a Sustrans-owned path I’m just wondering if the same conditions apply as for level crossings, tramways, and places which invite the public in – like pleasure parks. There the park owner/manager has to show that they are delivering a measures to demonstrate their duty of care to employees and non employees on ‘the premises’ under Health & Safety legislation. Generally though the removal of hazards wherever possible and when a hazard exists, the management of the risks. Incidents injuring a non employee come under Section 3, and a regime known as RIDDOR Reporting of Injuries, Diseases & Dangerous Occurrences Regulations requires a record to be made of any incident causing injuries serious enough to require hospital treatment and time off work.

        I recall a significant investigation when a lady on a hired bike on a Peak Park trail lost control on an incline and broke through a basic timber fence – square-on to the path, where a narrower replacement bridge spanned the A6 where the old railway bridge had been removed. She fell on to the road and died. All the bridges I worked on had lead-in rails to deflect any out of control rider away from the drop – usually a steep banking or vertical wing wall.

        So on the road an FoI request for your local Section 39 Investigations might get a lists of local crashes, with some very general analysis and… nothing gets done. We get repeats of the same type of crash at the same places 3 deaths in 2 years so far at Bow Roundabout, and the inherent hazard of 100% of the motor traffic driving right through the path taken by all the gullible cyclists lead to use CS2 (fortunately less than 40% as 60% use the much safer route over the flyover). All that prevents the next fatal crash here is the compliance of drivers & cyclists with the traffic signals – and a huge number of drivers do not.

        Might also be worth engaging with the local bike thieves to find out the best battery powered angle grinders that are easily carried around for resculpting metalwork, as well as cutting off bike stands and locks. Careful if its galvanised though – toxicity from fumes generated (H&S issue 🙂 )

  7. Mike Chalkley

    Hi All, – I’m more and more confused at how we can have ‘safety audits’ which point out when cyclists and pedestrians come into conflict but which completely fail to take into account the dangers we are reminded of weekly of mixing cyclists and 30mph+ traffic. People die all the time because these modes are mixed together and driver mistakes cause terrible accidents. I am considerably safer working on an oil rig than I am cycling on our roads.

    1. Kevin Hickman Post author

      I’ve been looking for the term that describes this psychological phenomenon where a person, and a society, can hold two inconsistent beliefs. I’ve come across ‘cognitive dissonance’ but I’m not sure that’s the right one.

  8. Christine Jones

    There’s a cycle bridge on NCN11 just south of Wicken Fen I’ve christened the bridge of doom. It takes two people to lift my dutch mummy bike up the steps with wheel rail and down the other side. Presumably it’s all that can be afforded. There’s none of these skillards but your point about design guidelines vs what the land owner and LA come up with stands.

  9. triptogenetica

    “Skillards” is a fantastic name for these.

    I’ve reproduced my comment on bollards on cycle paths from D Hembrow’s site, from 2013. I still cycle through 2 sets of Skillards almost every day. Since then I have added a child seat to the back of my bike, instead of having to tow the trailer. I can confirm that “the devil is in the details” – i.e. the Oxford NCN51 skillards seem ideally designed to restrict bikes with child seats. They’re too close together, and they have these metal ‘bulges’ (semicircular shoulders that stick out), just at handlebar and child seat height. Why do cycle planners do this?

    Post follows:

    Thank you! This is extremely helpful!

    Where I live – North Oxford – we have a very useful cycle and pedestrian link over the ring road, in the form of a dedicated bridge.

    Cyber Skillards on Cutteslowe Bridge, Oxford

    It’s on my daily commute. It’s also part of NCN route 51. But every single day, one thing about it gets to me – the bollards at either end, which are a daily exercise in stunt riding, being so close together.

    Imagine if a commute to work (by car) included a road with trees, parking meters, post boxes, or bollards exactly 1.9m apart, at intervals. That is, only slightly wider than your car / wing mirrors. That’s the situation here.

    I also sometimes tow my daughter in a bike trailer over this bridge.

    The clearance for its wheels between the bollards is literally a couple of cm, and if that wasn’t bad enough, you have to be sure to enter the bollards NOT in a straight line, but in a particular curve, as they are staggered and so the trailer needs to curve to clear both sets!

    To clarify – it is not possible to ride a bike, with or even without a bike trailer, in a straight line though this obstacle, on a straight bit of cycle route, at a dedicated bridge. The bollards make it an impossibility.

    If you look at the picture, there are FIVE bollards at the entrance to the bridge! They have additional metal at handlebar height, to make them even more of an obstruction.

    Why were five installed? Why not one, in the middle, on the dividing line between pedestrians and bikes? Or, if the bridge was too attractive to motorbikes etc, why not three?

    Which design standards were followed, in deciding how to place these bollards?

    Bollards number 1, 3, and 5 in the photo serve a useful purpose, closing off the route to motor traffic.

    Bollards 2 and 4 serve no useful purpose, making this national cycle route technically challenging to ride by bike, even more so with children in tow, and likely impossible with a trike, recumbent bicycle, or hand-cycle as used by many people with disabilities. They force cyclists to swerve, perhaps into the path of pedestrians.

    As such, how would we go about having them removed?



    19 August 2013 09:53

    1. Kevin Hickman Post author

      Thanks for posting James. I recall these from the Wheels for All taster sessions we ran in Cutteslowe Park a couple of years ago – very unfriendly indeed. The wooden skillards of Bath are bad enough, but cyber skillards are in another league entirely. Where’s the Doctor when you need him?

      Oxford City have just run a consultation on removing/repositioning barriers to allow/improve access for inclusive bikes, however, I don’t recall these being on the list.

      I’m working through the process for the audits carried out on this kind of infrastructure with the info provided by Tom, and when I know what we need to ask for I think we should put in an FoI request to Sustrans and the council(s) to find out how they came to be there. Then present a case for removal based on indirect discrimination.

      I should be coming into Oxford tonight, if the weather stays kind and I remember to charge my sonic screwdriver I’ll stop by and measure them up.

      1. triptogenetica

        Thank you! I could also tape measure them for you, later today. I’d be very surprised if they’re 1.5m, even centre to centre, and astounded if they’re much over 1m between those metal shoulders.

        If there’s a consultation on this, I’d happily make a statement / whatever is needed. They could take bollards 2 and 4 out quite easily, and maybe lop the shoulders off the others.

        To be honest, the price of scrap metal being what it is, I’ve been hoping a vigilante altruistic scrap merchant would sort this out for years. There must be a few quid tied up in those bollards.

  10. triptogenetica

    While we’re at it, the rest of NCN 51 into Oxford could do with a serious looking at.

    The section linking King’s Cross Road to Marston Ferry Road, and then further south on the old Charlbury Road near the Dragon School, is all a bit heavy on the guardrails. I will do a photo tour later on!

  11. triptogenetica

    I did a bit of measuring skillards this afternoon – as I thought, the devil is in the details:

    The Oxford skillards are 1.5m centre to centre, as previously suggested would be good practice.

    But include the metal ‘shoulders’, and the gap drops to just over 1.0m!

    Now I know why my bike handling has improved so much in the last few years. Photos aren’t the best, though!

  12. Kevin Hickman Post author

    I didn’t get there yesterday so thanks for the pic, it’s good enough.

    The barrier in your second pic *is* on the council’s list to be modified (this picture was part of the council’s consultation so I’m assuming there’s no problem with posting it)…
    Dragon Lane Barrier

    As you can see it’s an improvement. I asked for the evidence of the benefits of barriers used for speed reduction that supports keeping the additional bollard there. It raises again the issue of what spec engineers are working to when it comes to accessibility – how do we know what is and what isn’t being impeded/excluded? But the good news is the council have taken on board the issue of access for people using anything other than a standard two wheeler. Not sure we’re there yet with the issue of people who have limited manoeuvring skills/ability etc.

    My preferred solution there would be one bollard (removable for vehicular access if required) placed to ensure a 1.5m clear gap.

    1. triptogenetica

      Thanks Kevin!

      To be honest, on the whole route Cutteslowe to Oxford, there’s nothing as bad as the Ring Road bridge skillards. Nothing as ‘technical’ to ride through, I mean. In terms of lack of accessibility, probably the narrowest point is below:

      That’s at the South end of King’s Cross Road, at the entrance to the bike path. Maybe they give advance warning for a narrow point still to come, like at the entrance to a tunnel?

      Most importantly – this Consultation you mentioned – where / when was it? Does it discuss the skillards?

  13. triptogenetica

    Re that Oxford consultation – never mind, just found it by googling “oxford council barriers cycling” – I assume it is “Oxford Cycle City Minor Network Improvements”, dated 17th Jan 2013?

    Interesting some work (e.g. removing the awful University Parks cycle path barriers) has been done, yet other modest work has yet to happen.

    I can only see the summary doc on their site, the full results are missing! So I still can’t check if my skillards are in or out.

    1. Kevin Hickman Post author

      I believe that consultation was to gather comments and suggestions for minor changes (a small pot of money had been identified) and the changes to the barriers are a result of it. Not sure whether the consultation for the proposed changes to the barriers was a public one.

      I’m as sure as I can be that your cyber skillards on Cutteslowe bridge weren’t part of it though. Perhaps they weren’t identified by the ‘Minor Network Improvements’ consultation.

      1. triptogenetica

        Thank you. WHo in that case should I write to, to suggest it? Grinder Bill, perhaps? 😉

  14. Dave H (@BCCletts)

    Here’s another one – apparently already caused cyclist injury. Note how surface drops into loose soil off the end of the cattle grid, and cyclist have to turn on the grip to avoid the drop and fence.

    Would completely fail safety audit as built

    Current condition

    Cyclist injured here on 14/5/14 – note fence protecting against drop falling over*

    Note also that there is even a step between level of tarmac and brick edge laid around the cattle grid (with the holes in the bricks left open to collect water – freeze and then crumble – they don’t look to be engineering brick just common brick?). It is visible on the photographs, and thus greater then 6 mm, the height limit for a dropped kerb, above which the risk of tyre deflection (and bringing down a bike and rider) increases.

    I recall a detail which we took very seriously in the design of cycle routes in the 1980’s, namely the delivery of robust protection against cyclists dropping off an edge, following from the death of a cyclist who crashed through a timber fence, like that used here and fell down the drop, over the abutment of the former railway bridge, which was not used by the much narrower replacement used by the cycle route.

    The arrangements here could have been much safer had a bit more attention been paid to the alignment of the gate & grid relative to the path. As it stands there will need to be a significant building out at the top of the slope down to the former railway line, which presumably cannot be used at this point. Not that far from where I first worked for Sustrans on the Kennet & Avon Canal near Bradford.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s