A View from the Drawing Board - Turbogate

Another guest blog in our series from a traffic engineer, A View from the Drawing Board. Previous posts can be found here and here. This blog addresses some aspects of the plan to install a 'Turbo Roundabout' in Bedford. 


Much has been discussed about Bedford Borough Council's 'turbo-roundabout'. Others have already outlined the deficiency of the proposals (i.e. that it's a dangerous, disingenuous and fraudulent use of public money set aside for purposes this scheme makes no attempt to work toward). I would make some further observations. 

Firstly, the capacity benefit (for motor vehicles) of the 'turbo-roundabout'1 (as opposed to a conventional multi-lane roundabout) is achieved not only through the provision of multiple lanes, but also through their physical separation from entry to exit. The physical separation -

  • constrains the path of entering and circulating vehicles, and thus vehicle speeds, and 
  • limits the number of possible lines through the roundabout.

This is all aimed at minimising the gap in the circulating flow that drivers can safely and (from the drivers' point of view) comfortably emerge onto, and avoiding conflict (and thus disruption to the flow of circulating traffic) on the roundabout.2 3 4

Yes, it really, really is about squeezing as many cars as possible into the space (and, in the right place, with the right provisions for cyclists, this isn't necessarily a bad thing in itself).

It isn't clear from the drawing I've seen, but it appears this physical separation isn't provided. If it isn't, the junction simply isn't a turbo-roundabout (partial or otherwise). It's just a roundabout with road markings on it, and these aligned to give a far more generous alignment that a turbo roundabout would (and thus require larger gaps for traffic to emerge, thus reducing capacity)5. When this is added to the fact the Dutch have historically considered that roundabouts where cyclists remain on carriageway have less capacity than those with separate cycle tracks6, the question is begged – is the alleged benefit for motorists as much of a fraud as the alleged benefit for cyclists?

Secondly, I would note that, as I understand it, the raison d'être of turbo roundabouts in the Netherlands is to provide junctions with the safety benefits of Dutch single lane roundabouts, but with greater capacity, without enlarged geometry leading to increased speed and reduced inclination to give way.4 7

Whilst I fear my choice of words are tempting fate, if not certainty, in the case of this junction, the safety problems of increased speed and failure to give way posed to cyclists have been, er, done to death elsewhere, figuratively and literally (Hello Boris!).

What hasn't been mentioned is that the Dutch, as a consequence of the problems associated with generous geometry, now consider 'conventional' multi-lane roundabouts as relatively dangerous, even for the occupants of motor vehicles, and thus something to be avoided - hence turbo-roundabouts as a means of 'designing out' the problems (for motor vehicle occupants) of multi-lane roundabouts.

(Unless otherwise stated, subsequent references to diagrams, regulations or directions refer to the Traffic Signs Regulations and General Directions 2002 as amended)

Thirdly, and possibly most importantly, the designer's assertion that an annular cycle track with priority around the roundabout is ruled out by UK regulations is bullshit. The single dashed advisory give way line typically used at roundabouts cannot be used to this end8. However, direction 34(1)(b) specifically allows the use of mandatory double dashed lines at a roundabout without specifying where at a roundabout, and DfT Guidance indicates many possibilities of the use of this9. Presumably (although this isn't clear), the roundabout is regarded as a road in its own right (otherwise the meaning of the marking given by Regulation 25 makes no sense at a roundabout!). As with priority junctions (see last blog post), the requirement of Regulation 25(2) is for traffic crossing the line not to cross the line so as to endanger vehicle occupants on the major road, not just on the carriageway. Therefore, why couldn't cyclists on an annular cycle track be protected with give way lines in the same manner as I previously demonstrated they could be with major/minor priority junctions?

Of course, if this wasn't considered robust enough, putting the cycle crossing on a road hump allows another avenue for give way markings10. This has an added benefit of mitigating for the possibility of approaching drivers' view of crossing cyclists being obstructed by vehicles in adjacent lanes11, by controlling speeds, reducing braking distances and the consequences of any collision should a driver fail to stop in time. This is recommended by Dutch guidelines where cycle tracks cross on the level at multiple lane roundabouts 12.

Including a zebra crossing, and presuming (for clarity as much as anything) no road hump in this instance, the result would look something like this:

  • A - Give way line in accordance with Regulation 25(6) & Direction 34(1)(b).
  • B - Advisory give way line controlling entry/circulating conflict at circulating carriageway. (Could be mandatory double dashed line; this would seem to me to have little practical benefit and would require additional signing)
  • C - Blokmarkering and red surfacing provided under coloured surfacing loophole (see last blog post)
  • D - Controlled area of zebra crossing on cycle track side reduced to minimum length permitted
  • E - Belisha beacon
  • F - Upright 'give way' sign required by Direction 34(1)(b) (could include a diagram 962.1 'Cycle Track' plate).
  • G - Give way road marking required by Direction 17(1) (due to requirement for upright sign), placed outside of the controlled area of the pedestrian crossing.

It is, I grant you, somewhat messy. This has nothing to do with the cycle track crossing per se - it is due to the requirement for controlled areas (zig zags) at the zebra. It would be better if there was greater flexibility over the use of controlled areas, and if give way lines were permitted where zebra crossings were placed near roundabouts and priority junctions (such layouts can be very messy due to the zig zag lines regardless of provision for cyclists, and the minimum controlled area requirement can require that crossings are moved away from the desire lines where pedestrians are likely to cross in practice).

There is an additional issue where a road hump is provided; if any part of it overlaps the zebra crossing or its controlled area, the hump must be centred on the zebra crossing.13 This can be achieved by lengthening the plateau away from the roundabout; again, it would be better if the regulations only required that the crossing were on the plateau of the hump, to give greater design flexibility.

Finally, it is important not to lose sight of the fact that turbo-roundabouts are all about capacity for traffic. Due to the potential for cyclists to be masked by vehicles when crossing multiple lanes, and due to the increase time cyclists need to cross the wider arms, the risks and nuisance posed by motor traffic are greater than at single lane roundabouts. The Dutch prefer cycle routes are grade separated at turbo-roundabouts as a consequence. And there are risks associated with priority cycle tracks at roundabouts in any circumstance; the Dutch (in urban areas) accept it is necessary to accept this risk to meet sprovide adequate level of service in terms of ease of mobility for cyclists2.

Given these difficulties, a turbo-roundabout with annular cycle tracks wouldn't be my choice of Dutch infrastructure to introduce somewhere not used to it!

Actually, one final, postscript.

“Time we took road safety decisions away from politicians. If I’ve got cancer, I go to a doctor for a treatment plan, not the mayor. “ said user “platinum”, in response to http://aseasyasridingabike.wordpress.com/2014/02/24/the-mayors-vision-fo....

This raises an interesting point. Whats the difference between the politician and the doctor?

I would contend that a major part of the difference is that the doctor must be accredited by a professional body (the General Medical Council / GMC), and that accreditation is subject to the doctor maintaining professional standards.

I don't argue the GMC is perfect, but a doctor who, noting that some people in the Netherlands smoke tobacco, prescribes a patient two cigarettes in the morning to ease a violent cough would soon find themselves subject to disciplinary proceedings. Just sayin'.



I was wondering if the turbo roumdabout is similar to the schemes the TRL have been looking at? Assuming that the TRL solution would be best practice, of course..

TRL were tesing a proper 100% Dutch ordinary roundabout, of the type seen on town streets, which is nothing like a turbo roundabout which is designed for sitations where major motor routes cross each other. The Dutch would never expect cyclists to use a turbo roundabout, they'd usually provide grade-separated crossings for bicycles (underpasses or bridges) so that cyclists don't ever have to mix with fast-moving and busy motor traffic.

TRL tried both a roundabout with Dutch markings, and then various ways of achieving something similar using UK road markings. The aim was to find out how UK people understood the roundabout, and the different priorities involved, with various different types of road marking and signs.

The Dutch have much simpler road marking requirements than we have in the UK, and their "sharks' teeth" Give Way lines are directional, unlike our double-dashed ones.