Deriving visibility splays from a measured speed survey

The visibility splay at a new access is derived from the speed vehicles actually travel at, not from the posted limit. National street design guidance sets stopping sight distance from the 85th percentile wet weather speed measured on the road in question, and on constrained streets that figure is often well below the limit. A visibility splay measured speed survey is therefore the difference between an access that works and one that appears to need 20 metres of neighbour's front garden.

The corollary holds too. On a 30 mph road with a long straight approach, observed speeds can exceed the limit, and the splay required is longer than the standard table suggests.

Why the posted limit is the wrong input

A speed limit is a legal maximum. Stopping sight distance is a physical quantity built from driver perception and reaction time plus braking distance, and braking distance rises with the square of speed. Using a legal number to calculate a physical outcome gives the right answer only by coincidence.

Street design guidance for lower-speed roads works from measured speeds and a perception-reaction time appropriate to those speeds. The design manual used for higher-speed and strategic roads takes a more conservative approach. Which document governs depends on the road, and an access onto the strategic network follows a different route entirely.

Wet weather speeds are the relevant input because braking distances lengthen on a wet surface and drivers moderate speed accordingly. Where only dry-weather data exists, the guidance sets out how to adjust it. Do not invent a correction.

Running the survey so it is not rejected

The most common reason a highway officer rejects a speed survey is the method rather than the numbers.

1. Place the counters on the frontage itself, in both directions, clear of junctions, accesses, bends, parked car build-outs and traffic calming features that suppress speed locally.
2. Record for a continuous seven-day period covering a full weekly cycle, not a single weekday.
3. Capture speed for individual vehicles rather than a mean of aggregated bins, and report the 85th percentile in each direction separately.
4. Note the weather across the period, so the wet weather subset can be identified or the guidance correction applied.
5. Exclude periods where the road was closed, obstructed or subject to works, and say so.
6. Report sample size, equipment type and calibration, and appended raw data.

Radar and automatic counters both work. A short manual survey with a hand-held device does not, because it cannot show the distribution across a week and the operator's presence changes driver behaviour.

Where the road is new or the traffic pattern is about to change, a survey measures the wrong thing, and an agreed assumed speed is more defensible.

Timing matters as well. A survey run in the first week of the school holidays on a road serving a secondary school will understate term-time conditions, and an officer who knows the area will spot it. If the dates are awkward, say why in the covering note and offer a second period rather than waiting to be asked.

From speed to stopping sight distance to the splay

The splay is described by two dimensions.

The x distance is the set-back into the access, measured from the edge of the carriageway back along the centreline of the access to the driver's eye position. Street design guidance normally uses 2.4 metres for a driveway or minor access, with reductions available only in narrowly defined low-speed situations.

The y distance is the visibility available along the road, measured from that eye point to a point on the nearside channel line in each direction, and it is set equal to the stopping sight distance derived from the measured speed.

Both are measured as a clear envelope rather than a flat triangle. The area must be free of obstructions above a low threshold height, commonly taken as 600 millimetres, and the check applies in three dimensions, so a rising verge or a crest in the road can break a splay that works on plan.

Two further checks catch people out. The splay is measured to the nearside channel line rather than to the centre of the road, which is the stricter test. And on a bend, the envelope has to be swept rather than drawn as two straight lines.

Land control and how the splay is secured

A splay is only worth drawing if it can be kept clear for the life of the development.

Land within the application site is straightforward and is secured by condition. Land within the adopted highway is equally straightforward, because the highway authority controls it, provided the visibility is over the carriageway, verge or footway rather than over a private frontage beyond.

Third-party land is the problem, and it is worth testing early, before the layout is committed. Where the splay crosses a neighbour's garden, the options are an easement negotiated with that owner, a redesign moving the access, or a reduction in speeds through physical measures delivered under a highways agreement. A condition requiring the splay to be provided and kept clear, where the land is outside anyone's control, is unenforceable and should be resisted.

Hedges, walls and fences inside the splay need a maintenance obligation. Highway authorities can require overhanging vegetation to be cut back under the Highways Act 1980, but that power addresses obstruction of the highway rather than a private visibility envelope, so a planning condition is still the usual mechanism.

Presenting it for the highway authority

The drawing carries the case, and a splay drawing that answers the officer's questions in advance saves a round of consultation.

Show the x and y dimensions with the measured 85th percentile speed and the derived stopping sight distance annotated directly on the plan, with the source guidance named. Plot the envelope to the nearside channel line in each direction. Add a long section through the splay where levels vary, showing the 600 millimetre check against existing ground. Mark the land ownership boundary across the splay area, and show which parts sit in the adopted highway.

Then append the survey: equipment, dates, locations, sample sizes, directional 85th percentile speeds and the raw distribution.

Officers also look for what has been left off. A splay drawn across a bus stop cage, a loading bay or a parking space marked on the adopted highway is not a clear envelope in practice, and showing those features on the plan with a note on how they are dealt with avoids the assumption that they were overlooked. The same goes for a street tree in the verge, which will still be there in twenty years whatever the drawing says.

Illustrative example, for method only. A frontage on a residential through road with a 30 mph limit returns measured 85th percentile speeds of 24 mph northbound and 22 mph southbound. Applying the stopping sight distance relationship in the street design guidance to those speeds gives a y distance materially shorter than the figure for 30 mph, and the splay fits within the site frontage and the adopted verge. Submitting the same access on the posted limit would have required land in third-party ownership and would have been refused.

Where the access supports more than a handful of dwellings, the splay evidence sits inside a Transport Statement, and at larger scales inside a Transport Assessment. The threshold between the two is set out in our note on trip generation and the assessment threshold, and accesses onto the strategic network follow the separate route described in strategic road network access.

PlanScope drafts the transport documents from your site details and marks the survey inputs it cannot verify, so the speed data has to come from a real survey rather than an assumption.

This is general guidance, not legal advice. Check the position with your local planning authority or a qualified professional before relying on it.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.