Sunday, 19 August 2018

On Soft Ground

Building paths is no different to building motorways, it's just the order of magnitude which is different.

Motorway building attempts to be as economic as possible (relatively speaking) and this is very apparent when we consider the use of materials. Excavation, transportation and especially disposal. The motorway builders try and use as much "site won" material as possible in order to keep costs down. Topsoil is stripped to be reused for landscaping, subsoil of a useful quality is reused as bulk fill and any aggregates are used to make concrete and asphalt.

One problem which is often faced by engineers is dealing with ground with poor bearing capacity; that is ground which cannot cope with holding up the mass of road being built, let alone the traffic it's being built to carry. It would be economically viable to keep digging until decent ground is found as it could be many metres down and so some sort of improvement to the bearing capacity is required.

Even a humble path can find itself being built on poor ground and so rather than just spending the minimum knowing full well it's going to move and settle leading to cracking and a poor surface, there are some bits of technology out there which can help.

Some years ago now, I was involved in the construction of a couple of 'rural' paths. One was along a boggy old bridleway and one was through some woods. In both cases, the subsoil was pretty soft clay and especially through the woodland, we couldn't dig too deep for fear of damaging tree roots. Our way to provide a stable base was to use geotextile membranes and a cellular containment system.

The photograph above shows the boggy old bridleway which was pretty much impassible in the winter. Under the thin layer of topsoil was soft yellow clay. Even just building a path to take horse and human traffic would have taken some pretty hefty construction.

The first job was to roll out our geotextile membrane - essentially a synthetic cloth. In this case, we chose a non-woven version for superior performance. It's doing two jobs - first, it will stop the granular foundation to the path mixing into the clay below and it resists getting holed by stones below. On some parts of the path we didn't even bother clearing the topsoil as we wanted to raise the path out of the ground for better drainage.

With the geotextile installed and a thin layer of granular material installed and rolled (above) to provide a working platform (recycled concrete in this case) it was time for the next clever material to be installed, the cellular confinement system.

The system is essentially a plastic mesh which comes in rolls.  When the product is stretched out (above) it forms a grid of 'cells' into which granular material is placed and compacted. Notice the holes in the sides of each cell - they are there to allow sharp points from the granular material to lock into place. This particular version is made by Geoweb and was 75mm thick (i.e. the depth of the material when stretched out).

The way in which the granular material interacts with the cells means the thickness of material used can be far less than a conventional design. It's also helpful as the layer as a whole acts like a raft which can make it more forgiving over the odd soft spot. As can be seen on the section being tested above, you can run heavy kit over the system without a problem.

For this scheme, the client would not allow an asphalt surface and instead we had to use a self-binding gravel because it was considered to be more in keeping with the rural nature of the path. In my view, this was a mistake because self binding gravel requires a great deal of aftercare from a maintenance point of view to keep the edges weed-free and the surface compacted (especially with horse riding). However, nothing would have stopped us laying an asphalt surface - it wouldn't have need any edging or kerbs either.

As suggested above, the system can be used near trees as shown in the photograph above. Again, the client wanted a self-binding gravel surface and you can see how we picked a route through the woodland.


  1. That brings back memories - some 33 years, when I was doing just that with Sustrans on some of the worst subgrades imaginable including a railway cutting which had filled over a period with with thixotropic silt carried over due to the failure of the neighbouring quarry to maintain their settling pond. We nearly lost a 12T 360° tracked excavator in that....

    One key caveat on building dry/water-bound macadam is to pay great attention to the profile, to ensure run-off velocities and volumes are low and not focussed on the running surface. Most of my paths used either locally won alluvial material with a % of clay, laid without a blinding layer, or a graded basalt capped with a 6mm to dust, which, from one quarry especially had sufficient cementatious qualities that after a wet roll and weathering delivered a rock hard finish. All we had to do was keep all motor vehicles off the track, especially after a wet spell, and a keen eye could easily spot when tyres with those destructive sipes to flush out the binding fines, had travelled over the surface - just once.

    In many cases a quality drybound path is easier to maintain where settlement and tree root heave are likely - our ranger used to go out on his bike with a trailer-load of stone/dust and hand tools, to attend to minor repairs.

    I'm also impressed with the control you had on the contractors that kep the constriction work so well contained within the wayleave - we once were almost obliged to use the Council's DLO Roads team and I was appalled by the way the construction corridor spread out to churn up land (fortunately council-owned) 10-20 metres off the path line. Totally unnecessary.

    One question that does strike me is what planning was made for long-term (mechanised) maintenance. We planned for an annual pass - after seeds had set and birds fledged (usually late June-July) with a 3-4 person crew and a small tractor with a boom mounted flail, covering perhaps 1Km/day to the spec 1m path shoulders cut down to 'lawn' level, and a vertical clearance of all growth to at least 3 metres. As a moving work-site the tractor driver was under the control of one of the 2 cordon controllers on foot, and the 4th person had appropriate tools to trim off any larger branches, left after the flail had passed. After the cutting iy does look rough - but with summer growth, it soon goes back to green in a couple of weeks, and job done for another year. What really stymies this low cost regime is the daft placement of lighting columns, signs and seats (and large trees) in that shoulder zone....

    Closing thoughts - whacking down a drybound path like this using a wider sub base and raked up shoulders delivers a better draining path than one with timber edges or pin kerbs, as a fraction of the cost of using kerbs and tarmac, and delivered paths which immediately saw growing use, when waiting fior the funding plus the more complex access to lay tarmac would have sunk any prospects of building those early cycle routes at the rates we did, and still gets routes built in more remote locations where haulage costs would be crippling. The second though is on the way that a family with hereditary blindness was able to use paths with grass shoulders, as a tactile guidance system one foot on the path one on the grass to enjoy a walk without recourse to a dog, cane or companion.

  2. Nice job, and thanks for the explanation of the materials.

    I think there is a typo in your second paragraph. " It would be economically viable to keep digging until decent ground is found as it could be many metres down and so some sort of improvement to the bearing capacity is required." appeares to be missing a 'not'.