Ground Improvement Techniques Explained

A site with weak ground gives you three choices.

Pile through it to something competent. Design a structure stiff enough to bridge it. Or change the ground so it can carry what you need.

The third is ground improvement, and on the right site it costs a fraction of the alternatives.

Preloading and surcharge

The oldest method and frequently the cheapest.

Fill is placed over the area to a height that applies more pressure than the finished building will, and left. The soil consolidates under it, settlement happens now instead of later, and the fill is removed.

It works well in soft clays and it needs time, sometimes many months. Where the programme cannot absorb that, vertical drains are installed to shorten the drainage path and speed consolidation considerably.

The limitation is obvious: it needs space for the fill and time in the programme. Where both exist, nothing competes with it on cost.

Dynamic compaction

A heavy weight dropped repeatedly from height onto the ground surface, densifying loose material below.

Effective in loose granular fills and sands, to a depth that depends on the energy applied.

The constraint is that it generates significant vibration and noise. On a site with neighbours it is frequently unusable, and where it is used it needs [vibration monitoring](/blog/vibration-monitoring-on-construction-sites) and a dilapidation survey.

Vibro replacement and stone columns

A vibrating probe forms a hole, and crushed rock is compacted into it in stages, producing a column of dense stone through the weak layer.

The columns carry load, they stiffen the whole soil mass, and they drain, which accelerates consolidation.

This is a common answer for soft and variable ground under slabs, tanks and low rise structures, and it suits sites where piling every footing would be excessive.

Soil mixing and grouting

Binder is mixed into the ground in place to form columns or panels of treated material, or grout is injected to fill voids and bind loose soil.

Used where the ground is very soft, where contamination needs containing, or in a confined space where larger plant cannot work.

More expensive per cubic metre than the mechanical methods, and it works where they do not.

Engineered fill

The most routine form of ground improvement, and the one most often done poorly.

Unsuitable material is excavated and replaced with material placed in controlled layers and compacted to a specified density, with testing at each layer.

The failures are always the same: layers too thick, moisture content wrong, compaction not tested, or the testing done but the results never checked. Fill placed without control is the cause of a large share of the [differential settlement](/blog/differential-settlement-and-what-causes-it) problems that appear later.

Choosing

What the ground actually is, from a proper [geotechnical investigation](/blog/understanding-geotechnical-investigation-reports). Granular fills respond to densification. Soft clays need consolidation or reinforcement. The methods are not interchangeable.

How much settlement the structure can tolerate, which is a structural question as much as a geotechnical one.

Programme. Preloading is slow and cheap. Most alternatives are fast and expensive.

Neighbours. Vibration and noise rule out several methods on constrained urban sites.

Depth. Every technique has a practical depth limit. Past it, piling is the answer.

Verification is the whole point

Improvement that is not verified is an assumption.

Testing before, to establish the baseline. Testing after, to confirm the target was reached. Settlement monitoring during preloading, because the settlement curve tells you when consolidation is substantially complete and it is safe to build.

An improvement programme signed off without verification leaves the structural design resting on a number nobody confirmed.

The commercial case

On a site where piling every element would be the default, improvement can allow shallow footings and a slab on ground instead.

That is a large saving, and it is only available if the ground is investigated and the option is considered early, before the foundation system is assumed. It is another instance of [early engineering involvement](/blog/why-early-engineering-involvement-saves-money) paying for itself.

ACSES provides geotechnical engineering for development sites across Sydney. Send us a site.