Driven piles are faster and cheaper — until the soil says no. A practical comparison of solar mounting foundations and when each one is the right call.
On a 20 MW ground-mount plant you may install more than fifteen thousand foundations. A wrong choice of foundation system multiplied fifteen thousand times is not a detail — it is the project.
The three realistic options
1. Driven piles
A galvanised steel section driven directly into the ground by a pile driver. Fast — a good crew installs hundreds per day — with almost no concrete, no curing time and minimal site disturbance. This is the default choice on utility-scale plants wherever soil permits.
Works when: soil is drivable to the required depth, no shallow rock, low corrosivity, water table not aggressive.
Fails when: refusal on boulders or rock at shallow depth, very loose sand with inadequate lateral capacity, or highly corrosive soil that would eat galvanising well inside the 25-year design life.
2. Screw piles
A helical pile screwed into the ground. Handles mixed and denser strata better than driving, gives excellent uplift capacity, and installation torque provides a live indication of capacity as you go. Costs more per unit and installs slower.
3. RCC pedestals or cast-in-situ piles
Concrete foundations with an anchor plate for the mounting structure. Slow, concrete-intensive, and requires curing time — but the right answer for corrosive soils, high water table, rocky terrain and rooftops or ballast conditions where steel-in-ground is unsuitable.
Never decide without a pull-out test
Whatever the geotechnical report says, we install trial foundations across the site and test them: vertical pull-out, lateral load and compression. The results either confirm the design embedment or tell you to change it while it is still cheap to change. On sites with varying strata, we zone the layout and use different embedments — or different systems — across the plant.
Alignment is the hidden acceptance criterion
The MMS team will discover your foundation quality on day one. If top-of-pile levels and positions are out of tolerance, every table needs shimming, installation slows, and module tilt angles drift from design. We set out the array grid by total station, control level within tolerance, and run an independent alignment check ahead of the MMS crew so they install continuously rather than waiting.
Corrosion: the twenty-five year question
Soil resistivity, chloride and sulphate content decide how long galvanising survives below ground. Where the soil is aggressive, options are heavier galvanising, sacrificial thickness, protective sleeves at the ground line, or concrete encasement. Choosing the cheapest option on an aggressive site simply moves the cost to year twelve, with generation loss attached.
Rooftop is a different problem entirely
On rooftops the question is not bearing capacity but added load and waterproofing. Ballasted, non-penetrative systems avoid roof penetration but add dead load the structure must accept. Anchored pedestals need chemical anchors and meticulous waterproofing. Either way it starts with a structural assessment of the existing roof, not with a module layout.
Summary
Drive where you can, screw where driving refuses, cast where steel does not belong — and test before you commit to any of them.
