Before you select a foundation, you need to read the ground. Most solar project delays, change orders, and structural failures trace back to one skipped step: a proper geotechnical investigation. This guide covers what soil types you will encounter on solar sites, what they mean structurally, and exactly which tests to request before a single pile is specified.
Why the Ground Matters as Much as the Panels
A solar PV system is a 25-year investment in the sky, but its survival depends entirely on what happens underground. Wind uplift, frost heave, differential settlement, and soil corrosion all attack the foundation interface — the one point in the system that no module, inverter, or monitoring platform can compensate for.
Yet geotechnical investigation is still the most skipped step in solar EPC project development. The logic used to justify skipping it is always the same: "We've built on similar ground before." That logic works until it doesn't — and when it fails, it fails expensively.
Understanding what your soil actually is, and what tests extract that information, is the starting point for every reliable ground-mount installation.
The Soil Texture Triangle: Your First Orientation Tool
The Soil Texture Triangle (USDA classification) maps every soil based on the proportions of three particle types:
- Sand — coarse particles (0.05–2.0 mm), drains fast, low cohesion, good friction
- Silt — medium particles (0.002–0.05 mm), settles easily, unstable when saturated
- Clay — fine particles (<0.002 mm), high cohesion, expands and contracts with moisture
A soil sample's lab-determined percentages of sand, silt, and clay place it in one of 12 textural classes on the triangle — from pure sand in the bottom-left corner to pure clay at the apex. For solar EPC engineers, the triangle is a quick triage tool: where a soil plots on the triangle tells you immediately whether you're in easy C-post territory, challenging clay territory, or somewhere that needs more investigation.
For foundation design specifically, the USDA texture triangle is a starting point. The engineering classification system that matters most is USCS (Unified Soil Classification System, ASTM D2487), which adds plasticity and gradation data to produce the two-letter codes you will see throughout geotechnical reports.
The 6 Soil Types You Will Encounter on Solar Sites
1. Sandy / Gravel (USCS: GW, GP, SW, SP)
What it looks like: Loose to dense granular material, golden-tan to grey. Drains immediately when wet.
Structural behavior: Friction-dominant. Load transfers through particle-to-particle contact. Dense varieties (N > 20) are ideal for driven pile foundations — high passive resistance, predictable pull-out capacity.
Solar EPC implication: Best-case scenario for driven C-posts. Standard embedment depths (1.5–2.0 m) typically adequate. N-value and gradation curve from lab sieve analysis confirm whether you are in well-graded (GW/SW) or poorly-graded (GP/SP) territory — poorly graded loose sand requires deeper embedment or helical pile.
2. Sandy Loam / Silty Sand (USCS: SM, SC)
What it looks like: Mixed brown soil — sandy feel with some cohesion, holds shape when squeezed but crumbles easily.
Structural behavior: Friction plus mild cohesion. Medium bearing capacity (0.10–0.30 MPa). The most common productive agricultural soil globally — meaning this is what most solar farms in farmland and rural areas sit on.
Solar EPC implication: The default case for driven C-posts. 100×50×3.0C or 120×50×3.0C at 1.5–2.5 m embedment. Pull-out test confirms which section and depth combination achieves required uplift resistance. This is where the difference between 100×50 and 120×50 section choice matters most.
3. Soft Clay / Silt (USCS: CL, ML)
What it looks like: Dark grey to blue-grey, fine-grained, plastic when wet, hard when dry. Leaves a smooth smear when rubbed between fingers.
Structural behavior: Cohesion-dominant. Bearing capacity strongly dependent on moisture content — a site that performs well in dry season can lose 30–50% of cohesion when saturated. Consolidation settlement is a long-term risk.
Solar EPC implication: Borderline for driven C-post. N values of 4–10 require extended embedment (2.5–3.5 m) or a switch to helical pile. Always specify the 120×50×3.0C section. Critically: do not run the geo test in the dry season and assume dry-season values hold year-round.
4. Expansive Clay (USCS: CH, MH)
What it looks like: Dark brown to black, heavy, sticky. Cracks visibly on the surface during dry seasons. Known regionally as "black cotton soil" (South Asia, East Africa) or "expansive soil."
Structural behavior: Volume change with moisture is the defining hazard. Plasticity Index (PI) above 25 signals significant swelling and shrinkage cycles. A pile driven into expansive clay experiences cyclic uplift forces as the soil swells around the shaft — independent of wind loads. This fatigue accumulates over 25 years.
Solar EPC implication: C-posts are not recommended. Concrete spread footing or micropile is the standard response. If budget forces a pile solution, predrilling through the active zone plus grouting can isolate the pile from swelling forces — but this requires specialist geotechnical input.
5. Rock / Hardpan (USCS: GM, GC — refusal)
What it looks like: Solid grey to brown, requires a pick or drill to penetrate. SPT N > 50 (recorded as "refusal").
Structural behavior: Exceptionally high bearing capacity (0.5–10+ MPa). The structural challenge is penetration, not capacity.
Solar EPC implication: Driven C-posts cannot penetrate without predrilling. Options: predrill a starter hole through the hard layer, then drive into softer material below; or switch to micropile (drilled and grouted into rock). Micropile in competent rock achieves some of the highest pull-out capacities available, but requires a drilling rig and grout cure time.
6. Waterlogged / Organic (USCS: OL, OH, PT)
What it looks like: Dark, saturated, often with visible organic matter. Compresses easily underfoot. Peat is the extreme case — nearly pure organic material.
Structural behavior: Very low bearing capacity (< 0.05 MPa). High compressibility — the soil consolidates under load over time, causing settlement. Piles driven into organic soil have minimal friction resistance.
Solar EPC implication: No driven pile system is reliable. Concrete ballast (surface mount) or deep displacement piles reaching competent soil below the organic layer are the only viable options. These sites require specialist geotechnical and structural engineering input before any foundation commitment.
6 Geotechnical Tests to Request Before Foundation Design
1. Standard Penetration Test (SPT) — ASTM D1586
The SPT is the most widely used field test in geotechnical engineering. A thick-walled tube is driven into the borehole base using a 140 lb hammer dropped 30 inches. The number of blows per 300 mm penetration = the N-value.
Why it matters for solar: N-value is your first proxy for soil strength. It tells you immediately whether you are in C-post territory (N > 10) or not (N < 10). It also identifies driving refusal (N > 50) and the depth profile of soil layers across the site.
What to specify: Minimum one borehole per 0.5 MW, to a depth 1.5× your target pile embedment. On variable sites, increase density.
2. Soil Classification (Sieve Analysis + Atterberg Limits) — ASTM D6913 / D7928 / D4318
Lab analysis of particle size distribution (sieve analysis for coarse soils, hydrometer for fine-grained) and plasticity limits for clay soils.
Why it matters for solar: Produces the USCS two-letter code. Without this, you cannot reliably predict how the soil will behave under cyclic wind loading over 25 years. Atterberg limits are particularly critical — a Plasticity Index above 25 changes the entire foundation strategy.
3. Ultimate Bearing Capacity Test — ASTM D1143 / calculated from SPT
Either derived from SPT N-value correlations or directly measured using a plate load test. Expressed in MPa (ultimate) and reduced by a factor of safety (typically 2–3) to give allowable bearing capacity.
Why it matters for solar: Bearing capacity in tension (pull-out) governs solar pile design. Wind uplift is almost always the critical load case. A pile that passes compression checks but fails pull-out is the most common foundation design error in solar EPC.
4. Direct Shear Test (Friction Angle + Cohesion) — ASTM D3080
Lab test that applies shear force to a soil sample, measuring resistance at failure. Produces two Mohr-Coulomb parameters: friction angle (φ) for granular soils and cohesion (c) for clay soils.
Why it matters for solar: Friction angle directly controls passive soil resistance against the pile web — the mechanism that resists both pull-out and lateral loads. Higher φ = better C-post performance. Cohesion in clays must be tested at wet-season moisture content, not dry-season.
5. Soil Resistivity and pH — ASTM G57 / ASTM G51
Electrodes driven into the ground measure the in-situ electrical resistivity of the soil (ASTM G57). pH (ASTM G51), laboratory minimum resistivity (e.g. AASHTO T 288) and water-soluble chloride and sulfate (e.g. ASTM D4327) are measured on extracted samples.
Why it matters for solar: Low resistivity, low pH and high chloride or sulfate all point to corrosive ground, but published schemes draw the line in different places. In the NCHRP classification reproduced by the FHWA, soil below 2,000 Ω·cm is rated corrosive and below 700 Ω·cm very corrosive. FHWA/AASHTO criteria for non-aggressive select fill around steel reinforcement are resistivity above 3,000 Ω·cm, pH 5–10, chlorides below 100 ppm and sulfates below 200 ppm. Caltrans treats a site as corrosive where chloride is 500 ppm or more, sulfate 1,500 ppm or more, or pH 5.5 or less, and requires chloride and sulfate tests wherever minimum resistivity is 1,500 Ω·cm or less. European practice (EN 12501-2, DIN 50929-3) scores several soil properties together rather than applying one cut-off. Name the scheme in the specification, and compare like with like: these limits refer to laboratory minimum resistivity, which can be lower than an in-situ reading. How fast an under-protected post loses section depends on moisture and oxygen as much as on chemistry: long-term NBS exposures found little corrosion on steel piling driven into undisturbed soil, with more loss in fill and disturbed ground above the water table. Short solar posts sit largely in that upper, oxygenated zone. EN 1993-5 (Table 4-1) handles this by deducting a loss-of-thickness allowance per exposed face, chosen by soil category and design working life, from the section used in the structural check. Together these results set the post protection — ZAM-coated steel or hot-dip galvanizing, the coating mass and any sacrificial thickness — a material specification decision that cannot be reversed after procurement.
Sources: FHWA-NHI-09-087, Corrosion/Degradation of Soil Reinforcements for MSE Walls and Reinforced Soil Slopes (2009); Caltrans Corrosion Guidelines v3.2 (2025), Chapter 6; Romanoff, Corrosion of Steel Pilings in Soils, NBS Monograph 58 (1962); EN 12501-2; DIN 50929-3; EN 1993-5.
6. C-Post Pull-Out Test — ASTM D3689 (site-specific)
Not a standard lab test — this is a site-specific field test driven by solar mounting requirements. Drive sample piles of your target C-post sections (100×50×3.0C and 120×50×3.0C) at multiple embedment depths (1.5, 2.0, 2.5, 3.0, 3.5 m) using the same driver planned for production. Apply axial tension load incrementally and measure displacement to failure.
Why it matters for solar: This is the only test that directly measures what you actually need to know — how hard it is to pull your specific pile out of your specific soil. All other tests are inputs to a calculation. This test IS the answer. It cannot be substituted by manufacturer datasheets.
What a Complete Geo Report Should Contain
A geo report suitable for solar mounting foundation design should include:
| Item | Standard | What It Tells You |
|---|---|---|
| Borehole logs | ASTM D1586 | Soil layering and SPT N-values by depth |
| USCS classification | ASTM D2487 | Soil type per layer |
| Particle size distribution | ASTM D6913 / D7928 | Gradation curve, fines content |
| Atterberg limits | ASTM D4318 | Plasticity index for clay layers |
| Shear strength parameters | ASTM D3080 | φ and c for pile capacity calculation |
| Bearing capacity (allowable) | Derived | Foundation sizing input |
| Groundwater depth | Field measurement | Affects corrosion and frost risk |
| Soil resistivity, pH, chloride and sulfate | ASTM G57/G51, AASHTO T 288, ASTM D4327 | Corrosion environment classification |
| Frost depth (cold climates) | Local code / frost-depth data | Minimum pile embedment in frost zones |
| C-post pull-out test | ASTM D3689 | Direct pile capacity at your site |
A report missing more than two of these items is incomplete for foundation design purposes.
What Good Geo Data Enables
When you have a complete geo report, foundation design becomes a deterministic engineering process:
- Soil class known → Foundation type options narrowed to 1–2
- Bearing capacity and φ confirmed → Pile length and section calculated
- Pull-out test done → Required embedment depth confirmed against actual site data
- Resistivity, pH, chloride and sulfate tested → Steel coating specification locked in
The alternative — selecting foundation type without geo data — is speculation. It occasionally works. When it doesn't, it produces the most expensive problem in solar EPC: a foundation failure that requires mechanical extraction and reinstallation of every pile across the affected area.
Next: Matching Soil Type to Foundation Type
In the next article, we walk through exactly how each soil class maps to a specific foundation type — C-post sections and embedment depths, helical pile selection, micropile scenarios, and ballast systems. If you know your soil classification and bearing capacity, the foundation type selection becomes a clear decision tree.
Read Part 2: From Soil Test to Foundation Selection →
Solaracks provides foundation type recommendations, section selection, SAP2000 structural analysis, and load calculations within 24 hours of receiving your geo report. Contact us at info@solaracks.com.
