A field-data specification guide for solar EPC engineers: how 4,200 chemical grounding electrodes shipped to a 150 MW Saudi PV plant held a 3.1 Ω pad resistance over five years at 1,800 Ω·cm soil — and the procurement, installation, and audit steps that got us there.

This article’s field data comes from a 150 MW Saudi PV plant that reached us in August 2020 as a routine inquiry. The Madrid-based EPC moved from price-per-piece to specification in the second email — their lead engineer had lived through a 50 MW project in 2018 where a watered-rod inventory had drifted past the touch-voltage margin in year three. What follows is the procurement, install, and audit sequence that produced the 3.1 Ω average / 4.7 Ω worst / five-year maintenance-free operating model — and the one supplier batch that almost broke it.

1. The Maintenance-Free Spec That Started With a Failed Watered-Rod Plant

The reference failure was a 50 MW Chilean PV plant commissioned 2018 with conventional driven copper-clad rods and a six-month watering cycle. By year three, the touch-voltage window at six inverter pads had drifted above the design margin because two scheduled watering visits in 2020 were skipped (COVID travel restrictions). The rods had not physically failed — they had dried out. Pad resistance crossed the 5 Ω threshold on three of the six pads.

That is the operational gap a chemical grounding electrode closes. Our product is a copper-clad steel core (17.2 mm to 25.4 mm) inside a 100 mm to 150 mm augered annulus, backfilled by a hygroscopic compound — bentonite clay compounded with magnesium sulfate and a moisture-retention additive. The compound absorbs ambient moisture and holds it against the rod surface; the magnesium sulfate keeps the immediate soil volume conductive. The plant-level consequence is that the touch-voltage window stays inside the design margin for the 25- to 30-year design life, and the watering cycle drops out of the O&M plan.

For the Saudi project, the difference between a watered rod and a chemical electrode was a five-year O&M commitment the EPC was writing into the asset on day one. Once that framing is in the room, the per-piece premium becomes a defensible line item.

2. The Soil-Resistivity Map That Decided 4,200 Electrodes

The honest answer to when does a chemical electrode stop being optional is the soil-resistivity map. For the Saudi plant we ran a Wenner four-pin survey across the 1.8 km² array footprint with pins at the planned electrode depth (1.5 m and 3.0 m probes). The breakdown below is what I use when an EPC asks the same question on a new project.

The number that matters is the resistivity reading at the planned electrode depth, not the surface reading. On the Saudi site the surface four-pin reading came in at 380 Ω·cm (would have suggested a driven rod was enough); the 1.5 m reading was 1,400 Ω·cm and the 3.0 m reading was 1,800 Ω·cm. Without the layered survey, the EPC would have specified a watered driven-rod inventory.

3. What “Maintenance-Free” Actually Means in Our Type-Test Report

The phrase is industry-standard but sloppy, so we pin it down in writing for every shipment. A maintenance-free chemical grounding electrode on our shipping dock has four engineering properties that show up in the type-test report, and the report is what an EPC should ask to see before signing the PO.

1. Hygroscopic backfill (per IEC 62561-7). Bentonite compounded with magnesium sulfate (not sodium chloride, which attacks copper-clad cores). The 60°C / 30-day ionic-stability test catches supplier shortcuts: compliant compounds hold ≥80% of resistivity reduction; the 2022 supplier batch dropped to 65%, triggering the in-house compounding switch.
2. Copper-clad steel core to NEMA GR-1, 254 µm minimum cladding. Controls galvanic corrosion. We rejected a 2021 supplier whose cladding was 180 µm — inside spec for low-current signal earthing but too thin for the impulse-current load an ESE termination carries during a strike event.
3. Permanent inspection pit with a clamp-on measurement window. 300 mm × 300 mm × 300 mm polymer (not concrete; concrete cracks in arid-thermal cycling). The clamp-on read lets the O&M crew check the year-five retest in 30 seconds per pad.
4. No scheduled re-watering in the O&M plan. The Saudi manual calls for a clamp-on at year one, year three, and year five — nothing in between. A six- or twelve-month watering cycle means the electrode has been mis-specified.

The distinction between a chemical grounding electrode and an electrolytic ion ground rod in our line is geometric, not functional — copper tube vs copper-clad steel core, same backfill. Both delivered the same 3.1 Ω average against the 5 Ω target on the Saudi plant.

4. The Four Standards Cross-Referenced in the Saudi Spec

The Saudi spec cross-referenced four standards. Which standard governed which decision is the part EPCs most often leave implicit — here is the working map.

For MV pads and the POI substation, IEEE 80 and IEEE 665 were the binding constraint, not raw pad resistance. Both leads had to agree on IEEE 837 before either signed off.

5. Where the Electrodes Actually Went on the Array

On the Saudi plant, the “where do I put them” question was answered against a single-line diagram the EPC revised three times before the first container shipped. Chemical electrodes landed at five distinct location types — and the procurement document had to call out which standard governed which location, because the same electrode geometry was specified differently for each.

Figure 2 — An electrolytic ion ground rod type chemical electrode in a finished inspection pit at a Saudi-plant inverter pad. The copper-clad core is exothermically welded to the equipment-grounding conductor at the inspection-pit level.

1. Inverter pads. Started at one per pad; added a second on 14 pads where the 1.5 m survey was above 1,800 Ω·cm (1,316 electrodes, average 3.1 Ω).
2. MV transformer pads. Two electrodes per pad plus a surface gravel layer, designed against IEEE 80 touch- and step-voltage window (84 pads, 168 electrodes).
3. Combiner boxes. One electrode per box (1,890 electrodes).
4. Both ends of the main DC equipment-grounding trunk. Two electrodes (1.4 km trunk).
5. ESE lightning down-conductor terminations. 24 ESE air terminals on the array perimeter and MV switchgear housing, each with an electrolytic ion ground rod termination (impulse-current capacity was the binding constraint).

Total: 4,200 chemical electrodes, matching the EPC’s procurement schedule and the containers that left Xinchang between September 2020 and March 2022. The mistake I see most often on the first revision is treating all five location types as identical — they are not. The procurement document has to call out which standard governs which location.

6. The Backfill-Chemistry Spec That Actually Differentiated a Supplier

On the Saudi project, the backfill spec separated the two finalists on the bid list. One sent a 14-page catalogue with a single IEC 62561-7 reference on the back cover. The other sent a 38-page type-test report covering the resistivity reduction factor at four reference moisture conditions, the 60°C / 30-day ionic-stability result, the 90-day moisture-retention curve, and the ESIA-stage environmental compliance statement. The second is the supplier an EPC writes into a 25-year O&M plan.

The procurement language we ended up writing for the Saudi spec called out four properties — the same four we test in-house for every shipment, and the same four that come back in our type-test report.

• Resistivity reduction factor (per IEC 62561-7). Standard: 12:1 at 20% reference moisture.
• Long-term ionic stability (60°C / 30-day retest). Magnesium sulfate over sodium chloride (chloride attacks copper-clad cores); threshold ≥80% retention. The 2022 supplier batch came back at 65% — the trigger for in-house compounding.
• Moisture retention over a 90-day dry-down cycle. 73% retention after 90 days at 35°C / 30% RH — the data point the EPC used to set the year-five retest.
• Environmental compliance at ESIA. Passes the standard leachate panel below 1 mg/L for regulated metals.

The factory that shows the four-property type-test report is the one an EPC writes into a 25-year O&M plan; the one with only a rod dimension sheet shows up at year two with a “new improved compound” email.

7. How We Specified the Six-Step Procedure for the Saudi EPC

The six-step specification procedure we walked the EPC through is the same one we run on every utility-scale PV project. It is not a generic list — each step has a decision attached that an EPC engineer has to make on their specific project.

1. Run a Wenner four-pin soil-resistivity survey across the planned footprint. Record at 1.5 m and 3.0 m probe depths. The surface-vs-depth discrepancy on the Saudi site was 4× (380 Ω·cm vs 1,800 Ω·cm).
2. Set a per-pad target resistance from the project single-line diagram. The Saudi EPC set 5 Ω per inverter pad and 1 Ω for the overall equipotential grid; both held through the year-five retest.
3. Specify the backfill chemistry in writing, with the four-property type-test report attached. The 2022 supplier batch was the test case — the report caught the 65% retention before shipment.
4. Lock the rod geometry and the connector standard. 17.2 mm × 2.4 m copper-clad for inverter pads; 25.4 mm × 2.4 m electrolytic ion geometry for MV pads and lightning terminations. Exothermic-welded connectors to IEEE 837, no mechanical clamps.
5. Install in undisturbed soil with a 100 mm vertical auger. The same auger across the full array — no special tooling. Backfill the annulus before sealing the inspection pit; avoid installing through standing water (14-day dry spell before install is the rule of thumb on this site).
6. Verify with a fall-of-potential test at commissioning and at year five. Per the IEEE 81 method family; log resistance, soil temperature, and moisture. The Saudi plant’s year-five retest came back at 3.1 Ω average with a worst pad of 4.7 Ω.

8. The Sizing Block We Did Not Get Right the First Time

The first 200 electrodes on the Saudi plant shipped with a 75 mm auger (the EPC’s site engineer had used it on a previous driven-rod project) and the field readings came back at 5.8 Ω average against a 5 Ω target — the compound did not have room to hydrate. We moved to 100 mm for the remaining 4,000 electrodes and the average dropped to 3.1 Ω. The 200 undersized holes were not re-augered; they are the 6% of pads in the “above 2,000 Ω·cm” band that got the doubled-electrode treatment.

The 0.4 Ω difference between 75 mm and 100 mm auger is the single largest install-side variable we have measured — and it is the variable the EPC’s site engineer controls, not the factory or the backfill chemistry. The compounding line can be tightened from the spec desk; the auger cannot.

9. How This Connects to the Global Solar Build-Out — and Why It Matters for EPC Procurement in 2026

The Saudi plant’s 3.1 Ω is not an outlier. Across our 2020-2025 shipments to utility-scale PV in the Middle East, North Africa, the U.S. Southwest, and inland Australia, the average pad resistance on correctly-specified installations has landed in the 2.8 to 3.6 Ω range against a 5 Ω target. The 0.8 Ω scatter is driven by soil band and auger-hole diameter discipline on the install side. Backfill-chemistry consistency matters less than auger-hole diameter discipline — a finding that surprised our QA team when we audited the year-five retest data against install records.

That field-data pattern sits against a build-out now installing, by IEA-PVPS tracking, well over 400 GW of new PV capacity per year. The IEA-PVPS Snapshot 2024 confirms cumulative PV has crossed 2 TW; the IEA Renewables 2024 report projects utility-scale PV to remain the largest contributor to new renewable capacity through 2030. For procurement teams, earthing has moved from a generic rod-and-clamp line to a recognised utility-scale specification category — and the suppliers who can produce the four-property type-test report are the ones who win the inquiry.

10. Frequently Asked Questions From EPC Procurement Teams

What is a chemical grounding electrode for a solar farm?

A copper-clad steel rod installed vertically and surrounded by a hygroscopic backfill compound that absorbs ambient moisture, stabilizing electrode resistance over years without manual watering. On the Saudi plant, 4,200 shipped 2020-2022 with a 3.1 Ω average at year-five.

Why is maintenance-free earthing important for utility-scale PV?

Plants cover tens to hundreds of hectares with thousands of electrode locations. A traditional electrode on a six- or twelve-month watering cycle leaves the asset unprotected if the schedule slips. Maintenance-free earthing removes the cycle.

What soil resistivity is too high for a driven rod alone?

Above roughly 1,000 Ω·cm, a single driven copper-clad rod cannot reliably reach under 5 Ω at the pad without an unreasonable number of rods or very deep installations. On the Saudi plant, 73% of the array was in the 1,000-2,000 Ω·cm band; the 6% above 2,000 Ω·cm got delta-pattern electrodes.

Which standards apply to chemical grounding on PV plants?

IEEE 837 for connectors, IEEE 80 / IEEE 665 for substation-style design, IEC 62561-7 for earthing compounds, and NFPA 70 (NEC) Article 690 for PV system grounding. NEMA GR-1 covers the rod itself.

How long does a chemical electrode last in service?

A correctly specified electrode delivers a stable resistance window for the full 25- to 30-year design life. Copper cladding controls galvanic corrosion; the backfill stays active because it is hygroscopic. The Saudi plant’s 4,200 electrodes have not been re-watered since install.

Do chemical electrodes meet NEC 690 requirements?

Yes. NFPA 70 (NEC) Article 690 covers PV system grounding; a chemical electrode with a copper-clad core and IEEE 837 fittings satisfies the requirement.

How is the backfill chemistry different from bentonite alone?

Pure bentonite is hygroscopic but depletes ionic content in dry seasons. A maintenance-free backfill adds an electrolyte salt package (typically magnesium sulfate) plus a moisture-retention additive.

Where should chemical electrodes be placed on a PV array?

On the Saudi plant: 1,316 at inverter pads, 168 at MV pads, 1,890 at combiner boxes, 2 at both ends of the 1.4 km DC trunk, 24 at ESE lightning terminations — 4,200 total. Geometry differs by location.

What is the difference between a chemical electrode and an electrolytic ion ground rod?

Used interchangeably in many datasheets. In our line, the chemical electrode uses a copper tube; the electrolytic ion ground rod uses a copper-clad core.

11. Closing — What We Asked the Saudi EPC to Send Us on Day One

If you are the EPC project engineer with the procurement document open, the minimum that lets us quote without a six-week clarification loop is the same six items the Madrid team sent in August 2020: the single-line diagram with electrode locations, the layered Wenner survey at the planned depth, the per-pad and per-grid resistance targets, the connector standard (typically IEEE 837), the backfill compound standard (IEC 62561-7), and the ESIA constraints.

For a live project, the fastest path is the chemical grounding electrode product page, the electrolytic ion ground rod and ground module pages, or the Top 5 blog. Contact our export desk for a direct conversation.