Partner loginPartner portalSoiling-loss calculatorBook a free pilotFree pilot

A full soiling study for very large plants

A solar plant soiling study measures what soiling costs across your whole plant, block by block, and works out the cleaning programme that returns the most. It is built for very large plants, where soiling varies across the site and every full wash is a major job.

Scope a plant study Talk to an engineer
Wide drone view down a central access road through a large solar PV plant, with tracker rows stretching to the horizon
91%
of South African solar IPPs surveyed cleaned their pyranometers weekly, as IEC 61724 recommendsNaicker (2018), via SAPVIA; IEA PVPS (2022)
50–150 m³
of water for one full wet wash of a 10 MWp plant, at 1–3 L/m²NREL (2018); Naicker (2018)

Why a full study

On a very large plant, soiling is not one number. It varies across a site by block, row and season: see soiling monitoring. Every full wash is also a big job. At 1–3 L/m² (NREL 2018; Naicker 2018), one wet wash of a 10 MWp plant, with about 48,000–55,000 m² of modules, uses roughly 50–150 m³ of water. Scale that to hundreds of megawatts and a wrong interval is expensive in either direction: too many washes waste water and crew time, too few waste energy.

Why performance ratio alone isn’t enough. Temperature-corrected performance ratio (PR) can give a good estimate of soiling in dry seasons, as long as the plane-of-array irradiance sensor is kept clean (IEA PVPS 2022). IEC 61724 recommends cleaning it weekly. PR still mixes soiling with clipping, outages and degradation, though. A study measures soiling directly, then uses SCADA and PR to check it.

What the study covers

  1. Soiling measurement across the site. Soiling stations or sensors in the blocks that soil differently, running through a dry spell so soiling rates can be fitted. See soiling monitoring.
  2. SCADA and PR analysis. Inverter and string data before and after cleans and rain, how the irradiance sensors are cleaned, and the losses that look like soiling but aren’t.
  3. Contaminant and water. What is on the glass, and whether the cleaning water meets the module maker’s limits. See soiling assessment and water testing.
  4. Cleaning review. Method, equipment, crew time, water use and cost per clean today.
  5. Cleaning economics. The optimal interval or trigger, and whether a coating pays back on your site.
  6. Costed programme. A plant-wide Restore → Protect → Maintain plan: method, calendar, water treatment, equipment, crew and products, plus the measurement plan that keeps it honest.

How the optimal interval is set

The right interval depends on your tariff, soiling rate and what a clean costs, and it changes with the season. The calculator works it out for your plant.

What soiling is worth a year: a 10 MWp fixed-tilt plant near Upington
Soiling lossBid window 1R4.39/kWhBid window 4R1.10/kWhBid window 6R0.49/kWh
3%≈R2.6 M≈R0.66 M≈R0.29 M
5%≈R4.4 M≈R1.1 M≈R0.49 M

Energy value at the bid-window tariff, before cleaning costs; bid windows 1–4 in April-2020 rands. The plant makes about 19.9 GWh a year (Global Solar Atlas).

Why the study runs through the seasons

A short test can mislead, because the answer changes with the weather. The Vietnam case shows a coating paying more in the dry season than in the rainy season. In the peer-reviewed Green Energy Park trial in Morocco, Solar Wash Protect led in the dry season, while Industrial Glass Protect edged ahead in the rainy months of March and April (case study).

Position matters as much as season. At a Scatec-operated plant in Honduras, the test string sat nearer a road and was subject to much more dust than the two control strings (Case study: Chemitek (Portugal), with Scatec). A study that covers the blocks that soil differently, through a dry season and a wet one, tells you whether a coating or a new interval pays back across the whole year.

What you get

Deliverable What it tells you
Soiling map and rates How fast each part of the plant soils, and when
Loss in MWh and rand What soiling costs at your tariff
PR and SCADA review Whether your monitoring shows soiling correctly
Cleaning economics The optimal interval or trigger, and whether coatings pay back
Costed programme Method, calendar, water, equipment, crew and products for the whole plant
Measurement plan Which sensors stay, and the triggers your O&M team acts on

Starting small. A study can start with a free pilot on a few strings, which proves the method and chemistry before the plant-wide numbers are run. On smaller plants, a site soiling assessment answers most of the same questions in one visit.

Chemitek products for this

The full range

Running your own crews? You can order just these products from us, with or without our engineers.

Sources

  1. Naicker, survey of 11 South African solar IPPs (2018), hosted by SAPVIA
  2. IEA PVPS Task 13, Soiling Losses – Impact on the Performance of PV Plants (2022)
  3. Ilse et al., Joule 3:2303–2321 (2019)
Show all 11 sourcesShow fewer
  1. NREL, Best Practices for O&M of PV and Energy Storage Systems, 3rd ed. (2018)
  2. IPP Office, REIPPPP tariffs by bid window (September 2020)
  3. gov.za, Bid Window 5 announcement (8 December 2022)
  4. Global Solar Atlas, Upington
  5. Micheli et al., Energy 215 (2021)
  6. Case study: Chemitek (Portugal), with BIM AC Renewables: D-Solar Defendor, 330 MW plant, Vietnam (2020–21)
  7. Case study: Chemitek (Portugal), with Green Energy Park: Elamim et al., Heliyon 9, e16163 (2023)
  8. Case study: Chemitek (Portugal), with Scatec: Solar Wash Protect, Honduras

Full plant study

Common questions

Still have a question?

Ask one of our engineers. Tell us about your plant and we’ll answer it properly.

Send your question

What size of plant needs a full study?

Very large plants, where soiling varies across blocks and a full wash uses a lot of water and crew time. For smaller plants, a site soiling assessment and a pilot usually answer the same questions. Tell us your plant size and we'll advise.

What data do you need from us?

Inverter and string data from SCADA, irradiance and weather data, and your cleaning records: dates, method and water use. We also need the module datasheets and manuals, and your tariff or PPA price if you want the results in rand.

Can't we just use the performance ratio?

Partly. IEA PVPS (2022) notes that temperature-corrected performance ratio can give a good estimate of soiling in dry seasons, as long as the plane-of-array irradiance sensor is kept clean; IEC 61724 recommends cleaning it weekly. But PR also mixes soiling with clipping, outages and degradation. A soiling station measures soiling directly, and the study uses both.

Will you tell us to clean more often?

Not necessarily. The optimum depends on soiling rate, yield, tariff and cleaning cost, so a plant on a low tariff may be better off with fewer, better-timed cleans. In southern Spain, one well-timed summer clean raised profit by up to 3.6%, and extra cleans paid less (Micheli et al. 2021).

How long must soiling be measured?

Long enough to capture a dry spell with no rain or cleaning, so a soiling rate can be fitted: see soiling monitoring. A full dry season gives the most reliable programme.

Read next

Ready to maximise your asset performance?

Talk to an engineer about a free pilot, a soiling problem or a Chemitek order.

Scope a plant study Get in touch