The site and the test
SirajPower owns and maintains the plant, where cement dust had cured onto the glass. Its routine clean, every two weeks with water and a rotary brush, wasn’t removing it. Before treatment, the affected inverter was losing almost 7% of its output against a reference inverter whose modules had no cement on them and were cleaned with water only. Chemitek’s report doesn’t state the plant’s type or capacity.
- Test area: 288 panels on inverter 6, cleaned with Chemitek Inorganic Removal Agent (CRA). Inverter 7, free of cement and cleaned with water, was the reference.
- Dates: 1 May to 5 June 2022 (35 days), with the CRA clean on 1 June.
- Dilution: 1:2, each 5 L of CRA in 10 L of water, as Chemitek recommended for heavy contamination.
- Dose: 500–800 mL of diluted solution per panel.
- Method: spray on, then brush and rinse off with water a few seconds later. The product must never dry on the module; if it does, the clean has to start again.
- Measure: daily energy and daily performance of inverter 6 against inverter 7, both under the same irradiation and ambient temperature.
Results
- After the 11 May clean−6.77%
- After the 24 May clean−6.68%
- Before CRA, 27–31 May−6.59%
- After CRA, 1–5 June−2.06%
Routine cleans were water and rotary brush; the CRA clean was one application at 1:2 (Chemitek case).
| Cleaning (2022) | Method | Average daily gap to the reference inverter |
|---|---|---|
| Routine clean, 11 May | Water and rotary brush, both inverters | −6.77% |
| Routine clean, 24 May | Water and rotary brush, both inverters | −6.68% |
| Before CRA, 27–31 May | – | −6.59% |
| CRA clean, 1 June (1–5 June) | CRA at 1:2: spray, brush, rinse | −2.06% |
Two routine cleans left the gap about where it was. One CRA clean narrowed it from −6.59% to −2.06%, about 4.5 percentage points; Chemitek describes this as recovering almost 5% of output. The O&M team reported that CRA broke the cement’s bond to the glass and emulsified the particles, and that the crew got faster with practice. It decided to use CRA across the whole contaminated plant, then protect it with Solar Wash Protect or Antistatic Solar Armor to reduce cement adhesion and avoid repeat CRA cleans.
How to read it. The after-treatment data cover five days, so the report shows the recovery, not how long it lasted. A −2.06% gap remained, and the report doesn’t say whether that is residual cement or a normal difference between the two arrays. A longer measurement period, ideally with a soiling sensor, would answer both.
What this means for a South African plant
- Heat. Hot glass dries the product fast, and Chemitek advises against cleaning hot modules. In Dubai the crew rinsed within seconds. In South Africa that means early-morning, late-afternoon or overcast work, especially in summer and in the arid Northern Cape, where parts get under 70 mm of rain a year (SAEON).
- Water. Cleaning more often with water won’t fix cured cement: two fortnightly cleans left the gap about where it was. The fix was a chemical remover, not more water.
- Contaminant. Cement and limestone dust cure the same way wherever they settle. See Chemitek’s Chile carport case and Israel case, and our North West cement-dust pilot.
- Quantities. 500–800 mL per panel at 1:2 suited heavy contamination here. Chemitek’s advice is to test a small area first and to leave the product on longer rather than mix it stronger. Our cement-dust page covers the method, and a pilot sets the dose for your deposits.
