What soiling actually costs a Highveld array

Soiling is the only fault on a solar installation that never announces itself. Inverters do not alarm on it, monitoring platforms rarely flag it, and because it accumulates gradually there is no day on which anybody notices the array is producing less.

Across four Highveld commercial installations monitored through a full dry season, annualised soiling losses on uncoated arrays ran between 5.2% and 7.1%. The upper end belonged to a site adjacent to an unpaved access road — unsurprising, but worth stating, because proximity to dust sources matters far more than most owners assume.

The pattern within each quarter was consistent. Output recovered to near clean-state immediately after a professional clean, then declined roughly linearly for six to eight weeks before flattening as the deposit reached a stable thickness. That flattening is the dangerous part: the array settles into a degraded steady state and stays there until the next clean.

What coating changes is the slope of that decline, not the starting point. A coated array loses ground more slowly and recovers more fully from ordinary rain, so the area under the curve — which is what you actually get paid for — is substantially larger.