Coupled outputs

Downwash cone: rotor airflow accelerates downward, merges into a turbulent jet/wall-jet interaction near the panel, then spreads radially across the glass. This creates drag, shear, turbulence, a pressure gradient, and edge roll-off. The exact stand-off is treated as a controlled operating envelope, not a fixed physics constant.

Ion plume: the emitter injects ions into the rotor flow so the airflow that transports the ions is also the airflow that later may transport particles. The emitter can be a corona discharge element or a dielectric-barrier-discharge device; both are shown as ion sources riding inside the aerodynamic wake.

Particle-size regimes

Adhesion forces holding dust down

F_vdW is the molecular adhesion baseline and can govern micrometre-scale residual fine-dust contact. F_img is an image-charge contribution for charged dust near a field-bearing or conducting boundary; because it scales with particle charge, partial neutralisation may reduce this component. F_es is shown in crimson as a surface-field-dependent contribution that can matter under particular module, surface-potential, moisture and contamination conditions. F_cap appears when local surface moisture, dew, brine or humidity is high enough to form meniscus bridges.

High-value target: in arid regions, the visible loose upper dust layer may move while a finer bound film remains. That residual fine film is the regime AIr·ION is designed to address, provided it is still loose/fresh mineral dust rather than cemented gypsum, salt crust, biofilm or oil-bound contamination.

Mobilisation forces moving and clearing dust

F_aero,t and τ are interpreted primarily as surface-parallel wall-jet drag and boundary-layer shear acting on rolling or sliding thresholds. Re represents turbulent bursts and disaggregation. ∇P is treated as a pressure-gradient and wall-flow transport contribution. F_EHD is an optional ionic-wind / EHD contribution to be measured or bounded, not assumed to dominate rotor flow. F_g helps only after mobilisation, especially on tilted panels.

Bipolar vs unipolar ionisation

Bipolar mode emits both polarities and is used to reduce net particle charge across mixed-polarity dust. Because image-charge adhesion depends strongly on net charge, partial neutralisation may reduce the adhesion budget.

Unipolar mode is sign- and boundary-condition-dependent: one selected polarity may create a favourable Coulomb assist state relative to the local surface field, or it may be ineffective or adverse. The separate violet F_coul arrow is therefore a conditional diagnostic path, not a guaranteed release mode.

Surface-state window

Humidity remains a useful anchor variable, but it is not the whole operating model. AIr·ION is most relevant when the dust is dry to weakly water-mediated: fine mineral particles, weak capillary bonding, no brine film, no cemented crust, and charge effects still relevant. Relative-humidity bands are only heuristics. Panel temperature, dew-point margin, wet–dry history, wind, dust chemistry, surface coating and electrostatic memory can shift the effective boundary.

Environmental surface-state variables

The useful site question is not only “what is the average RH?” but “what is the particle–surface state during cleaning?” Log panel temperature, dew/fog/rain history, wind at panel height, dust chemistry, particle size, coating/roughness, tilt/tracker state and live/off/grounding preconditioning. Phoenix- or Bikaner-type arid conditions are plausible candidates for favourable surface states, while local dust chemistry, dew history and panel condition determine the actual response.

Four charge reservoirs

Useful variables to think in

What happens when ions arrive

Bipolar operation mainly reduces net particle charge. Because the image-charge term depends strongly on particle charge, even partial neutralisation can collapse a large part of the electrostatic adhesion without requiring perfect discharge.

Unipolar operation mainly drives the particle toward a selected charge state. If the selected polarity and local field are favourable, the particle experiences a conditional favourable Coulomb term. This is why the figure separates the purple neutralisation arrow from the violet conditional Coulomb assist arrow.

Not all ions survive. Some recombine in the plume, some attach to other particles, and some are diluted by turbulence. That is why delivered ion flux at the surface matters more than nameplate emitter voltage alone.

Can the charge be “overcome”?

The right question is not whether the system overwhelms every force in absolute terms. The practical target is whether ion exposure changes the residual-response metric relative to matched airflow-only treatment under the same airflow, standoff, exposure time and surface state.

Most relevant regime: fresh or moderately adhered dry mineral dust, especially the measured residual fine-dust fraction where airflow alone may struggle. Conditional regime: mixed dust in a weakly water-mediated surface state, with RH used only as a heuristic anchor. Limitation regime: high-RH capillary states, hygroscopic coastal salts, cemented gypsum crusts, biofilm/EPS and oily contamination.

The term “how much charge the wind can entail” is better translated as how much ion current / ion flux the rotor-carried plume can deliver to the panel. Air itself is not a charge reservoir in the same way a conductor is; the actionable quantity is charge transported per unit time to the surface.

The stubborn arid fine layer

Yes — this is the intended target. If a loose upper dust layer moves but a finer grey/brown film remains, that residual film is usually the small-particle regime where Van der Waals, image-charge and surface-field and image-charge adhesion can matter over gravity and inertia.

The caveat is material state. AIr·ION is most relevant for fresh or moderately aged mineral dust. It is less likely to solve cemented gypsum, salt crusts, EPS/biofilm or oily organics without another cleaning mode.

Panel and string configuration

Panel type and wiring are relevant only when they change the local electric field or leakage behaviour at the glass surface. Important variables include string voltage, module bias, frame grounding, inverter topology, leakage/PID tendency, glass/ARC dielectric properties, residual charge after shutdown, and whether the panel is live during cleaning.

They do not replace the basic dust-adhesion physics: Van der Waals contact still exists even on an unpowered panel. But they can materially change F_es, field-assisted ion drift and the polarity choice for unipolar mode. Treat E_s as a measurement variable, not a fixed assumption.

Ion plume / footprint

The rotor wake does not carry a perfectly collimated “beam”. It entrains, dilutes and spreads the ions. The important quantity is the delivered ion footprint at the panel plane, at the real standoff distance and rotor throttle.

Ion plume control

The system should not rely only on nameplate emitter voltage. Control logic can estimate delivered ion current density or surface response at the panel plane, then adapt ion current, polarity, duty cycle, standoff, traverse speed or repeated passes to maintain a target delivered dose.

A useful control target is D_i ≈ J_panel · W_ion / v_t. If delivered flux is low, the system can slow down, reduce standoff, increase duty cycle, repeat a pass or switch mode rather than simply increasing high voltage.

Charge numbers: order of magnitude

For a small isolated dust particle, a rough capacitance estimate is C_p ≈ 4πϵ₀r. This is only a scaling aid: small particles can shift potential with very small absolute charge, but real PV dust is irregular, multi-contact and affected by humidity, surface potential and dielectric boundaries.

This is why small particles can be electrostatically important even with tiny absolute charge. The bottleneck is usually not whether charge can be generated, but whether enough useful ions reach and attach to the right particle-surface interfaces.

Formula caution

The image-charge square relationship should be treated as scaling intuition, not a narrow exact model for every panel. In the ideal conductor case, the image-charge component is strongly dependent on particle charge and often written as proportional to q_p². Real PV modules add glass thickness, dielectric layers, ARC coatings, panel stack geometry, humidity, dew-point margin, wet–dry history, particle shape, roughness and contact state.

A careful technical statement is: ion exposure reduces a charge-dependent electrostatic adhesion component until the remaining adhesion can fall below the aerodynamic removal force.