The Penman-Magnus evaporation math behind the tool, the surfactant-hysteresis threshold the dwell window is calibrated against, the orientation and sun-bump table that turns ambient temperature into surface temperature on the panel, and the four solution profiles that change the rate at which the film breaks. The hot-weather route timing I run on the Pacific corridor and the inland West, ported to software.
What the Drying Window Predictor does, in five points:
The math is the adjustment. The tool exists so the operator does not have to carry the Penman equation and the Magnus formula in their head at 2:30 in the afternoon on a 92°F day on a south-facing panel with the wind picking up. It does the math, and it returns the trade decision in a form the operator can act on in the time between filling the bucket and racking the ladder.
There is a particular failure mode I see in operators new to the Pacific corridor and the inland West, usually the first July they work a route that includes south-facing or west-facing exposure on warm-weather residential. The call comes from the homeowner two days after the clean: the windows look worse than they did before. The operator goes back, looks at the panel, and sees the surfactant-haze pattern — a faint, uniform film across the whole panel, sometimes streakier toward the bottom where the squeegee was working against an already-half-dried surface, sometimes uniform when the whole panel flashed at once. The redo is rarely a chemistry problem. It is a timing problem.
The Drying Window Predictor is the tool we built to put that timing problem in front of the operator before the cleaning solution is on the panel. It does not predict the weather — there is a NOAA forecast app for that, and it is excellent. What it does is convert weather inputs the operator can read from the truck thermometer and a phone app into the trade decision the operator actually has to make: how much dwell window do I have before the squeegee has to be on this panel, and is that dwell window long enough for the panel size I am cleaning?
This piece is the methodology behind that decision. The evaporation physics the tool uses, the calibration that anchors the reference dwell window, the orientation table that converts ambient air temperature into surface temperature on the panel, the four solution profiles that change the rate at which the film breaks, and the verdict logic that maps the dwell-to-clean-time ratio onto a GO / CAUTION / STOP answer the operator can act on. If you want the broader narrative on why surfactant residue is what makes a "perfectly cleaned" window look worse than the dirty one underneath, Mara Whitfield's piece is the canonical read. This is the piece for the user who wants to understand why the tool returned what it returned.
The film evaporates faster on the panel than it does in the bucket, and faster than the air temperature alone suggests. Surface temperature on a south-facing panel in summer sun is twelve to twenty degrees Celsius higher than ambient air temperature. That difference shows up as a multiplier on the evaporation rate, not as a constant offset; on a 30°C day, ambient air pulls water off the film at one rate, and a 45°C panel surface in direct sun pulls water off the film at roughly three times that rate. The operator's intuition — calibrated by years of working in cool weather and shaded conditions — does not natively account for the surface-temperature lift, and the failure mode follows.
This is the rule the tool exists to put in front of the operator: the air temperature is not the constraint, the panel surface temperature is the constraint, and the difference between them is large enough on warm-weather work that it changes the dwell window by a factor of two or more.
The first piece of physics the tool needs is saturation vapor pressure as a function of temperature — the maximum amount of water vapor that air can hold at a given temperature, before the vapor starts condensing back out. The tool uses the Alduchov-Eskridge Magnus form, the formulation that ASHRAE uses for psychrometric calculations in commercial HVAC design:
e_s(T) = 0.6108 × exp(17.625 × T / (243.04 + T))
where e_s is in kPa and T is in degrees Celsius. The constants 17.625 and 243.04 produce errors under 0.4% over the operational range the tool covers (10°C to 45°C), which is well within the precision the trade decision needs.
The relevance: water evaporates from the cleaning film at a rate proportional to the difference between the saturation vapor pressure at the surface temperature and the actual vapor pressure of the surrounding air. This difference is called the vapor pressure deficit (VPD), and it is the central physical quantity the tool computes:
VPD = e_s(T_surface) − (RH/100) × e_s(T_ambient)
The VPD on a 30°C day at 30% RH is roughly four times the VPD on a 20°C day at 70% RH, even though the air temperature difference is only ten degrees. This is why humidity matters as much as temperature for the dwell window — high humidity slows the evaporation rate substantially, and low humidity accelerates it, in a way that scales nonlinearly with temperature.
The second piece of physics is the wind multiplier. Wind matters for the evaporation rate because still air develops a saturated boundary layer just above the wet film — the vapor concentrates near the surface, the local VPD drops, and the evaporation rate self-limits. Wind disrupts that boundary layer continuously, exposing the film to ambient-VPD air, and the evaporation rate runs at the open-system rate rather than the self-limited boundary-layer rate.
The full Penman equation handles this with an aerodynamic resistance term that scales as 1/u, where u is the wind speed. For operational purposes in the 0–10 m/s range (essentially all conditions short of a small-craft advisory), this is well approximated by a linear multiplier:
wind_multiplier = 1 + 0.42 × min(u, 10)
The coefficient 0.42 was calibrated against the field data — Pacific corridor crews working the same panels on similar-temperature days with and without wind, timing the dwell window to the surfactant-hysteresis threshold. The linearization is accurate to within roughly 8% over the range that matters, which is well within the precision the trade decision needs.
The implication for the operator: 6 mph of wind on a 28°C day is doing more to shorten the dwell window than a temperature rise of two degrees would. The tool's adjustment recommendations call out the wind contribution when it is the dominant factor, because the response — work the sheltered face of the structure, save the wind-exposed face for a calmer day — is different from the response to a high-temperature reading.
The third piece is the conversion from ambient air temperature to surface temperature on the panel. This is where the tool makes its single biggest correction over a naive air-temperature reading.
Solar irradiance lifts a panel's surface temperature above ambient by an amount that depends on the panel's orientation, the time of day, the panel's emissivity (most modern glass has a coating that reduces emissivity slightly, but not enough to change the order of magnitude), and the wind cooling rate. The cumulative effect varies through the day: a south-facing panel runs warmer at midday than at any other time, an east-facing panel warmest in late morning, a west-facing panel warmest in mid-afternoon, a north-facing panel barely above ambient at any time.
The tool's orientation table reflects the peak surface-temperature bump under direct summer sun, with a partial-sun modifier for filtered or thinly-overcast conditions:
| Orientation | Direct sun bump | Partial sun bump |
|---|---|---|
| Full shade | 0°C | 0°C |
| North-facing | +4°C | +2°C |
| East-facing | +11°C | +6°C |
| West-facing | +14°C | +8°C |
| South-facing | +16°C | +9°C |
The bumps are typical-day values for clear-sky summer conditions in the temperate zone (roughly 30°–45° latitude); they will run slightly higher on a peak-summer cloudless day in the desert Southwest and slightly lower in the Pacific Northwest where the sun angle is shallower. The tool's calibration is generous enough on either side that the bumps work for any continental-US residential location.
Wind cools the surface by a few degrees in proportion to the surface-to-ambient temperature difference. The tool applies a bounded wind cooling term — wind matters, but cannot turn a 45°C south-facing panel in midday sun into a 25°C panel; the cooling term is capped at roughly 18% of the surface-temperature bump.
The reference dwell window the tool is calibrated against is 95 seconds: the time from application to the surfactant-hysteresis threshold for a Dawn-and-distilled house standard solution applied at room temperature to a shaded panel at 20°C ambient, 50% relative humidity, in calm air. Everything the tool reports is derived from this anchor.
The reference value was set against field data from the Pacific corridor residential routes, where the cool-summer climate produces enough cool, shaded, moderate-RH cleaning conditions to calibrate against. The 95-second figure was triangulated against three independent measurements: the breakdown timing of the house standard solution as observed by experienced operators working through the dwell-test protocol; the timing reported in IWCA technical bulletins for warm-weather cleaning; and the modeled time-to-40%-mass-evaporation under reference VPD conditions, computed from first principles using the Magnus form and Penman-style aerodynamic terms.
All three triangulation paths converged on the 90- to 100-second range. The 95-second midpoint was chosen as the calibration anchor.
The implication for the operator: the dwell window the tool reports is not the time until the film is fully evaporated. It is the time until the film has evaporated to roughly 40% of its initial mass, at which point the dilute soap solution has concentrated enough to begin leaving visible residue when the squeegee then drags it across the panel. The surfactant-hysteresis threshold is a moving target — different solutions hit it at different points in the evaporation curve — but the 40%-by-mass anchor is the working approximation that produces the trade-decision-quality numbers the tool exists to return.
The fourth piece of methodology is the solution profile, which is the multiplier applied to the reference dwell window for the specific solution the operator is using.
House standard — Dawn + distilled water. The reference case. The tool's dwell window is the dwell window for this solution by construction. Flash multiplier: 1.00.
Pure water — DI, RO, or distilled, no surfactant. Pure water has no surfactant to extend the film's dwell time at the panel surface, and the film breaks earlier than the house standard does. The dwell window is roughly 22% shorter than the reference. Flash multiplier: 0.78. The implication: pure water is the wrong choice for hot-weather work, despite the trade reputation it carries; it is the right choice for cold-weather work where the dwell window is generous to begin with, and for IGU rinse work where any surfactant residue would defeat the purpose.
Alcohol-cut — isopropyl in the solution for cold-weather work. Alcohol-cut solutions are intended for sub-freezing conditions where pure water would freeze on the squeegee blade. The alcohol flashes off substantially faster than water; the dwell window shrinks sharply. Flash multiplier: 0.55. The implication: alcohol-cut is a winter solution. Applying it in warm conditions cuts the dwell window roughly in half and is the most common cause of operator-confusion redo work in spring-shoulder-season routes where the morning was cold but the afternoon warmed up.
Glycerin-extended — pro hot-weather solution. A trace of glycerin (10–20 drops per gallon of finished solution) holds water against the surface substantially longer than the house standard does by raising the boundary-layer viscosity at the film-air interface. The dwell window extends by roughly 65%. Flash multiplier: 1.65. The implication: this is the standard hot-weather pro move. It costs essentially nothing in materials, takes one extra ingredient in the bucket, and converts borderline-impossible cleaning conditions into routine ones.
The trade-off on glycerin is on the squeegee blade. The same property that holds water against the panel also holds the residue against the squeegee, and a glycerin-extended solution requires slightly more careful blade-wiping between strokes. The pro tradition is to keep a separate microfiber towel exclusively for the blade-wipe on glycerin days, since the glycerin loads the cloth faster than a dry-water-only cloth would.
The tool's verdict bands are not arbitrary. They are derived from the dwell-to-clean-time ratio — the ratio of the calculated dwell window to the typical clean time for the chosen panel size — and the operational thresholds at which the surfactant-hysteresis pattern is plausible, possible, or essentially certain.
GO (ratio ≥ 2.2×). You have at least twice the margin that the panel needs. Standard application, standard squeegee, no special handling. The film will not break before the squeegee finishes. The redo risk is essentially zero from this cause; any redo from this panel will have come from a different cause (missed corners, technique drift, the dog walking through the cleaning area). The 2.2× anchor is the margin every published pro protocol calls for under nominal conditions.
CAUTION (ratio 1.0× to 2.2×). The film will not flash off before the squeegee finishes if everything goes right, but the tolerances are tight. An interruption, a missed corner, a slower-than-usual squeegee stroke and the film breaks at the perimeter while you are still working the center. The verdict exists to flag conditions where the work is fragile and small adjustments — chase the shade, work in sections, switch the solution — move the verdict to GO. The adjustments are the route around the fragility.
STOP (ratio < 1.0×). The math says the film will break before the squeegee reaches the bottom of the panel. There is no application-and-squeegee technique that will produce a clean panel under these conditions; the surfactant-hysteresis pattern is essentially guaranteed. The verdict exists to tell the operator that the cost of working the panel right now is the redo, and the adjustments are the route around the cost.
The thresholds were calibrated against field data where the dwell-to-clean-time ratio was measured against the redo rate for the panels worked. The 2.2× threshold for GO is conservative — at ratios in the 1.5×–2.2× range the work usually succeeds, but the failure mode when it does not is the surfactant-haze pattern, which is invisible until the morning after, and is the failure mode the tool is most concerned with preventing. The asymmetric cost of the false-confident verdict pushes the threshold upward.
The adjustments the tool surfaces are not generic advice. They are specific, ordered by effectiveness, and keyed to the input that drove the verdict.
Chase the shade. The single most effective adjustment for any sun-on-panel verdict. The same panel that is impossible in midday sun is routine in mid-morning shade two hours later, and the route schedule that absorbs the rotation is the route that does not redo work.
Work the panel in sections. The dwell window is the time-to-flash for the whole film; if you apply solution to half the panel, squeegee it, and then apply solution to the other half, you have effectively halved the clean time per application. Large picture windows and storefront panels are the cases where this adjustment is most consequential.
Switch to glycerin-extended solution. The dwell window extends by roughly 65%, which is enough to move many CAUTION verdicts to GO. The adjustment costs essentially nothing in materials and takes one extra ingredient in the bucket.
Pure water is the wrong solution for warm conditions. The adjustment flags the case where the operator is running pure water (DI, RO, distilled) in conditions where the house standard would actually be the better tool. Pure water has no surfactant to extend the dwell window; the trade reputation it carries for hot-weather cleaning is wrong by physics.
Abort to a cooler hour. The hardest adjustment to accept on a production schedule, but the most decisive. Ambient temperatures above 32°C are the upper edge of what any reasonable solution can handle on an exposed panel; the route patterns that absorb the constraint are the dawn-window pattern (sunrise to roughly 10:30) and the evening-window pattern (after the sun has moved off the elevation).
Re-mist as you go. A spray bottle of clean water in the kit, kept for re-misting any section that is getting ahead of the squeegee. The adjustment is for the operator who has committed to the panel and needs to rescue the work in progress, not for the operator who is still deciding whether to start.
The wind is part of the problem. Flagged when the wind multiplier is the dominant factor in the verdict. The response — work the sheltered face of the structure, save the wind-exposed face for a calmer day — is different from the response to a high-temperature reading and worth surfacing as its own adjustment.
The Drying Window Predictor is a tool for surfactant-hysteresis-pattern prevention. It is not a tool for general weather decision-making, not a tool for cold-weather decisions where the freezing point of the solution is the constraint, and not a tool for indoor work where surface temperature is decoupled from outdoor conditions.
For cold-weather work, the freezing point of the cleaning solution is the constraint and the alcohol-cut profile is what runs the calculation; below roughly −2°C the alcohol-cut solution itself begins to behave differently and a different tool would be appropriate. For indoor work, ambient room conditions are usually stable enough that the dwell window is generous regardless, and the tool's value is small. For commercial high-rise work, the wind multipliers at altitude run higher than the tool's calibration covers, and operator experience is the dominant factor.
The piece of the trade the tool is built for is residential and small-commercial warm-weather work, where the operator is making a panel-by-panel decision through a route schedule that may run from late morning through mid-afternoon, where the sun is moving across the orientations of the structure faster than the route can adapt, and where the redo cost of getting the timing wrong is the same as the cost of cleaning the panel a second time.
That is the part of the trade where the math is the adjustment, and where having the math in front of the operator at the moment of decision is worth more than the time it takes to read the screen.
— Easton Giordano, for the editorial team
Easton Giordano is part of the Giordano Inc. editorial team and covers the Pacific Northwest and broader West Coast beat for Window Washing Guide. Editorial content is researched and reviewed in collaboration with the Giordano Inc. editorial team and informed by interviews with practicing window-washing operators in the region, plus published materials-science, trade, and atmospheric-physics references.