The “◯ W/m²” on a switchable film datasheet is not a constant of that film. The same film draws different power depending on how many hertz the controller drives it at. The reason is simple: electrically, switchable film is a capacitor.
Published 20 August 2026 · Miraigen Inc.
The short answer
| Change this | Power scales | Example |
|---|---|---|
| Drive frequency | Roughly linearly | Double it and power roughly doubles |
| Drive voltage | With the square | 1.2× the voltage is about 1.44× the power |
| Area | Linearly | Double the area, double the power |
Of these three, frequency is the one the controller decides, not the film. So the same film draws different power depending on which controller is driving it. Size a power supply straight from a catalogue W/m² figure and you can end up well short — or paying for far more headroom than you need.
Switchable film is a thin active layer between two plastic sheets, each carrying a transparent conductor. Electrically that is a parallel-plate capacitor with a dielectric between the plates — nothing more exotic.
A capacitor passes no direct current. Apply an alternating voltage and it charges and discharges every time the polarity reverses. Each of those charge movements dissipates a small amount of heat in the material and in the resistance of the electrodes. One cycle’s worth is tiny, but it happens as many times per second as the frequency — so doubling the frequency roughly doubles the heat. That is all “power scales with frequency” means.
Written out:
P ≈ 2πf · C · V² · tanδ
f is the drive frequency, C the capacitance of the film (proportional to area), V the drive voltage, and tanδ the loss tangent, a coefficient set by the material. All three rows of the table above come straight out of this one expression.
Strictly, tanδ itself varies somewhat with frequency, so the relationship is not exactly linear. “Roughly proportional” is a sound working assumption for sizing, but if you are changing frequency by a large factor, measure rather than extrapolate.
In principle DC would remove the charge–discharge cycle altogether and the power draw would be very small. In practice, you must not drive switchable film with DC.
The active layer contains traces of ionic impurity left over from the materials. Hold a field in one direction and those ions drift to one electrode, where they set up a field of their own that opposes the applied one. The result is that the effective voltage across the layer falls while the applied voltage stays the same. What the customer sees is a panel that was clear when it was commissioned slowly going hazy. Electrode and material degradation follow.
So switchable film is driven with AC, with the DC component kept at zero — the same reason liquid-crystal displays are AC-driven. “Use DC to save power” was never on the table.
That is the right direction, and it does lower the power. But there is a floor.
How clear the film appears is set by the magnitude of the voltage across it. Because the drive is AC, that voltage passes through zero twice per cycle. Drive fast enough and the eye integrates it into a steady state. Drive slowly enough and the periodic dip becomes visible flicker — and on something the size of a partition or a window, it is hard to ignore.
The range actually in use is wide: roughly 40 Hz up to several hundred hertz. Mains frequency, 50–60 Hz, is by far the most common, and our own drivable-area guidance is quoted at 60 Hz. Some products deliberately run higher to suppress flicker or sharpen the switching response, and our controller’s drive frequency is adjustable too. As above, that costs power. Treat the frequency as a value someone has already traded off between flicker and consumption.
This is where the physics turns into a purchasing problem. A W/m² figure with no stated drive frequency cannot be compared with another one.
Suppose one film is quoted at 3 W/m² measured at 50 Hz, and another at 3.5 W/m² measured at 60 Hz. On the numbers alone the first looks like the efficient choice. Put the first on the same 60 Hz basis and it becomes 3.6 W/m² — and the ranking reverses.
This is not a hypothetical. One Japanese manufacturer’s product page quotes 5 W/m² for both its normal-mode and reverse-mode film; read the notes and the measurement conditions are 40 Hz for one and 60 Hz for the other. Even within a single table from a single supplier, the figures have to be put on a common basis before they mean anything.
In fairness, that footnote was there to be read. It is the manufacturer’s own technical documentation that tends to state conditions; distributor and installer literature more often carries a bare “3–5 W/m²”. Where the conditions are missing, they are worth asking for.
When you are selecting a film, four things are worth pinning down about its power figure:
1. Supply capacity and heat
The higher the frequency, the more current the same area of film draws. What makes a shortfall hard to spot is that a drive circuit running short of current does not stop — its output voltage sags instead. And above its threshold a film’s transmittance climbs with voltage and approaches full clarity — it does not snap to fully clear at one particular voltage. So a voltage below target does not fail to switch the film; it leaves it short of fully clear. And because the film changes so little across the saturated region, the shortfall stays invisible until it is fairly large. The symptom reads as a film quality problem rather than a power problem.
It also depends on how much area is being driven at that moment. Current demand scales with the total area under drive, so a supply that is comfortable one zone at a time can fall short the instant several zones are switched together. That is why this kind of fault tends to appear only on certain operations, and only after installation.
2. How much area one unit can drive
A controller has a finite output. Since power scales with frequency, the higher the frequency, the less area one unit can drive. Which also means a drivable-area figure quoted without a frequency is not a specification.
3. How large an area can be driven uniformly
The same “it is a capacitor” fact shows up in a second place. The transparent conductor has real sheet resistance, and current flowing through it means the voltage falls the further you get from the busbar. The far end of the panel sits at a different transmittance from the near end, and the panel looks uneven.
That current also rises with frequency, so the higher the frequency, the shorter the distance you can feed uniformly. The distance enters as a square root, so quadrupling the frequency roughly halves the feedable distance. The same sheet resistance is what makes zones interfere with each other — we work that through in why zones interfere in segmented PDLC film.
We offer an evaluation kit so you can check multi-zone control on the switchable film you already work with. We are happy to go through power draw and drivable area for a specific project.
Contact us