Every few weeks a buyer asks for it: can the levitating lamp be dimmable? Stepless, ideally. It is a fair question — every desk lamp on the market dims. But a levitating lamp is not a desk lamp, and the honest answer is: we can, we have, and we mostly recommend you don’t. Here is why.
The power budget is the whole story
A levitating lamp is powered across an air gap. The receiver in the floater lives on whatever the wireless link can deliver — and as we broke down in the lamp-power piece, that budget is small. The industry norm for a 18–20 mm gap is under 100 mW at the receiver. Our tuned link — LC optimization plus a variable-frequency pulse scheme — pushes it to about 500 mW on the products where the geometry allows.
Keep those two numbers in mind. Everything below is just arithmetic on top of them.
Three steps nobody can see
Dimming resolution is only useful if the eye can resolve the steps. And the eye is logarithmic — a 20% power change is barely a just-noticeable difference, and even a halving of power looks like far less than “half as bright”. We documented the perception side of this in our gamma-correction piece: 50% PWM reads as about 70% bright. Dimming curves can be engineered beautifully; the problem here is upstream of the curve.
Split a 500 mW budget into three levels and you get steps of roughly 165 mW each — before driver losses. At the light levels these lamps operate at, the perceived jump between adjacent steps is small enough that users hunt for the difference and half the time conclude the button did nothing.
This is not theory. We shipped it. An early moon lamp of ours had a three-step dimming mode — tap once per level. On the bench it worked exactly as designed. In customers’ hands, the feedback was a shrug: the steps were hard to tell apart, the feature didn’t photograph, didn’t demo, and didn’t sell. The non-dimming versions outsold it. We took the lesson.
The industry’s brute-force fix eats the lamp
The common workaround is to bolt on a remote-control module — an RF or IR receiver with a standby MCU, listening for commands around the clock. It works, in the sense that the lamp dims when you press the button. What the spec sheet does not say is what the receiver costs you:
| Budget item | Typical draw | On a 100 mW link |
|---|---|---|
| RF/IR receiver + MCU, always listening | 30–50 mW | 30–50% of the budget |
| Left for the LED | the rest | ~half the brightness, gone |
On a wired lamp those 40 mW are noise. On a levitating lamp they are half of everything you have. Gain the remote, lose the light — the dimming feature literally dims the lamp. We have torn down competitor units that ship exactly this trade, and you can see it from across the room: the dimmable one is the dim one.
Stepless dimming? We built that too
For completeness: yes, true stepless dimming over the power link is possible, and we have done it — an FSK scheme that rides the command bits on the power carrier itself, so the receiver decodes dimming commands without any RF front-end at all. We developed it for a Canadian retail group’s display program; it works, and the standby cost is negligible.
And it still didn’t matter. The physics problem is not the control channel — it is the perceptual one. Slide from 100% to 60% on a lamp whose absolute output is modest to begin with, and the room barely changes. Users turned the dial, squinted, and asked if anything happened. A feature nobody can perceive is not a feature, it is firmware. We shelved it as a standard offering and kept it in the toolbox for projects where the budget allows it to matter.
What we recommend instead: change the color, not the level
Within the same power constraint, there is a change the eye cannot miss: color. Switching a lamp from warm white to red to blue to rainbow is a categorical, photographable, demo-able event — no squinting required. Same watts, completely different room.
Our current color-switching link takes the FSK lesson further: commands ride the power carrier, the receiver logic draws microamps, and the switching itself is essentially zero-loss — so the lamp keeps full brightness in every color. No remote module, no standby tax, no dim lamp. The zero-power switching architecture is written up in Zero-Power Color Switching and the ultra-low-power bulb version in its companion piece.
This is now standard across our levitating bulbs and moon lamps, and it is the feature customers actually demo to their customers. If you are speccing a levitating lamp and the brief says “dimmable”, my advice after building it three different ways: take the colors instead.
Engineering notes from our bench. Power figures are typical measured values for our link geometry — for design reference, not product specifications or delivery commitments.
Related reading: Why Lamp Power Is Small · Gamma-Corrected Dimming · Zero-Power Color Switching · Ultra-Low-Power Color Switching for Bulbs.