The Moon Learned to Glow Through Two Millimetres of PLA

It’s 11:47 AM and I’m writing this from the kitchen table instead of the office because I couldn’t wait. The telescope is still pointing out the back window from last night’s imaging run, the laptop has seventeen browser tabs open about libration, and there’s a test print of a single lunar disc sitting on the windowsill catching the March light.

Hold it up to the window. Look at it.

You already know about lithophanes — you figured out the pipeline back in January with the RF waterfall prints, freezing an hour of the 2-metre band into plastic ridges that glow when backlit. Same principle applies here: thickness encodes brightness, thin areas transmit light, thick areas block it. You’ve done this before.

But the moon is different, and I want you to remember why.

A spectrogram has clean horizontal structure — time flows down, frequency spreads across, signals burn bright against the noise floor. The moon is a sphere that won’t stay still. I spent an hour last night learning the word libration, which comes from the Latin for balance scale — the same root as Libra — and describes how the moon appears to tip and nod over the course of a month. Its elliptical orbit, its axial tilt, the changing perspective as Earth rotates beneath it. All of this means that we see about 59% of the lunar surface over time, not the 50% you’d expect from strict tidal locking.

Every night’s moon is geometrically unique. Not just a different phase, but a different angle, a different wobble, a slightly different slice of Mare Orientale peeking around the limb. NASA publishes this data hourly — sub-Earth longitude, sub-Earth latitude, position angle — and you can pull it as JSON from the Goddard Space Flight Center. The specificity is dizzying. There’s something wonderful about the precision: at 22:00 UTC on March 14th, 2026, the moon’s sub-Earth point was at exactly this longitude, this latitude. I photographed that exact moon. Tomorrow’s will be different.

The first print — that disc on the windowsill — already taught me something. I shot the moon last night at 29% illumination, waning crescent, terminator cutting cleanly across Mare Crisium and the eastern limb. In the photograph, the craters look sharp. In the lithophane, they vanished into a uniform grey mush.

The problem is albedo. Lunar reflectivity is brutally narrow: the maria average 7%, the highlands maybe 18%. That’s the entire range. A linear height-map compresses this into the middle of the thickness gradient, and everything washes out. I spent an hour adjusting gamma correction until Tycho’s ray system finally emerged, white streaks radiating across the southern highlands, visible at last because I stretched the contrast hard enough to separate a few percentage points of reflectivity into distinguishable plastic thickness.

The terminator is where the detail lives. Full moon photographs look flat because the sun is directly behind the camera, no shadows, everything evenly lit. But at quarter phase, the terminator throws long shadows across every crater rim, every mountain range, every boulder field. The 3D relief of the surface becomes visible through shadowing. When you convert that to a lithophane height map, you’re encoding shadow depth as plastic thickness. The craters pop. The maria sink. Tycho glows.

Here’s what I want you to know, future me: this is going to become a calendar.

Twelve strips, one per month, each strip holding thirty-odd lunar discs arranged left to right. Clear nights only — some months will have gaps, blank spaces where clouds or exhaustion intervened. Each disc printed from that night’s photograph, corrected for libration, capturing the specific geometry of that specific sky. Hang them in a row. Backlight them. Watch the year’s celestial rhythm emerge in transmitted light.

I don’t know yet how to handle the ragged edge of missed nights. I don’t know if I’ll maintain the discipline to image every clear evening through November. The printer is already humming on a test strip, seven phases at 0.12mm layer height, and I’m watching it stack Tycho crater at 40mm/s, the nozzle tracing curves it doesn’t understand.

The moon tonight will be slightly thinner, slightly different in angle, slightly wobbled on its invisible axis. I want to catch it.

— slepp March 14, 2026, 11:47 AM