CNC Milling Clock Dials, Gear Trains, and Escapements for Mechanical Clock Making
Wooden gear clocks occupy an odd and satisfying corner of the maker world: the gear trains are large, visible, and mechanically honest in a way a quartz movement never is, and a desktop CNC router like the Wolfpawn 4040 Pro is genuinely capable of cutting the gear train and dial accurately enough to run. A full mechanical escapement is a different story, and this guide is honest about where desktop CNC tolerances stop being enough. What follows covers gear math for a time-keeping train, machining technique for clean wooden gears, dial engraving, and where to stop and buy rather than build.
Gear train basics for timekeeping
A clock's gear train exists to turn one input rotation (from a weight, spring, or in a powered build, a slow synchronous motor) into the correct relative speeds for hour, minute, and second hands. The core relationship that has to hold exactly is the minute-to-hour ratio: the minute hand must turn exactly 12 times for every single rotation of the hour hand, which in practice is built from a chain of smaller gear pairs whose combined ratio multiplies out to 12:1 — commonly split across two or three gear pairs rather than one enormous single-stage reduction, both for mechanical practicality and to keep individual gears a sane size.
- Module/pitch — defines tooth size relative to gear diameter; match module consistently across every gear that meshes with another, since two gears with different modules simply won't mesh correctly regardless of tooth count.
- Tooth profile — involute profiles (the standard in metal gearing) work in wood, but many wooden-gear-clock designers prefer cycloidal profiles, which distribute contact forces differently and tend to run more quietly and wear better in low-precision wood-on-wood meshing at the pitch diameters typical of these clocks.
- Backlash allowance — wood swells and shrinks with humidity more than metal or acrylic; build in slightly more clearance between meshing teeth than you would for a metal gear train, or the clock will bind on a humid day.
Generating gear files
Free wooden-clock gear generator tools (widely available online, aimed specifically at this hobby) will output tooth profiles for a given module, tooth count, and pressure angle as DXF or SVG, ready to import into CAM software. Fusion 360 and Vectric both also have gear-generation plugins/add-ins if you'd rather design the whole train inside your existing CAM package rather than round-tripping through a separate tool.
Machining the gear train
- Cut gears from Baltic birch plywood rather than solid wood where possible — cross-laminated plies resist the seasonal warping and grain-direction weakness that make solid wood gears less dimensionally stable over time.
- Use a small-diameter end mill (1/16" or smaller is typical) to resolve the tooth gaps on anything but the largest, lowest-tooth-count gears — a standard 1/8" bit simply can't cut into the root of most clock gear tooth profiles without gouging adjacent teeth.
- Run small cutters at higher spindle RPM and lower feed rate than you would for a typical plywood sign or box joint job; small-diameter bits are far more fragile and will snap under the same feed rate that's perfectly safe for a 1/4" bit.
- Cut gears thicker than one pass depth in multiple shallow passes rather than one deep plunge, both to protect the small bit and to get a cleaner tooth edge.
- Always cut a single test gear pair at your intended module and tooth count before committing to a full train — verify the two mesh smoothly by hand before cutting the remaining six or eight gears in that train.
Dial and face work
A clock dial is a good use for V-carving: numerals and hour marks cut as V-grooves in MDF or hardwood read cleanly with nothing more than a light wash of paint or wax rubbed into the grooves and wiped from the surface, the same technique used elsewhere on this site for V-carved signage. A two-tone laminate blank (dark core, light face or vice versa) gives high-contrast numerals without any paint at all, straight off the router.
Where CNC gearing stops and precision horology starts
Be realistic about what a router-cut wooden gear train can and can't do. The gear train itself — the part that visibly turns and looks impressive — is well within a desktop CNC router's capability. A genuine mechanical escapement (the anchor-and-pendulum or similar mechanism that actually regulates timekeeping to the second) depends on tolerances and material properties a plywood router cut cannot reliably hit: hardened steel or brass pivots, jeweled or precisely bushed bearing points, and an escapement geometry that's unforgiving of the backlash a wood gear train tolerates just fine. The practical path most successful wooden gear clock builds take is a hybrid one: CNC-cut gear train and case for the visual mechanism, paired with either a purchased clock movement/escapement module driving the display gears, or a separately sourced precision escapement kit rather than attempting to machine one from scratch on a router.
Reference gear counts for an 8-day movement
StageTypical functionApproximate ratio contribution Mainspring/weight drum to first wheelPrimary power transmissionVaries by design Center wheel trainMinute hand reference rotation1:1 (defines "one hour" per full turn convention) Minute-to-hour reductionHour hand drive12:1 total, split across 2–3 gear pairsExact tooth counts vary significantly between published wooden clock designs — use a proven, published gear train design rather than deriving ratios from scratch for a first build, and treat the table above as a conceptual map of what each stage is doing rather than a drop-in spec.
Safety notes
- Small-diameter end mills are brittle and can snap during a crash or an overly aggressive feed rate — keep eye protection on and stand slightly to the side of the cutting path when running fine detail passes.
- Test cuts at reduced feed rate before committing to a full gear train pass are cheaper than a broken bit and a ruined blank.
A CNC-cut gear train clock is a legitimately impressive shop project and a good excuse to get comfortable with small-diameter tooling and gear math, as long as the final build plan accounts honestly for where a router's capability ends and where a sourced movement or escapement needs to take over.