Doubling the number of revolutions on one shaft means using a gear pair where the driven gear has half the tooth count of the driver, so the output shaft turns twice for every input revolution. This solves the practical problem of a prime mover turning slower than the tool or process needs. A small pinion meshing with a gear twice its tooth count gives a clean 1:2 step-up. The tradeoff is straightforward — you double speed and halve available torque, minus mesh losses of around 2-3% per stage.
Doubling Shaft Revolutions Interactive Calculator
Vary the driver teeth, pinion teeth, and input revolutions to see the step-up ratio, output revolutions, and torque tradeoff.
Equation Used
The driver and pinion share the same tooth pitch, so the output speed ratio is set by tooth count: Z1 divided by Z2. A 40-tooth driver turning a 20-tooth pinion gives 40/20 = 2, so one input revolution produces two output revolutions.
- Same module and tooth pitch on both gears.
- External spur gears mesh without slip.
- Speed ratio is based on tooth count only.
- Torque factor is ideal and ignores mesh losses.
How the Doubling the Number of Revolutions on One Shaft Actually Works
The mechanism is a step-up gear pair. The driver gear sits on the input shaft and carries twice the tooth count of the driven gear on the output shaft. Because both gears must share the same tooth pitch and the same module to mesh cleanly, the smaller driven gear has half the pitch diameter. Each tooth of the driver advances one tooth on the driven, so by the time the input completes one full turn the output has rotated twice. That is the entire kinematic story — tooth count ratio sets the speed ratio, full stop.
Why design it this way rather than belt it up? Spur gears give you a positive, slip-free ratio that holds under shock load. A 40-tooth driver running a 20-tooth pinion will deliver exactly 2.000 output revs per input rev, every time. A belt or chain drift under load and you cannot guarantee phase. If the application needs the output shaft to stay phased with the input — a cam timing shaft, a counter, an indexing head — you want gears.
Tolerances matter more than people expect. Centre-distance error of more than about 0.05 mm on a module-1 pair will tighten the backlash window and you will hear the mesh whining. Too loose and the teeth hammer at every torque reversal — that hammering shows up as pitting on the pressure flank within a few hundred hours. Tooth-count error is impossible (you either have 20 teeth or you don't), but profile error from a worn hob or a miscut shaper produces an output that wobbles in instantaneous velocity even though average speed is exactly 2×. If you put a tachometer on the output and see ±3% ripple at tooth-mesh frequency, that's profile error, not ratio error.
Key Components
- Driver gear: The larger gear, fixed to the input shaft. Tooth count is exactly twice the driven gear — for example 40 teeth driving 20. Bore-to-shaft fit should be H7/k6 transition or tighter; a sloppy bore lets the gear orbit on the shaft and corrupts the ratio with backlash that varies through each revolution.
- Driven (output) pinion: The smaller gear with half the tooth count. It carries the doubled output speed but only half the input torque. Face width usually matches the driver, but tooth bending stress is higher here because each tooth meshes twice as often per input revolution — so this is the gear that fails first under fatigue.
- Centre-distance fixture (housing or plate): Holds the two shaft centres at a fixed distance equal to (Ddriver + Ddriven) / 2. For a module-1 pair of 40 and 20 teeth that is 30.000 mm — and you want it within ±0.03 mm. Drift outside that and you get either binding or backlash.
- Idler gear (optional): Inserted between driver and driven when you need the output to spin the same direction as the input. An idler does not change the ratio — only tooth counts of the input and final output gears matter for the 1:2 result.
- Bearings on the output shaft: These see double the input RPM and so accumulate fatigue cycles twice as fast. A 6202 deep-groove bearing rated for 18,000 hours at the input speed only delivers 9,000 hours on the doubled output. Size accordingly.
Who Uses the Doubling the Number of Revolutions on One Shaft
Speed-doubling gear pairs show up anywhere a motor turns slower than the working tool, but the designer wants the rigidity and timing of gears rather than a belt. They are particularly common in machine-tool spindles, fan drives, instrumentation, and any timed counter or indexer where the output must stay phased to the input. The mechanism is cheap, compact, and predictable — those are the three reasons it survives in modern designs against more exotic alternatives.
- Machine tools: Speed-doubling overdrive heads on watchmaker lathes — the Schaublin 70 accessory head and similar Bergeon-style attachments use a 1:2 step-up to push a small grinding spindle from 3,000 to 6,000 RPM off a standard headstock.
- Automotive: Distributor drives on older inline engines — the camshaft turns at half crank speed, but the distributor or oil pump shaft is geared 1:2 off the cam to bring it back to crank speed.
- Industrial fans: Belt-replacement step-ups on Howden centrifugal blowers where a 1750 RPM motor needs to spin a 3500 RPM impeller and the customer specifies gear drive for sealed-environment service.
- Mechanical counters: Decade-counter drives in Veeder-Root and Hengstler totalisers, where the units wheel must turn twice for every input pulse from a half-speed cam.
- Robotics and lab gear: Centrifuge rotor drives in benchtop units like the Eppendorf 5424 family — the motor runs at moderate speed and a single gear pair brings the rotor up to the spec'd 15,000 RPM.
- Printing: Folder-cylinder drives on Heidelberg sheet-fed presses where the impression cylinder runs at half speed and a 1:2 step-up returns the gripper-bar shaft to sheet-feed cadence.
The Formula Behind the Doubling the Number of Revolutions on One Shaft
The formula is trivial in form but has real consequences across the operating range. At low input speeds — say 100 RPM on a hand-cranked apparatus — the doubled output of 200 RPM is gentle, mesh losses are negligible, and you can run the pair dry-lubed. Push the input to 3,000 RPM and the output hits 6,000 RPM, where pitch-line velocity climbs above 6 m/s on a typical module-1 pair and you need oil mist or splash lubrication or the teeth will scuff. The sweet spot for a small spur pair sits around 1,500-2,500 RPM input — fast enough to be useful, slow enough that windage and noise stay tame.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Nout | Output shaft rotational speed | rev/min (RPM) | RPM |
| Nin | Input shaft rotational speed | rev/min (RPM) | RPM |
| Zdriver | Tooth count of driver gear (the larger one for a step-up) | teeth | teeth |
| Zdriven | Tooth count of driven gear (the smaller pinion) | teeth | teeth |
| Tout | Output torque after the step-up | N·m | lb·ft |
Worked Example: Doubling the Number of Revolutions on One Shaft in a textile loom heald-frame cam shaft
You are designing the cam-shaft drive on a Picanol OmniPlus rapier loom retrofit. The main motor pulley turns the bottom shaft at a nominal 600 RPM. The heald-frame cam stack needs to run at twice that speed — 1,200 RPM — to give the dobby head time to lift each frame within the pick cycle. You decide on a single spur pair: a 48-tooth driver on the bottom shaft meshing a 24-tooth pinion on the cam shaft, module 2.0, both 20 mm face width.
Given
- Nin = 600 RPM (nominal)
- Zdriver = 48 teeth
- Zdriven = 24 teeth
- Module = 2.0 mm
- Face width = 20 mm
Solution
Step 1 — confirm the ratio from tooth counts:
Step 2 — at nominal 600 RPM bottom-shaft speed, calculate cam-shaft output:
That is exactly the design target. The dobby gets a clean 50 ms per pick to actuate, which is the working sweet spot for the OmniPlus mechanism.
Step 3 — at the low end of the loom's operating range, the bottom shaft might idle at 300 RPM during thread-up:
At 600 RPM cam speed the heald frames lift slowly enough that the operator can watch each shed open and check warp tension visually — that is exactly why thread-up runs at half speed.
Step 4 — push the loom to its top rated speed of 750 RPM bottom shaft:
1,500 RPM on the cam shaft puts pitch-line velocity at v = π × (24 × 2.0 / 1000) × 1500 / 60 ≈ 3.77 m/s. That is fine for splash-lube spur gears but you will hear the mesh frequency clearly — 600 Hz for a 24-tooth pinion at 1,500 RPM — and any centre-distance error above 0.05 mm produces audible whine.
Step 5 — torque on the cam shaft is half of bottom-shaft torque, minus mesh losses:
Result
The cam shaft turns at the nominal 1,200 RPM with the 48:24 pair, exactly the design target for the dobby timing window. Across the operating range, the cam shaft tracks from 600 RPM at thread-up speed to 1,500 RPM flat-out — the 1,200 RPM nominal sits comfortably in the middle of that band and is where the loom should spend most of its production time. If you put a tachometer on the cam shaft and read 1,180 RPM instead of 1,200, the input belt is slipping — gears don't lose ratio, belts do. If you read 1,200 RPM but the cam timing drifts by a few degrees relative to the bottom shaft over a shift, the pinion is loose on its taper or the keyway is rolled, and the gear is creeping on the shaft. If you hear a rising whine at exactly tooth-mesh frequency that wasn't there at install, check centre distance — a housing bore that has crept 0.1 mm under thermal cycling will produce that signature within the first 200 hours.
Choosing the Doubling the Number of Revolutions on One Shaft: Pros and Cons
A 1:2 spur pair is the obvious choice when you need timed, rigid speed-doubling on parallel shafts. But it is not the only option, and for some applications a belt or a planetary stage wins on cost, noise, or packaging. Compare on the dimensions that actually matter to the build.
| Property | 1:2 spur gear pair | 1:2 timing belt drive | 1:2 planetary stage |
|---|---|---|---|
| Speed ratio precision | Exact, slip-free, ±0 over life | ±0.5% short-term, drifts as belt stretches | Exact, slip-free |
| Practical speed limit | ~6,000 RPM output before lubrication becomes critical | ~10,000 RPM output, limited by belt centrifugal stress | ~15,000 RPM output with proper bearing selection |
| Efficiency per stage | 97-98% | 95-98% when new, drops as belt ages | 95-97% |
| Cost (small batch, module-2 pair) | $30-80 for the pair | $15-40 belt + 2 pulleys | $120-300 for a packaged unit |
| Service life under continuous duty | 20,000+ hours with adequate lube | 3,000-8,000 hours, belt is wear item | 10,000-20,000 hours |
| Noise at 1,200 RPM output | 65-75 dBA, tonal at mesh frequency | 55-65 dBA, broadband | 70-80 dBA, multi-tonal |
| Phase preservation under shock | Perfect — teeth cannot slip | Belt jumps teeth above ~3× rated torque | Perfect — no slip path |
| Packaging on parallel shafts | Compact, requires precise centre distance | Tolerant of centre-distance drift | Coaxial only, not parallel-shaft |
Frequently Asked Questions About Doubling the Number of Revolutions on One Shaft
If the long-term average ratio drifts off 2.000, the gears themselves are not the problem — tooth counts are integers and cannot lie. What you are seeing is encoder slip on one of the shafts, almost always the input. A grub-screw-mounted encoder hub creeps under cyclic torque reversal, and over a few thousand revolutions you accumulate a small angular offset that shows up as a fractional ratio error.
Switch to a clamping or taper-bushed encoder hub and the measurement will lock onto 2.000 exactly. If the error persists with a clamped hub, your encoder counts per rev divides unevenly into your sample window — that is a sampling artefact, not a real ratio error.
Yes — any tooth-count pair where the driver is exactly twice the driven gives the same 1:2 ratio. 50:25, 60:30, 80:40 all produce the identical kinematic result. What changes is the centre distance and the tooth bending stress.
Bigger pairs sit further apart and have larger teeth, so they handle more torque but eat more space. For a given module, doubling tooth counts roughly doubles centre distance. Check that your housing accommodates the new C/L before you commit. Also watch hunting tooth wear — a 50:25 ratio means tooth #1 of the pinion meshes the same two driver teeth every revolution, accelerating localised wear; a 49:24 pair (not exactly 2:1, but close) would distribute wear more evenly. For a true 2:1 you accept the hunting penalty.
For a single 1:2 step-up at 6,000 RPM output, a spur pair wins on cost, simplicity, and serviceability. Planetary stages earn their cost when you need higher ratios (>4:1) in a coaxial package or when you need balanced radial loads on the output shaft.
The decision flips if shaft alignment matters — planetary is coaxial, spur is offset. If your driven equipment must sit on the same axis as the motor (a centrifuge rotor, a coaxial spindle), planetary is the only option. If you have any freedom to offset the output by 30-50 mm, spur is cheaper, quieter, and easier to inspect.
You are seeing tooth-mesh vibration, and 24 is the tooth count of your pinion. Each tooth engagement is a small stiffness step — the mesh stiffness rises and falls as one pair, then two pairs, then one pair share the load. That excitation hits the output shaft at Zdriven × Nout Hz.
If the amplitude is small (under 1% of nominal speed) it is normal and unavoidable with spur gears — the cure is helical gears, which keep contact ratio above 2 and smooth out the stiffness ripple. If the amplitude is larger than that, check tooth profile error first, then centre distance. A profile error of 10 µm on a module-2 tooth produces visibly larger ripple than a clean-cut tooth.
Mechanically yes, the gears do not care which shaft drives. But the bearings, lubrication, and seals were sized for the original direction of energy flow, and reversing it changes which gear sees high-speed loading.
The bigger issue is efficiency at light load. A step-up runs efficiently because the driven (smaller) gear gets the full transmitted force on a small pitch radius — high tangential force, normal sliding velocity. Run the same pair as a reduction and the now-output big gear sees lower tangential force at higher pitch radius, which is fine, but at very light loads the static friction and seal drag eat a larger fraction of input power. Below ~10% rated load, efficiency drops from 97% to as low as 80%. For a one-off direction reversal it is fine; for continuous duty in the new direction, re-spec the bearings.
For module-2 spur gears at this speed, target 0.08-0.15 mm of total backlash measured at the pitch line. Tighter than 0.05 mm and the pair binds when the housing warms up — gear teeth grow with temperature and the steel housing grows less, so you lose clearance. Looser than 0.20 mm and you get audible clatter on every torque reversal, plus accelerated pitting on the loaded flank.
Set centre distance such that nominal backlash falls in that window at 20 °C ambient and recheck after a warm-up run. If the pair runs in oil bath at 60-70 °C steady state, dial the cold backlash toward the upper end of the range to leave room for thermal growth.
References & Further Reading
- Wikipedia contributors. Gear train. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
