Gear Ratio Calculator (Multi-Stage)
Gear ratio calculator. Single, two-stage, and three-stage compound gear trains. Output speed, torque multiplication, mechanical advantage, overall efficiency.
Gear Ratio
| Output speed | — |
| Output torque | — |
| Mechanical advantage | — |
| Overall efficiency | — |
| Output direction vs input | — |
How to use the gear ratio calculator
Choose number of stages
1 stage: simple gear pair. 2 stages: two pairs in series — total ratio = product of stage ratios. 3 stages: gearbox-style. Multi-stage allows higher total ratios with reasonable individual stage ratios — design texts put the practical ceiling for a single external spur pair at about 10:1 before the pinion becomes impractically small against the gear (Shigley, Ch. 13, gear trains).
Enter input speed + torque
Input speed in rpm. The 1750 rpm default is a 4-pole induction motor on a 60 Hz supply (synchronous 1800 rpm less slip); on the 50 Hz networks of Singapore, Malaysia, Indonesia, Thailand and Vietnam the same motor runs at about 1440–1470 rpm, so enter the nameplate speed. Input torque in N·m. Power = torque × angular speed, so a gear train trades speed for torque at constant power less the efficiency losses.
Set stage teeth counts
Driver teeth = the gear on the input side of the stage. Driven teeth = the gear on its output side. Ratio per stage = driven/driver. The inputs accept 6 teeth and up, but a standard 20° full-depth pinion undercuts below 17–18 teeth when it meshes with a much larger gear (the interference limit derived in Shigley, Ch. 13), so real spur pinions rarely go below that.
Set efficiency
Per-stage efficiency is the fraction of input power that survives one mesh. Well-lubricated spur and helical meshes lose only a few per cent; a worm set loses far more, anywhere from over half to about a tenth of its input depending on lead angle and friction coefficient (Shigley gives the worm efficiency equation in Ch. 13). Total efficiency = (stage efficiency)^stages, so a 3-stage train at 95% per stage is 0.95³ ≈ 86% overall. The 95% default is deliberately conservative for spur gearing.
Read output speed + torque
Output speed = input / total ratio. Output torque = input × total ratio × overall efficiency. A 4:1 reduction at 95% efficiency turns 1750 rpm + 10 N·m → 437 rpm + 38 N·m. Higher ratio = lower speed + higher torque. Useful for motor-to-load matching.
Gear ratios — trading speed for torque
Gear ratios are how mechanical engineering matches motor speed to load requirements. A typical AC induction motor runs at 1750 rpm — far too fast for most applications. A gearbox reduces speed and multiplies torque proportionally: a 10:1 reduction takes 1750 rpm at 10 N·m to 175 rpm at ~95 N·m (after efficiency losses). The fundamental relationship: power is conserved (minus losses), speed and torque trade off. Multi-stage and strain-wave gearboxes enable the large reductions that robot joints and indexing drives need; a bicycle's chainring and sprocket are a single stage that steps up rather than down, so a 50-tooth chainring driving a 12-tooth sprocket gives 4.17 wheel turns per pedal turn.
Mechanical advantage
The mechanical advantage this page reports is the ideal one: it equals the overall ratio, and the torque line applies the efficiency on top of it. A 4:1 reduction provides 4× ideal mechanical advantage — for static loads (holding a weight) and dynamic ones (accelerating a mass) alike — at the price of 4× slower output. Worm drives reach high ratios in one stage but pay for it in sliding friction, which is why their efficiency depends so strongly on lead angle; spur, helical and bevel meshes roll rather than slide and lose only a few per cent each.
Gear ratios are the universal mechanical translator. Speed-fast motor + slow load? Reduce. Slow input + fast output needed? Step up. Power is conserved (minus losses); speed and torque just trade.
Compound vs simple
Simple gearing is one stage, and design texts stop it at about 10:1 — beyond that the gear dwarfs the pinion and the pinion's tooth count falls toward the interference limit. Compound (multi-stage) gearing achieves higher overall ratios with sensible individual gears: the ratios multiply, so three 5:1 stages give 125:1. The drawback: each stage adds a mesh loss, adds its own backlash, and needs more bearings and housing. Planetary arrangements pack a high ratio into a compact coaxial volume, which is why automotive automatic transmissions are built from them. This page models external spur-type meshes only; a planetary set has its own ratio formula.
Gearmotors on a 50 Hz grid
A gearmotor's input speed is set by the mains frequency and the motor's pole count: synchronous speed = 120 × frequency ÷ poles. Most of ASEAN — Singapore, Malaysia, Indonesia, Thailand and Vietnam — runs on 50 Hz, so a 4-pole induction motor turns at a synchronous 1500 rpm and about 1440–1470 rpm under load; the Philippines, like North America, is 60 Hz, where the same motor gives the 1750 rpm this page defaults to. Reading a reducer's ratio off a 60 Hz catalogue for a 50 Hz site therefore under-states the output speed by a sixth — enter the nameplate speed rather than the default.
Teeth ratios, compounding stages and the 10:1 single-mesh ceiling
Ratio = N_driven / N_driver. Output speed = Input / Ratio.
Output torque = Input × Ratio × efficiency. Speed↓ → Torque↑.
Spur and helical meshes roll and lose a few per cent; worm sets slide and their efficiency depends on lead angle.
Single external pair: about 10:1 is the textbook ceiling. Higher needs compound stages.
Planetary gears: compact high ratios. Used in automotive automatics.
Each external mesh reverses rotation. Even stages: same direction; odd stages: reversed.
Interference: a 20° full-depth pinion needs ≥ 18 teeth against a rack, ≥ 13 against an equal gear.
Bicycle: 0.5:1 to 4:1 to match cadence (~80-100 rpm) with desired wheel speed.
Power = Torque × angular speed. Gearing trades speed for torque at conserved power (minus losses).
Backlash: free play between gear teeth. Critical for precision positioning + servo control.
Frequently asked questions
Design texts put the practical ceiling for a single external spur pair at about 10:1: beyond that the gear becomes very large relative to the pinion, and shrinking the pinion runs into the interference limit (17–18 teeth for a standard 20° pinion against a much larger gear). For higher overall ratios, use multiple stages — the ratios multiply, so a 3-stage gearbox of 5:1 per stage is 125:1 — at the cost of one mesh loss and one backlash contribution per stage.
A worm set gives a large ratio in one compact stage and, at low lead angles, will not back-drive. Its efficiency is set by the lead angle and the friction coefficient between worm and wheel (Shigley, Ch. 13), and can be well under half at low lead angles — the lost power appears as heat. A multi-stage spur or helical train is bulkier but loses only a few per cent per mesh. Enter the worm stage's own efficiency here rather than the spur default if you are modelling one.
Backlash = free angular play between gear teeth. Cumulative through multi-stage. For positioning applications (CNC, robotics): use anti-backlash gears or strain-wave (harmonic) gears with zero backlash. For typical power transmission: backlash is fine.
Helical: angled teeth. Smoother, quieter, higher load capacity. But generates axial thrust requiring thrust bearings. Spur: straight teeth, simple, no axial loads. For high speed or high power: helical. For simple machinery: spur.
For a 100 kW industrial drive at 95% efficiency: 5 kW dissipated as heat. Needs oil cooling or housing thermal management. Higher efficiency = lower running cost over years. 1% efficiency on 100 kW = 1 kW = 24 kWh/day saved.
Two external gears in mesh turn in opposite directions, so each stage flips the rotation: one or three stages reverse the output relative to the input, two stages restore it. The page assumes every stage is an external spur-type mesh. An idler gear, an internal (ring) gear or a planetary set changes the direction rule without changing the ratio, so if your train contains one, read the direction line accordingly.
Ratio = front chainring teeth / rear sprocket teeth. 50/12 = 4.17:1 (high gear, fast). 28/32 = 0.875:1 (climbing gear, slow). Combined with wheel circumference for distance per pedal rotation. Used by cyclists for cadence + power optimization.
Yes — for pulleys: ratio = driven diameter / driver diameter, with the same speed and torque relationships, so you can enter the two diameters in the teeth boxes. Two differences: a belt on two pulleys does not reverse direction (the direction line here assumes gears), and a belt creeps and can slip under overload, so its real ratio is slightly lower than the diameter ratio and its efficiency a little below a gear mesh.
A compact reducer built from an elliptical wave generator, a thin flexible spline and a rigid circular spline; the ratio comes from the two-tooth difference between the splines, which is why one stage reaches ratios well beyond anything a spur pair can. It has near-zero backlash, which is what robot joints and precision positioners buy it for, and an efficiency below a spur mesh that falls further at low speed and light load. Its ratio does not follow the driven/driver rule this page uses.
Budynas and Nisbett, Shigley's Mechanical Engineering Design (11th ed., 2020), Chapter 13 — gear trains and the train value, interference and minimum tooth numbers, worm efficiency. Norton, Design of Machinery (6th ed., 2020), Chapter 9, for compound and planetary trains. ANSI/AGMA 1012-G05 for gear nomenclature and ISO 1328-1:2013 for accuracy grades.
Method & sources
How it computes
Gear-train value for one to three external meshes: overall ratio = Π(driven teeth / driver teeth); output speed = input rpm / ratio; output torque = input torque × ratio × η^n, with one per-stage efficiency η applied n times; ideal mechanical advantage = ratio; direction reverses at each external mesh, so an odd stage count reverses the output. Pulleys may be entered by diameter in place of teeth. Planetary, internal-gear and worm geometry are not modelled.
What this tool implements
- Train value e = product of (driving teeth / driven teeth) as defined in Shigley Ch. 13 and ANSI/AGMA 1012-G05; this page reports its reciprocal as the reduction ratio.
- Efficiency compounds multiplicatively (η^stages) with a single per-stage figure; the 95 % default is a conservative spur/helical value and must be replaced for a worm stage, whose efficiency is a function of lead angle and friction coefficient.
- Direction rule assumes external spur-type meshes only — no idlers, internal (ring) gears, planetary sets or belts.
- Ratio guidance quoted on the page — about 10:1 per external pair; 18-tooth minimum for a 20° full-depth pinion against a rack and 13 against an equal gear — is textbook interference theory (Shigley Ch. 13) and is not enforced by the calculator, which accepts 6 teeth.
Sources
- Budynas RG, Nisbett JK. Shigley's Mechanical Engineering Design. 11th ed. New York: McGraw-Hill Education; 2020. Chapter 13, Gears — General: interference and minimum tooth numbers, worm-gear efficiency, gear trains and the train value.
- Norton RL. Design of Machinery: An Introduction to the Synthesis and Analysis of Mechanisms and Machines. 6th ed. New York: McGraw-Hill Education; 2020. Chapter 9, Gear trains — simple, compound, reverted and planetary trains.
- ANSI/AGMA 1012-G05. Gear Nomenclature, Definitions of Terms with Symbols. Alexandria, VA: American Gear Manufacturers Association; 2005.
- Chapman SJ. Electric Machinery Fundamentals. 5th ed. New York: McGraw-Hill; 2012. Chapter 6, Induction motors — synchronous speed n_sync = 120 f/P and slip, behind the 1750 rpm (60 Hz, 4-pole) default and the 1500 rpm 50 Hz equivalent quoted on the page.
What can make this go out of date
- None at runtime — ratios follow from tooth counts. The 1750 rpm default is a 60 Hz, 4-pole motor figure; on the 50 Hz networks of Singapore, Malaysia, Indonesia, Thailand and Vietnam the same motor runs at about 1440–1470 rpm, and the page now tells the reader to enter the nameplate speed.
Pick up where you left off
Stored only in this browser — never sent to our servers.