Gear ratio: trading speed for strength (like bicycle gears)

You feel gear ratio before you know the term. A bicycle suddenly becomes easy to climb when you shift into a big rear sprocket. A hand winch turns slowly but lifts a heavy load without drama. A small electric motor can stir thick slurry only because a gearbox reduces speed and multiplies torque before the motion reaches the shaft. All of those machines are making the same bargain: give up output speed, gain output turning force.

Gear ratio is the size relationship between a driving member and a driven member, usually expressed through tooth counts, pitch diameters, or rotational speeds. For ordinary gears with the same module, tooth count is the most convenient measure because diameter grows directly with the number of teeth. If the driven gear has more teeth than the driver, the output turns more slowly and usually produces more torque. If the driven gear has fewer teeth, the output turns faster but torque falls.

The beginner mistake is to treat ratio as a mysterious gearbox label. In reality, ratio is just conservation of power wearing mechanical clothes. Neglecting losses, power in equals power out. Since P = T x omega, a lower output angular speed must be balanced by a higher output torque, and a higher output speed must be balanced by a lower output torque.

What ratio really means

For a simple gear pair, engineers often define the reduction ratio as:

i = Z_driven / Z_driver = N_driver / N_driven

Here Z is tooth count and N is rotational speed. If a 20-tooth pinion drives a 60-tooth gear, the ratio is i = 60 / 20 = 3. That means the output speed is one-third of the input speed. The trade is not free. The output torque rises by about the same factor, after allowing for efficiency:

T_out ~= T_in x i x eta

where eta is efficiency written as a decimal. A good spur-gear mesh may be very efficient, but it is never perfect. Friction, churning, seal drag, and bearing losses consume some power.

Bicycles use the same physics but people often quote the ratio the other way round, as front chainring teeth divided by rear sprocket teeth. That convention tells a rider how many rear-wheel turns are produced per crank turn, ignoring wheel size and losses. So a cyclist talking about a "4.0 ratio" is describing a fast gear, while a gearbox designer talking about a "4:1 ratio" usually means a speed reduction. The machine is the same. Only the reporting convention changed.

Figure 1: A small pinion driving a larger gear is a classic reduction stage. The large gear turns more slowly, but it can deliver more torque at its shaft.

The governing physics in plain language

Imagine the tooth force acting at the pitch circle, the imaginary circle where the pair behaves as if it were rolling. The tooth force is nearly the same for both gears at the point of contact because action and reaction must balance. What changes is the lever arm. The larger gear has a larger pitch radius, so the same tangential force produces a larger torque:

T = F_t x r

If radius triples, torque triples for the same tangential force. But that larger pitch circle also has farther to travel for one revolution, so the large gear cannot spin as many revolutions per minute as the small one while the teeth stay meshed. That is why ratio is best understood as geometry plus conservation of power, not as a magic force generator.

A gear train does not create power. It repackages the same power into a different combination of speed and torque.

For chain drives and sprockets, the same idea appears through chain speed. The chain carries a nearly common tension from one sprocket to the other. A larger driving sprocket pulls more chain per revolution, so it turns the driven sprocket more times per crank revolution when the driven sprocket is smaller. That gives speed. Put the chain on a larger rear sprocket instead, and the rear wheel receives more torque for each unit of chain tension difference, but wheel speed falls for the same cadence.

Worked example 1: why bicycle climbing gears feel easier

A rider can comfortably maintain 90 rpm cadence with about 32 N.m average crank torque during a short effort. Compare two rear-sprocket choices with the same 36-tooth front chainring and a 2.10 m wheel circumference.

Case A, cruising gear: 36/12

Case B, climbing gear: 36/30

For a chain drive, rear-wheel speed relative to crank speed is approximately:

N_wheel = N_crank x Z_front / Z_rear

In Case A:

N_wheel = 90 x 36 / 12 = 270 rpm

Road speed is:

v = circumference x rpm x 60 / 1000

v = 2.10 x 270 x 60 / 1000 = 34.0 km/h

The ideal torque multiplication from crank to rear sprocket is the inverse of the speed multiplication:

T_rear / T_crank ~= Z_rear / Z_front = 12 / 36 = 0.333

So the rear sprocket torque before wheel-radius effects is roughly:

T_rear ~= 32 x 0.333 = 10.7 N.m

Now Case B:

N_wheel = 90 x 36 / 30 = 108 rpm

v = 2.10 x 108 x 60 / 1000 = 13.6 km/h

And the torque ratio is:

T_rear / T_crank ~= 30 / 36 = 0.833

T_rear ~= 32 x 0.833 = 26.7 N.m

The climbing gear cuts road speed to about forty percent of the cruising gear at the same cadence, but it more than doubles the torque delivered to the rear sprocket. That is why the rider can keep pedaling smoothly up a hill instead of standing on the pedals and stalling. The gear did not make the rider stronger. It changed the bargain between speed and force.

Figure 2: Bicycle gearing makes the speed versus torque trade visible. A smaller front chainring or larger rear sprocket reduces wheel speed for the same cadence but makes climbing easier.

Worked example 2: a reducer for a mixer shaft

A small industrial mixer uses a motor running at 1440 rpm and delivering 2.8 kW. The first reduction stage is a 24-tooth pinion driving a 72-tooth gear. Assume stage efficiency eta = 0.97. Find output speed and output torque for that stage.

First calculate the reduction ratio:

i = Z_driven / Z_driver = 72 / 24 = 3

Output speed is:

N_out = N_in / i = 1440 / 3 = 480 rpm

Now compute input torque from power:

T_in = P x 60 / (2 x pi x N)

T_in = 2800 x 60 / (2 x pi x 1440) = 18.6 N.m

Ideal output torque would be:

T_out,ideal = 18.6 x 3 = 55.8 N.m

Including efficiency:

T_out = 55.8 x 0.97 = 54.1 N.m

The shaft now turns at one-third the motor speed while delivering almost three times the torque. That is a useful trade for a mixer because the blades need turning force against thick fluid more than they need raw rpm.

Worked example 3: why a hand winch feels slow but powerful

A hand winch has a small 16-tooth pinion on the crank shaft driving a 64-tooth gear on the cable drum. The operator turns the crank at 40 rpm and applies about 12 N.m at the handle. Assume mechanical efficiency of the gear stage and bearings combined is 90%. Estimate drum speed and drum torque.

The ratio is:

i = 64 / 16 = 4

So drum speed is:

N_drum = 40 / 4 = 10 rpm

Output torque is:

T_drum = 12 x 4 x 0.90 = 43.2 N.m

The drum turns only one quarter as fast as the handle, but torque rises from 12 N.m to about 43 N.m. If the drum radius is small enough, that extra torque becomes a useful line pull. The operator experiences the result as "slow but manageable," which is exactly what a lifting device is supposed to feel like.

Figure 3: A geared winch deliberately sacrifices drum speed so a modest hand torque can lift a heavier load through a larger drum torque.

Assumptions and where simple ratio math stops helping

Ratio equations assume rigid parts, correct tooth geometry, and steady load. Real machines bend, heat up, vibrate, and wear. If shafts deflect, one side of a gear face may carry more load than the simple ratio math suggests. If backlash is too small, thermal growth can make the mesh noisy and hot. If efficiency is poor because lubrication is wrong, the expected torque rise will not actually reach the output shaft.

Simple ratio math also does not tell you whether the teeth are strong enough, whether the bearings can carry the radial load, or whether the housing is stiff enough to hold alignment. Ratio chooses the motion trade. Strength design still has to check bending stress, contact stress, scuffing risk, shaft shear, key capacity, and bearing life.

Common failure modes and what a bad ratio choice looks like

• Motor overload: the chosen ratio is too fast, so the motor cannot produce enough output torque and current rises sharply.
• Stalling on hills or at startup: common in vehicles and conveyors when the ratio favors speed more than launch torque.
• Overheating gearbox: a very large reduction in too few stages can raise sliding losses, tooth load, and oil churning.
• Tooth pitting or bending fatigue: the ratio may be kinematically correct but still produces tooth forces beyond what the selected module and face width can survive.
• Unusable machine speed: the ratio gives great torque, but the output is so slow that productivity collapses.
• Poor controllability: too much ratio spread between available gears can make a machine jump from "too heavy" to "too spinny" with no comfortable middle ground.

Practical rules of thumb

• When the machine struggles to start under load, look first for more reduction before assuming you need a bigger motor.
• For a single spur-gear stage, very large reductions are possible but usually become easier to package and lubricate when split across multiple stages.
• On bicycles, lower gears are for keeping cadence alive under high resistance, not for making the rider produce more power.
• Choose ratio together with wheel, drum, or pulley size. Output torque alone means little until it acts through a radius to create usable force.
• If the ratio is fixed, the real tuning knobs left are motor speed, tooth count combination, and the diameter of the part receiving the torque.
• Always check whether the ratio convention being quoted is a reduction ratio, a speed ratio, or a front-to-rear tooth-count ratio. Teams waste time when those are mixed carelessly.

How standards treat ratio decisions

Standards usually do not tell you one perfect ratio. They give the framework for making a defensible choice. ISO 21771 organizes gear terminology and geometry so tooth counts, pitch circles, and reference quantities are discussed consistently. ISO 6336 and comparable AGMA methods are then used after ratio selection to check whether the chosen gears survive bending and contact stress. ISO 1328 covers accuracy grades, which matter because a mathematically correct ratio can still run badly if pitch and profile errors are poor.

In vehicle or lifting applications, the ratio decision is also influenced by duty, allowable startup load, controllability, and safety margins in the rest of the system. Standards push engineers to translate "this feels about right" into a documented combination of speed requirement, torque requirement, service factor, and tooth-strength verification.

Engineering judgment: ratio is a design choice, not a number to memorize

Good engineers do not ask only, "What ratio fits?" They ask, "What must the machine feel like at the output?" A conveyor should start cleanly without jerking. A bicycle should let the rider hold cadence across hills and flats. A mixer should not stall in thick product. A hoist should lift confidently without becoming dangerously fast. Ratio is how you tune that behavior.

If you remember one thing, remember this: gear ratio is the language engineers use to trade speed for torque on purpose. Once you see that bargain clearly, gearboxes, bicycles, winches, and reducers all stop looking like separate subjects. For a related primer, read Gears: wheels with teeth that never slip or continue through the EnggTools articles library.