Gear Trains, Compound Reducers, and Efficiency — aka “How to Stack Gears Without Accidentally Building a Space Heater”

If cams were our choreography and mobility was the dance floor, gear trains are the hype squad that multiply your moves and sometimes change the song's tempo.

---

Why This Matters (And Why Your Motor Is Screaming)

In Power Transmission, we picked the parts: belts, chains, gears, bearings, springs, brakes, clutches — the whole Avengers cast. Then in mechanisms, we learned how links behave and how cams push followers without emotional support. Now we combine that kinematic wisdom with power flow: chaining gears into trains that hit the exact speed, torque, and direction we want — while staying efficient, quiet-ish, and not melting.

This session: simple and compound gear trains, two- and three-stage reducers, planetary teasers, and the harsh reality of efficiency. By the end, you’ll know how to:

• Compute overall ratios like a legend
• Distribute ratios across stages to shrink your gearbox
• Predict losses so your design doesn’t become an oven
• Avoid the classic “oops, it spins backwards” moment

---

1) Kinematics 101: Gears Gossip, But Math Keeps Receipts

The Core Ratio Truths

• For an external mesh of two gears (pinion p driving gear g):
  • Speed ratio: i_pg = ω_p / ω_g = N_g / N_p
  • Direction flips once for each external mesh (parity matters!)
• Overall train value (input to output):
  • i_total = (ω_in / ω_out) = Π (N_driven / N_driver) × sign
  • sign = (-1)^(# of external meshes); internal meshes do not flip direction.

Idlers are like middle managers: they don’t change the final decision (ratio), but they add meetings (losses) and sometimes reverse the vibe (direction if count is odd).

Simple vs. Compound vs. Reverted

• Simple gear train: one gear per shaft. Ratios typically limited (≤ ~10:1 total) and long center distances for big ratios.
• Compound train: multiple gears share shafts (co-axially locked). Big ratios achievable in compact space.
• Reverted train: input and output are coaxial. Geometry condition: sums of the mating tooth counts match stage-to-stage so center distances align.
  • Condition for 2-stage reverted: (N1 + N2) = (N3 + N4) with 1–2 and 3–4 as the two meshes.

Planetary (Epicyclic) Cameo

The epicyclic is the Swiss Army gearbox: coaxial, compact, high ratio per stage, and delightfully non-intuitive until it isn’t.

• Willis’ formula for a simple planetary (sun s, ring r, carrier c):
  • (ω_s − ω_c) / (ω_r − ω_c) = − Z_r / Z_s
• Example: Ring fixed (ω_r=0), input sun, output carrier:
  • ω_s / ω_c = 1 + Z_r / Z_s → Cute reduction in one tidy package.

---

2) Designing Compound Reducers Without Crying

We often need big drops (e.g., 1800 rpm to 50 rpm). Single-stage spur ratios beyond ~10:1 get awkward: tiny pinions, undercut, huge gears, noisy meshes.

Ratio Distribution Rule of Thumb

• For n stages with total ratio R, pick roughly equal per-stage ratios: r ≈ R^(1/n).
• Why? Balances torque growth, keeps pinions robust, and evens pitch line velocities (noise control).

Worked Mini-Example: Two-Stage, 36:1 Spur Reducer

• Goal: 1 kW motor at 1800 rpm → ~50 rpm output (36:1 as demo). Target quiet-ish.
• Choose two equal stages: each ≈ 6:1.
• Teeth choices (20° PA, avoid undercut — keep pinions ≥ 18 teeth):
  • Stage 1: N1:N2 = 20:120 (i1=6)
  • Stage 2: N3:N4 = 24:144 (i2=6)
  • Overall: i_total = 36

Sizing sprinkles (pick module later):

• Input torque: T_in ≈ 9550·P(kW)/n(rpm) = 9550·1/1800 ≈ 5.3 N·m
• Assume each stage efficiency η_stage ≈ 0.96 (mesh + bearings)
• Output torque: T_out ≈ T_in · R · η_total = 5.3 · 36 · (0.96^2) ≈ 5.3 · 36 · 0.922 ≈ 176 N·m

Tooth forces escalate downstream — design your shafts and bearings for the later stages:

• Assume module m = 2 mm → pitch diameters: d1=40 mm, d3=48 mm
• Tangential force at stage 1 mesh (pinion): F_t1 = 2·T_1 / d1 ≈ 2·5.3 / 0.04 ≈ 265 N
• Torque after stage 1: T_shaft2 ≈ 5.3 · 6 · 0.96 ≈ 30.5 N·m
• Stage 2 pinion tangential force: F_t2 = 2·30.5 / 0.048 ≈ 1271 N (big jump!)

Pitch line velocity check (noise loves speed):

• v = π·d·n / 60. Keep spur meshes ideally under ~10–15 m/s for civilian eardrums.

Pro move: equal ratios per stage keep tooth sizes moderate and mesh speeds similar, which means lower noise and saner bearing loads.

---

3) Efficiency: The Myth of Free Torque Ends Here

Every mesh shears off a bit of your soul and your watts.

Typical Per-Mesh/Per-Stage Numbers

• Spur/Helical mesh: 0.97–0.99 (helical often slightly better under load due to smoother contact)
• Bearings per shaft pair: 0.99-ish (depends on preload, seals, speed)
• Worm gears: 0.40–0.90 (lead angle is destiny)
• Hypoid/Bevel: 0.90–0.97
• Planetary stage: per mesh ~0.98, with multiple meshes in parallel → stage ~0.95–0.98

Total efficiency multiplies:

• η_total = Π η_stage (include meshes and bearing groups)
• Idler gears don’t change ratio, but they add extra meshes and bearing losses: more idlers → lower η_total.

Heat Is Coming

Power lost becomes heat:

• P_loss = P_in · (1 − η_total)
• Your housing must dump this heat: fins, oil splash vs. forced lube, maybe a fan. If it’s too hot to touch, your grease is writing a resignation letter.

Bonus: Worm Magic (and Mischief)

• Small lead angles can be self-locking: good for holding loads; bad for efficiency.
• If you need backdrivability (robot arms), avoid extremely low lead angles.

---

4) Dynamics and Vibrations: When Ratios Meet Reality

You met motion profiles with cams; now gears add periodic stiffness and transmission error (TE). Translation: even perfect gears sing at the mesh frequency.

• Mesh frequency: f_m = (N_teeth) · (rpm/60) for each mesh. Multiply by harmonics if TE is spicy.
• Compound trains stack excitations from all stages — avoid placing structural resonances near mesh harmonics.
• Backlash: necessary for lubrication and thermal growth, but too much → rattle; too little → heat and scuffing.
• Helicals reduce noise via overlap ratio; beware axial thrust (use paired helicals or bearings that lift).
• Keep pitch line velocities balanced across stages to tame noise and wear.

---

5) Quick Compare: What Train To Take?

Type	Packaging	Typical Ratio/Stage	Direction	Notes
Simple (with idlers)	Long	≤ ~10:1 total	Flips per external mesh	Cheap, easy, low η if many idlers
Compound (spur/helical)	Compact	4–8:1 per stage	Flips per external mesh	Workhorse of reducers
Reverted compound	Coaxial in/out	4–8:1 per stage	As above	Gearbox-friendly geometry
Planetary	Very compact	3–10:1 per stage	Depends on hold/input	High power density, great η
Worm	Very compact	10–60:1 per stage	No flip	Quiet, high reduction, lower η

---

6) Micro-Workflow: Laying Out a Reducer

Given: P_in, n_in, R_target, duty, noise target, packaging.
1) Pick train family (compound spur/helical? planetary? worm?)
2) Choose stages n and distribute ratios: r ≈ R_target^(1/n)
3) Select tooth counts: pinions ≥ 18–20 teeth (20° PA) to avoid undercut
4) Choose module/DP to satisfy bending & pitting via AGMA/ISO checks
5) Compute torques per shaft (T grows each stage · η)
6) Size shafts/bearings for F_t, radial, and axial (helical) loads
7) Estimate η_stage and η_total; check P_loss and thermal limits
8) Check mesh frequencies vs. structure; tweak ratios or tooth counts to dodge resonances
9) Iterate with material, heat treatment, and lubrication choices

---

Lightning Example: Idler Reality Check

Three external meshes total: driver → idler1 → idler2 → driven.

• Ratio: i = (N_driven/N_driver) (idlers cancel out).
• Direction: 3 external meshes → odd → output flips relative to input.
• Efficiency hit: multiply by idler mesh losses (ouch).

Moral: Use idlers for layout constraints, not for fun. They’re the decorative pillows of gear design.

---

Key Takeaways (Tape These To Your Brain)

• Overall ratio is the product of tooth count ratios and a direction sign from mesh parity.
• Compound reducers win on packaging and control of tooth sizes — distribute ratios evenly.
• Efficiency multiplies; a few percent per stage adds up. Budget for heat.
• Dynamics matter: mesh frequency is forever; balance stage speeds and avoid resonances.
• Idlers don’t change ratios, only your maintenance schedule.

Final Insight

Gears are contracts between kinematics and thermodynamics: every perfect ratio comes with a small tax in friction and noise. Great designs don’t fight this — they budget for it, spread the work across stages, and keep the whole train in tune.