An Epicyclic Train (form 4) is a planetary gear arrangement where the ring gear is held fixed, the planet carrier acts as the input, and the sun gear delivers the output — producing a step-up speed ratio with reversed sense relative to a form 1 reduction. Aerospace gearbox designers and robotics drivetrain engineers rely on form 4 when a compact input shaft needs to spin a faster output, such as a generator drive or a centrifugal pump head. The fixed ring forces planets to roll around it, which carries the sun faster than the carrier. Typical step-up ratios sit between 1:3 and 1:11.
Epicyclic Train Form 4 Interactive Calculator
Vary the fixed-ring to sun ratio and carrier input speed to see the Form 4 planetary step-up speed and torque tradeoff.
Equation Used
For a Form 4 epicyclic train, the ring is fixed, the carrier is driven, and the sun is the output. Willis' relation reduces to a step-up speed ratio of 1 + N_ring/N_sun, so a larger ring-to-sun ratio spins the sun faster while reducing ideal output torque by the same factor.
- Ring gear is fixed to the housing.
- Carrier is the input and sun gear is the output.
- Ideal gearing with no friction or compliance losses.
- Ring-to-sun value is the effective tooth-count ratio N_ring/N_sun.
The Epicyclic Train (form 4) in Action
Form 4 starts with the same three elements every planetary gearbox uses — sun, planets, ring — but reassigns which member does what. The ring gear is bolted to the housing and never rotates. The carrier (the cage that holds the planet pins) is the input. The sun gear at the centre is the output. Drive the carrier and the planets are forced to walk around the inside of the fixed ring; because each planet is also meshed with the sun, that rolling motion spins the sun faster than the carrier. The ratio depends only on tooth counts, and you compute it from the Willis equation.
Why build it this way? You get a compact step-up in a single stage without a long external gear pair. The whole assembly stays coaxial — input and output share the same centreline — which matters for generator drives, fan hubs, and centrifugal pumps where you cannot offset the shafts. The fixed ring also means torque reacts straight into the housing through the ring teeth, not through a separate idler structure.
Get the tolerances wrong and form 4 punishes you fast. Planet pin position must hold within ±0.02 mm of the theoretical pitch circle on a 60 mm carrier — drift beyond that and load sharing across the planets collapses, throwing 80%+ of the torque onto one planet. Tooth-count combinations must satisfy the assembly condition (sun + ring) divisible by the number of planets, or the planets simply will not slot in at equal spacing. The most common field failures we see are: ring-bolt loosening (the ring is supposed to be fixed and isn't), planet-bearing seizure from undersized needle rollers, and sun-shaft torsional fatigue because the output spins fast and small misalignments stack into bending stress.
Key Components
- Fixed Ring Gear (Annulus): Internal-toothed gear bolted rigidly to the housing. Provides the reaction torque path. Pitch-diameter concentricity to the carrier bore must hold within 0.05 mm or planet mesh load distribution skews and tooth pitting starts at the loaded end within 500 hours.
- Planet Carrier (Input): Cage holding 3 to 5 planet pins on a common pitch circle. Drives the assembly. Pin parallelism to the central axis must stay below 0.01 mm over the pin length — beyond that the planets cock under load and the bearings see edge loading.
- Planet Gears: Idler gears that mesh simultaneously with the fixed ring on the outside and the sun on the inside. Typical count is 3 for high torque, 4-5 for higher speed and lower per-tooth load. Tooth count Zp is set by Zp = (Zr − Zs) / 2.
- Sun Gear (Output): Central external-toothed gear. Spins faster than the carrier. Sun-shaft runout matters here — keep it under 0.015 mm TIR or you will hear gear whine above 3000 RPM output speed.
- Planet Bearings: Needle rollers or bushings inside each planet. Carry the orbital and rolling reaction loads. L10 life scales with load cubed, so a 20% overload cuts life roughly in half.
Industries That Rely on the Epicyclic Train (form 4)
Form 4 shows up wherever a slow input shaft needs a fast coaxial output in a compact space. It is not the dominant planetary form (that is form 1, ring fixed with sun input and carrier output for reduction), but it earns its place in step-up applications where a parallel-shaft gearbox would be too bulky or too misaligned with the driven load.
- Aerospace auxiliary power: Step-up drive between a low-speed accessory pad and a high-speed permanent-magnet generator on a Honeywell 36-150 APU, where the carrier input runs at engine accessory speed and the sun output spins the generator rotor.
- Wind turbine pitch systems: Compact step-up unit inside the hub of a Vestas V90 pitch drive feedback module, converting slow blade-root rotation into encoder-shaft speed for position sensing.
- Marine propulsion: Step-up box on a Volvo Penta IPS pod drive auxiliary alternator pickoff, taking carrier input from the prop shaft and stepping up to charge-rate alternator RPM.
- Industrial centrifugal pumps: Coaxial step-up between a 4-pole 1750 RPM motor and the impeller of a Sulzer MBN multistage pump where the impeller needs 5000+ RPM but the motor frame is fixed.
- Hand-crank generators and emergency power: BaylisBrand Eton wind-up radio drives use a form 4 stage to convert ~120 RPM crank input into 1000+ RPM dynamo speed inside a casing smaller than a deck of cards.
- Test rig drives: Step-up gearhead on a Magtrol DSP6001 dynamometer adapter for spinning small high-speed rotors from a slower driving servomotor.
The Formula Behind the Epicyclic Train (form 4)
The form 4 ratio tells you how much faster the sun output spins relative to the carrier input, given the tooth counts of the sun and ring. At the low end of the practical range — say a 60-tooth ring and a 30-tooth sun — you get a 1:3 step-up, which is gentle and forgiving on bearings. At the high end, a 90-tooth ring with a 12-tooth sun gives roughly 1:8.5, but the small sun pinion sees brutal contact stress and fatigue life drops sharply. The sweet spot for most industrial form 4 builds sits between 1:4 and 1:6, where tooth bending stress, planet count flexibility, and sun-bearing life all stay in the comfortable zone.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| nsun | Output speed of the sun gear | rev/min (RPM) | RPM |
| ncarrier | Input speed of the planet carrier | rev/min (RPM) | RPM |
| Zring | Tooth count of the fixed ring gear (annulus) | teeth | teeth |
| Zsun | Tooth count of the sun gear | teeth | teeth |
Worked Example: Epicyclic Train (form 4) in a small wind-up emergency dynamo
You are sizing the form 4 step-up stage inside a hand-crank emergency dynamo similar to an Eton FRX3 weather radio. The crank turns the carrier through a small bevel pair at a comfortable human cranking rate, and you need the sun output spinning fast enough to drive a 6V DC dynamo at useful charge current. The ring has 72 teeth, the sun has 18 teeth, and the planets fill in at 27 teeth each (3 planets, equally spaced). Cranking speed varies with the user — a child cranks slowly, an adult cranks briskly.
Given
- Zring = 72 teeth
- Zsun = 18 teeth
- Zplanet = 27 teeth
- ncarrier,nominal = 120 RPM
Solution
Step 1 — compute the form 4 step-up ratio from tooth counts:
So one carrier turn produces 5 sun turns. Step 2 — at nominal cranking speed of 120 RPM (a relaxed adult cadence), the sun output is:
That is the design point — fast enough to wake the dynamo into useful output, slow enough that the user does not feel like they are wrestling the crank. Step 3 — at the low end of the typical operating range, a child or a tired user cranking at 60 RPM:
At 300 RPM the dynamo is below its useful charge threshold — you will see the LED dim and the radio module starve. The user feels they are doing work but nothing is happening. Step 4 — at the high end, an adult cranking hard at 200 RPM in an emergency:
1000 RPM at the sun is where you start to hear gear whine through the plastic case, and the planet needle bearings (typically 3 mm rollers in this size class) start running near their thermal limit if the cranking continues for more than a couple of minutes. The sweet spot is clearly the 100-140 RPM cranking band, giving 500-700 RPM at the sun.
Result
Nominal sun output is 600 RPM at a 120 RPM crank input — exactly where the dynamo wants to live for steady charging. At the 60 RPM low end the sun spins at 300 RPM and the dynamo essentially produces nothing useful, which is why every commercial wind-up radio has a minimum-cranking-rate sticker on it. At the 200 RPM high end you get 1000 RPM but pay for it in audible gear whine and planet-bearing heat. If your measured sun speed comes in below predicted, the most likely causes are: (1) the ring not actually being fixed — a loose ring-retention clip lets the ring back-rotate and steals output speed, (2) excessive backlash from a worn injection-moulded sun pinion (common in cheap consumer units after a few hundred cycles), or (3) a slipping crank-to-carrier coupling where the bevel input pin has rounded off.
When to Use a Epicyclic Train (form 4) and When Not To
Form 4 is one of four kinematic configurations of the same three-element planetary set. The decision between them comes down to whether you need a step-up or step-down, how much ratio you need, and whether your input and output shafts must be coaxial. Here is how form 4 stacks up against the two most common alternatives engineers actually consider.
| Property | Epicyclic form 4 (ring fixed, carrier in, sun out) | Epicyclic form 1 (ring fixed, sun in, carrier out) | Parallel-shaft spur gear pair |
|---|---|---|---|
| Typical ratio range | 1:3 to 1:11 step-up | 3:1 to 11:1 reduction | 1:5 to 5:1 either direction |
| Shaft arrangement | Coaxial input/output | Coaxial input/output | Offset by centre distance |
| Torque density | High — 3-5 planets share load | High — 3-5 planets share load | Low — single mesh point |
| Output speed range | Up to 15,000 RPM in well-built units | Typically capped at 6000 RPM input | Up to 10,000 RPM with precision grinding |
| Bearing life at full rated load | 3000-8000 hours typical | 8000-20,000 hours typical | 10,000-30,000 hours typical |
| Relative cost (same ratio) | Medium-high — ground tooth set, needle bearings | Medium-high — same hardware | Low — two gears, two bearings |
| Sensitivity to misalignment | Self-centring on planets, but sun runout critical | Self-centring, more forgiving | Very sensitive — center distance ±0.05 mm |
| Best application fit | Compact coaxial step-up (generators, dynamos) | Compact coaxial reduction (servo gearheads) | Offset shafts, simple drives |
Frequently Asked Questions About Epicyclic Train (form 4)
Both configurations share identical hardware — the difference is purely which member you drive and which you take output from. In form 1 the sun is the input and the carrier is the output, so the planets walk slowly around the fixed ring while the sun spins fast — a reduction. In form 4 you flip it: drive the carrier, take output from the sun, and now the same kinematics that produced reduction become a step-up because the sun is now downstream of the slow carrier and gets spun by the planets rolling on the fixed ring.
The ratios are reciprocal. A form 1 unit at 5:1 reduction becomes a 1:5 step-up if you re-plumb it as form 4. Same gears, same housing, same bearings.
Mechanically yes, but it is a bad idea in most cases. Servo gearheads are designed for reduction duty — the input bearing is sized for high-RPM, low-torque, and the output bearing is sized for low-RPM, high-torque. Run it backwards and you are now putting high RPM through a bearing built for slow rotation, and the seal package was never qualified for that surface speed. You will overheat the output-side seal within hours.
The other issue is efficiency. Planetary gearheads are typically 92-97% efficient in their designed direction; reverse-driven they often drop to 85-90% because the tooth contact pattern was optimised for one load direction. If you need a step-up, buy a unit specified as a step-up.
Form 4 ratio is rigid kinematically — tooth counts cannot lie. If your measured ratio under load is lower than the geometric ratio, you have either backlash being taken up dynamically (output lags input by a constant phase, but average ratio still matches over multiple revolutions) or the ring is not fully fixed.
The most common culprit is the ring retention. If the ring is held by a press fit or a snap ring instead of bolted, it can micro-rotate under torque pulses, and that rotation subtracts directly from your output speed. Put a paint mark across the ring-to-housing joint and run the unit — if the mark shears, your ring is moving. Bolt it down properly with a torqued flange and the measured ratio will snap back to 5.0.
Three planets is the default for high-torque step-down applications because three points always define a stable plane and load-sharing is automatically equalised by the carrier floating slightly. For form 4 step-up duty you often want 4 or 5 planets, because the higher output speed means each planet sees more cycles per minute and you want to reduce the per-planet load to extend bearing life.
The constraint is the assembly condition: (Zsun + Zring) must be divisible by the planet count for equal angular spacing. So with Zsun = 18 and Zring = 72, the sum is 90, which divides cleanly by 3, 5, 6, 9, and 10 — but not by 4. If you want 4 planets you have to re-pick tooth counts.
Two effects that are silent at low speed become loud above 3000 RPM at the sun. First, planet pin runout that is invisible statically becomes a periodic excitation at planet-pass frequency — typically 200-500 Hz at 5000 RPM sun speed — and the housing acts as a sound box. Second, the sun pinion is small (often 12-20 teeth) and its tooth-mesh frequency lands in the audible range exactly where the human ear is most sensitive, around 2-4 kHz.
Quick check: spin the unit and listen with the ring bolts loose by half a turn. If the noise drops markedly, your ring resonance is the issue and you need a damping shim. If it stays the same, the noise is coming from sun-pinion mesh and you need a finer tooth quality grade — go from AGMA Q8 to Q10 and the whine drops 6-8 dB.
Yes, and that turns it into a form 1 reduction. Form 4 has no inherent self-locking — efficiency in either direction is similar (low 90s percent), so torque applied at the sun will spin the carrier slowly. This matters for safety: if your form 4 step-up is driving a flywheel, the flywheel can back-drive the input crank when you let go, and a hand-crank dynamo without a one-way clutch will whip the handle around at high speed.
The fix is a sprag clutch or ratchet between the input and the carrier. Every commercial wind-up dynamo has one — feel the crank when you stop turning, it should free-wheel forward but never spin backwards.
References & Further Reading
- Wikipedia contributors. Epicyclic gearing. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
