A skew worm and wheel gear is a power transmission set where a helical worm drives a mating wheel on a shaft that is neither parallel nor perpendicular to the worm — the axes cross at an angle other than 90°. The worm itself is the key component, acting as a screw that pushes the wheel teeth one tooth per revolution. We use this layout when packaging or chassis constraints force an off-square shaft angle but you still need the high reduction and self-locking behaviour of a worm drive. Ratios of 20:1 to 100:1 in a single stage are typical, with efficiencies of 40-85% depending on lead angle.
Skew Worm and Wheel Gear Interactive Calculator
Vary wheel teeth, worm starts, and shaft crossing angle to see the reduction ratio, output speed fraction, and animated skew-worm mesh.
Equation Used
The ideal worm reduction is the number of wheel teeth divided by the number of worm starts. A 1-start worm advances the wheel by one tooth per worm revolution; more starts increase output speed and reduce the reduction ratio. The crossing angle is shown because it is central to skew-worm packaging, but the basic tooth-count ratio is unchanged.
- Ideal kinematic ratio; friction, backlash, and elastic deflection are not included.
- Worm starts are treated as an integer slider from 1 to 4.
- Crossing angle changes the teaching diagram and skew offset KPI, but not the ideal tooth-count ratio.
How the Skew Worm and Wheel Gear Works
The worm is a helical screw, usually with 1 to 4 starts, that meshes with a wheel cut to match. In a standard worm set the shafts cross at exactly 90°. In a skew worm and wheel set, that crossing angle is something else — 60°, 75°, 105°, whatever the housing geometry demands. The wheel teeth are cut at an angle that compensates, so the helix of the worm and the helix of the wheel still align along the line of contact. If you get this angle wrong by even half a degree, you lose line contact and drop to point contact, which spikes Hertzian stress on the wheel face and chews the bronze in a few hundred hours.
The worm gear ratio is set by the number of wheel teeth divided by the number of worm starts. A single-start worm meshing a 40-tooth wheel gives 40:1. The lead angle of the worm — the angle the helix makes with a plane perpendicular to the worm axis — controls efficiency and self-locking. Below about 5° lead, the drive is irreversibly self-locking: the wheel cannot back-drive the worm. Above 10° you start to lose the self-lock but pick up efficiency, and above 25° you are firmly in non-self-locking territory. For a skew set the effective lead angle shifts because of the shaft offset, so you have to recompute it for the actual crossing angle, not assume the catalogue figure.
Failure modes here are predictable. Pitting on the wheel flank means contact stress is too high — usually because the centre distance was machined oversize and contact dropped from a line to a few teeth. Scoring on the worm thread means the EP additive in the oil broke down, often because the bulk oil ran above 95°C. A whining howl that gets louder with load is almost always a misaligned crossing angle — the teeth are sliding instead of rolling.
Key Components
- Worm (screw): Hardened and ground steel screw, typically 58-62 HRC on the thread flank with surface finish Ra ≤ 0.4 µm. Drives the wheel by sliding contact along the helix. Number of starts (1-4) sets the ratio when paired with the wheel tooth count.
- Worm wheel: Bronze ring, usually centrifugally cast CuSn12 or CuAl10Fe5Ni5, bolted or shrunk onto a steel hub. The bronze is sacrificial — it wears in to match the worm. Tooth-cutting hob must match the actual shaft crossing angle within ±0.1°.
- Output shaft and bearings: Carries the wheel and reacts the high axial thrust that the skew geometry generates. Tapered roller bearings preloaded to 30-50 µm of axial stiffness are standard, because a soft bearing stack lets the wheel walk and ruins the contact pattern.
- Worm shaft and thrust bearing: Takes the reaction thrust from the wheel pushing back on the worm. On skew sets this thrust has both an axial and a radial component, so the front bearing usually pairs an angular-contact ball bearing with a deep-groove for radial support.
- Gear housing: Cast iron or aluminium box that fixes the shaft crossing angle. Bore-to-bore squareness must be held within 0.05 mm over 100 mm, otherwise the contact pattern shifts off the tooth centre and edge-loads the wheel.
- Lubricant: ISO VG 460 to VG 680 mineral or polyglycol gear oil. Polyglycol runs 5-10°C cooler than mineral at the same load, which matters because skew sets generate more heat per kW than orthogonal worm sets due to the extra sliding component.
Industries That Rely on the Skew Worm and Wheel Gear
Skew worm and wheel sets show up wherever the housing has to fit around an awkward shaft angle and you still want a single-stage high reduction. They are not the first choice when shafts can cross at 90° — a standard worm is cheaper and more efficient there. They earn their keep when the input motor sits at an angle dictated by something else: a chassis rail, a conveyor frame, a hand-crank position. Because the worm self-locks at low lead angles, you also see them anywhere a load must hold position when the motor is unpowered, like a screw jack or a sluice gate.
- Material handling: Drive head on a Hytrol EZLogic accumulation conveyor where the gearmotor mounts at 75° to the roller axis to clear a structural beam
- Water control: Hand-wheel actuator on a Rodney Hunt fabricated slide gate at a municipal wastewater plant, using a 60:1 skew worm set to hold the gate against head pressure with no brake
- Food processing: Mixer paddle drive on a Hayssen Unifiller depositor where the paddle shaft enters the bowl at 70° to the motor
- Agriculture: Header drive on a Krone EasyCut disc mower conditioner where the PTO shaft and the cutterbar drive cross at roughly 85° due to the folding frame geometry
- Theatre rigging: Manual fly-line winch on a JR Clancy SureTarget hoist where a hand crank at chest height drives a drum with its axis horizontal across the catwalk
- Heavy machinery: Steering box on a Caterpillar 988 wheel loader, using a skew worm set to convert the steering column angle into the drag link motion
The Formula Behind the Skew Worm and Wheel Gear
The most useful single formula here is the gear ratio combined with the output torque relationship. Ratio is set by geometry alone — number of wheel teeth over number of worm starts — but the torque you actually deliver depends on efficiency, and efficiency on a skew set is sensitive to lead angle. At the low end of the typical lead-angle range (around 3-5°), efficiency drops to 40-50% but you get firm self-locking. At the nominal sweet spot (8-12° lead) you sit at roughly 70-80% efficiency with marginal self-lock. Push the lead above 20° and efficiency climbs past 85% but the wheel can now back-drive the worm, which kills the use case for a holding application.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Tout | Output torque at the wheel shaft | N·m | lbf·ft |
| Tin | Input torque at the worm shaft | N·m | lbf·ft |
| i | Gear ratio = Nwheel / Zstarts | dimensionless | dimensionless |
| η | Mesh efficiency (function of lead angle and friction coefficient) | dimensionless | dimensionless |
| Nwheel | Number of teeth on the worm wheel | count | count |
| Zstarts | Number of starts on the worm | count | count |
Worked Example: Skew Worm and Wheel Gear in a hospital laundry tumbler door winch
You are sizing the skew worm and wheel reducer that drives a counter-balanced loading door on a Milnor 76039 CBW continuous batch washer at a hospital laundry in Hamilton Ontario. The door weighs 180 kg and rises vertically on a cable drum. The motor sits at a 70° angle to the drum shaft because of a steam-pipe routing conflict. You are running a 0.55 kW gearmotor with 3.5 N·m at the worm shaft, into a 1-start worm and a 50-tooth bronze wheel. You need to confirm the door will lift and that the drive will hold the door open without a brake.
Given
- Tin = 3.5 N·m
- Zstarts = 1 count
- Nwheel = 50 count
- Lead angle (nominal) = 9 degrees
- Drum radius = 0.075 m
Solution
Step 1 — compute the gear ratio from tooth count and starts:
Step 2 — at the nominal 9° lead angle on a fresh, well-lubricated set, efficiency lands around 0.75. Compute output torque:
Convert to cable tension at the 0.075 m drum radius: F = 131.25 / 0.075 = 1,750 N, which is 178 kgf. The door weighs 180 kg, so at nominal you are right on the line — a fresh drive lifts the door, a worn one stalls. That is the sweet spot: enough margin to operate, not enough margin to hide a problem.
Step 3 — at the low end of the typical lead-angle range (5°), the worm is heavily self-locking and efficiency drops to about 0.50:
That will not lift the 180 kg door at all — the motor stalls at the start of the cycle. Step 4 — at the high end (15° lead), efficiency climbs to roughly 0.85 but you lose self-lock:
The door now lifts comfortably, but the moment you cut motor power the door's weight back-drives the worm and the door drops on the operator's head. That is why the 9° nominal is the design choice — it lifts and it holds.
Result
Nominal output is 131 N·m at the wheel, giving roughly 178 kgf of cable tension to lift a 180 kg door. In practice the operator sees the door rise smoothly, takes about 4 seconds to clear the loading window, and stays put when power is cut — exactly what you want from a manually-loaded washer. The 5°-lead variant stalls before the door moves and the 15°-lead variant back-drives the moment you let go, so the 9° design is genuinely the only viable point in the range. If you measure cable tension below 1,500 N on a working drive, suspect three things first: shaft crossing-angle error of more than 0.5° (look for a polished band on only one side of the wheel teeth), gear oil viscosity below ISO VG 320 (warm oil from a hot wash cycle thins out and friction climbs), or worm thread wear past 0.15 mm tooth-thickness loss measured with a pin gauge.
Skew Worm and Wheel Gear vs Alternatives
A skew worm and wheel is one of three realistic choices when shafts cross at a non-90° angle. The decision usually comes down to whether you need self-locking, how much efficiency you can sacrifice, and how much you can spend on the gearbox. Here is how it stacks up against the two most common alternatives.
| Property | Skew Worm and Wheel | Hypoid Gear Set | Crossed Helical Gears |
|---|---|---|---|
| Single-stage ratio range | 20:1 to 100:1 | 3:1 to 15:1 | 1:1 to 5:1 |
| Mesh efficiency at nominal load | 40-85% (lead-angle dependent) | 90-96% | 70-90% |
| Self-locking capability | Yes below ~5° lead angle | No | No |
| Load capacity per unit volume | High — line contact on bronze wheel | Very high — multiple teeth in mesh | Low — point contact |
| Typical service life under continuous duty | 8,000-20,000 hours (bronze wear limit) | 20,000-40,000 hours | 2,000-5,000 hours |
| Cost for a 50:1 ratio at 130 N·m output | $300-700 USD | Not achievable in single stage | Not achievable in single stage |
| Heat generation per kW transmitted | High — needs oil cooling above 5 kW | Moderate | Moderate to high |
| Best application fit | High-reduction holding drives at odd shaft angles | Automotive rear axles, high-power right-angle drives | Light-duty motion transfer between skew shafts |
Frequently Asked Questions About Skew Worm and Wheel Gear
The catalogue figure assumes a 90° crossing angle. On a skew set the sliding velocity at the mesh is higher than on an orthogonal worm at the same input RPM, because the relative velocity vector picks up a component along the wheel's axis. More sliding means more friction work, which becomes heat.
Check the actual shaft crossing angle against the gearbox nameplate. If it differs by more than 2°, you are running outside the thermal envelope the maker designed for. Either drop the duty cycle to about 60% of nominal or fit an external oil cooler — a small plate cooler rated for 1-2 kW of heat rejection will usually pull the sump back into range.
Static self-locking and dynamic self-locking are different things. An 8° lead angle is self-locking statically — once the load stops, it stays stopped. But while the worm is still rotating and decelerating, vibration and inertia can walk the wheel backward by a fraction of a tooth before friction grips. If your application sees impact loads or stops abruptly, you will see this creep.
The fix is either to drop the lead angle to 4-5° (cost: efficiency falls to about 50%) or to fit a separate holding brake on the worm shaft. The latter is what most lift and gate drives actually do, despite the marketing claim that the worm 'doesn't need a brake'.
If the ratio is above 30:1 and the package is tight, the skew worm wins on parts count and cost. A bevel-plus-worm stack is two gearboxes, two oil sumps, and an extra coupling, which roughly doubles the price and adds a failure point.
If the ratio is below 20:1 or efficiency above 80% is non-negotiable, go with the bevel-plus-helical or a hypoid. The skew worm's efficiency penalty at low ratios makes it a poor choice — you are paying the heat tax without getting the reduction benefit.
Coat 4-6 wheel teeth with engineer's blue, hand-rotate the worm two full turns under light braking load on the wheel, then inspect. A correct pattern shows a contact band centred on the tooth flank, covering 60-80% of the face width, slightly biased toward the leaving side of the wheel.
If the pattern sits hard against one end of the tooth, your shaft crossing angle is off — shim the worm bearing housing to walk the worm along its axis until the pattern centres. If the pattern is a small dot rather than a band, the centre distance is wrong and no amount of axial adjustment will fix it; you have to re-machine the housing or scrap the wheel.
Three causes account for nearly all premature bronze wear on skew sets, and they are different from the failure modes that appear during initial commissioning. First, water contamination in the oil — even 0.5% water by volume cuts the EP additive's effectiveness in half and the bronze sees metal-to-metal contact at the addendum. Second, an undersized breather that lets the sump pressurise during heat-up and forces oil out past the seals, leaving the upper flank dry. Third, a worm hardness below 55 HRC — soft worms wear themselves into a slightly larger profile that hammers the bronze on every tooth pass.
Pull an oil sample, send it for water and particle count, and pop the inspection cover to micrometer the worm thread. One of those three will be your culprit.
Only if the lead angle is above roughly 12° and you accept that efficiency will be 10-15% lower than in the forward direction. Below 12° the wheel physically cannot drive the worm — the friction angle exceeds the lead angle and the mesh locks.
Even when geometrically possible, it is rarely a good idea. Worm sets are designed assuming the worm is the input, so the bearing arrangement and thrust paths are wrong for reverse running. The output shaft bearing will see thrust loads it was never specified for, and you will eat that bearing in months. Use a helical or planetary set if you need a speed increaser.
References & Further Reading
- Wikipedia contributors. Worm drive. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
