A Globoid Spiral Gear Wheel is a gear set where the worm is shaped like an hourglass and wraps around a matching curved worm wheel, so multiple teeth carry load at once instead of just one or two. The hourglass-profiled worm is the key part — its waisted body conforms to the wheel's curvature and spreads contact across a wide arc. This geometry exists to handle torque levels that would crush a standard cylindrical worm. You see it on heavy mixer drives, ship-deck winches, and Cone Drive reducers pushing 50,000 lb-ft and beyond.
Globoid Spiral Gear Wheels Interactive Calculator
Vary torque, teeth in mesh, cylindrical contact count, and wrap angle to see tooth load sharing and stress reduction.
Equation Used
This calculator estimates how a globoid or double-enveloping worm drive spreads output torque across multiple teeth. The worked diagram shows 7 teeth in mesh over a 65 deg wrap angle; dividing torque by the number of engaged teeth gives the approximate load carried by each tooth.
- Output torque is evenly shared by the teeth simultaneously in mesh.
- Cylindrical worm comparison uses the selected 1 to 2 teeth in contact.
- Relative surface stress follows the article statement that stress falls roughly with the square of contact area.
- Center distance, housing rigidity, lubrication, and material limits are assumed acceptable.
Inside the Globoid Spiral Gear Wheels
A standard cylindrical worm meshes its wheel along essentially a single point of contact at any instant. That point carries every pound of load you push through the box. A globoid spiral gear wheel — also called a double-enveloping or hourglass worm drive — solves that by curving the worm itself into an hourglass shape that wraps the worm wheel's throat. Now you have anywhere from 6 to 11 teeth in mesh simultaneously, depending on ratio and lead angle. Load per tooth drops by a factor of 5 or more. Surface stress on the bronze wheel falls roughly with the square of contact area, so a globoidal worm drive in the same envelope as a cylindrical one will carry 2 to 3 times the torque before pitting starts.
The geometry only works if both members are cut on machines built for it. The worm is generated on a Hindley-style hob path where the cutter sweeps an arc matching the wheel's pitch radius. The wheel is hobbed with a cutter that mimics the finished worm. Centre distance has to be held within roughly ±0.05 mm on a 200 mm box — go wider and the end teeth disengage, narrower and the middle teeth bind and overheat. If you notice a hot spot at the worm's waist after running in, your centre distance is tight. If you see wear concentrated at the worm tips and nothing in the middle, you're running too far apart.
The other failure mode you'll see in the field is mounting deflection. Globoidal worm drives are stiffness-sensitive. A housing that flexes 0.1 mm under load will shift contact off the designed pattern and you'll lose half your tooth count from the mesh. Cone Drive's published install spec calls for housing rigidity such that worm-shaft deflection stays under 0.025 mm at rated torque. Miss that and you're running effectively a cylindrical worm with extra cost.
Key Components
- Hourglass (Globoidal) Worm: The driving member, waisted in the middle so its pitch line follows an arc of radius equal to the wheel's pitch radius. Made from case-hardened 8620 or 4140 steel, ground to a surface finish of 0.4 µm Ra or better. The waist depth is typically 8-15% of the worm's max OD.
- Throated Worm Wheel: The driven gear, with a concave throat that wraps the worm. Almost always cut from centrifugally cast tin bronze (SAE 65 or CuSn12) for its galling resistance against steel. Tooth count usually 30-80 for ratios from 30:1 up to 100:1 in a single stage.
- Centre Distance Spacer / Shim Pack: Sets the worm-to-wheel centre distance to within ±0.05 mm. Standard practice is a ground shim pack between bearing carrier and housing, adjusted during assembly using a contact-pattern check with engineer's blue.
- Tapered Roller Bearings (worm shaft): Carry the high axial thrust generated by the worm's lead angle — typically 8° to 30°. Preloaded to roughly 5-10 µm axial play to keep the worm from walking under load reversal. Worm-shaft deflection must stay below 0.025 mm at rated torque.
- Bronze Wheel Hub & Steel Centre: On larger boxes the bronze rim is shrunk or bolted onto a steel centre to save material cost. Interference fit is typically 0.0008 × hub diameter, with axial pins or bolts as a safety against rim slip under shock load.
Industries That Rely on the Globoid Spiral Gear Wheels
Globoidal worm drives show up wherever you need brutal torque density in a compact box, self-locking action so the load can't backdrive, and a single-stage ratio that a helical gearbox would need 3 stages to match. They cost more than a cylindrical worm — typically 1.8 to 2.5x — and they're harder to manufacture, so engineers reach for them when the application is genuinely demanding. The Cone Drive name has become almost synonymous with the type since FJ Cone's 1925 patents.
- Marine: Steering gear and capstan drives on commercial vessels — Cone Drive MHU-series boxes drive anchor windlasses on offshore supply vessels at ratios from 40:1 to 80:1.
- Solar Tracking: Single-axis and dual-axis solar tracker slew drives, like the Kinematics Manufacturing SE-series, where the self-locking globoid worm holds panels against wind load without a brake.
- Heavy Industrial Mixing: Top-entry agitator drives on chemical reactors — Lightnin and Chemineer mixer gearboxes use double-enveloping worm stages for the slow-shaft output at 30-90 RPM.
- Stage & Theatre Rigging: Powered batten hoists in performing arts venues, where the self-locking globoid drive holds scenery aloft without dependence on a brake.
- Defence & Radar: Azimuth and elevation drives on tracking radars and naval gun mounts, where backlash under 2 arc-minutes and shock-load survival both matter.
- Mining & Bulk Handling: Bucket-wheel reclaimer slew drives and apron feeder drives, where the box sees impact loading and must survive grit ingress.
The Formula Behind the Globoid Spiral Gear Wheels
The number that decides whether a globoidal worm drive earns its cost premium is the simultaneous tooth contact count — how many teeth share the load at any instant. At the low end of the practical range, around 3 teeth in mesh, you're not getting much benefit over a well-cut cylindrical worm. The sweet spot sits around 6 to 9 teeth in mesh, which is where you get the 2-3x torque uplift that justifies the build. Push the geometry too far — wrap angle past 90° trying to chase 11+ teeth — and the worm ends start undercutting, manufacturing tolerances stack up, and you lose contact on the outer teeth anyway. The formula below estimates teeth in mesh from the worm's wrap angle and the wheel's tooth pitch.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| nc | Number of worm-wheel teeth in simultaneous contact with the hourglass worm | teeth (dimensionless) | teeth (dimensionless) |
| θwrap | Wrap angle of the hourglass worm around the wheel's pitch circle | degrees | degrees |
| Z2 | Total tooth count of the worm wheel | teeth (dimensionless) | teeth (dimensionless) |
Worked Example: Globoid Spiral Gear Wheels in a cement plant kiln-feed screw conveyor drive
Sizing the globoidal worm reduction on the head-shaft drive of a 400 mm-diameter kiln-feed screw conveyor at a Lehigh Hanson cement plant. The conveyor runs at 45 RPM output from a 1750 RPM 4-pole motor — that's a 38.9:1 reduction in a single stage. You're specifying a Cone Drive HU40-style box and need to confirm the simultaneous tooth contact justifies the globoid premium over a cylindrical worm box of the same centre distance.
Given
- Z2 = 39 teeth
- θwrap (nominal) = 65 degrees
- Centre distance = 152 mm
- Worm starts = 1 —
Solution
Step 1 — at the nominal wrap angle of 65°, calculate teeth in simultaneous mesh:
That's right in the sweet spot. 7 teeth sharing load means each tooth carries about 14% of the total tangential force, versus 100% on a single-point cylindrical worm. The bronze wheel will run cool and pitting life extends to 25,000+ hours at rated load.
Step 2 — at the low end of the practical wrap range, 35° (typical of a budget globoid box or a worn-in unit with end-tooth disengagement):
At under 4 teeth in mesh you've lost most of the globoid advantage — you're paying 2x the price of a cylindrical worm for maybe 1.3x the torque capacity. If your application allows it, a well-made cylindrical worm with a wider face would be a better buy here.
Step 3 — at the high end, 90° wrap (about as far as practical manufacturing pushes):
Roughly 10 teeth in mesh sounds great on paper, but at 90° wrap the worm tips become very thin and prone to chipping under shock load, and the end teeth need micron-level centre-distance accuracy or they'll lift out of mesh. For a cement plant where dust ingress and shock loads from wet-feed slugs are routine, 90° wrap is too aggressive — 65° is the right design choice.
Result
At the nominal 65° wrap angle, you get roughly 7 teeth in simultaneous mesh, which is the design target for a Cone Drive HU40 in this conveyor application. That tooth count means each tooth sees about one-seventh of the tangential force, which keeps bronze wheel surface stress well below the pitting threshold and supports a 25,000-hour L10 life at rated torque. Compare that to the 35° low-end case (under 4 teeth in mesh, marginal globoid benefit) and the 90° high-end case (nearly 10 teeth but fragile and dust-sensitive) — 65° is the engineering sweet spot. If during run-in you measure a contact pattern showing only the centre 3-4 teeth marking with engineer's blue, the most likely causes are: (1) worm-shaft tapered roller bearings under-preloaded, letting the worm walk axially under load, (2) housing centre distance sitting 0.08 mm or more above nominal which lifts the end teeth out of mesh, or (3) the worm's axial position offset from the wheel centreline by more than 0.10 mm — adjust the shim pack between the bearing carrier and housing to bring the pattern back to a full 7-tooth mark.
When to Use a Globoid Spiral Gear Wheels and When Not To
Globoidal worm drives compete with cylindrical worm drives on cost and with helical-bevel multi-stage reducers on ratio and torque density. The right choice depends on whether you actually need the load-sharing benefit, whether self-locking is required, and how tight your envelope is.
| Property | Globoid Spiral (Double-Enveloping) Worm | Cylindrical Worm Drive | 2-Stage Helical-Bevel Reducer |
|---|---|---|---|
| Single-stage ratio range | 10:1 to 100:1 | 5:1 to 100:1 | 8:1 to 25:1 per stage |
| Torque density (per unit volume) | High — 2-3x cylindrical | Moderate baseline | Moderate to high |
| Efficiency at rated load | 55-90% (lead-angle dependent) | 50-85% | 94-97% |
| Self-locking under back-drive | Yes, below ~6° lead | Yes, below ~6° lead | No — always back-drives |
| Manufacturing cost (relative) | 1.8x to 2.5x | 1.0x baseline | 1.5x to 2.0x |
| Centre-distance tolerance sensitivity | ±0.05 mm critical | ±0.15 mm acceptable | ±0.10 mm acceptable |
| Typical L10 life at rated load | 20,000-30,000 hr | 10,000-15,000 hr | 30,000-50,000 hr |
| Best application fit | High-shock heavy industry, slew drives | General industrial, conveyors | Continuous-duty, high-efficiency drives |
Frequently Asked Questions About Globoid Spiral Gear Wheels
This is almost always a centre-distance issue. Globoid drives are designed for a specific centre distance to ±0.05 mm. If yours is sitting tight — even by 0.08 mm — the middle teeth bind under load and the friction generates heat at the worm's waist rather than spreading it across all 7+ contact teeth.
Pull the inspection cover, run the box at no-load, and feel the worm housing along its length after 10 minutes. If the centre is markedly hotter than the ends, remove a 0.05 mm shim from the bearing carrier and recheck. The other common cause is the wrong oil — globoidal worm boxes generally want ISO VG 460 or 680 compounded gear oil with EP additives, not the AGMA 4 mineral oil that's fine in cylindrical worms.
Three questions decide it. First, is your duty cycle continuous and your shock load factor above 1.5? If yes, the globoid's load sharing pays for itself in bearing life. Second, is your envelope constrained to the point where you can't fit a wider-faced cylindrical worm? Third, does your application see frequent load reversals that would walk a cylindrical worm out of position?
If you answered no to all three, buy the cylindrical worm and put the saved money into a better motor or a brake. The globoid premium only earns out under sustained heavy or shock-loaded service.
It is normal — and in fact desirable up to a point. The bronze wheel work-hardens and conforms to the steel worm during break-in, so the contact pattern at hour 200 should be wider and more even than at hour 1. What you don't want is the pattern shifting toward one end of the wheel, which indicates worm-shaft axial walk from under-preloaded tapered rollers or progressive housing distortion under thermal load.
Mark the bronze wheel with engineer's blue at commissioning, again at 50 hours, and again at 500 hours. A symmetric pattern that grows in width is healthy. An asymmetric pattern that creeps toward one face means you have a mechanical problem to chase before the wheel pits.
In theory yes, in practice almost never successfully. The cylindrical worm housing wasn't built to the stiffness spec a globoid needs — Cone Drive calls for worm-shaft deflection under 0.025 mm at rated torque, and a generic cylindrical-worm housing typically flexes 0.05-0.10 mm. Your globoid will lose its end-tooth contact under load and behave like an expensive cylindrical worm.
The bearings also matter. Cylindrical worm boxes often run angular contact ball bearings; globoids generally need tapered rollers preloaded to 5-10 µm axial play to handle the higher thrust without letting the worm walk. If you really want the globoid benefit, replace the housing or accept that you're paying for capacity you won't get.
Three usual suspects. First, oil viscosity — if you're running ISO VG 680 in an application the manufacturer rated with VG 320, churning losses at low speed can swallow 8-10 percentage points. Second, oil level — globoid boxes are sensitive to over-filling because the hourglass worm sweeps a lot of oil. Drop the level to the bottom of the worm waist and recheck.
Third, and most often missed: the manufacturer's efficiency curve is for fully run-in gears at rated load. If you're testing at 30% load on a freshly assembled box, friction is dominated by static rather than rolling components and you'll be 10-15 points below the published curve until break-in completes around 200-500 hours.
Stay at or below 5° lead angle for guaranteed static self-locking with a typical steel-on-bronze friction coefficient of 0.08-0.10. Between 5° and 7° you're in the marginal zone — the box will hold load at rest but a vibration source can cause it to creep, which is why you'll see stage rigging and solar trackers spec'd at 4° or below.
The cost is efficiency: a 4° lead angle worm runs around 55-65% efficient, versus 80-88% at 20° lead. If your application has a brake or a holding clutch elsewhere in the drivetrain, spec the higher lead angle and recover the efficiency. If the globoid is your only thing keeping the load from falling, stay below 5° and accept the heat.
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.
