A Friction-clutch Wheel is a pair of pressed-together wheels that transmit torque purely through frictional contact at their rims, slipping when the load exceeds a preset threshold. You see them across packaging, printing, and capping machinery as a built-in overload defence. They work because a normal force squeezes the two wheels together, and the available drive torque equals that normal force times the coefficient of friction times the contact radius. The outcome is a drive that protects shafts, gearheads, and product from jam damage — a Bosch capping head, for instance, will slip its friction clutch rather than crush a misfed bottle.
Friction-clutch Wheels Interactive Calculator
Vary friction, spring normal force, contact radius, and load torque to see the slip torque threshold and whether the clutch transmits or slips.
Equation Used
The friction clutch slips when the load torque is greater than the available rim-friction torque. The threshold is the coefficient of friction mu multiplied by the spring normal force Fn and the contact radius r in metres.
- Dry Coulomb friction at the wheel rim.
- Normal force is evenly distributed across the contact patch.
- Contact radius is converted from mm to m for torque in N*m.
- Slip begins when demanded torque exceeds the calculated threshold.
Inside the Friction-clutch Wheels
A Friction-clutch Wheel works on Coulomb friction. You press one wheel against another with a calibrated normal force — usually from a spring stack, a Belleville washer pack, or an air cylinder — and torque transfers through the contact patch. The driving wheel rotates, friction at the rim drags the driven wheel along, and as long as the demanded torque stays below μ × Fn × r, the pair rotates as one unit. Push the load past that threshold and the rim slips. That slip is the whole point — it caps the torque downstream so a jam never breaks a shaft, strips a gear, or tears a label web.
The geometry has to be exact or the clutch lies to you. The rim faces must be parallel within roughly 0.05 mm across the contact width, otherwise the normal force concentrates on one edge and you get a hot spot, glazing, and a slip torque that drifts 20-30% over a few hours of running. The friction face — typically leather, cork-rubber, sintered bronze, or a moulded resin liner — must stay clean and dry. A drop of cutting oil or silicone spray on the rim will drop the coefficient of friction from around 0.4 down below 0.15, and your overload protection silently re-rates itself to a third of design.
Failure modes are predictable. Glazing from chronic micro-slip is the most common — the surface goes shiny, μ drops, and the clutch starts slipping under normal load. Spring fatigue is next: a coil spring that loses 8% of its preload over 2 years of cycling will quietly shift the trip torque. And contamination — oil mist from a nearby chain drive, fines from a label-paper stack — kills friction faces faster than wear ever does.
Key Components
- Driving Wheel: Rotates on the input shaft and presses against the driven wheel. Rim is usually steel or cast iron, ground to a parallelism tolerance under 0.05 mm. Width sets the contact patch length and therefore the heat-dissipation area during slip events.
- Driven Wheel with Friction Liner: Carries the friction face — leather, cork-rubber, sintered bronze, or moulded phenolic. Liner thickness is typically 3-6 mm. Coefficient of friction sits around 0.3-0.5 dry; drop a wet contaminant on it and you can lose half of that.
- Normal-force Element: Spring stack, Belleville washer pack, or pneumatic cylinder that sets the clamp load. A typical light-duty packaging clutch runs 200-800 N of preload. The element must hold preload within ±5% across its service life or trip torque drifts.
- Adjustment Nut or Pressure Regulator: Lets the technician tune slip torque. On spring-loaded designs it's a castle nut with locking pin; on pneumatic, a precision regulator at 1-6 bar. One full turn of a typical M16 adjustment nut shifts trip torque by roughly 15-25%.
- Hub and Bearing: Allows the driven wheel to slip relative to its shaft (or to the driving wheel) during overload. Needle bearings or bronze bushings handle the relative rotation; the bearing must survive thousands of slip events without seizing.
Where the Friction-clutch Wheels Is Used
Friction-clutch Wheels show up wherever a drive needs to fail safely instead of breaking something. The reason they're chosen over a shear pin or an electronic torque limiter is that they reset themselves automatically — slip happens, the jam clears, the line keeps running, no operator has to swap a sacrificial part. They're also cheap, repairable in the field, and tolerant of dust, vibration, and the kind of wash-down environments that murder electronics.
- Bottle Capping: Bosch Packaging KHS Innofill capping heads use friction-clutch wheels on each spindle so a cross-threaded cap slips the clutch instead of stripping the bottle neck or the chuck.
- Printing: Heidelberg Speedmaster sheet-fed offset presses run friction-driven feed rollers where slip protects the sheet from tearing if a jam forms in the gripper bar.
- Conveyor Drives: Hytrol EZLogic accumulation conveyors use roller-pressure friction clutches at each zone so accumulated cartons stop their own driving roller without stalling the main motor.
- Label Application: Avery Dennison 64-05 wipe-down label applicators use a friction roller drive on the take-up spindle to keep web tension stable as roll diameter shrinks.
- Winding Machines: Schärer Schweiter Mettler textile winders use friction-clutch wheels between the drum and the package to slip when yarn tension spikes, protecting the yarn from breakage.
- Hand Power Tools: Hilti TE 70-AVR rotary hammers include a slip clutch on the output spindle so a stuck bit slips instead of wrenching the operator's wrist.
The Formula Behind the Friction-clutch Wheels
The slip torque of a Friction-clutch Wheel is what you actually care about — it's the trip point above which the clutch protects the rest of the drive. The formula tells you what torque the contact will pass before slipping. At the low end of typical operating range — say 100 N of clamp force on a 30 mm radius rim with a worn leather face — you might be looking at trip torques as low as 1 Nm, which is fine for a label spindle but useless for a capper. At the nominal middle of the range you sit in the 5-15 Nm band where most packaging-line friction clutches operate. Crank the spring up to 1500 N and you're up around 25-30 Nm, but past that point the friction face wears in hours, not months. The sweet spot is wherever your trip torque sits about 30-40% above your worst-case running torque.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Tslip | Torque at which the clutch begins to slip | N·m | lbf·ft |
| μ | Coefficient of friction at the rim contact (dimensionless) | — | — |
| Fn | Normal force pressing the wheels together | N | lbf |
| r | Effective contact radius from the shaft centre to the rim contact | m | ft |
Worked Example: Friction-clutch Wheels in a pharmaceutical bottle-capping spindle
You are sizing the friction-clutch wheel inside a single capping spindle on a Groninger KFL 1040 pharmaceutical capper that runs 80 child-resistant closures per minute onto 30 mL HDPE bottles. The cap supplier specifies a 1.8 Nm torque-to-failure on the closure, so the clutch must trip cleanly below that. The driven wheel has a cork-rubber liner with μ ≈ 0.45 dry, an effective contact radius of 25 mm, and a Belleville washer stack you can preload from 50 N to 200 N via the adjustment nut.
Given
- μ = 0.45 dimensionless
- r = 0.025 m
- Fn,low = 50 N
- Fn,nom = 120 N
- Fn,high = 200 N
- Tcap,max = 1.8 N·m
Solution
Step 1 — at the nominal preload of 120 N, calculate the slip torque the clutch will deliver:
That sits comfortably 25% below the 1.8 N·m cap-failure threshold, so a properly threaded cap reaches its target seating torque while a cross-threaded one slips the clutch before it cracks the closure. This is exactly where you want a child-resistant cap clutch to live.
Step 2 — at the low end of typical adjustment, 50 N of preload:
0.56 N·m is too low for a CRC closure — the clutch will slip before the cap is fully seated, and you'll see loose-cap rejects on the downstream torque-test station within the first 100 bottles. Useful only for very light friction-fit closures or as a starting point during commissioning.
Step 3 — at the high end, 200 N preload:
2.25 N·m exceeds the 1.8 N·m cap-failure threshold. At this setting a jam will crack the closure before the clutch trips — exactly the failure you bought the clutch to prevent. This is past the safe envelope.
Result
Set the Belleville stack to roughly 120 N preload to deliver a nominal slip torque of 1. 35 N·m — high enough to seat the cap cleanly, low enough to protect the closure on a cross-thread. The low-end (50 N → 0.56 N·m) is the under-tightening regime where you'll fail the downstream torque test. The high-end (200 N → 2.25 N·m) is the cap-cracking regime. Your sweet spot is the middle third of the adjustment range. If you measure trip torque well below 1.35 N·m on a calibrated torque tester, the three usual suspects are: (1) cork-rubber liner contaminated with silicone spray from a nearby filler, dropping μ from 0.45 toward 0.15, (2) Belleville stack installed in the wrong alternating pattern, halving the effective spring rate, or (3) the adjustment nut backed off by vibration because the locking pin or thread-locker was missed during assembly.
When to Use a Friction-clutch Wheels and When Not To
Friction-clutch Wheels compete with a few other ways to limit drive torque. The choice usually comes down to how often you expect overload events, how repeatable the trip torque needs to be, and whether the line can tolerate downtime to reset a sacrificial element. Here's how they stack up against the two most common alternatives.
| Property | Friction-clutch Wheel | Shear-pin Coupling | Magnetic Hysteresis Clutch |
|---|---|---|---|
| Trip-torque repeatability | ±10-15% over service life | ±5% but single-use | ±2-3% across millions of cycles |
| Reset after overload | Automatic, instant | Manual pin replacement, 2-10 min downtime | Automatic, instant |
| Typical torque range | 0.5 to 500 N·m | 5 to 50,000 N·m | 0.05 to 100 N·m |
| Maintenance interval | Liner inspection every 6-12 months | Pin stock check per shift | Effectively maintenance-free |
| Typical unit cost | $40 to $400 | $5 pin, $80 hub | $300 to $2,000 |
| Tolerance to contamination | Poor — oil kills μ | Excellent | Excellent — non-contact |
| Best application fit | Packaging and capping lines | Heavy gearbox protection on crushers and conveyors | Tension control on film and fibre winders |
Frequently Asked Questions About Friction-clutch Wheels
The friction face is heating up and its coefficient of friction is dropping. Cork-rubber and leather liners both lose 15-25% of μ between 20°C and 80°C. If the clutch is sitting next to a shrink tunnel or a sealing jaw, ambient soak-through alone can do it. More commonly, micro-slip during normal running is generating its own heat, and you've got a thermal runaway.
Check whether the clutch is slipping during normal operation by marking the rim with a paint pen and watching after a 50-cycle run. Any rotation between the marks means you're sized too tight to running torque — bump the trip torque up so it sits at least 30-40% above peak running load.
The catalogue μ value almost always assumes a clean, broken-in, dry friction face. A brand-new liner often runs 20-30% below its rated μ for the first few hundred cycles until the surface burnishes in. New cork-rubber or new leather is the usual culprit.
The other common cause is contact-radius error. If your liner has a slight crown or the wheels aren't parallel, the effective r is smaller than the geometric r. Run the clutch through a deliberate 200-cycle break-in at light load before you trust the measurement.
Magnetic hysteresis wins for film winding. Tension repeatability on a friction clutch drifts as the liner wears — you'll see torque variation of 10-15% over a roll, which prints as visible tension bands on a clear film. A hysteresis clutch holds ±2-3% across millions of cycles with no contact wear.
The exception is high-torque, low-cost lines where the film is thick enough to tolerate the variation. Below about 5 N·m and on premium optical or capacitor films, hysteresis is the right call. Above 50 N·m or on craft-paper webs, friction is fine and a tenth of the price.
Chatter means the static and dynamic friction coefficients are too far apart, so the clutch alternates between gripping and slipping. The face has usually glazed — a shiny, hard surface where μstatic is much higher than μdynamic. You'll hear it as a low-frequency growl rather than a clean slip whoosh.
Pull the wheel and lightly scuff the friction face with 120-grit emery cloth in a radial pattern, then blow it clean. If chatter returns within a week the clamp force is too high relative to running torque — back off the preload until normal running is well below the slip threshold.
You can, but only between the servo gearbox output and a passive load — never inside the position-feedback path. A friction clutch breaks the rigid coupling the servo encoder relies on, and any slip event puts the load and the encoder out of sync. After a slip, the servo thinks it's at position X but the load is actually at X minus the slip angle.
If you need overload protection on a servo, put the clutch downstream of the load encoder, or use a torque-monitoring servo and let the drive itself fault out at the trip torque. That keeps position integrity intact.
Belleville stacks have a non-linear load-deflection curve, especially when stacked in series-parallel combinations. A quarter turn near the flat part of the curve barely changes preload; the same quarter turn near the steep part can shift trip torque 20%. If you're adjusting blind by counting flats, you're chasing a moving target.
The fix is to set preload by measuring slip torque on a calibrated torque wrench through the output shaft, not by counting nut turns. Document the torque value, not the nut position, and re-verify after the first 100 cycles because the stack settles.
References & Further Reading
- Wikipedia contributors. Friction disk shock absorber and clutch mechanisms — see Clutch article. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
