An Angular Wiper is a motion-control mechanism that converts continuous rotary input into a controlled angular oscillation of a follower arm, typically with a programmed dwell at one or both ends of the swing. You see them everywhere in packaging machinery — bottle escapements, label applicators, and carton flap-folders all run on angular wipers. A driving cam or crank pushes a pivoted arm through a fixed sweep angle, then holds it stationary while the input keeps turning. The result is precise, repeatable indexing without the cost or backlash of a servo.
Angular Wipers Interactive Calculator
Vary the roller-to-pivot arm length and geometry error to see the resulting angular sweep shift in an angular wiper.
Equation Used
This calculator uses the article geometry example: a small roller-to-pivot length error creates an angular sweep shift. The shift is calculated as the arctangent of the error divided by the arm length, then converted to degrees.
- Geometry error is small compared with the arm length.
- The arm is treated as rigid with no bushing slop or compliance.
- Offset is entered as a positive magnitude.
Operating Principle of the Angular Wipers
An Angular Wiper takes a constant-speed input shaft and produces a follower arm that swings through a defined arc, pauses, swings back, and pauses again — all from one continuous rotation. The classic build uses a cam plate with a contoured slot or rib, and a pivoted lever carrying a roller follower that rides the contour. As the cam rotates, the radial position of the contour changes, which forces the lever to rotate around its pivot. Where the contour holds a constant radius, the lever sits still — that's your dwell. Where the contour ramps in or out, the lever sweeps. Sweep angles of 30° to 90° are typical, and dwell fractions of 30% to 60% of the cycle are common in indexing applications.
The geometry is unforgiving. If the roller-to-pivot distance is off by 0.5 mm on a 100 mm arm, your sweep angle shifts by roughly 0.3° at the tip — enough to miss a bottle neck on an escapement. The contour transitions matter even more. A sharp corner where the dwell meets the rise gives you infinite acceleration on paper and a hammered roller in practice. Modified sine or modified trapezoidal cam profiles are what real machines use, because they keep peak acceleration bounded and the follower stays in contact with the cam at speeds up to 300 RPM input.
Failure modes are predictable. The most common is roller follower wear — once the roller develops a flat from skidding, the dwell loses its sharp edge and the indexed product starts arriving early or late. Next is pivot bushing slop, which lets the arm rattle through the dwell phase and adds 1-2° of position uncertainty at the tip. The third is spring-return failure on systems that use a one-sided cam — without consistent spring preload of around 15-25 N, the follower lifts off the cam at high speed and the whole motion goes erratic.
Key Components
- Drive Cam (Wiper Plate): The contoured plate that defines the motion profile. Typical thickness 8-15 mm, hardened to 58-62 HRC on the working surface. The contour transitions must be modified sine or cycloidal — never linear ramps into dwells, or the follower will hammer.
- Roller Follower: A small bearing-mounted roller that rides the cam contour. Diameter is usually 12-25 mm. The roller bore must match the stud with a slip fit of H7/g6 — too tight and it won't spin freely, too loose and it skids and develops a flat within hours.
- Wiper Arm (Lever): The pivoted arm that carries the follower at one end and the working tool at the other. Lever ratios of 1:1 to 1:3 are typical. Stiffness matters — any flex shows up as position error at the tool tip, so 6061-T6 aluminum or 4140 steel is standard.
- Pivot Bearing: A pair of needle or ball bearings supporting the arm. Radial play above 0.05 mm translates directly into output position error. For high-cycle applications above 1 million cycles, sealed needle bearings outperform plain bushings every time.
- Return Spring (one-sided cams only): Holds the follower against the cam contour. Preload is typically 15-25 N, sized so the follower never lifts off at maximum input speed. Spring rate matters — too stiff and you waste motor torque, too soft and you get follower bounce.
Industries That Rely on the Angular Wipers
Angular Wipers show up wherever a machine needs a precise, repeatable swing with built-in dwell and the budget rules out servos. Packaging is the biggest user, but you also find them in textile machinery, glass-bottle handling, and assembly lines where mechanical timing is more reliable than electronic timing. The reason is simple — a properly sized wiper cam with a hardened contour will run 50 million cycles without measurable wear, and you can't say that about a servo-driven equivalent at the same cost.
- Packaging Machinery: Bottle escapement gates on Krones and KHS filling lines — an angular wiper releases one bottle per cycle from a queue into the filling carousel.
- Label Application: Wipe-down arms on Label-Aire pressure-sensitive label applicators, where the wiper sweeps a foam pad across a passing container to seat the label.
- Carton Erecting: Flap-folding arms on Bosch Cartoning Machines that swing inward to tuck the minor flaps before glue application.
- Textile Machinery: Shuttle-driving levers on traditional Picanol weaving looms — the wiper arm propels the shuttle across the warp at a precise moment in the loom cycle.
- Glass Container Handling: Take-out arms on Emhart Glass IS machines that swing freshly-formed bottles from the blow mold to the conveyor.
- Assembly Automation: Indexing arms on rotary-table assembly machines from companies like Mikron and Hirata, where each station needs a tool stroke timed to the table dwell.
The Formula Behind the Angular Wipers
The core calculation for an Angular Wiper is the relationship between cam contour displacement, lever ratio, and follower arm sweep angle. At the low end of the typical range — small contour displacements of 5-10 mm with long arms — you get fine sweeps of a few degrees, useful for delicate operations like label wiping. At the nominal middle range you hit the design sweet spot of 30-60° sweep, which covers most packaging applications. Push the contour displacement past the geometric limit and the cam profile gets so steep the follower lifts off, regardless of spring preload. The formula tells you where you are in that operating window before you cut metal.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| θsweep | Total angular sweep of the wiper arm tip | degrees (°) | degrees (°) |
| hcam | Radial displacement of the cam contour (rise from dwell to peak) | mm | in |
| Larm | Distance from arm pivot to roller follower centre | mm | in |
| Nin | Input cam rotation speed | RPM | RPM |
| tdwell | Dwell duration per cycle | seconds | seconds |
Worked Example: Angular Wipers in a pharmaceutical blister-pack feeder
You are designing the angular wiper that ejects filled blister cards from the sealing station onto an outfeed belt on a pharmaceutical packaging line running 80 cards per minute. The wiper arm is 120 mm long from pivot to roller centre, the cam contour rise is 25 mm, and the cam runs at 80 RPM matched to the indexer. You need to verify the sweep angle clears the card edge, and you need to know how the motion behaves if the line speeds up to 120 RPM or slows to 40 RPM during a recipe change.
Given
- Larm = 120 mm
- hcam = 25 mm
- Nin (nominal) = 80 RPM
- tdwell fraction = 40 % of cycle
Solution
Step 1 — calculate the nominal sweep angle from the cam rise and arm length:
That sweep is what actually clears the blister card off the sealing platen — at 120 mm arm length, 11.9° at the tip works out to roughly 25 mm of linear motion at the working end, matching the cam rise as expected.
Step 2 — find the cycle time and dwell duration at the nominal 80 RPM:
0.30 seconds is plenty of dwell time for the upstream sealer to release the card before the wiper begins its return stroke.
Step 3 — at the low-end speed of 40 RPM (line slowdown during recipe change):
The card sits on the platen for twice as long, which is harmless — slower is always safer for an angular wiper because peak follower acceleration drops with the square of input speed.
Step 4 — at the high-end speed of 120 RPM:
0.20 seconds of dwell is tight but workable. The bigger concern at 120 RPM is follower acceleration — peak acceleration on a modified-sine cam scales with N2, so going from 80 to 120 RPM multiplies acceleration by 2.25. If your return spring preload was sized for 80 RPM you will see follower lift-off above roughly 110 RPM and the card will get launched instead of wiped.
Result
The wiper produces a nominal 11. 9° sweep at the arm tip with a 0.30 s dwell at 80 RPM — clean, controlled, and well within the geometric envelope of the cam. At 40 RPM the dwell stretches to 0.60 s and the motion feels almost lazy, while at 120 RPM the dwell collapses to 0.20 s and follower acceleration climbs 2.25× over nominal, putting you on the edge of follower lift-off. The sweet spot sits between 60 and 100 RPM. If your measured sweep comes back as 10° or 14° instead of 11.9°, the three usual suspects are: arm-length tolerance (a 5 mm error on Larm changes sweep by about 0.5°), cam wear flattening the contour peak by 0.5-1.0 mm, or a worn pivot bearing letting the arm rotate eccentrically and shorten the effective lever.
When to Use a Angular Wipers and When Not To
Angular Wipers compete with two other intermittent-motion options when an engineer needs a swinging arm with dwell — Geneva drives and servo-driven oscillators. Each wins in a different operating envelope, so the choice comes down to speed, precision, programmability, and cycle life.
| Property | Angular Wiper | Geneva Drive | Servo-Driven Oscillator |
|---|---|---|---|
| Typical input speed | 50-300 RPM | 30-200 RPM | 0-3000 RPM |
| Position repeatability | ±0.2° (good cam, new bearings) | ±0.05° (locked dwell) | ±0.01° (encoder feedback) |
| Sweep angle flexibility | Fixed by cam geometry | Fixed by slot count (60°, 90°, 120°) | Fully programmable |
| Dwell fraction | 20-70% (cam-defined) | Fixed at (n-2)/2n per cycle | Programmable 0-100% |
| Cycle life before rebuild | 50M+ cycles | 10-20M cycles | Limited by motor bearings, ~20-50M |
| Capital cost (typical mid-size) | $200-600 | $400-900 | $2,500-6,000 |
| Suited for | Packaging escapements, label wiping, flap folding | Indexing turrets with locked dwell | Multi-recipe or variable-pattern motion |
Frequently Asked Questions About Angular Wipers
You've got a contour problem at the dwell-to-rise transition. A true dwell needs the cam contour to be a perfect concentric arc relative to the cam centre — any deviation of even 0.1 mm in radial position over the dwell zone shows up as arm creep. Two real-world causes dominate.
First, the cam was machined with a CNC tool path that approximated the dwell arc with chord segments instead of true circular interpolation. You'll see this on cams cut from a CAM package set up with loose chord tolerance. Second, the cam has worn unevenly — high-cycle wear concentrates at the rise transitions and bleeds slightly into the dwell zone, rounding the corner.
Diagnostic check: mount a dial indicator on the follower stud and rotate the cam by hand through the dwell zone. Anything more than 0.02 mm of indicated movement means the contour is the problem, not the bearings.
Preload alone doesn't tell you whether the spring keeps up with the follower at speed. What matters is whether the spring force exceeds the follower's mass times peak negative acceleration at every point in the cycle.
At 250 RPM on a modified-sine cam with 25 mm rise, peak follower deceleration easily hits 200-400 m/s². If your follower assembly weighs 0.15 kg, you need the spring to provide at least 60 N of force at maximum extension — not preload. Preload is what you measure at minimum extension, which is the lowest force in the cycle. Recalculate using Fspring = m × apeak with a 1.5× safety factor, then back-solve for the required spring rate and preload. You'll usually find you need either a stiffer spring or a lighter follower.
For a 6-station turret with a fixed 60° index, the Geneva drive is almost always the better choice. Geneva mechanisms give you a perfectly locked dwell — the output shaft is mechanically constrained, not held by spring force against a contour — and the dwell fraction at 6 slots is exactly 66.7% per cycle, which lines up with most assembly station timing.
An Angular Wiper makes sense when your sweep angle isn't a clean fraction of 360°, when you need a non-uniform dwell distribution, or when the output is a swinging arm rather than a rotating turret. If the turret might change to 4 or 8 stations later, neither cam-based option helps — that's when you spec a servo indexer.
The rule we use is that the pressure angle — the angle between the cam contour normal and the follower motion direction — should never exceed 30° for translating followers, or 35° for swinging-arm followers like an angular wiper. Above that, the side load on the follower stud climbs fast and you start eating bushings.
The quick check: cam base radius should be at least hcam / tan(30°) for the rise segment occupying the smallest cam arc. For our 25 mm rise example, if the rise occupies 60° of cam rotation, you need a base radius of at least 24 mm, so a cam diameter of 50 mm minimum. In practice we go 1.5× that to give room for the dwell and return segments. Skimping on cam diameter is the single most common rookie mistake in wiper design.
Thermal expansion in the arm. A 120 mm aluminum 6061 arm grows about 0.03 mm per °C — so a 20°C rise in machine temperature lengthens the effective lever by 0.06 mm and shifts your sweep angle by roughly 0.03°. That's small, but it compounds with bearing thermal growth and any thermal expansion in the cam itself.
If timing drift is critical, two fixes work. Switch the arm material to Invar or 4140 steel — 4140 expands at roughly one-third the rate of aluminum and steel cams pair well with steel arms because they grow together. Or, if you're stuck with aluminum, design the system to reach thermal equilibrium during a 10-minute warm-up cycle and set the timing hot, not cold.
Only if your duty cycle is genuinely light. Plain bronze bushings work fine up to maybe 500,000 cycles at moderate side loads, after which they wear oval and you start seeing 1-2° of arm wobble at the tip. Needle bearings will run 20-50 million cycles in the same application before showing measurable play.
The hidden cost of a plain bushing isn't the bushing itself — it's the downtime when it fails. A packaging line running 60 cycles per minute hits 500,000 cycles in about 140 hours, or two weeks of single-shift operation. We've yet to meet a maintenance manager who will trade a $40 needle bearing for a $4 bushing if the line stops every two weeks.
References & Further Reading
- Wikipedia contributors. Cam (mechanism). Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
