A knitting machine cam mechanism is a set of profiled steel plates that push and pull latch needles up and down their tracks as the cam box (or cylinder) sweeps past them. Modern circular machines like the Mayer & Cie Relanit run cam carriers at 1.2 m/s linear speed and produce 30+ stitches per needle per second. The cams convert rotary or linear carriage motion into the precise vertical needle stroke that forms knit, tuck or miss loops. Without them, no fabric leaves a Stoll, Shima Seiki or Santoni machine.
Knitting Machine Cam Mechanism Interactive Calculator
Vary cam-box speed and stitch-cam angle to see needle velocity, cam slope, and butt-fatigue severity on the animated cam-track diagram.
Equation Used
The cam box moves horizontally at v_cam. On the stitch cam ramp, that horizontal motion is converted into vertical needle motion according to the ramp angle theta, so the peak needle velocity is v_needle = v_cam x tan(theta). The fatigue index compares the current velocity energy level to the 1.2 m/s, 47 deg reference case.
- Stitch cam angle is measured from horizontal.
- Needle butt follows the cam ramp without slip.
- Peak vertical velocity occurs at the steepest stitch cam ramp.
- Clearance, friction, and needle rebound are not included.
How the Knitting Machine Cam Mechanism Works
The needles sit in vertical slots in the needle bed or cylinder, each one carrying a small protruding heel called the needle butt. As the cam box travels past, the butt rides through a profiled channel — the cam track — cut between a raising cam, a stitch cam, and an upper guard cam. The track shape decides everything. A full lift to clearing height pulls the needle up far enough that the old loop drops behind the open latch, then the stitch cam pushes the needle back down through the yarn to form a new loop. If you only lift halfway, the latch never clears, and you get a tuck stitch instead. Hold the needle still, and you get a miss.
The geometry is unforgiving. The angle of the stitch cam — typically 45° to 50° on a circular sock machine — sets how hard the butt slams into the descending ramp. Too steep and you spike butt-breakage rates; we've seen 47° cams shed butts inside 200 hours when run at 1.2 m/s carriage speed. Too shallow and the loop length goes inconsistent because the descent rate doesn't match yarn feed. The clearance between butt and track also matters: 0.05 to 0.10 mm is standard. Open it to 0.20 mm and the needle rattles, the loops go uneven, and you start dropping stitches at the cam exit. Tighten it below 0.03 mm and butt friction climbs, the cam overheats and you'll see galling on the stitch cam face within a shift.
Most failures trace back to three things — a worn stitch cam (visible as a polished groove and rounded entry edge), bent needle butts from a previous jam, or oil starvation that lets steel-on-steel galling start. On a flat knitter the cam box reverses every traverse, so the cams must be symmetric or switchable; on a circular machine the cams are fixed and the cylinder rotates one direction, which is why circular machines hit higher speeds.
Key Components
- Raising Cam: Lifts the needle butt from rest position up to either tuck height (about 4 mm above rest) or full clearing height (8-10 mm above rest, depending on needle gauge). The selection of partial or full lift is what gives the machine its knit/tuck/miss capability.
- Stitch Cam (Cam Drawing): Pulls the needle back down through the yarn after clearing, forming the new loop. The vertical position of this cam sets stitch length directly — moving it 0.1 mm changes loop length measurably. On Stoll CMS machines this cam is electronically positioned via stepper motors.
- Upper Guard Cam: Caps the cam track from above, holding the butt down through the lift sequence. It also defines the upper boundary of the knitting zone. Guard cam wear lets needles fly out of track during high-speed running, leading to needle smash.
- Needle Butt: The protruding heel on the side of every latch needle that rides the cam track. Standard butt thickness is 0.95 to 1.05 mm on E18 gauge needles. A bent butt is the single most common cause of cam jams.
- Cam Box (or Cam Carrier): The housing that holds the cam plates and travels across the needle bed on a flat machine, or stays fixed while the cylinder rotates on a circular machine. On a Shima Seiki SVR123 the cam box weighs around 12 kg and traverses at up to 1.6 m/s.
- Selector Jacks: On patterning machines, selector jacks (or selectors) push individual needle butts into either the knit, tuck, or miss path under electronic control. A Santoni SM8-TOP2 uses piezo selectors that switch in under 1 ms per needle.
Industries That Rely on the Knitting Machine Cam Mechanism
Cam mechanisms appear in every weft-knitting machine ever built — from a 1589 William Lee stocking frame replica to a modern 84-feeder Mayer & Cie circular sock machine running at 1500 RPM cylinder speed. Whenever you need to drive a population of needles through identical synchronised strokes, cams are how you do it.
- Apparel knitwear: Stoll CMS 530 HP flat knitting machines producing fully-fashioned sweater panels for Hugo Boss — uses electronically-positioned stitch cams to vary loop length course-by-course.
- Hosiery and socks: Lonati GL616 single-cylinder sock machines running 84 feeders at 1500 RPM cylinder speed, with fixed cams and rotating needle cylinder.
- Seamless underwear and activewear: Santoni SM8-TOP2 seamless body-size machines for Spanx and Wolford — 8 feeders, electronic needle selection feeding cams via piezoelectric selectors.
- Technical textiles: Karl Mayer warp-knit machines and Stoll machines producing 3D-knit composite preforms for Adidas Futurecraft 4D shoe uppers.
- Heritage and museum machines: Restored Cotton's Patent fully-fashioned frames at the Knitting Heritage Centre in Ruddington, Nottinghamshire, where the cams still run on 1920s phosphor-bronze tracks.
- Industrial fabrics: Mayer & Cie Relanit 3.2 II circular machines producing terry-loop fabric for towel manufacturers, where stitch cam timing controls the loop pile height to within 0.2 mm.
The Formula Behind the Knitting Machine Cam Mechanism
The single most useful number to compute on a cam mechanism is the needle butt velocity at the steepest point of the stitch cam — that is what determines butt fatigue life and noise. At low cam-box speeds (0.3 m/s, typical for a museum machine or a hand-cranked sample knitter) the butt forces are negligible and you can run any reasonable cam angle. At nominal industrial speeds (1.0 to 1.2 m/s) the butt-cam interaction governs everything and you must size the cam angle carefully. Push to the high end (1.6 m/s on a Shima SVR123) and even a 1° error in cam angle starts shedding butts within a few production hours. The formula below relates cam-box linear speed, cam angle, and the resulting vertical needle velocity — the number that actually does the work and breaks the parts.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| vneedle | Vertical velocity of the needle as the butt rides the stitch cam ramp | m/s | in/s |
| vcam | Linear speed of the cam box across the needle bed (or equivalent tangential cylinder speed on a circular machine) | m/s | in/s |
| θ | Angle of the stitch cam descent ramp measured from horizontal | degrees | degrees |
| Fbutt | Lateral force on the butt (derived: F = mneedle × vneedle2 / s where s is ramp length) | N | lbf |
Worked Example: Knitting Machine Cam Mechanism in a Stoll CMS 530 HP flat knitting machine
A technical knitwear plant in Reutlingen, Germany is recommissioning a Stoll CMS 530 HP flat knitting machine producing E14 gauge merino sweater panels for an outdoor brand. The cam box traverses the 50-inch needle bed at 1.0 m/s nominal, with a stitch cam angle of 47°. They want to know the peak vertical needle velocity at three operating points — slow start-up, nominal production, and the machine's rated maximum traverse — and whether the 47° cam will survive the high end.
Given
- vcam,nom = 1.0 m/s
- vcam,low = 0.3 m/s
- vcam,high = 1.6 m/s
- θ = 47 degrees
- needle gauge = E14 needles/inch
Solution
Step 1 — at nominal 1.0 m/s cam-box speed and 47° stitch cam, compute peak needle velocity:
This is the production sweet spot. The needle accelerates from rest to 1.07 m/s and back to rest in under 8 ms, the latch swings cleanly, and butt-cam contact stress sits within the fatigue limit of the through-hardened tool steel cam (typically D2 at 60-62 HRC).
Step 2 — at the low end of the operating range, 0.3 m/s during pattern start-up or slow-knit hem zones:
At this speed the cam is barely loaded. You could run a 60° cam angle here without consequence — which is why slow-knit machines from the 1950s used very steep cams that would self-destruct on a modern fast machine. The downside is throughput: 0.3 m/s traverse on a 50-inch bed gives one course every 1.27 seconds, which is fine for a 12-stitch hem but disastrous for production economics on a body panel.
Step 3 — at the high end, 1.6 m/s rated maximum traverse:
Now the butt-on-cam contact pressure roughly triples relative to nominal because the kinetic energy scales with v2. At 1.72 m/s peak needle velocity, a 47° cam will start spalling at the descent entry edge inside 400 to 600 production hours. Stoll's own recommendation is to drop the cam angle to 43° for sustained operation above 1.4 m/s — at 43° the high-end needle velocity drops to 1.49 m/s and butt-shedding rates fall by an order of magnitude.
Result
Peak needle velocity at nominal 1. 0 m/s cam-box speed is 1.07 m/s. In practice this is the speed at which a skilled mechanic can still hear an even, rhythmic 'tick-tick-tick' from the cam box — a smooth note means clean butt-cam tracking. At 0.3 m/s start-up the needle barely whispers at 0.32 m/s; at 1.6 m/s rated max the needle hits 1.72 m/s and the cam box turns into a rattling complaint within hours if the cam angle isn't dropped. If you measure butt velocity in the field (high-speed video at 5000 fps works) and the number comes back lower than predicted, check three things: (1) cam-track wear has rounded the descent entry edge — easy to spot as a polished radius where there should be a sharp corner — which softens the actual ramp angle below the nominal 47°; (2) cam-box loose on its guide rails so it lifts slightly off the needle bed during traverse, reducing effective butt engagement; (3) the wrong stitch cam fitted from a parts bin — Stoll alone supplies cams in 43°, 45°, 47° and 50° variants and they look near-identical at a glance.
When to Use a Knitting Machine Cam Mechanism and When Not To
A cam-driven needle bed is the dominant solution in weft knitting, but it isn't the only way to drive needles. Warp knitters use bar-mounted needles driven by linkage trains, jacquard knitters add selection layers on top of the cam track, and a few experimental machines drive each needle individually with a linear actuator. Each approach trades off speed, pattern flexibility, cost, and maintenance load.
| Property | Cam Mechanism (weft knitting) | Linkage-driven Bar (warp knitting) | Individual Linear Actuator per Needle |
|---|---|---|---|
| Maximum needle stroke rate | 30-50 strokes/s per needle | 20-30 strokes/s per needle | 5-10 strokes/s per needle |
| Stitch length accuracy | ±0.05 mm with electronic stitch cam positioning | ±0.02 mm (geometrically fixed) | ±0.1 mm (actuator repeatability limited) |
| Cost per needle position | $2-8 (mechanical) to $40 (electronic selection) | $1-3 (passive) | $80-200 (per-needle servo) |
| Pattern flexibility | Excellent — knit/tuck/miss per needle per course | Poor — fixed pattern set by linkage | Theoretically unlimited but commercially impractical |
| Wear life (hours to cam/butt rebuild) | 3000-8000 hr typical | 10000+ hr (no wear surface) | 20000+ hr (no contact wear) |
| Best application fit | Sweaters, socks, seamless garments, technical knits | Tricot, lace, warp-knit composites | Research machines, ultra-low-volume custom knits |
| Industry adoption | Dominant — Stoll, Shima Seiki, Santoni, Mayer & Cie, Lonati | Dominant in warp knitting — Karl Mayer, Liba | Negligible — academic prototypes only |
Frequently Asked Questions About Knitting Machine Cam Mechanism
Almost always a cam-box rail wear issue, not a cam profile issue. The cam box rides on hardened linear rails and as the bearings wear, the box tilts slightly when it decelerates at the end of stroke. The right-hand end is usually worse because that's where the box reverses against the take-up spring tension on most CMS machines.
Check rail-to-box clearance with a feeler gauge at both ends — anything above 0.15 mm and the cams are no longer parallel to the needle bed, so the butt engagement depth varies across the cam track. The needles at the trailing end of the cam see reduced lift, which gives intermittent dropped stitches. Re-shim the rails or rebuild the linear bearings before you go anywhere near the cams themselves.
Drop to 43° anytime your sustained cam-box speed exceeds 1.4 m/s, or anytime you're running coarse-gauge needles (E5 to E8) where butt mass is higher and inertial loading on the cam ramp is correspondingly worse. The trade-off is loop-length resolution: a shallower cam means the stitch cam moves through a longer vertical travel for the same horizontal cam-box motion, which slightly reduces the precision of electronic loop-length control.
For E14 to E18 fine-gauge fashion knitting at 1.0-1.2 m/s, 47° is the right choice. For E7 woollen outerwear at 1.5 m/s, go 43°. For anything above 1.6 m/s you should be looking at 40° and accepting the loop-length precision hit.
800 hours to first visible polishing is on the early side but not catastrophic, provided the polishing is uniform across the cam face. Uniform polish is just the running-in pattern of the butts and the cam will stabilise. What you're looking for is a localised polished groove or a rounded leading edge on the descent ramp — that's spalling-precursor wear and means the cam is heading for failure within the next 1000-2000 hours.
The likely root cause at 800 hours is oil starvation. Check your cam-track lubrication mist density — Stoll specifies 0.15 to 0.25 ml per cam box per hour of mist oil. Below that, you get mixed-mode lubrication and the polishing accelerates into pitting.
Three reasons stack up. First, the dial reads cam position not loop length — the relationship is linear but the calibration drifts as the cam wears, because a worn cam pulls the needle slightly less far down for the same dial setting. Second, yarn input tension affects loop length measurably; a 5 cN tension change shifts loop length by 0.1-0.2 mm depending on yarn count. Third, fabric relaxation: the loop you measure off the machine is 5-15% shorter than the loop on the needle, so you're not actually measuring what the cam set.
If you want to calibrate the dial accurately, knit a test course at a known cam setting, measure courses-per-cm on a relaxed swatch after 24 hours, and back-calculate. Re-do this calibration every 500 hours or after any cam change.
You can run tuck on every course indefinitely from the cam's perspective — the butt only sees a partial lift, which is gentler than a full knit stroke. The problem is the needle latches, not the cams. A tuck stitch leaves the old loop on the needle while a new yarn enters the hook, so on consecutive tucks you accumulate loops on the needle until the latch can't close around them. Most latch needles tolerate 3 accumulated loops before the latch jams; some heavy-duty needles handle 5.
If you're seeing 'every-course tuck' as a design requirement, you almost certainly want a tuck-and-knit alternation or a different fabric structure entirely. Talk to your needle supplier about long-latch needles before you commit to the design.
The cam absolutely still does everything — the selector only chooses which of three parallel cam paths the needle butt enters. On a Santoni SM8-TOP2 each needle position has a piezo-actuated selector jack that flicks the butt either fully into the knit raising path, partially into the tuck raising path, or leaves it in the miss path. Once the butt is in a path, conventional cam geometry takes over for the rest of the stroke.
The reason this matters operationally: a selector failure looks like a cam failure (missed needle, dropped stitch in one column) but the fix is completely different. Always probe the selector first with a piezo driver test before you pull the cam box.
References & Further Reading
- Wikipedia contributors. Knitting machine. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
