A cognate linkage is a second four-bar mechanism that traces the exact same coupler curve as a first four-bar, but with different link lengths and a different ground pivot location. The motion principle relies on the Roberts-Chebyshev theorem, which proves that every four-bar coupler curve is generated by three distinct cognate linkages, not just one. Designers use cognates to relocate pivots out of the way of obstacles, repackage a mechanism inside a tighter envelope, or convert a curve-tracing problem into a more buildable geometry. The outcome is geometric freedom — same path, different machine, no compromise on the curve.
Cognate Linkage Interactive Calculator
Vary the coupler-triangle ratio, reference side, and hole tolerance to see the scaled similar-triangle dimensions used for cognate linkage synthesis.
Equation Used
The Roberts-Chebyshev cognate construction preserves the coupler-triangle proportions. This calculator scales an a:b:c triangle so side c equals the chosen reference side, then reports the resulting side lengths and the hole-location tolerance as a percent of that reference.
- Cognate coupler triangles preserve the original side ratio exactly.
- The reference side is mapped to ratio side c.
- Only geometric scaling is calculated, not dynamic bearing loads.
- Tolerance percent is relative to the reference link length.
How the Cognate Linkage Works
The Roberts-Chebyshev theorem says any coupler point on a four-bar linkage has two siblings — two other four-bar linkages whose own coupler points trace the identical curve. That gives you triple generation: three mechanisms, one shared path. The way you find them is by drawing the Cayley diagram. You take the original four-bar, build similar triangles on each link using the coupler triangle's proportions, and the diagram reveals the link lengths and pivot positions of the two cognates falling out of the construction.
The geometry is exact, not approximate. If the original coupler triangle has sides in the ratio 3:4:5 with the coupler point at the apex, both cognates inherit that same ratio in their own coupler triangles — just scaled and rotated. Get the similar-triangle proportions wrong by even 1% and the cognate curve drifts off the original path by a visible amount over a full cycle. We've seen builders try to eyeball the Cayley construction and end up with a curve that looks right at the extremes but wanders 2-3 mm at the midpoint of the stroke, which is enough to ruin a film-advance mechanism or a straight-line approximator.
Failure modes are almost always geometric, not mechanical. The cognate links carry the same forces as any four-bar, so bearing loads and link stresses follow standard four-bar analysis. What kills cognate designs is sloppy synthesis — pivots placed at calculated positions but with ±0.5 mm hole-location tolerance, which compounds across three links and shifts the coupler point off the theoretical curve. Hold pivot-to-pivot centre distance to ±0.05 mm on a 100 mm link if you want the cognate curve to overlay the original within a draftsman's line width.
Key Components
- Original four-bar linkage: The starting mechanism — ground link, two side links, and a coupler link with a designated coupler point. Defines the curve that the cognates will replicate. Coupler triangle proportions, typically expressed as the ratio of its three sides plus the coupler-point location, fully determine what the cognates look like.
- Coupler triangle: The triangular plate (real or virtual) on the coupler link with the coupler point at one vertex. The shape of this triangle propagates into both cognate linkages as similar triangles. A 60-80-40 mm coupler triangle in the original becomes a scaled 60-80-40 ratio in each cognate, just sized differently.
- Cayley diagram: The graphical construction that exposes the two cognates from the original. You build parallelograms and similar triangles off the original linkage; the cognate ground pivots and link lengths read directly off the diagram. Most CAD packages won't do this for you — you draw it by hand or scripted in Python.
- Cognate ground pivots: Two new fixed pivot locations, one for each cognate. They sit at calculated positions relative to the original two ground pivots. Locating these to ±0.05 mm on a 100 mm-scale build is what separates a true cognate from a near-miss approximation.
- Coupler point: The single point on each coupler link whose path is the shared curve. All three mechanisms — original plus two cognates — have their own coupler point, and all three points trace the identical trajectory in space when the mechanisms run in sync.
Who Uses the Cognate Linkage
Cognates earn their keep when the original four-bar works kinematically but fails packaging-wise. You need the curve, but the pivots are in the wrong place — they collide with a frame member, or they sit outside the machine envelope, or they're awkward to mount. A cognate gives you the same curve with pivots somewhere else entirely. That's the practical reason designers reach for them. The other reason is straight-line motion: Roberts and Chebyshev both used cognate analysis to derive their famous straight-line approximators, and modern designers still use the same approach when synthesising new ones.
- Film and camera equipment: Geneva-replacement film-advance mechanisms in 16 mm projectors where the intermittent claw must trace a precise D-shaped path but the original four-bar pivots interfere with the film gate. A cognate relocates the pivots behind the gate plate.
- Walking machines and legged robots: Theo Jansen Strandbeest leg redesigns where students at TU Delft have used cognate analysis to find alternative pivot layouts that fit inside a narrower hip frame while keeping Jansen's original foot path intact.
- Industrial sewing machines: Needle-bar drive mechanisms in heavy-duty machines like the Juki LU-2810 where the coupler point traces the needle's elliptical path. Cognates allow the drive pivots to be moved clear of the bobbin assembly.
- Straight-line motion generators: Hoekens and Chebyshev linkage designs in school engineering kits like the Tamiya Mechanical Linkage Kit, where cognate variants are taught as proof that one curve has multiple buildable mechanisms.
- Aerospace deployment mechanisms: Satellite antenna deployment four-bars where launch-volume constraints force pivot relocation. Engineers at firms like MDA Space have used cognate substitutions to fit deployment paths inside fairing envelopes without altering the deployed antenna trajectory.
- Automotive convertible tops: Folding hardtop linkages on cars like the Mercedes SLK where the coupler curve must clear the rear deck precisely, but cognate analysis finds alternative pivot layouts that simplify the folded-stack height.
The Formula Behind the Cognate Linkage
The core relationship for a Roberts cognate is the similar-triangle proportionality between the original coupler triangle and each cognate's coupler triangle. The ratio determines the scaled link lengths of the cognate. At the low end of useful scale ratios (around 0.3-0.5), the cognate becomes a compact secondary mechanism — handy for tight packaging but with magnified angular velocities at the coupler that can stress small bearings. The nominal sweet spot sits around a scale ratio of 0.7-1.0, where the cognate is similar in size to the original and runs at comparable speeds. At the high end (1.5-2.5), the cognate sprawls — pivots end up far from the original ground line, and you gain re-routing flexibility at the cost of a larger footprint.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Lcog | Length of the corresponding link in the cognate linkage | mm | in |
| Lorig | Length of the link in the original four-bar | mm | in |
| aorig | Side of the original coupler triangle adjacent to the coupler point | mm | in |
| acog | Corresponding side of the cognate's coupler triangle (similar-triangle scaling) | mm | in |
| k | Scale ratio acog / aorig — controls overall cognate size | dimensionless | dimensionless |
Worked Example: Cognate Linkage in a vinyl record cutting lathe cutter-head linkage
You are designing the cutter-head suspension linkage for a Neumann VMS-70-style vinyl record cutting lathe at a small mastering studio in Berlin. The original four-bar has a ground link of 120 mm, input crank of 40 mm, coupler of 100 mm, output rocker of 80 mm, with a coupler triangle of sides 100-60-50 mm and the coupler point at the 60-50 vertex. The cutter must trace a slightly curved compliance path, but the original input pivot collides with the helium feed line on the cutter head. You need a Roberts cognate that relocates that pivot 35 mm to the side without changing the cutter path.
Given
- Lground,orig = 120 mm
- Lcrank,orig = 40 mm
- Lcoupler,orig = 100 mm
- Lrocker,orig = 80 mm
- aorig = 100 mm
- k (nominal scale) = 0.8 dimensionless
Solution
Step 1 — at the nominal scale ratio k = 0.8 (a cognate similar in size to the original), compute the cognate input crank length using the Roberts proportionality:
Step 2 — apply the same scale to the coupler and rocker links to get the full nominal cognate geometry:
This nominal cognate fits comfortably inside the cutter-head envelope and runs at angular velocities within 25% of the original — the bearings see comparable loading, and a standard SKF 6000-series radial bearing handles the input pivot without resizing.
Step 3 — at the low end of the practical range, k = 0.4 (compact cognate):
A 16 mm crank is tiny — you save space but the angular velocity at the coupler point increases by roughly 1/k, so the cognate input crank spins effectively 2.5× faster than the original to trace the same curve at the same cutter speed. That stresses miniature bearings and amplifies any pivot slop. At the high end, k = 1.6:
The 64 mm cognate sprawls — it gives you maximum freedom to relocate the input pivot away from the helium line, but the cognate ground pivot now sits 80-100 mm outside the original ground line, which forces a larger baseplate. For this lathe build, k ≈ 0.8 is the buildable answer.
Result
The nominal cognate uses a 32 mm crank, 80 mm coupler, and 64 mm rocker, with the input pivot relocated 35 mm clear of the helium line — and the cutter point traces the identical compliance curve as the original four-bar. At k = 0.4 the cognate is compact but spins 2.5× faster at the input, which pushes miniature bearings near their limit; at k = 1.6 the cognate gives huge re-routing freedom but eats 80-100 mm of extra baseplate width. The k = 0.8 sweet spot balances size, speed, and packaging. If your built cognate traces a curve that drifts more than 0.1 mm from the original path, the three usual culprits are: (1) coupler-triangle similarity broken because the cognate coupler-point hole was drilled at the wrong vertex angle — check with a height gauge against a CAD overlay, (2) ground-pivot centre distance off by more than ±0.05 mm so the cognate's effective ground link is wrong, or (3) link-length cumulative tolerance stack-up exceeding ±0.1 mm across the four cognate links, which compounds at the coupler point.
When to Use a Cognate Linkage and When Not To
Cognate substitution competes against two other approaches when you need a particular coupler curve in an awkward envelope: redesigning the original four-bar with new dimensions, or switching to a cam-follower system that can trace any curve at all. Each route has its place, and the decision usually comes down to how much packaging freedom you need versus how much synthesis effort you can stomach.
| Property | Cognate Linkage | Redesigned Four-Bar | Cam and Follower |
|---|---|---|---|
| Curve fidelity | Exact match to original (theoretical) | Approximate — depends on synthesis effort | Exact to cam profile, ±0.02 mm typical |
| Synthesis effort | Moderate — Cayley diagram or scripted construction | High — full Burmester or atlas-based synthesis | Low for the curve, high for the cam profile machining |
| Packaging flexibility | High — three pivot layouts available for one curve | Medium — limited by Grashof and curve constraints | Very high — cam can sit anywhere |
| Cost (parts and machining) | Same as original four-bar — 4 links, 4 pivots | Same as original — 4 links, 4 pivots | 2-3× higher — precision cam profile machining |
| Maximum operating speed | Up to 1500-2000 RPM at the input crank | Up to 1500-2000 RPM at the input crank | Limited by follower dynamics, typically 600-1200 RPM |
| Wear and maintenance interval | 10,000+ hours on greased bushings | 10,000+ hours on greased bushings | 2,000-5,000 hours — cam surface wear dominates |
| Best application fit | Pivot relocation around obstacles in tight machines | Greenfield design with no envelope constraints | Arbitrary non-four-bar curves and dwells |
Frequently Asked Questions About Cognate Linkage
That signature — endpoints correct, middle wandering — almost always means your coupler-triangle similarity is broken. The Cayley construction requires the cognate coupler triangle to be similar to the original, meaning all three angles match, not just one or two side ratios.
Check your coupler-point location with a CMM or against a 1:1 CAD print. If the coupler point sits 0.5 mm off the correct vertex of a 60 mm triangle, that's nearly a degree of angular error, and it shows up as a 1-2 mm midstroke deviation on a 100 mm-scale linkage. Re-fixture and re-drill that hole to ±0.05 mm and the curves will overlay cleanly.
You get three mechanisms total — the original plus two cognates — and the choice between the two cognates comes down to which one puts the ground pivots where you actually want them. Draw the full Cayley diagram, mark all six possible ground pivot locations (two for each linkage), and overlay your machine envelope.
Pick the cognate whose pivots clear obstacles, sit on existing structure you can mount to, and keep input torque demand reasonable. If both cognates are buildable, prefer the one with a scale ratio closer to 1.0 — it'll have angular velocities and bearing loads similar to the original, so you can reuse spec calculations.
Yes, the Roberts-Chebyshev theorem makes no Grashof requirement. Cognates exist for crank-rockers, double-rockers, and double-cranks alike. What does change is the rotational behaviour of the cognate links — a non-Grashof original produces non-Grashof cognates, so if your input link only oscillates rather than fully rotating, both cognates will inherit that same constraint.
The trap is assuming the cognate's input link rotates the same way as the original's. It doesn't always. Run a quick range-of-motion check on the cognate input before committing to a motor sizing — sometimes you discover the cognate's input is a rocker even when the original was a crank.
Heat at the input bearing means higher angular velocity, higher load, or both. If you scaled the cognate down (k less than 1.0), its input crank spins faster than the original to trace the curve at the same coupler-point speed — angular velocity scales roughly as 1/k. A k = 0.5 cognate runs its input at 2× the original RPM, doubling viscous drag and bearing heat generation.
Two fixes: increase the bearing size class (move from a 6000 to a 6200-series), or accept a larger cognate (k closer to 1.0) and lose some packaging benefit. Don't try to fix it with thinner grease — that just accelerates wear.
Sometimes, yes. If the original four-bar has poor transmission angles in part of its cycle — say, dipping below 30° near top-dead-centre — one of the cognates may have better transmission angles at the equivalent point in its cycle, because the link geometry is different even though the coupler curve is identical.
This is a real reason to do cognate analysis even when packaging isn't a concern. Generate all three mechanisms, compute transmission angles across a full cycle for each, and pick the one whose worst-case angle is highest. We've seen 15-20° improvements in minimum transmission angle just by switching to a cognate, which translates directly into lower input torque and smoother running.
Cognate synthesis sits in a niche between standard linkage simulation (which most CAD packages handle) and full kinematic synthesis (which requires specialist tools like SAM, GIM, or Linkage). Mainstream packages like SolidWorks and Fusion 360 will simulate a four-bar you've drawn, but they won't construct cognates for you.
The practical workaround is to script the Cayley construction in Python with a parametric CAD plugin, or use one of the academic synthesis tools mentioned above. For a one-off design it's faster to draw the Cayley diagram by hand on paper, measure off the cognate dimensions, and rebuild in CAD as a fresh sketch.
References & Further Reading
- Wikipedia contributors. Roberts–Chebyshev theorem. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
