Machine Kinematics
solved, then drawn
Field

Linkages

Four bars and four pins is the smallest interesting machine there is. Everything about which link turns, how hard it pushes and where it jams follows from the four lengths.
A four-bar at 60°, solved. Ground 4, crank 1, coupler 3.5, rocker 3. Every joint position here is the output of a Newton–Raphson solve on the loop-closure equations, converged to 0.0e+0 — not a placement that looked right. Grashof's condition classifies these lengths as a crank rocker, and sweeping the crank through 360° confirms it: 120 of 120 positions assemble. The transmission angle at this instant is 66.9°.

Four bars and four pins

The four-bar linkage positioned by solving its loop-closure equations rather than by placing its links, why the solve has two solutions, and what the four lengths determine.

 Grashof's classification, predicted and then swept. Four sets of link lengths. For each, Grashof's condition predicts from the lengths alone whether the input can rotate a full turn, and the solver then attempts all 180 positions and reports how many assembled. The prediction and the measurement agree in every case, which is what licenses quoting the classification for a mechanism nobody has swept.

Grashof, predicted and then swept

Grashof's condition as a prediction from four lengths, tested by attempting every input angle, the four classifications it produces, and what the inversion changes.

 The transmission angle through one turn. μ is the angle at B between coupler and rocker, computed from each solved position rather than from a formula. It runs from 54.3° to 100.3° for these lengths. The shaded band is the usual design rule — keep μ between 40° and 140° — and this linkage stays inside it throughout. The rule is about geometry alone: nothing here knows about friction, and a mechanism with a comfortable μ can still be a poor machine.

The transmission angle

The transmission angle of a four-bar computed from solved positions, the 40°–140° design rule and what it is a rule about, and why a linkage that passes Grashof can still be a poor machine.

 Slider-crank at 50°. Crank 1, connecting rod 3. The slider's travel is 2.0000 — exactly twice the crank throw, which is the one thing about this mechanism that does not depend on the rod length. Everything else does: the rod length decides how far the piston's motion departs from a sine wave, and that departure is the second harmonic every engine balancer has to deal with.

The slider-crank

The slider-crank as a four-bar with one joint changed, why the stroke is twice the crank radius, and how the connecting rod's length controls the departure from harmonic motion.

 Where the output stops, and why the return is quicker. A crank-rocker's output reaches an extreme exactly when the crank and coupler line up — stretched out, so O₂ to B is 4.50, or folded back, so it is 2.50. Nothing about the rocker enters the condition, which is why the limits can be written down rather than searched for. Those two crank angles are 40.8° and 228.5°, so the crank spends 187.7° going one way and 172.3° coming back while the rocker covers the same 40.0° both times. The ratio is 1.0894 predicted and 1.0894 measured over 7200 swept positions — a shaper cuts on the slow stroke and returns on the fast one, and this number is what the proportions are chosen to get.

The return stroke is quicker

Limit positions of a crank-rocker from the collinearity condition, the time ratio predicted and then measured over a full sweep, and why a shaper cuts on the slow stroke.

 One chain, four mechanisms. The same four bars and the same four pins in every panel. What changes is which link is bolted to the bench, and that is not a property of the chain — it is a decision about where the bench is. The four mechanisms are crank rocker, double crank, crank rocker, double rocker: one input turns fully in some and rocks in others, and what each one is for is different. What cannot change is the shape of the closed loop, and the two diagonals measure that without reference to which link is held still: swept independently, all four visit the same locus of diagonal pairs to within 2.9e-3, half the sampling resolution. This is why the Whitworth quick-return and the oscillating-cylinder engine are not merely similar to a slider-crank; they are one.

One chain, four mechanisms

Kinematic inversion, the four mechanisms one four-bar chain produces, and the diagonals that no choice of frame can change — swept independently for each inversion and compared.

 Watt's chain. Six links and 7 pin joints, so Grübler counts 3(6 − 1) − 2(7) = 1 and the rank of the Jacobian measures 1. Two of the six links must carry three joints; they are shaded. Here they share a joint, which makes this a Watt chain. That adjacency is the entire classification, and it is read off the graph rather than off the picture.

Six bars, and what the extra dyad buys

The two six-bar chains, classified from the joint graph rather than from a table, and the one thing a four-bar cannot do at all.

 A dwell is a measurement, not a stop. The Stephenson six-bar's output against a full turn of the crank, with the four-bar it is built on for comparison. Inside a band of ±1° the six-bar's output holds still for 145.8° of crank and the four-bar's for 41.9°. The dwell comes from a stretch of the coupler curve that fits a circle of radius 3.006 to within 2.96e-3, and the arm from the coupler point is that radius. Nothing here stops; it moves less than the band.

A dwell made from a curve

A Stephenson six-bar that holds its output still without a cam, built by finding the most nearly circular stretch of a coupler curve and measuring what it is worth.

 Grashof's classification, predicted and then swept. Four sets of link lengths. For each, Grashof's condition predicts from the lengths alone whether the input can rotate a full turn, and the solver then attempts all 180 positions and reports how many assembled. The prediction and the measurement agree in every case, which is what licenses quoting the classification for a mechanism nobody has swept.

Grashof is a shape test

Why Grashof's condition survives a scaling, what that means for measuring a four-bar, and why the classification should never be quoted without the distance to the boundary.

 Eight four-bars, one from each region the three signed sums cut. One linkage from each of the eight sign patterns of T₁ = g + c − a − b, T₂ = g + b − a − c and T₃ = b + c − a − g, each solved at a crank angle in the middle of its range. The thick arc round the left pivot is where the input's pin can go and the arc round the right pivot is where the output's can: crank-rocker + + +, input 100% of a turn and output 22%; double crank − − +, input 100% of a turn and output 100%; rocker-crank − + −, input 22% of a turn and output 100%; double rocker + − −, input 23% of a turn and output 21%; 0–π rocker + + −, input 76% of a turn and output 72%; π–π rocker + − +, input 84% of a turn and output 51%; π–0 rocker − + +, input 84% of a turn and output 85%; 0–0 rocker − − −, input 67% of a turn and output 91%. Grashof's condition names four of these and calls the other four one thing; the arcs show that the four triple rockers differ in which way along the ground line each rocker swings through.

Eight kinds of four-bar

The four-bar classified by the signs of three sums of its lengths: eight regions of length space, a census with no exception, two routes to what each grounded link can do, and why the walls between the regions are the change points.

 Every pivot on these arcs gives a 60° swing at a time ratio of 1.2. The rocker's pivot O₄ is fixed and its two limit positions B₁ and B₂ are 60° apart. At a limit the crank and coupler are in line, so the crank pivot sees the chord B₁B₂ at the angle δ = 180°(Q − 1)/(Q + 1) = 16.36°, and the points that see a chord at a fixed angle are arcs of two circles. The thick stretches are the 828 sampled pivots that give a crank-rocker with exactly this swing and ratio; the rest of each circle gives a linkage of another kind. The linkage drawn is the member whose worst transmission angle is largest, at both limits: ground 1.222, crank 0.478, coupler 1.130 and rocker 1, with its worst transmission angle 40.32°.

A swing and a time ratio

The classical crank-rocker design problem solved as a construction: the arc of crank pivots a swing and a time ratio allow, the member that keeps the best transmission angle, and the measured limit on the time ratio for each swing.

 Four offset slider-cranks, one from each region the limit leaves. One slider-crank from each region of T₂ = b − a + e and T₃ = b − a − e, with the slide the dashed vertical line a distance e from the crank's pivot and the ground line dashed across. The thick arc round the pivot is where the crank's pin can go and the thick stretch of the slide is where the slider can: crank-rocker + +, crank 1, rod 3.5, offset 0.8, the crank reaching 100% of a turn and the slider between 2.37 and 4.43 on each side of it; double rocker − −, crank 3.5, rod 1.2, offset 0.6, the crank reaching 22% of a turn and the slider between 2.22 and 4.66 on each side of it; 0–π rocker + −, crank 2, rod 2.5, offset 1.6, the crank reaching 65% of a turn and the slider running from −4.21 to 4.21 through the ground line; π–π rocker − +, crank 2, rod 2.5, offset −1.6, the crank reaching 65% of a turn and the slider running from −4.21 to 4.21 through the ground line. These are the four of the eight four-bar kinds in which T₁ is positive.

Four kinds of slider-crank

The offset slider-crank classified as the limit of the eight kinds of four-bar: the two sums that survive, the four regions they cut, a census and two routes to what each crank and slider can do, the slices of crank and rod length at an offset, and the length at which each vanishing four-bar kind folds onto a surviving one.

 A drag link driving a crank-rocker at a phase of 284.8°. A drag link, ground 1, crank 2.5, coupler 2.25, output 1.5, whose output crank carries the crank of a crank-rocker with a 60° swing, ground 1.2223, crank 0.4783, coupler 1.1298, output 1, turned 284.8° ahead of it, drawn at an input angle of 40°. The two cranks on the middle pivot are one rigid part. The thick arc at the right is the rocker's swing. At this phase the whole machine returns 2.71 times as fast as it works; the crank-rocker alone, driven at constant speed, returns 1.2 times as fast.

A drag link ahead of a crank-rocker

A six-bar made of a drag link driving the crank of a crank-rocker: the two-stage time ratio composed from the crank-rocker's limit positions and the drag link's output table, checked against the swept six-bar, its dependence on the phase between the stages, and a search of 821 drag links for pairs that reach a ratio of 2 with both stages keeping 40°.

 The chain's whole configuration space, drawn on its two angles. A slider-crank asks one thing of its two angles: a cos θ + b cos φ = e, with θ the crank's angle and φ the rod's. Each panel is the square of those two angles from −π to π, with the curve that equation cuts out, followed by arclength on the equation alone. The number under each panel is how many whole turns θ, φ and φ − θ make around a circuit. A curve that crosses the square from side to side carries a turn in that angle; a closed loop inside the square carries none. The two regions where a member turns have two circuits each and the two where none does have one, which is what the four-bar regions these came from predict.

Three rotations, and four benches

The slider-crank chain read as a chain: its whole configuration space as one equation on two angles, the six relative rotations of four links with one identically nought, winding numbers followed round every circuit over fifteen hundred chains, and the table of which of the four inversions can be driven continuously in each region — three of four, or none.

 How evenly the rocker is driven through the working stroke. The rocker's speed through the working stroke, divided by its mean over the stroke, against the fraction of the stroke's duration, for a crank-rocker with a 60° swing and a time ratio of 1.2 of its own, ground 1.2223, crank 0.4783, coupler 1.1298, output 1. Every curve must start and end at nought, because the rocker stops to reverse; what differs is the middle. The number after each name is the fastest speed over the slowest while the rocker covers the central 80% of its swing. Driven directly at constant speed the crank-rocker gives 1.91 at a time ratio of 1.20. The drag link that gives the highest time ratio, 2.71, gives 2.77: a hump in the middle of the cut. The most even design that still reaches 2, a drag link of ground 1, crank 5, coupler 4.5, output 2.25, gives 1.21 at a ratio of 2.03 — more even than the crank-rocker alone. Dragging moves that design's phase.

A quick return that cuts evenly

The rocker speed through the working stroke of a drag link driving a crank-rocker, measured over the central 80% of the swing as the ratio of fastest to slowest; the design with the highest time ratio against the most even design reaching a ratio of 2, over every drag link on the grid and every phase angle; the phase at which each specification peaks; the frontier of evenness against required ratio; and the three crank-rockers scored alone and with each drag link.

 A six-bar followed round one circuit, with the links that turn against the frame coloured. A Watt six-bar — a four-bar of ground 4, crank 3.17, coupler 3.10 and rocker 1.90, with a dyad of 3.83 and 3.03 hung from its rocker to a third ground pivot — followed once round one of its 4 circuits. Links drawn in ink come back to where they started; links drawn in colour make a whole turn against the frame. On this circuit that is rocker + dyad link + output, and they turn together, so every one of the fifteen relative rotations between pairs of links is either nought or that one turn. The pins' paths are dashed.

Fifteen pairs and one cut

Relative rotations in Watt and Stephenson six-bar chains: link windings followed round every circuit of a census of random six-bars, the fifteen pair rotations as differences of six integers, the cut between turning and returning links, drivable inversions as the links touching the cut (0, 3, 4 or 5 of 6), chains whose circuits disagree, and a Stephenson circuit with three winding levels.

 How evenly the ram is driven through the working stroke. The ram's speed through the working stroke of an offset slider-crank with a time ratio of 1.2 of its own, crank 1, rod 2.719, offset 0.847, divided by its mean over the stroke, against the fraction of the stroke's duration. The number after each name is the ratio of the fastest to the slowest ram speed while the ram covers the central 80% of its stroke. Driven directly at constant speed the slide gives 2.02. The drag link that gives the highest time ratio, 2.71, gives 2.60. The most even design that still reaches 2, a drag link of 4.75, 4, 2.25 on a ground of 1, gives 1.19 at a ratio of 2.03 — the same pattern the shaper's rocker showed, measured on a ram. Dragging moves that design's phase.

The first stage does not know what it drives

A drag link ahead of an offset slider-crank, with the ram's speed measured over the central 80% of its stroke; the single-stage quick return each machine gives before its transmission angle falls to 40°; the highest-ratio and most even designs over 821 drag links and every phase angle between the stages; the working-stroke shapes of rocker and ram at equal own ratios; six drag links tried on six outputs; the ram speed split into its two factors; and the phase shift between rocker and ram against the angle between their stroke starts.

 Shortening a centred slide's rod evens the dragged ram until the rod is about two cranks long, and then it stops helping. A centred slider-crank with crank 1 and its rod swept from 20 cranks down to 1.305, where its worst transmission angle reaches 40°, plotted against crank ÷ rod. For each rod, the ratio of the ram's fastest to slowest speed across the central 80% of its working stroke: driven directly, which worsens from 1.701 to 2.752; and with the most even of 821 drag links, at whichever phase angle between the stages suits it, that still gives a combined time ratio of 2, which improves from 1.435 to 1.246 at a rod of 2.0 (worst angle 60°) and then worsens again to 1.401 at the 40° limit. Dashed: the best design found on any offset slide, 1.181, at an offset of 0.6 and a worst angle of 65°.

A centred slide has a best rod, and an offset beats it

A centred slider-crank behind a drag link at a combined time ratio of 2, with its rod swept from 20 cranks to the 40° limit: the most even cut over 821 drag links and every angle between the stages at fifteen rods; the ram's speed through the stroke at three of them; the own speed peak's position and height; the ram's speed split into the crank's and the lever's; and offset slides at four offsets and four transmission angles.

 A centred slide's time ratio is its drag link's split of half a turn. The output angle of a drag link against its input angle over one turn: ground 1, input crank 4.019, coupler 3.920, output crank 1.339, the double crank that splits half a turn of its output most unevenly while keeping a worst transmission angle of 40.0°. A centred slider-crank driven by its output reaches its two dead centres at output angles exactly half a turn apart, whatever its rod, so its forward and return strokes take the input angles between the two dotted levels and outside them. At a phase of 66.3° between the stages they are 104.5° and 255.5° of input, a time ratio of 2.443. The largest over every phase angle is 2.443, against 2.304 for the best link of the quarter-step grid: no rod can raise it.

A centred ram cannot be timed past its drag link

A drag link ahead of a slider-crank, asked for combined time ratios from 1.6 to 2.5: the drag link's output against its input with a centred slide's two dead centres half a turn apart; the largest combined ratio on the grid against the slide's worst transmission angle at five offsets; the centred slide's most even train against crank ÷ rod at four ratios; the best centred and best offset trains against the ratio; how many trains reach each ratio; and the ram's speed through the stroke at a ratio of 2.2.

 A ratio of 3 costs both stages forty degrees. The largest quick-return ratio of a shaper's drive against the worst transmission angle it must keep, from 30° to 60°: an offset slider-crank alone, 1.448, 1.360, 1.286, 1.224, 1.172, 1.128, 1.092; a drag link alone ahead of a centred slide, 2.867, 2.632, 2.443, 2.284, 2.130, 1.968, 1.809; and the two in series, both stages keeping the angle, 3.975, 3.427, 3.006, 2.696, 2.432, 2.184, 1.957, against the product of the separate ceilings, 4.153, 3.580, 3.143, 2.796, 2.496, 2.221, 1.976. The train reaches 3 at 40.0°. Marked: 40°, where the train reaches 3.006.

A ratio of three costs forty degrees

The quick-return frontier of a drag link ahead of an offset slider-crank: the largest ratio against the worst transmission angle each stage keeps, for each stage alone, for both in series and for the product; the train's ceiling as a share of the product; the slide's largest angle at a ratio of 3 against the drag link's; the best train at each angle; the drag link's thirteen searched summits at 40°; and the ram's motion through a turn at a ratio of 3.

 Five degrees below forty buy back an even cut at a ratio of three. At each worst transmission angle kept by both stages of a drag link ahead of an offset slide: the evenness of the cut, the ram's fastest speed over its slowest across the middle four fifths of the working stroke, for the train with the largest time ratio at that angle (dashed), which reaches 3.98, 3.75, 3.52, 3.43, 3.33, 3.24, 3.16, 3.08, 3.01 and cuts at 6.11, 5.21, 4.70, 4.35, 4.14, 3.90, 3.65, 3.46, 3.25; and for the most even train that still reaches a ratio of 3 (solid): 1.22, 1.26, 1.32, 1.42, 1.64, 1.90, 2.53, 2.73, 2.97, at 30, 32, 34, 35, 36, 37, 38, 39, 40°. The shaded band is 1.12 to 1.47, the best ratio-2 trains of the earlier essays.

Five degrees buy an even cut at a ratio of three

A drag link ahead of an offset slider-crank held at a quick-return ratio of 3: the evenness of the cut against the worst transmission angle kept in both stages, for the train with the largest ratio and for the most even train; the ram's speed across its cut at five angles; the ram's position through a turn for the even trains at 35° and 40°; the drag link's three lengths along the even frontier; and the trains in a table.

All essays