Lecture 1: Analysis of Mechanisms

[For Illustration Only]

Mechanical design is concerned with machines that convert, transmit, and control motion under real constraints of force, speed, size, and durability. Even very early engines required engineers to manage large moving components, repetitive loading, and useful power output, so the basic questions of mechanism analysis have been central to engineering for a long time.

Modern engines show how demanding mechanical design becomes when high power must be delivered from a compact package. Large rotational speeds, high piston accelerations, elevated temperatures, and tight packaging all act together, so links, joints, bearings, shafts, and surrounding structures must be designed for both performance and survivability.

At the highest-performance end of engineering, mechanism geometry and inertia effects become even more critical. Extremely high speed and acceleration produce severe dynamic loads, which means that small geometric decisions can strongly influence stress, vibration, efficiency, and service life.

Mechanical design is not limited to engines. The same underlying ideas appear in vehicles, robotic systems, aerospace hardware, consumer products, pumps, compressors, and other machines wherever motion must be generated, constrained, transmitted, or used to produce useful actuation.

The main topics are degree of freedom, types of motion, links and joints, kinematic diagrams, and mobility calculation. Together, these ideas provide a language for describing planar mechanisms, reducing real hardware to an analysable form, and deciding whether a proposed assembly can move as intended.

Mobility, or degree of freedom, is the number of independent parameters required to define the position of a body or mechanism at an instant. In mechanism design, the same quantity can also be understood as the number of independent inputs needed to produce a predictable motion.

This idea is built on the rigid-body assumption, in which deformation is neglected so that motion can be described through geometry and constraint rather than elasticity. Once links are idealized as rigid, the central question becomes how joints and supports remove possible motions.

A useful way to understand degree of freedom is to compare systems with $1$, $2$, and $3$ independent motions. A one-degree-of-freedom system has a single input variable that determines configuration, while higher-degree systems require additional independent coordinates to describe their motion fully.

Independent motions must be genuinely separate. If one coordinate can be found from another because a joint or a support imposes a constraint, it is not an additional degree of freedom.

A rigid body in a plane has $3$ degrees of freedom, typically two translations and one rotation, while a rigid body in space has $6$ degrees of freedom, typically three translations and three rotations. For a planar body these are often written as motion in $x$, motion in $y$, and rotation by an angle $\theta$.

This distinction matters because most of the course examples we consider are planar, so their mobility counting follows the planar case rather than the full spatial one. Every unconstrained planar link begins with three possible motions before joints and the ground remove some of them.

Real systems often combine translation and rotation in ways that depend strongly on their constraints. A vehicle confined to a plane, a ship moving with roll and pitch, and a spatial body moving freely in three dimensions all have different motion because their environments and supports remove some coordinates while allowing others.

The constraint, not merely the shape of the body, determines the available motion. The same rigid body can have very different degrees of freedom when it is free, guided in a slot, pinned to ground, or brought into contact with another body.

Planar rigid-body motion can be described as pure rotation, translation, or general complex motion. Pure rotation occurs when one point remains fixed relative to the reference frame and all other points move on circular arcs about it. Translation occurs when all points move along parallel paths, so the body’s orientation does not change even though its position does.

Complex motion combines simultaneous translation and rotation, so both position and orientation change together. This is the usual case for coupler links in mechanisms. Although the motion may appear complicated, it can still be analyzed through the same rigid-body principles once the grounded points, moving links, and joint constraints have been identified clearly.

A link is a rigid body that contains at least two nodes where connections to other links may be made. A node is a potential connection point, so the number of nodes determines how many kinematic relationships that link can participate in.

Classification by link order follows naturally from the number of nodes: binary links have two, ternary links have three, and quaternary links have four. Link order becomes important in both mechanism description and mechanism synthesis because it controls both connectivity and joint count.

A joint is the connection between two or more links that allows relative motion, or potential relative motion, between them. The term kinematic pair is commonly used because the analysis of mechanisms depends on the relative constraint imposed at the connection rather than on the detailed hardware alone.

In practice, a joint is understood by asking which relative motions remain possible after contact is made. The allowed motion is what defines the pair.

Joints may be classified by the number of degrees of freedom they permit between the connected links. In planar mechanism theory, a one-degree-of-freedom joint is often called a full joint, while a two-degree-of-freedom joint is called a half joint.

A full joint removes two relative motions between planar links, while a half joint removes only one. This difference is the reason the two joint types appear with different coefficients in mobility equations.

Different joint geometries admit different combinations of translation and rotation. Revolute joints permit angular motion about one axis, prismatic joints permit linear motion along one axis, and more complex joints such as cylindrical, helical, planar, or spherical joints allow additional combinations.

A pin-in-slot is a useful example because it leaves both sliding and relative rotation available, so it has two relative degrees of freedom. A simple pin joint leaves only rotation, and a slider leaves only translation.

Joint order is defined as the number of links joined at a connection minus one. This classification distinguishes an ordinary connection between two links from multiple joints where three or more links meet at the same location.

In counting, a pin joining three links is not treated as a single ordinary pin joint. It behaves like two first-order joints sharing one center, so multiple joints must be represented carefully in the kinematic diagram before mobility is calculated.

More generally, if $n$ links meet at one joint center, the connection contributes the equivalent of $n-1$ first-order joints. This is a counting rule, not a statement that there are physically separate pins.

A mechanism is a system arranged to transmit motion in a predetermined manner. A machine extends that idea by transmitting both motion and energy so that useful work is performed. The distinction is important because some assemblies are analysed mainly for motion generation, while others must also be evaluated for power flow, force transmission, and work output.

The boundary between mechanism and machine is not purely visual. An assembly may contain links and joints in either case, but the classification depends on whether the arrangement is considered mainly as a motion-transforming structure or as a system that delivers useful energy transfer and work through its motion.

The same linkage may therefore be studied in two ways. Kinematic analysis asks how motion is constrained and transmitted, while machine analysis also asks how forces, torque, power, and actuation are transmitted through the same geometry.

Kinematic analysis begins by replacing a physical machine with a simplified schematic of links and joints. A good kinematic diagram removes unnecessary shape detail while preserving the connectivity, joint types, and relative motions that determine how the mechanism behaves.

Actual contour and kinematic contour are not always the same. The shape drawn in the diagram is chosen to express motion and connectivity clearly, not to reproduce the manufactured outline.

Fasteners, covers, decorative parts, and other features that do not affect the kinematic constraint structure are normally omitted. What remains is only the set of rigid bodies and joints that determine motion.

A complicated physical device can often be analysed only after it has been reduced to its kinematic structure. The important features are the grounded links, moving links, pivot locations, and the points where forces are applied. Once those have been identified, the mechanism can be studied without carrying along the full product geometry.

Bringing all links into a common plane is often part of this abstraction. Even if the actual hardware is offset in depth, its motion may still be treated as planar if the joints constrain the relative motion to one effective plane.

Drawing a kinematic diagram requires identifying the ground link, the moving links, the joint types, and the path by which input motion is transformed into output motion. In a foot-operated pump, the mechanism converts the motion of the pedal into reciprocating motion of the pumping element through a constrained linkage arrangement.

If the diagram is inaccurate, the mobility count will also be wrong. Clear numbering of links and clear marking of shared joints are essential because the later calculation depends entirely on the abstracted model rather than on the appearance of the product.

A practical method is to identify the fixed frame first, then mark each distinct moving rigid body once, and only then classify the joints between them. This reduces the risk of double-counting parts that belong to the same rigid link.

Mobility is calculated using Gruebler’s equation for planar mechanisms, \(M = 3(L - 1) - 2J_1 - J_2,\) where $M$ is the degree of freedom, $L$ is the number of links, $J_1$ is the number of one-degree-of-freedom joints, and $J_2$ is the number of two-degree-of-freedom joints. The ground is counted as one link, so the fixed frame or base must be included in the link count.

This expression can be understood by starting with $3$ possible motions for each planar link and then subtracting the motions removed by joints and grounding. A full joint removes two relative motions, a half joint removes one, and fixing one link to ground removes the three motions that link would otherwise possess.

When $M=1$, one input is sufficient to determine the entire configuration. When $M>1$, the chain still has mobility, but more than one independent input or coordinate is needed to specify its position completely.

The sign of $M$ determines the basic nature of the assembly. If $M>0$, at least one independent motion is available and the assembly behaves as a mechanism. If $M=0$, no motion is available and the assembly behaves as a structure. If $M<0$, the constraints are excessive and the assembly is preloaded, so internal stresses are introduced during construction or closure.

Exact constraint corresponds to the case where the number of constraints is just sufficient for the intended behaviour. When more constraints than necessary are imposed, the system becomes sensitive to manufacturing error or geometric inconsistency because closure may require elastic deformation.

Mobility calculation depends first on correct counting. The number of links, the number of full joints, and the number of half joints must all be identified accurately before the resulting value of $M$ can be interpreted. Small counting errors can change a mechanism into a structure on paper, even when the physical intention is clear.

Any external body that becomes kinematically engaged may need to be counted as an additional link. If a gripped object, contact surface, or slider block constrains motion, it is no longer merely an external load; it has become part of the kinematic model.

This also means that mobility can change when contact is created or removed. A device may have one degree of freedom before contact and become a zero-degree-of-freedom structure after engaging another body if the new contact introduces sufficient additional constraints.

For $L=8$, $J_1=10$, and $J_2=0$, \(M = 3(8-1) - 2(10) - 0 = 1.\) This result corresponds to a mechanism with one independent input. Many practical planar machines are designed this way because a single prescribed motion is sufficient to determine the configuration of the entire assembly.

Multiple joints do not invalidate the method, but they must be counted correctly. A pin shared by three links contributes more constraint information than a simple two-link pin joint and must be represented accordingly in the link-and-joint count.

When a half joint is present, the mobility count must include its weaker constraint contribution explicitly. With $L=6$, $J_1=7$, and $J_2=1$, \(M = 3(6-1) - 2(7) - 1 = 0,\) so the assembly is classified as a structure rather than as a moving mechanism.

This example shows why a half joint is not counted the same way as a full joint and why multiple joints often make the bookkeeping look unusual. Fractional totals can appear in intermediate joint counts when equivalent first-order joints are used to represent a shared connection.

Worked mobility determination requires direct recognition of grounded links, full joints, half joints, and multiple joints from the geometry of the chain. The physical appearance of a mechanism may vary considerably, but the counting logic remains the same.

The safest sequence is to draw or imagine the kinematic diagram first, count links second, classify joints third, and substitute into the mobility equation last. Performing the count directly from a complicated product image is more error-prone than counting from the abstracted mechanism.

For example (a), the counts $L=6$, $J_1=7$, and $J_2=1$ lead to \(M = 3(6-1) - 2(7) - 1 = 0,\) so the assembly is a structure. Apparent complexity does not guarantee mobility; the decisive factor is the balance between available motion and imposed constraint.

The count also illustrates how a multiple joint can coexist with an overall zero-degree-of-freedom structure. Counting must follow constraint logic rather than intuition about how complicated the figure looks.

Examples (b) and (c) both evaluate to a single degree of freedom. For (b), $L=3$, $J_1=2$, and $J_2=1$ give $M=1$; for (c), $L=4$, $J_1=4$, and $J_2=0$ also give $M=1$. Different layouts can therefore be kinematically equivalent even when their shapes and joint arrangements are not visually similar.

Equivalent mobility does not imply identical motion paths or identical force transmission. It means only that the number of independent inputs is the same.

Mechanism analysis in planar systems rests on a connected chain of ideas. Degree of freedom defines how much independent motion exists, motion types describe how bodies move, links and joints define the physical constraint structure, kinematic diagrams abstract the system, and mobility equations convert that structure into a quantitative result.

Understanding mobility requires both conceptual definition and repeated calculation. The terminology of links, joints, and motion must remain consistent across worked examples, otherwise the final degree-of-freedom count loses physical meaning.

In practice, the most common sources of error are omitting the ground link, misclassifying a half joint as a full joint, and overlooking the special treatment required by a multiple joint or by a constrained external body.

Contemporary robotic systems still rely on the same underlying concepts. Whether the system is an industrial manipulator, a mobile platform, or a legged machine, its motion is still determined by the arrangement of rigid links, joints, constraints, and actuation.

[For Illustration Only]

[For Illustration Only]

[For Illustration Only]

[For Illustration Only]

The second part extends basic mobility analysis into number synthesis, paradoxes, and linkage transformation. The emphasis shifts from defining mechanism structure to reasoning about how that structure can be generated, modified, or misinterpreted.

Number synthesis is the process of determining how many links and joints, and of what order, are required to produce a mechanism with a specified degree of freedom. This requires consistent counting of binary, ternary, quaternary, and higher-order links according to the number of nodes each link contains.

The purpose is not to solve dimensions immediately, but to identify which connection patterns are even possible before detailed geometry is chosen.

Mechanism synthesis can be organized through a counting framework: \(L = B + T + Q + P + H,\) \(J = \frac{2B + 3T + 4Q + 5P + 6H}{2},\) and, for planar full-joint chains, \(M = 3(L - 1) - 2J.\) Combining these relations gives \(L - 3 - M = T + 2Q + 3P + 4H,\) and these relations identify feasible combinations of link orders for a desired degree of freedom and show how joint count is tied directly to link order distribution.

For planar chains containing only full joints, mobility and link count have opposite parity: odd mobility corresponds to an even number of links, while even mobility corresponds to an odd number of links. That observation helps narrow the set of feasible linkage structures before any geometry is drawn.

Gruebler’s equation is a structural count, not a complete geometric test. Because it ignores detailed dimensions and special configurations, it can predict mobility that is not actually realized. Mechanism paradoxes arise when link proportions or geometric coincidences impose additional constraints, or preserve motion in unexpected ways, that are invisible to the counting formula alone.

Several transformation rules are useful when comparing related planar chains. Replacing a revolute joint in a loop by a prismatic joint can leave mobility unchanged, provided sufficient rotational joints remain. Replacing a full joint by a half joint increases mobility by $1$, while removing a link decreases mobility by $1$.

These rules are useful because they separate kinematic equivalence from physical appearance. Two mechanisms can look different yet impose the same number of constraints.

Further transformation rules show how combined changes may preserve mobility. Replacing a full joint by a half joint and then removing a link can leave the degree of freedom unchanged. Shrinking a higher-order link by coalescing nodes can create multiple joints without necessarily changing mobility, although complete shrinkage is equivalent to link removal and therefore reduces mobility.

This is one reason higher-order links must be recognized correctly. Their role in a chain is not determined by shape alone, but by the number and arrangement of nodes that survive the abstraction.

A Grashof crank-rocker and a Grashof slider-crank can share the same mobility even though one contains only revolute pairs and the other contains a prismatic pair. The key point is that equivalent constraint structure, rather than identical physical appearance, governs the degree of freedom.

Applying the relevant transformation sequence changes the description of links and joints, but the mechanism remains equivalent in mobility: \(M = 1,\) so the same single independent motion is retained after transformation.

[For Illustration Only]

[For Illustration Only]

[For Illustration Only]

Knowledge Check