2.1.2 · D5Analytical Mechanics
Question bank — Generalized coordinates — choosing them, degrees of freedom
True or false — justify
A generalized coordinate must have units of length.
False. A coordinate need only be independent and complete — an angle (radians), a ratio, or even a charge qualifies. See Configuration space and phase space where axes routinely mix dimensions.
Every holonomic constraint removes exactly one degree of freedom.
True, provided it is independent of the others. An equation lets you solve for one variable in terms of the rest, killing one number.
A time-dependent (rheonomic) constraint adds a degree of freedom because time can vary.
False. Time is a parameter, not a coordinate you may freely dial; the constraint still removes one DOF. See Constraints — holonomic vs non-holonomic.
Writing three separate constraint equations always removes three degrees of freedom.
False. Only independent equations count; if one follows from the others it removes nothing.
A free rigid body in space has 6 degrees of freedom regardless of how many atoms it contains.
True. Rigidity fixes all inter-particle distances, leaving 3 for position and 3 for orientation — see Rigid body kinematics — Euler angles.
The number of generalized coordinates always equals .
False in general. That formula holds for holonomic systems; non-holonomic (velocity) constraints can restrict motion without reducing the count of position coordinates.
A rigid diatomic molecule modelled as two point atoms has 6 DOF.
False — it has 5. The bond-length equation removes one, and spinning about the bond axis moves nothing for point atoms, so that "rotation" is not a DOF.
Choosing for a pendulum instead of changes the number of degrees of freedom.
False. DOF is a property of the system, not the coordinate choice; both describe a 1-DOF system, but bakes in the constraint so it is cleaner.
Two beads on a straight wire (in 3D) have 2 degrees of freedom.
True. Each bead's confinement to the axis is two equations (), so and .
Adding a spring between two constrained beads changes their degrees of freedom.
False. A spring is a force, not a constraint; it changes dynamics and energy but leaves the count of coordinates untouched.
Spot the error
"A bead on a rotating wire has 2 DOF because its position needs and the wire's angle."
The wire's angle is forced to be — it is not free. Only is independent, so the system has 1 DOF; the angle is set by time.
"The pendulum constraint is , and the ground is at , so and ."
The ground plays no role for a swinging bob — there is no constraint. Only the rod length constrains it, so and .
"A disk rolling without slipping on a line loses one DOF, so use ."
Pure rolling is typically non-holonomic — it constrains (a velocity relation) that cannot be integrated to . The configuration space keeps both and ; see Constraints — holonomic vs non-holonomic.
"A rigid body needs coordinates minus rigidity constraints, and , so it has infinitely many DOF."
The constraints also grow without bound and cancel all but 6. The correct count is 6, found by direct reasoning (position + orientation), not by naive subtraction.
"For a molecule of three non-collinear point atoms, (three bond lengths) so ."
That is actually correct for three general fixed distances — (3 translation + 3 rotation). The trap is assuming it "should" be 5 like the dumbbell; three non-collinear atoms have a genuine third rotational DOF.
"Since appears in , time is one of the generalized coordinates."
Time appears explicitly only because the constraint is rheonomic; is the independent parameter of evolution, never a you vary at fixed instant.
Why questions
Why can two atoms of a dumbbell not use a third orientation angle?
Rotating two point atoms about the line joining them produces an identical configuration — the "spin" angle is physically invisible, so it is not an independent coordinate.
Why is baking the constraint into the coordinate (like ) considered the "smart" choice?
Because the constraint equation then holds automatically for every value of the coordinate, so it disappears and you work only with truly free variables — feeding cleanly into Lagrangian mechanics — the Lagrangian L = T - V.
Why must we check independence before subtracting constraints?
A dependent constraint carries no new information; subtracting it double-counts a restriction and gives a wrong (too small) DOF.
Why do we count for a rigid body instead of term by term?
Because the number of rigidity relations is enormous and highly redundant; reasoning "where + how oriented" directly yields the 6 independent numbers without wrestling the redundancy.
Why does a rheonomic constraint still remove a DOF even though depends on ?
At any fixed instant it is still one equation relating the coordinates, so it pins one variable; time only shifts which configuration is allowed, not how many are free.
Why is the choice of center-of-mass and separation preferred for two spring-linked beads?
The spring energy depends only on , so the equations decouple — evolves freely while carries all the interaction, simplifying the Euler–Lagrange equations.
Why does a non-holonomic constraint break the tidy counting?
It restricts velocities (differential relations) that can't be integrated into position equations, so it forbids certain motions without forbidding any configuration — the position-coordinate count stays high.
Edge cases
What is the DOF of a single free particle in 3D with no constraints?
3 — the raw ; there is nothing to subtract, .
What is the DOF of a particle pinned to a single fixed point (all three coordinates fixed)?
0. Three independent constraints fully lock it: , a degenerate "system" that cannot move.
A bead constrained to a surface (one equation ) in 3D — how many DOF?
2. One holonomic equation removes one number from three, leaving a 2-parameter surface to roam.
What if you write as a "second" pendulum constraint?
It is algebraically the same equation scaled — not independent — so it removes nothing; still .
Can a system have zero generalized coordinates yet still be a valid configuration?
Yes — a fully constrained (rigidly fixed) system has ; it has exactly one allowed configuration and no freedom to move.
Two coincident particles forced to occupy the same point (, three equations) — DOF?
. They move together as one point; the three equality constraints collapse six numbers to three.
A pendulum whose length grows as — how many DOF?
Still 1. The constraint is rheonomic but holonomic, removing one DOF; the angle remains the sole free coordinate.
Connections
- Constraints — holonomic vs non-holonomic
- Configuration space and phase space
- Lagrangian mechanics — the Lagrangian L = T - V
- Euler–Lagrange equations
- Rigid body kinematics — Euler angles
- Principle of virtual work and d'Alembert's principle