Home / Courses / GATE Physics / Practice / Phase space, fixed points and stability

Phase space, fixed points and stability

Classical mechanics (GATE and NET) · 67 questions and templates for GATE Physics, in order from warm-up to top rank. Check each answer, then read the full solution.

Loading questions… (they need JavaScript; nothing you type is sent anywhere)

Written questions with solutions

The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.

  1. MCQWarm-up

    For the one-dimensional flow x˙=f(x), a fixed point x∗ (where f(x∗)=0) is stable if
    1. f(x∗)>0
    2. f′(x∗)<0
    3. f′(x∗)>0
    4. f′′(x∗)=0
    Answer and solution

    Answer: B

    Near x∗, write x=x∗+η: η˙≈f′(x∗)η, so η∝ef′(x∗)t. It decays when f′(x∗)<0: the flow points towards x∗ from both sides. If f′(x∗)=0 the linear test is inconclusive.
  2. MCQWarm-up

    The Jacobian at a fixed point of a planar system has two real, negative, distinct eigenvalues. The fixed point is
    1. a saddle
    2. an unstable node
    3. a stable node
    4. a centre
    Answer and solution

    Answer: C

    Both eigen-directions decay exponentially without rotation, so every nearby trajectory runs straight into the point: a stable node. Opposite signs give a saddle, both positive an unstable node, and purely imaginary eigenvalues a centre.
  3. MCQWarm-up

    The linearisation at a fixed point has eigenvalues λ=±2i. For the linear system, the fixed point is
    1. a stable spiral
    2. a saddle
    3. an unstable spiral
    4. a centre
    Answer and solution

    Answer: D

    Purely imaginary eigenvalues give e±2it: closed elliptical orbits circling the point, neither growing nor decaying. For the full nonlinear system a linear centre may become a weak spiral; conservative (Hamiltonian) systems keep it a true centre.
  4. NATWarm-up

    The flow x˙=x2−4 has two fixed points. The stable one is at x= ______.
    Answer and solution

    Answer: -2

    Fixed points: x=±2. f′(x)=2x: at x=−2, f′=−4<0, stable; at x=2, f′=4>0, unstable. Picture: f<0 between the roots (flow to the left) and f>0 outside, so trajectories gather at −2.
  5. MCQWarm-up

    Two distinct trajectories in the phase space of an autonomous system x→˙=f→(x→) (smooth f→) can never cross. Why?
    1. A crossing point would have two different futures, violating uniqueness of solutions
    2. Energy is always conserved
    3. Phase space is always two-dimensional
    4. All trajectories are closed
    Answer and solution

    Answer: A

    Through each point the velocity f→(x→) is fixed, and the existence-uniqueness theorem gives exactly one solution from each starting point. Two trajectories through one point would be the same trajectory. (They can approach the same fixed point as t→∞, which is not a crossing.)
  6. MCQWarm-up

    The phase-space trajectories (x,p) of an undamped harmonic oscillator are
    1. spirals into the origin
    2. ellipses centred on the origin
    3. straight lines through the origin
    4. hyperbolas
    Answer and solution

    Answer: B

    Energy conservation, p22m+12kx2=E, is the equation of an ellipse with semi-axes 2E/k and 2mE. Damping turns the ellipses into inward spirals; hyperbolas belong to an inverted oscillator (a saddle).
  7. NATWarm-up

    The flow x˙=sin⁡x has exactly one stable fixed point between x=1 and x=5. It is at x= ______ (to two decimal places).
    Answer and solution

    Answer: 3.14 (±0.01)

    Fixed points: sin⁡x=0, so x=0,π,2π,…; only π lies between 1 and 5. Stability: f′(x)=cos⁡x, and f′(π)=−1<0, so it is stable. (Just below π, sin⁡x>0 pushes x up; just above, sin⁡x<0 pushes it down.)

    Why people get this wrong: x=0 and 2π have f′=+1: they are unstable, and they lie outside the range anyway.

  8. MCQStandard

    The fixed point at the origin of x˙=x+2y, y˙=3x+y is
    1. a saddle
    2. a stable node
    3. a centre
    4. an unstable spiral
    Answer and solution

    Answer: A

    A=(1231): Δ=1−6=−5<0. A negative determinant means real eigenvalues of opposite sign (λ=1±6), whatever the trace: a saddle.
  9. NATStandard

    For x→˙=Ax→ in the plane, with τ=trA and Δ=det⁡A: the eigenvalues are λ=12(τ±τ2−4Δ). For A=(−14−2−1), the eigenvalues are −1±iω. The value of ω is ______.
    Answer and solution

    Answer: 2.828 (±0.01)

    τ=−2, Δ=1+8=9, τ2−4Δ=−32. So λ=12(−2±i32)=−1±22i, ω=2.828: a stable spiral turning at angular frequency 22.
  10. MCQStandard

    For the undamped pendulum θ¨=−ω02sin⁡θ, the fixed point (θ,θ˙)=(π,0) (pendulum upright) is
    1. a centre
    2. a stable spiral
    3. a saddle
    4. an unstable node
    Answer and solution

    Answer: C

    Put θ=π+ϕ: sin⁡θ≈−ϕ, so ϕ¨=+ω02ϕ, eigenvalues ±ω0, opposite signs: a saddle. Its incoming and outgoing branches form the separatrix that divides swinging from whirling motion. The hanging position θ=0 is a centre.
  11. MCQStandard

    For the underdamped oscillator x¨+2γx˙+ω02x=0 with 0<γ<ω0, the origin of the (x,x˙) plane is
    1. a stable spiral
    2. a stable node
    3. a centre
    4. a saddle
    Answer and solution

    Answer: A

    Eigenvalues λ=−γ±iω02−γ2: negative real part and non-zero imaginary part, so trajectories spiral inwards. Overdamping (γ>ω0) gives a stable node, critical damping a degenerate node.
  12. NATStandard

    A particle of unit mass moves in V(x)=x4−2x2. The angular frequency of small oscillations about either minimum is ______.
    Answer and solution

    Answer: 2.828 (±0.01)

    V′(x)=4x3−4x=0 at x=0,±1. V′′(x)=12x2−4: at x=±1, V′′=8>0 (minima, centres); at x=0, V′′=−4 (a maximum, a saddle). ω=V′′/m=8≈2.83.
  13. MSQStandard

    For a Hamiltonian flow in phase space, which statements are correct?
    1. The volume of any region of phase space is preserved as it moves
    2. There are no asymptotically stable fixed points or attracting limit cycles
    3. The divergence of the phase-space velocity field is zero
    4. Two trajectories can merge into one after a finite time
    Answer and solution

    Answer: A, B, C

    ∇⋅(q˙,p˙)=∂2H/∂q∂p−∂2H/∂p∂q=0, so volumes are preserved (Liouville). An attractor would shrink a ball of initial conditions onto a point or curve, which a volume-preserving flow cannot do. Trajectories never merge (uniqueness).
  14. MCQStandard

    For a particle in a one-dimensional potential V(x), a fixed point at a local maximum of V is, in the (x,p) plane,
    1. a centre
    2. a stable node
    3. a stable spiral
    4. a saddle
    Answer and solution

    Answer: D

    Linearising mx¨=−V′(x) about x∗ gives mη¨=−V′′(x∗)η. At a maximum V′′<0, so η grows or decays as e±|V′′|/mt: a saddle. At a minimum V′′>0 gives a centre. Without friction there are no nodes or spirals.
  15. NATStandard

    The flow x˙=4x−x3 has three fixed points. The number of them that are stable is ______.
    Answer and solution

    Answer: 2

    Fixed points x=0,±2. f′(x)=4−3x2: f′(0)=4>0 (unstable), f′(±2)=−8<0 (stable). So 2 stable points; this is the supercritical pitchfork above threshold.
  16. MCQStandard

    As r passes through 0, the flow x˙=r−x2 undergoes
    1. a pitchfork bifurcation
    2. a saddle-node bifurcation
    3. a transcritical bifurcation
    4. a Hopf bifurcation
    Answer and solution

    Answer: B

    For r>0 there are two fixed points x=±r (the positive one stable, the negative one unstable); they approach each other, merge at r=0 and disappear for r<0. Fixed points created or destroyed in pairs: saddle-node (fold) bifurcation.
  17. MCQStandard

    The fixed point x=0 of the flow x˙=x2 is
    1. stable
    2. unstable from both sides
    3. half-stable: it attracts from one side and repels on the other
    4. a centre
    Answer and solution

    Answer: C

    For x<0, x˙=x2>0, so x rises towards 0; for x>0, x˙>0 too, so x moves away. The point attracts from the left and repels on the right. Here f′(0)=0, so linear analysis is silent and the sign of f on each side decides.

    Why people get this wrong: f′(0)=0 does not mean "stable": the next term in the expansion decides.

  18. MSQStandard

    For a smooth one-dimensional autonomous flow x˙=f(x), which statements are correct?
    1. No solution can oscillate
    2. Every solution is monotonic in time (or constant)
    3. If all fixed points are non-degenerate (f′≠0 there), neighbouring fixed points have opposite stability
    4. A solution can overshoot a stable fixed point and then come back to it
    Answer and solution

    Answer: A, B, C

    x˙ has the sign of f(x), and x cannot cross a zero of f (uniqueness), so each solution moves one way only: no oscillation, no overshoot. With simple zeros, f changes sign at each fixed point, so the slopes f′ alternate in sign along the line: stable and unstable points alternate.

    Why people get this wrong: Overshoot and oscillation need at least two dimensions (for example, position and velocity).

  19. NATExam level

    The system x˙=y, y˙=−x+x3 has a fixed point at (1,0). The positive eigenvalue of the Jacobian there is ______.
    Answer and solution

    Answer: 1.414 (±0.005)

    Jacobian (01−1+3x20); at x=1 it is (0120), so λ2=2, λ=±2: a saddle. (The potential is V=x2/2−x4/4, with a maximum at x=±1.)
  20. MCQExam level

    The origin of x˙=−x+y, y˙=−x−y is
    1. a stable node
    2. a stable spiral
    3. a centre
    4. a saddle
    Answer and solution

    Answer: B

    τ=−2, Δ=(−1)(−1)−(1)(−1)=2. τ2−4Δ=4−8=−4<0: complex eigenvalues −1±i with negative real part, a stable spiral. In polar form r˙=−r, θ˙=−1: it spirals clockwise into the origin.
  21. NATExam level

    For x¨+bx˙+9x=0, the value of b at which the origin changes from a stable spiral to a stable node is ______.
    Answer and solution

    Answer: 6

    Eigenvalues λ=12(−b±b2−36). They are complex for b<6 (spiral) and real for b>6 (node). At b=6 (critical damping) they coincide: a degenerate node.
  22. MCQExam level

    As r increases through 0, the flow x˙=rx−x2
    1. loses both its fixed points
    2. gains two new fixed points
    3. keeps two fixed points, x=0 and x=r, which exchange stability
    4. develops a limit cycle
    Answer and solution

    Answer: C

    Fixed points x=0 and x=r for every r. f′(x)=r−2x: f′(0)=r and f′(r)=−r. For r<0, x=0 is stable; for r>0, x=r is stable. They pass through each other at r=0 and swap stability: a transcritical bifurcation (as at a laser threshold).
  23. NATExam level

    A pendulum with ω0=g/l=3 rad/s moves on its separatrix (it just reaches the upright position). Its angular speed as it passes the lowest point is ______ rad/s.
    Answer and solution

    Answer: 6 rad/s (±0.01)

    Per unit ml2, E=12θ˙2−ω02cos⁡θ. On the separatrix E equals the energy at the top, +ω02. At the bottom: 12θ˙2−ω02=ω02, so θ˙=2ω0=6 rad/s. The separatrix is θ˙=±2ω0cos⁡(θ/2).
  24. NATExam level

    For x→˙=Ax→ with A=(2112), the larger eigenvalue is ______.
    Answer and solution

    Answer: 3

    τ=4, Δ=3, λ=12(4±16−12)=3,1, with eigenvectors (1,1) and (1,−1). Both positive: an unstable node, with trajectories leaving fastest along (1,1).
  25. MCQExam level

    For a critically damped oscillator, the fixed point at the origin of the (x,x˙) plane is
    1. a stable spiral
    2. a centre
    3. a degenerate stable node (a repeated eigenvalue with one eigenvector)
    4. a saddle
    Answer and solution

    Answer: C

    At critical damping λ=−ω0 twice, but the matrix (01−ω02−2ω0) has only one eigenvector, (1,−ω0). Solutions (c1+c2t)e−ω0t approach the origin tangent to that line without spiralling: a degenerate (improper) node.
  26. NATExam level

    For H=12p2+12x2−13x3, the energy of the separatrix (the orbit through the saddle point) is ______.
    Answer and solution

    Answer: 0.1667 (±0.001)

    V=x2/2−x3/3, V′=x−x2=0 at x=0 (minimum, centre) and x=1 (maximum, saddle). The separatrix passes through the saddle with p=0: E=V(1)=12−13=16. Orbits with E<1/6 near the origin are closed; above it, the particle escapes.
  27. MCQExam level

    For the damped pendulum x˙=y, y˙=−sin⁡x−0.5y, the fixed points (0,0) and (π,0) are respectively
    1. a stable spiral and a saddle
    2. a centre and a saddle
    3. a stable node and an unstable spiral
    4. a saddle and a stable spiral
    Answer and solution

    Answer: A

    Jacobian (01−cos⁡x−0.5). At (0,0): trace −0.5, determinant 1, and τ2−4Δ=0.25−4<0: complex eigenvalues with negative real part, a stable spiral. At (π,0): −cos⁡π=1, determinant −1<0: a saddle (the inverted pendulum).

    Why people get this wrong: Without damping the bottom would be a centre; any damping turns it into a stable spiral.

  28. MCQExam level

    A planar gradient system 𝐱˙=−∇V(𝐱), with V smooth,
    1. always has at least one limit cycle
    2. cannot have closed orbits
    3. has an antisymmetric Jacobian at fixed points
    4. conserves V along trajectories
    Answer and solution

    Answer: B

    Along a trajectory dVdt=∇V⋅𝐱˙=−|∇V|2≤0, with equality only at fixed points. On a closed orbit V would have to return to its starting value while strictly decreasing, which is impossible. The Jacobian −∂i∂jV is symmetric (real eigenvalues: nodes and saddles, never spirals), and V decreases rather than staying constant.

    Why people get this wrong: Gradient flows are the opposite of Hamiltonian flows: V always falls, while H is conserved.

  29. MSQExam level

    Which statements about the divergence of the phase-space flow are true?
    1. For x¨+γx˙+ω2x=0, written as x˙=v, v˙=−ω2x−γv, the divergence is −γ
    2. For that oscillator, a phase-space area shrinks as e−2γt
    3. For the Lorenz system the divergence is −(σ+1+b)
    4. For any Hamiltonian system the divergence is zero
    Answer and solution

    Answer: A, C, D

    ∇⋅𝐟=∂xx˙+∂vv˙=0−γ, and areas evolve as dA/dt=(∇⋅𝐟)A, so A∝e−γt. Lorenz: ∂x[σ(y−x)]+∂y[rx−y−xz]+∂z[xy−bz]=−σ−1−b. Hamiltonian flows: ∂q∂pH−∂p∂qH=0.

    Why people get this wrong: The energy decays like e−γt too, but the amplitude like e−γt/2; the area rate is the trace of the Jacobian, not twice it.

  30. MCQExam level

    For x˙=−y+x(x2+y2), y˙=x+y(x2+y2), the origin is
    1. an unstable spiral, although the linearisation predicts a centre
    2. a centre, as the linearisation predicts
    3. a stable spiral
    4. a saddle
    Answer and solution

    Answer: A

    The linear part x˙=−y, y˙=x has eigenvalues ±i: a centre. But in polar form, rr˙=xx˙+yy˙=r2⋅r2, so r˙=r3>0, while θ˙=1. Orbits spiral slowly outwards: an unstable spiral. Linearisation is reliable only when no eigenvalue has zero real part (a hyperbolic point); a linear centre can turn into either kind of spiral.

    Why people get this wrong: A linear centre is the borderline case where small nonlinear terms decide the stability.

← All topics · Timed mock test →