Flow Lab

Syllabus / Module 12: Applications: Flight & Aerospace

Lesson 12.5

Trim, Linearization & Longitudinal Modes

Trimmed flight, small-perturbation equations, phugoid and short-period modes as eigenvalues, static stability margin.

Written byClaude Opus

🌊
Why this matters for simulation: Why a paper plane porpoises (phugoid!).

By the end of this lesson you can…

  • Linearise the 12.4 model about trim with a finite-difference Jacobian and read the entries of the longitudinal matrix $A$.
  • Read each eigenvalue of a $4\times4$ $A$ as a mode (real part decay, imaginary part oscillation) and compute $\omega_n$, $\zeta$, period and time to half.
  • Identify the short period and phugoid, and approximate the short period with a $2\times2$ block.
  • Derive Lanchester's phugoid period $\sqrt2\,\pi V/g$ and explain, then correct, its $23.6\%$ gap.
  • Link static margin to short-period frequency and predict the eigenvalues as the centre of gravity moves to the neutral point.
  • Judge when the linear model stops matching the nonlinear one, against handling-qualities limits.
On this page
·
  1. Trim: the point we linearise about
  2. Small perturbations: from thirteen ODEs to x˙=Ax\dot{\mathbf x} = A\mathbf xx˙=Ax
  3. The Jacobian by finite differences
  4. Eigenvalues as modes
  5. The two modes of this aircraft
  6. The short period
  7. Worked example: the short period by hand
  8. The phugoid and Lanchester’s approximation
  9. Static margin and the short-period frequency
  10. How small is small?
  11. Handling qualities
  12. Computing it
  13. Concept checks
  14. Problems
  15. Key Takeaways

12.4 6-DOF Aircraft Equations of Motion ended with a pitch kick and two motions on two clocks: the nose bobbed and settled in about a second, then the aircraft spent half a minute trading height for speed. The simulator shows them but does not explain them. Linearised about trim, the same model becomes x˙=Ax\dot{\mathbf x} = A\mathbf x with four states, and the four eigenvalues of AA give both motions, periods and damping, before any time stepping. It is also how designers check a simulator and how control engineers start an autopilot.

Tools: 7.7 Nonlinear Systems & Linearization, 7.4 Complex & Repeated Eigenvalues, 1.7 Complex Eigenvalues & Rotation.

▶ Video Trim, the small-perturbation matrix, eigenvalues as modes, Lanchester's phugoid, and the static margin moving the short periodcaptions only
Transcript
  1. Last lesson ended with a pitch kick and two motions on two clocks. The nose bobbed and settled in about a second. Then the aircraft spent half a minute trading height for speed. Linearised about trim, the same model has four eigenvalues, and they give both motions before any time stepping.
  2. Everything starts at trim, an equilibrium where every derivative is zero except steady motion along the path. At 50 metres per second Newton's method gave alpha equal to theta, 2.6606 degrees, elevator minus 2.7452, thrust 931.2 newtons, weight 10 791. Expand about any other point and a constant leftover term makes the modes drift.
  3. Write the state as trim plus a small change and expand to first order; the Jacobian J does the work. Two facts cut thirteen states to four. At wings-level trim the lateral states appear only in products, so the lateral and longitudinal blocks decouple. And position drops out. That leaves delta u, delta w, delta q and delta theta.
  4. A simulator programmer does not differentiate forces by hand. Nudge each state up and down inside the same derivative function the simulator flies, and divide by 2 h. The steps are 10 to the minus 4 for u and w, and 10 to the minus 5 for q and theta, so the errors sit near 10 to the minus 8.
  5. Here is the matrix, with columns u, w, q and theta. The entry in row 2, column 3 is the forward speed 49.94610, the rotating frame term. Row 3, column 2 is minus 0.27371: the pitch stiffness, from C m alpha. Row 3, column 3 is minus 2.86316 per second, the pitch damping. Each entry is a physical sensitivity.
  6. Now the geometry. If A only stretches a direction v, a start along v stays on that line and decays as e to the lambda t. Any start is a mix of the four, each a mode. A complex pair makes one spiral: the real part is the decay rate, the imaginary part the turning rate. From them come the natural frequency, zeta, the period and the time to half.
  7. The four eigenvalues form two pairs. The short period, minus 2.6647 plus or minus 3.6964 i, has zeta 0.5848 and a period of 1.70 seconds: the angle of attack changes at nearly constant speed. The phugoid, minus 0.0126 plus or minus 0.2246 i, has zeta 0.0561 and a period of 27.98 seconds: speed and height trade at nearly constant angle of attack. They differ twentyfold, so the system is mildly stiff.
  8. Over a second the speed hardly changes, so drop u and theta. Omega n squared is Z w times M q minus u zero times M w, and twice zeta omega n is minus the sum of Z w and M q. By hand that gives 4.5533 and 0.5853, good to 0.1 percent. The spring is the weathercock action of C m alpha; the dashpot is pitch damping.
  9. Lanchester, in 1908, held the angle of attack fixed and kept only speed and path angle. Gravity slows a climb, and lift grows as speed squared, so a speed excess curves the path. The period is root 2 pi V over g: 22.64 seconds. The full model's 27.98 seconds is 23.6 percent longer, a gap of 5.33 seconds. What is the cause?
  10. Remove effects one at a time. Without drag the period barely moves: 27.85 seconds. Without pitch damping it falls to 22.70, almost exactly Lanchester. The nose must rotate with the curving path, and its damping moment is balanced by a lower angle of attack, so the lift curving the path is weaker. That softens the spring by 1 plus kappa, 0.5101, stretching the period by 1.2289 to 27.83 seconds.
  11. Now the static margin slider. The short period spring is the margin, but the real part has no C m alpha in it, so the pair moves vertically. At C m alpha minus 3, the imaginary part is 6.79 and zeta falls to 0.365. At minus 0.1 it is 1.23 and zeta rises to 0.908. At the neutral point the pair splits onto the real axis, and past it a positive root appears: the nose simply diverges.
  12. How small is small? A first-order series should err by the disturbance squared. A pitch kick of 0.2 radians per second leaves gaps of 0.017 degrees and 0.14 metres of altitude. Three times the kick, 0.6, gives 0.158 degrees and 1.27 metres: nine times the gap. The linear model is for small disturbances.
  13. Handling qualities turn pilot ratings into eigenvalue bounds. From memory, a Level 1 aircraft wants a short period zeta of about 0.35 to 1.30, and a phugoid zeta of at least 0.04. Ours, 0.585 and 0.056, pass, though the derivatives are assumed values.
  14. To recap. Linearise about trim with a central-difference Jacobian. Each eigenvalue is a mode: the real part decays, the imaginary part turns. The short period is minus 2.6647 plus or minus 3.6964 i, and the phugoid takes 27.98 seconds. Lanchester's 22.64 is 23.6 percent short, through pitch damping rather than drag.
  15. Next, lateral modes: the other block of four, v, p, r and phi. It has three modes: a fast roll subsidence, a slow spiral that may diverge, and the Dutch roll wobble that brought the first yaw dampers.

Trim: the point we linearise about

A trim state is an equilibrium: every derivative is zero except steady motion along the path. 12.4 found level trim at 5050 m/s by Newton’s method: α=θ=2.6606∘\alpha = \theta = 2.6606^\circ, δe=−2.7452∘\delta_e = -2.7452^\circ, T=931.2T = 931.2 N, with W=10 791W = 10\,791 N. A Taylor expansion needs this exact base point: about a state that is not an equilibrium a constant leftover term appears, and the “modes” drift for reasons unrelated to stability.

Small perturbations: from thirteen ODEs to x˙=Ax\dot{\mathbf x} = A\mathbf x

Write the state as trim plus a small change, s=s0+Δs\mathbf s = \mathbf s_0 + \Delta\mathbf s, and expand to first order (7.7 Nonlinear Systems & Linearization):

Δs˙≈J Δs,Jij=∂fi/∂sj at s0,\Delta\dot{\mathbf s} \approx J\,\Delta\mathbf s, \qquad J_{ij} = \partial f_i/\partial s_j \text{ at } \mathbf s_0 ,

since f(s0)=0\mathbf f(\mathbf s_0) = 0 (apart from the steady x˙\dot x, which feeds nothing back). Two facts cut thirteen states to four.

Longitudinal and lateral decouple. At wings-level trim v=p=r=ϕ=0v = p = r = \phi = 0. In the uu, ww, qq equations these appear only in products (rvrv, pvpv, prpr) or squares (β2\beta^2 inside VV), or through cos⁡ϕ\cos\phi, so their first derivatives vanish. The Jacobian splits into a longitudinal and a lateral block; 12.6 takes the lateral one.

Position drops out. Nothing depends on xx, yy, ψ\psi, and with sea-level density everywhere hh does not feed back. (A real atmosphere adds a weak fifth state.) That leaves x=(Δu,Δw,Δq,Δθ)\mathbf x = (\Delta u, \Delta w, \Delta q, \Delta\theta) and x˙=Ax\dot{\mathbf x} = A\mathbf x; altitude is an output, h˙=usin⁡θ−wcos⁡θ\dot h = u\sin\theta - w\cos\theta.

The Jacobian by finite differences

A simulator programmer does not differentiate forces by hand. They nudge each state in the same derivative function the simulator flies:

A:,j≈f(x0+hej)−f(x0−hej)2h,A_{:,j} \approx \frac{\mathbf f(\mathbf x_0 + h\mathbf e_j) - \mathbf f(\mathbf x_0 - h\mathbf e_j)}{2h},

with h=10−4h = 10^{-4} for uu, ww and 10−510^{-5} for qq, θ\theta. The sim builds 12.4’s 13-state vector (quaternion (cos⁡θ2,0,sin⁡θ2,0)(\cos\tfrac\theta2, 0, \sin\tfrac\theta2, 0)), calls its deriv, and keeps u˙\dot u, w˙\dot w, q˙\dot q; the last row is θ˙=q\dot\theta = q, exact at ϕ=0\phi = 0. Columns u,w,q,θu, w, q, \theta:

A=[−0.024760.19580−2.32100−9.79942−0.27778−2.4666449.94610−0.455380.01272−0.27371−2.8631600010].A = \begin{bmatrix} -0.02476 & 0.19580 & -2.32100 & -9.79942\\ -0.27778 & -2.46664 & 49.94610 & -0.45538\\ 0.01272 & -0.27371 & -2.86316 & 0\\ 0 & 0 & 1 & 0 \end{bmatrix}.

Reading the entries. Each is a physical sensitivity.

  • A13=−w0A_{13} = -w_0, A23=u0A_{23} = u_0: 12.4’s rotating-frame terms −qw-qw, +qu+qu.
  • A14=−gcos⁡θ0A_{14} = -g\cos\theta_0, A24=−gsin⁡θ0A_{24} = -g\sin\theta_0: nose up tips the weight backward.
  • A22=−2.467A_{22} = -2.467 s−1^{-1}: more ww means more α\alpha and more lift opposing ww; mostly qˉSCLα/(mV)=2.441\bar qSC_{L\alpha}/(mV) = 2.441 s−1^{-1}.
  • A32=−0.2737A_{32} = -0.2737: CmαC_{m\alpha}, divided by VV because Δα≈Δw/V\Delta\alpha \approx \Delta w/V. A33=−2.863A_{33} = -2.863 s−1^{-1}: pitch damping, CmqC_{mq}.
  • A11=−0.0248A_{11} = -0.0248 s−1^{-1}: drag rises with speed; small, as drag is a tenth of lift.
  • A31=0.0127A_{31} = 0.0127: raising uu lowers α=arctan⁡(w/u)\alpha = \arctan(w/u); a body-axis artefact.

Eigenvalues as modes

Geometrically: suppose v\mathbf v is a direction in the 4-D state space that AA only stretches, Av=λvA\mathbf v = \lambda\mathbf v. Start exactly along v\mathbf v; then x˙=λx\dot{\mathbf x} = \lambda\mathbf x, the state stays on that line, and x(t)=eλtv\mathbf x(t) = e^{\lambda t}\mathbf v. Any start is a mix of the four eigenvectors, each evolving on its own:

x(t)=∑kckeλktvk.\mathbf x(t) = \textstyle\sum_k c_k e^{\lambda_k t}\mathbf v_k .

Each term is a mode. Complex eigenvalues come in pairs σ±iωd\sigma \pm i\omega_d, and a pair makes one real motion, eσt(acos⁡ωdt−bsin⁡ωdt)e^{\sigma t}(\mathbf a\cos\omega_dt - \mathbf b\sin\omega_dt) with v=a+ib\mathbf v = \mathbf a + i\mathbf b (7.4 Complex & Repeated Eigenvalues): a spiral in the plane of a\mathbf a and b\mathbf b. The real part is the decay rate; the imaginary part is the turning rate. Then

ωn=∣λ∣,ζ=−σ/∣λ∣,T=2π/ωd,t1/2=ln⁡2/(−σ).\omega_n = |\lambda|, \qquad \zeta = -\sigma/|\lambda|, \qquad T = 2\pi/\omega_d, \qquad t_{1/2} = \ln 2/(-\sigma).

ζ\zeta is the cosine of the angle from λ\lambda to the negative real axis: on the axis no oscillation, on the imaginary axis no damping.

The two modes of this aircraft

The characteristic polynomial is λ4+5.3546λ3+20.9488λ2+0.7940λ+1.0505\lambda^4 + 5.3546\lambda^3 + 20.9488\lambda^2 + 0.7940\lambda + 1.0505, with roots (sim readout, matched by numpy in the lab cell):

modeλ\lambda (1/s)ωn\omega_n (rad/s)ζ\zetaperiod (s)t1/2t_{1/2} (s)
short period−2.6647±3.6964i-2.6647 \pm 3.6964i4.55674.55670.58480.58481.701.700.260.26
phugoid−0.0126±0.2246i-0.0126 \pm 0.2246i0.22490.22490.05610.056127.9827.9854.8954.89

The phugoid period is exactly the altitude-peak spacing 12.4 measured on the nonlinear model (peaks at 7.267.26 and 35.2435.24 s). The pairs differ in size twentyfold: mildly stiff (4.5 Stability Regions & Stiffness).

Mode shapes (sim readout, per radian of Δθ\Delta\theta): the short period has ∣Δw∣=61.6|\Delta w| = 61.6 m/s but ∣Δu∣=1.5|\Delta u| = 1.5 m/s, an angle-of-attack change at nearly constant speed. The phugoid has ∣Δu∣=43.4|\Delta u| = 43.4 m/s and ∣Δw∣=0.33|\Delta w| = 0.33 m/s, a speed change at nearly constant angle of attack, uu leading θ\theta by 96.5∘96.5^\circ. These are the sim’s “pure mode” disturbances.

The short period

Over a second the speed hardly changes, so drop uu, and drop θ\theta (it enters w˙\dot w only through gsin⁡θ0g\sin\theta_0). The (w,q)(w, q) block remains:

[w˙q˙]=[Zwu0MwMq][wq],ωn2=ZwMq−u0Mw,2ζωn=−(Zw+Mq).\begin{bmatrix}\dot w\\ \dot q\end{bmatrix} = \begin{bmatrix} Z_w & u_0\\ M_w & M_q\end{bmatrix}\begin{bmatrix} w\\ q\end{bmatrix}, \qquad \omega_n^2 = Z_wM_q - u_0M_w, \quad 2\zeta\omega_n = -(Z_w + M_q).

Worked example: the short period by hand

With Zw=−2.4666Z_w = -2.4666, u0=49.946u_0 = 49.946, Mw=−0.27371M_w = -0.27371, Mq=−2.8632M_q = -2.8632:

  1. ZwMq=7.0624Z_wM_q = 7.0624, −u0Mw=13.6706-u_0M_w = 13.6706, so ωn=20.733=4.5533\omega_n = \sqrt{20.733} = 4.5533 rad/s.
  2. 2ζωn=5.32982\zeta\omega_n = 5.3298: σ=−2.6649\sigma = -2.6649, ζ=0.5853\zeta = 0.5853.
  3. ωd=ωn2−σ2=3.6921\omega_d = \sqrt{\omega_n^2 - \sigma^2} = 3.6921 rad/s.

The 4×44\times4 gives 4.55674.5567 and 0.58480.5848: good to 0.1%0.1\%. The spring is mostly −u0Mw-u_0M_w, the weathercock action of CmαC_{m\alpha}; the dashpot is CmqC_{mq} plus lift opposing ww.

Reconciling 12.4’s pitch peak. The kick sets q0=0.2q_0 = 0.2 rad/s, w=0w = 0, so q(t)=eσt(q0cos⁡ωdt+Bsin⁡ωdt)q(t) = e^{\sigma t}(q_0\cos\omega_dt + B\sin\omega_dt) with B=(Mq−σ)q0/ωd=−0.01074B = (M_q - \sigma)q_0/\omega_d = -0.01074. Pitch peaks where q=0q = 0: tan⁡ωdt=18.62\tan\omega_dt = 18.62, t=0.411t = 0.411 s. Both sim runs peak at 0.410.41 s, 2.045∘2.045^\circ above trim: 12.4’s 4.71∘4.71^\circ.

The phugoid and Lanchester’s approximation

Lanchester (1908, from memory) held the angle of attack fixed, let thrust cancel drag, and kept only speed VV and path angle γ\gamma. Gravity slows a climb, V˙=−gγ\dot V = -g\gamma; lift at fixed CLC_L grows as V2V^2, so a speed excess curves the path up:

mVγ˙=2WVΔV  ⟹  ΔV¨=−2g2V2ΔV,TL=2 πVg=22.64 s.mV\dot\gamma = \frac{2W}{V}\Delta V \;\Longrightarrow\; \Delta\ddot V = -\frac{2g^2}{V^2}\Delta V, \qquad T_{\text{L}} = \frac{\sqrt2\,\pi V}{g} = 22.64\ \text{s}.

The full model’s 27.9827.98 s is 23.6%23.6\% longer (ratio 1.23551.2355, gap 5.335.33 s). The sim’s model menu removes effects one at a time, re-trims and re-linearises:

modelphugoid period (s)
full27.9827.98
drag removed (CD0=0C_{D0} = 0, no induced drag)27.8527.85
pitch damping removed (Cmq=0C_{mq} = 0)22.7022.70
Lanchester22.6422.64

Drag barely moves the period. Its job is damping: Lanchester with drag gives ζ≈1/(2 L/D)=1/(2⋅11.55)=0.061\zeta \approx 1/(\sqrt2\,L/D) = 1/(\sqrt2\cdot11.55) = 0.061, near the full 0.0560.056, and without drag the phugoid here is slightly unstable (σ=+0.0012\sigma = +0.0012).

The mechanism is pitch damping. The path keeps curving, so the nose must rotate with it, q≈γ˙q \approx \dot\gamma, and rotating costs a damping moment Cmqqc/2VC_{mq}qc/2V. The aircraft balances it by flying at a slightly different angle of attack, CmαΔα+Cmqqc/2V=0C_{m\alpha}\Delta\alpha + C_{mq}qc/2V = 0:

Δα=−τγ˙,τ=Cmqc2VCmα=12.4⋅1.5100⋅0.89=0.209 s.\Delta\alpha = -\tau\dot\gamma, \qquad \tau = \frac{C_{mq}c}{2VC_{m\alpha}} = \frac{12.4\cdot1.5}{100\cdot0.89} = 0.209\ \text{s}.

While the path curves up, α\alpha is lower and the lift curving it is smaller. In the normal equation:

mVγ˙(1+κ)=2WVΔV,κ=qˉSCLαmVτ=2.441⋅0.209=0.5101.mV\dot\gamma(1 + \kappa) = \frac{2W}{V}\Delta V, \qquad \kappa = \frac{\bar qSC_{L\alpha}}{mV}\tau = 2.441\cdot0.209 = 0.5101 .

The spring softens by 1+κ1 + \kappa, the period grows by 1.5101=1.2289\sqrt{1.5101} = 1.2289: 27.8327.83 s, leaving 0.150.15 s for drag and dropped terms. As qˉ∝V2\bar q \propto V^2 and τ∝1/V\tau \propto 1/V, κ\kappa is speed-independent; the trim speed menu gives ratio 1.2381.238 at both 4040 m/s (22.4322.43 against 18.1218.12 s) and 7070 m/s (39.2539.25 against 31.7031.70 s).

Static margin and the short-period frequency

Extra lift acts −Cmα/CLα=0.1675-C_{m\alpha}/C_{L\alpha} = 0.1675 chord behind the centre of gravity, at the neutral point; that distance is the static margin KnK_n, and Cmα=−CLαKnC_{m\alpha} = -C_{L\alpha}K_n (holding CLαC_{L\alpha} and CmqC_{mq} fixed). The short-period spring is −u0Mw≈qˉScCLαKn/Iyy-u_0M_w \approx \bar qScC_{L\alpha}K_n/I_{yy}, so

ωsp2≈81.785 Kn+7.062 s−2,2ζωsp=5.330 s−1.\omega_{sp}^2 \approx 81.785\,K_n + 7.062\ \text{s}^{-2}, \qquad 2\zeta\omega_{sp} = 5.330\ \text{s}^{-1}.

On the slider:

  • The pair moves vertically. σ=−(Zw+Mq)/2\sigma = -(Z_w + M_q)/2 has no CmαC_{m\alpha}: the real part stays near −2.66-2.66 (−2.6628-2.6628 at Cmα=−3C_{m\alpha} = -3, −2.6544-2.6544 at −0.01-0.01) while ωd\omega_d falls from 6.796.79 to 3.703.70 (−0.89-0.89) to 1.231.23 (−0.1-0.1).
  • Less margin, more damping ratio: ζ=0.365\zeta = 0.365, 0.5850.585, 0.9080.908 at those three. Forward centre of gravity: a stiff, quick, lightly damped nose; aft: sluggish and well damped.
  • At the neutral point the pair splits. A few thousandths from Cmα=0C_{m\alpha} = 0 both pairs fall onto the real axis; at 00 a root sits at λ=0\lambda = 0; beyond, a positive real root (+0.073+0.073 s−1^{-1} at +0.05+0.05). Not a growing oscillation: the nose simply diverges, doubling every 9.59.5 s. That is static instability in eigenvalue language.

The phugoid lengthens too, through τ∝1/Cmα\tau \propto 1/C_{m\alpha}: 36.2136.21 s at −0.3-0.3, 54.2254.22 s at −0.1-0.1. Lanchester is the stiff-pitch limit.

◆ Interactive The 12.4 aircraft linearised about trim: A matrix, eigenvalues in the complex plane with a static-margin slider, and the nonlinear flight beside the linear model after a disturbance

Loading simulation…

How small is small?

A first-order Taylor series should err by the disturbance squared. The sim’s “nonlinear − linear” readout over 60 s:

pitch kickmax Δθ\Delta\theta gapmax ΔV\Delta V gapmax Δh\Delta h gap
0.20.2 rad/s0.017∘0.017^\circ0.0150.015 m/s0.140.14 m
0.60.6 rad/s0.158∘0.158^\circ0.1370.137 m/s1.271.27 m

Three times the kick, nine times the gap. At 0.60.6 rad/s (a hard yank) altitude is off by over a metre in 1515. Nonlinear terms also couple the modes: from the pure short-period vector (Δw=2\Delta w = 2 m/s), the linear altitude stays within 0.090.09 m, but the nonlinear one reaches 0.380.38 m at 13.813.8 s, half a phugoid cycle later.

Handling qualities

Pilots rate aircraft on the Cooper-Harper scale (1 excellent, 10 uncontrollable), and specifications turn good ratings into eigenvalue bounds. From memory, MIL-F-8785C asks Level 1 aircraft for a short-period ζ\zeta of about 0.350.35 to 1.301.30 in precision tasks and landing, and a phugoid ζ\zeta of at least 0.040.04. Ours, 0.5850.585 and 0.0560.056, pass. The sim’s dashed rays are ζ=0.35\zeta = 0.35; at Cmα=−3.30C_{m\alpha} = -3.30 (static margin 0.620.62) the pair sits on them. Too much margin is a problem too: a stiff, under-damped nose needing large elevator.

Computing it

🐍 PythonLab: finite-difference Jacobian, eigenvalues, the 2x2 short period and Lanchester with pitch damping
⌘/Ctrl + Enter

This is the 4-state restriction of 12.4’s model (v=p=r=ϕ=0v = p = r = \phi = 0), so it reproduces the sim’s matrix to the printed digits. Set Cmq = 0 (trim does not change, since q=0q = 0 there) and the phugoid falls to 22.7022.70 s, next to Lanchester’s 22.6422.64 s.

🧪 Coding exerciseExercise: modes from a longitudinal A matrix, and the phugoid against Lanchester
⌘/Ctrl + Enter

Graded: the short period (ωn=4.557\omega_n = 4.557 rad/s, ζ=0.585\zeta = 0.585) and phugoid (27.9827.98 s, half-time 54.954.9 s) from the 5-digit matrix; the sorting on a second matrix with known eigenvalues; and Lanchester’s 22.6422.64 s with the full model 1.23551.2355 times longer.

Concept checks

✅ Concept check

A longitudinal $A$ has eigenvalues $-3.1 \pm 4.2i$ and $-0.02 \pm 0.19i$. Which pair is the phugoid?

✅ Concept check

Lanchester's phugoid period at $V = 50$ m/s with $g = 9.81$ m/s$^2$, in seconds?

✅ Concept check

Why is this aircraft's phugoid period 23.6% longer than Lanchester's estimate?

✅ Concept check

The centre of gravity moves aft toward the neutral point ($C_{m\alpha}$ toward 0). What happens to the short-period eigenvalue pair?

✅ Concept check

The phugoid has $\lambda = -0.01263 \pm 0.22458i$. About how many seconds until its amplitude halves?

Problems

✏️ Pen & paperThe pitch peak from two states●●○

Use the (w,q)(w, q) block with Zw=−2.4666Z_w = -2.4666, u0=49.946u_0 = 49.946, Mw=−0.27371M_w = -0.27371, Mq=−2.8632M_q = -2.8632. (a) Find σ\sigma and ωd\omega_d. (b) After a kick q(0)=0.2q(0) = 0.2 rad/s, w(0)=0w(0) = 0, when does the pitch angle first peak? (c) Why before a quarter of the damped period?

Show worked solution

(a) σ=(Zw+Mq)/2=−2.6649\sigma = (Z_w + M_q)/2 = -2.6649 s−1^{-1}; ωn2=7.0624+13.6706=20.733\omega_n^2 = 7.0624 + 13.6706 = 20.733, so ωd=20.733−7.1017=3.6921\omega_d = \sqrt{20.733 - 7.1017} = 3.6921 rad/s.

(b) q(t)=eσt(q0cos⁡ωdt+Bsin⁡ωdt)q(t) = e^{\sigma t}(q_0\cos\omega_dt + B\sin\omega_dt). With w=0w = 0, q˙(0)=Mqq0\dot q(0) = M_qq_0, and differentiating gives q˙(0)=σq0+ωdB\dot q(0) = \sigma q_0 + \omega_dB, so B=(Mq−σ)q0/ωd=−0.01074B = (M_q - \sigma)q_0/\omega_d = -0.01074. θ\theta peaks where q=0q = 0: tan⁡ωdt=−q0/B=18.62\tan\omega_dt = -q_0/B = 18.62, ωdt=1.5171\omega_dt = 1.5171, t=0.411t = 0.411 s. The sim shows 0.410.41 s for both models.

(c) A pure decaying cosine would reach zero at π/(2ωd)=0.425\pi/(2\omega_d) = 0.425 s. Pitch damping makes qq fall faster from the start (B<0B \lt 0), so it gets there sooner.

✏️ Pen & paperLanchester with and without pitch damping●●●

(a) From V˙=−gγ\dot V = -g\gamma and mVγ˙=2WΔV/VmV\dot\gamma = 2W\Delta V/V, derive T=2 πV/gT = \sqrt2\,\pi V/g. (b) With quasi-steady pitch balance, show Δα=−τq\Delta\alpha = -\tau q, τ=Cmqc/(2VCmα)\tau = C_{mq}c/(2VC_{m\alpha}), and that the period becomes T1+κT\sqrt{1 + \kappa}, κ=qˉSCLατ/(mV)\kappa = \bar qSC_{L\alpha}\tau/(mV). (c) Evaluate both at 5050 m/s. (d) Show κ\kappa does not depend on VV.

Show worked solution

(a) ΔV¨=−gγ˙=−(2g2/V2)ΔV\Delta\ddot V = -g\dot\gamma = -(2g^2/V^2)\Delta V (using W=mgW = mg), so ω=2 g/V\omega = \sqrt2\,g/V and T=2π/ωT = 2\pi/\omega.

(b) No pitch acceleration (the short period has settled): CmαΔα+Cmqqc/2V=0C_{m\alpha}\Delta\alpha + C_{mq}qc/2V = 0, so Δα=−τq\Delta\alpha = -\tau q. With α\alpha nearly constant, q≈γ˙q \approx \dot\gamma. The extra lift −qˉSCLατγ˙-\bar qSC_{L\alpha}\tau\dot\gamma moves to the left: mVγ˙(1+κ)=2WΔV/VmV\dot\gamma(1 + \kappa) = 2W\Delta V/V, so ω2\omega^2 is divided by 1+κ1 + \kappa.

(c) T=22.64T = 22.64 s. τ=0.2090\tau = 0.2090 s, qˉSCLα/(mV)=25 265.6⋅5.313/55 000=2.4407\bar qSC_{L\alpha}/(mV) = 25\,265.6\cdot5.313/55\,000 = 2.4407, κ=0.5101\kappa = 0.5101, T=22.64⋅1.2289=27.83T = 22.64\cdot1.2289 = 27.83 s: 0.5%0.5\% under the full 27.9827.98 s, where Lanchester alone is 19%19\% under.

(d) qˉ∝V2\bar q \propto V^2, τ∝1/V\tau \propto 1/V, so qˉτ/V\bar q\tau/V is constant. The sim’s ratio: 1.2381.238, 1.23551.2355, 1.2361.236, 1.2381.238 at 4040, 5050, 6060, 7070 m/s.

✏️ Pen & paperMoving the centre of gravity aft●●○

Ballast moves the centre of gravity aft to a static margin of 0.050.05 chord. (a) Find CmαC_{m\alpha}. (b) Predict the short-period ωn\omega_n and ζ\zeta. (c) Predict the phugoid period with the pitch-damping correction. (The sim gives ωn=3.340\omega_n = 3.340, ζ=0.797\zeta = 0.797, T=37.61T = 37.61 s.)

Show worked solution

(a) Cmα=−5.313⋅0.05=−0.2657C_{m\alpha} = -5.313\cdot0.05 = -0.2657.

(b) ωn2=81.785⋅0.05+7.062=11.151\omega_n^2 = 81.785\cdot0.05 + 7.062 = 11.151, ωn=3.339\omega_n = 3.339 rad/s; ζ=5.330/(2⋅3.339)=0.798\zeta = 5.330/(2\cdot3.339) = 0.798: slower, better damped.

(c) τ=12.4⋅1.5/(100⋅0.2657)=0.700\tau = 12.4\cdot1.5/(100\cdot0.2657) = 0.700 s, κ=1.709\kappa = 1.709, T=22.642.709=37.27T = 22.64\sqrt{2.709} = 37.27 s, within 1%1\%. Less margin, more lag of α\alpha behind the path, a longer phugoid.

Key Takeaways

  • Linearise about trim; a central-difference Jacobian of the simulator’s own derivative function gives AA with no hand differentiation.
  • At wings-level trim (u,w,q,θ)(u, w, q, \theta) decouple from the lateral states; position and (constant density) altitude drop out.
  • Each eigenvalue is a mode: real part decay rate, imaginary part turning rate; ωn=∣λ∣\omega_n = |\lambda|, ζ=−σ/∣λ∣\zeta = -\sigma/|\lambda|, T=2π/ωdT = 2\pi/\omega_d.
  • Here: short period −2.6647±3.6964i-2.6647 \pm 3.6964i (ζ=0.585\zeta = 0.585), phugoid −0.0126±0.2246i-0.0126 \pm 0.2246i (27.9827.98 s), matching 12.4’s 0.410.41 s pitch peak and 27.9827.98 s altitude cycle.
  • The (w,q)(w, q) block gives the short period to 0.1%0.1\%; its spring is static margin, ωsp2≈81.8Kn+7.06\omega_{sp}^2 \approx 81.8K_n + 7.06, its real part independent of it.
  • Lanchester’s 22.6422.64 s is 23.6%23.6\% short, through pitch damping, not drag; 1+κ\sqrt{1 + \kappa} recovers 27.8327.83 s.
  • Past the neutral point a positive real root makes the nose diverge.
  • Next, 12.6 Lateral-Directional Modes: the other block, (v,p,r,ϕ)(v, p, r, \phi), and its three modes: fast roll subsidence, a slow spiral that may diverge, and the Dutch roll wobble that brought the first yaw dampers.
Finished this lesson?

Mark it complete to track your progress (saved in this browser). Reading every section does this for you.

Where this leads

These later lessons build directly on this one: