3.4.11Rocket Flight Mechanics
Dynamic stability — pitch - yaw damping derivatives
1,947 words9 min readdifficulty · medium1 backlinks
WHAT are we describing?
When a rocket pitches (nose up/down) or yaws (nose left/right), every surface behind the center of gravity (CG) sweeps sideways through the air. That sideways sweep creates a local angle of attack purely from the rotation itself, which produces a restoring aerodynamic force — and hence a torque opposing the rotation.
HOW: derive the damping moment from first principles
Step 1 — A rotating rocket gives each point a local vertical velocity. Take the CG as origin, measured aft. A point a distance behind the CG, when the body pitches at rate , moves downward (relative to the airstream) with speed Why this step? Rigid-body rotation: velocity of a point = . A pitch rate about the CG lifts the tail as it drops the nose, so a point at (behind CG) gains a transverse velocity .
Step 2 — That transverse velocity is a local angle of attack. The airflow arrives at speed axially, plus this new transverse , so that element feels an extra angle of attack Why this step? Angle of attack is just transverse/axial for small angles.
Step 3 — Extra angle of attack → extra local normal force. An element with local lift-curve slope and local area contribution generates an incremental normal force
=\tfrac12\rho V^2\,c_{n_\alpha}(x)\,\frac{qx}{V}\,dA.$$ *Why this step?* Force = dynamic pressure $\times$ effective area $\times$ (slope $\times$ angle). Standard linear aero. **Step 4 — Each force at moment arm $x$ makes a torque; integrate.** The restoring torque (moment about CG) is $$M=-\int x\,dN=-\tfrac12\rho V^2\frac{q}{V}\int c_{n_\alpha}(x)\,x^2\,dA .$$ *Why the minus sign?* The force from $\Delta\alpha$ points to *reduce* the motion (a downward-moving tail is pushed up), so the resulting moment opposes $q$. **The integral carries $x^2$**: this is the geometric heart of damping — surfaces far behind the CG dominate, because they both sweep faster *and* have a longer lever arm. **Step 5 — Non-dimensionalize.** Define $C_m=M/(\tfrac12\rho V^2 S d)$ and $\hat q = qd/(2V)$. Substituting, $$\boxed{\,C_{m_q}=\frac{\partial C_m}{\partial \hat q} = -\frac{2}{S\,d^2}\int c_{n_\alpha}(x)\,x^2\,dA\, }$$ > [!formula] Result you must remember > $$C_{m_q}=-\frac{2}{S d^2}\int c_{n_\alpha}\,x^2\,dA \;<\;0 \quad\text{(damping)}$$ > **Sign is negative → it always opposes rotation.** Magnitude scales with the **second moment** of aerodynamic area about the CG. Move fins farther back → $x$ up → damping up (fast). --- ## HOW damping enters the equation of motion Combine the *static* restoring moment ($C_{m_\alpha}\alpha$, spring) with the *damping* moment ($C_{m_q}\hat q$, dashpot). Linearised pitch dynamics: $$I_y\ddot\theta - \tfrac12\rho V^2 S d\left(C_{m_\alpha}\,\theta + C_{m_q}\frac{d}{2V}\dot\theta\right)=0.$$ This is a **damped harmonic oscillator** $\ddot\theta + 2\zeta\omega_n\dot\theta + \omega_n^2\theta=0$ with $$\omega_n=\sqrt{\frac{-\,\tfrac12\rho V^2 S d\,C_{m_\alpha}}{I_y}},\qquad \zeta=\frac{-\tfrac14\rho V S d^2\,C_{m_q}}{2 I_y\,\omega_n}.$$ *Why this matters:* $\omega_n$ needs $C_{m_\alpha}<0$ (static stability); $\zeta>0$ needs $C_{m_q}<0$ (damping). **You need BOTH** to actually calm down. ![[3.4.11-Dynamic-stability-—-pitch-yaw-damping-derivatives.png]] --- ## Worked examples > [!example] 1. Single tail-heavy fin set > A rocket has all its aerodynamic area concentrated at the fins, $x_f = 0.6\,$m behind CG, with $c_{n_\alpha}A_f = 8\text{ m}^2$ (per rad). Reference $S=0.008\text{ m}^2$, $d=0.1$ m. Find $C_{m_q}$. > > $C_{m_q}=-\dfrac{2}{Sd^2}\,(c_{n_\alpha}A_f)\,x_f^2 = -\dfrac{2}{0.008\times 0.01}\times 8 \times 0.36$ > **Why this step?** With area lumped at one station the integral is just value $\times x_f^2$. > $=-\dfrac{2\times8\times0.36}{8\times10^{-5}} = -7.2\times10^{4}.$ Strongly damped — good. > [!example] 2. Doubling fin distance > Move the same fins to $x_f=1.2$ m. Since $C_{m_q}\propto x_f^2$, damping **quadruples**: $\approx -2.9\times10^{5}$. > **Why this step?** The $x^2$ in the integral means moving fins twice as far back multiplies damping by four, not two. This is the single most important design lever. > [!example] 3. Is the oscillation actually decaying? > Given $\zeta=0.06>0$, the response $\propto e^{-\zeta\omega_n t}\cos(\omega_d t)$ decays. If a design error flipped the fins to *ahead* of CG, both $x^2$-weighting still gives a moment but now $C_{m_\alpha}>0$: $\omega_n^2<0$, roots real & positive → pure **divergence**, no oscillation at all. > **Why this step?** Damping only helps a *statically stable* body; on an unstable one it can't save you. --- ## Common mistakes > [!mistake] "Static stability guarantees the rocket is safe." > **Why it feels right:** A statically stable rocket *does* start turning back toward the wind, which looks stabilizing. **The flaw:** it overshoots and oscillates; if $C_{m_q}$ is too weak (or, at high altitude, air density $\rho$ too low), the oscillation barely damps and can be excited into a tumble. **Fix:** always check $\zeta>0$ *and* that $\zeta$ is large enough (typically want $\zeta \gtrsim 0.05$–$0.1$ across the flight envelope). > [!mistake] "Damping scales with fin distance $x$." > **Why it feels right:** Longer lever arm = more torque, so surely linear. **The flaw:** the *velocity* each surface sweeps also grows with $x$ (Step 1), so torque $\propto x \times x = x^2$. **Fix:** remember the **second moment** — damping goes as distance *squared*. > [!mistake] "$C_{m_q}$ is a constant of the rocket." > **Why it feels right:** It's a fixed geometry number in wind-tunnel tables. **The flaw:** the *physical damping moment* $=\tfrac12\rho V S d^2 C_{m_q}\cdot(\text{rate})$ depends on $\rho$ and $V$. As the rocket climbs, $\rho\downarrow$ and damping weakens dramatically. **Fix:** evaluate $\zeta$ at every altitude, not just at launch. --- ## #flashcards/physics What physical effect does $C_{m_q}$ quantify? ::: The aerodynamic moment opposing pitch *rate* — the rotational shock absorber that damps oscillations. Why does damping scale as $x^2$ (second moment of area)? ::: A surface at distance $x$ sweeps transversely at speed $qx$ (∝x) AND acts on a lever arm $x$ (∝x); torque ∝ $x·x = x^2$. What sign must $C_{m_q}$ have for stability, and what does it mean? ::: Negative; the moment always opposes the rotation rate, removing energy from the oscillation. Static stability requires which derivative negative? ::: $C_{m_\alpha}<0$ (restoring moment vs angle of attack). Which two conditions together give a decaying oscillation? ::: $C_{m_\alpha}<0$ (gives $\omega_n^2>0$) AND $C_{m_q}<0$ (gives $\zeta>0$). How does the damping ratio change as the rocket gains altitude, and why? ::: It decreases, because the physical damping moment $\propto \rho V$ and air density $\rho$ falls with altitude. Why is $C_{n_r}=C_{m_q}$ for a rocket? ::: Axial symmetry — pitch and yaw are geometrically identical planes. Definition of the non-dimensional pitch rate $\hat q$? ::: $\hat q = qd/(2V)$, pitch rate scaled by chord-time. If you move fins from 0.6 m to 1.2 m behind CG, damping changes by what factor? ::: ×4 (because ∝ $x^2$). --- > [!recall]- Feynman: explain to a 12-year-old > Imagine spinning on an office chair with your arms out. The air pushes back on your hands and slows you down — the faster you spin, the harder it pushes back. A rocket's fins are like those arms: when the rocket wobbles, the fins sweep through the air and get pushed in the direction that *stops* the wobble. Fins that are far back at the tail sweep through the air really fast AND have a long "handle," so they're double-good at stopping wobbles. If the fins are too small or the air is too thin (way up high), the wobble takes a long time to stop — and that's dangerous. > [!mnemonic] Remember it > **"Far Fins Fight Faster, Force ∝ x²."** And for the two rules: **"Spring says which way (α), Dashpot says how fast (q)."** You need both the spring ($C_{m_\alpha}$) and the dashpot ($C_{m_q}$). ## Connections - [[Static Stability — Center of Pressure & Margin]] — supplies $C_{m_\alpha}$; needed together with $C_{m_q}$. - [[Damped Harmonic Oscillator]] — the ODE template ($\zeta$, $\omega_n$) all this reduces to. - [[Fin Design & Sizing]] — how geometry sets the $x^2$ integral. - [[Atmospheric Density vs Altitude]] — why $\rho\downarrow$ erodes damping in flight. - [[Barrowman Equations]] — the standard way the area distribution $c_{n_\alpha}(x)$ is computed. - [[Moments of Inertia of a Rocket]] — the $I_y$ that sets $\omega_n$. ## 🖼️ Concept Map ```mermaid flowchart TD SS[Static stability] -->|causes overshoot| OSC[Oscillation after snap-back] OSC -->|question| DS[Dynamic stability] DS -->|needs| DM[Damping moment] DM -->|quantified by| DD[Damping derivatives Cmq Cnr] DD -->|defined as| DEF[dCm / d q-hat] Q[Pitch rate q] -->|rigid-body rotation| WL[Local velocity w = q x] WL -->|divide by V| DA[Local angle of attack q x / V] DA -->|linear aero| DN[Incremental normal force dN] DN -->|times arm x, integrate| M[Restoring torque M] M -->|opposes rotation| DM DD -->|axisymmetry| SYM[Cnr equals Cmq] DD -->|stable if| NEG[Cmq less than 0] NEG -->|oscillations| DECAY[Oscillations die out] ``` ## 🔊 Hinglish (regional understanding) > [!intuition]- Hinglish mein samjho > Dekho, sirf **static stability** hona kaafi nahi hai. Static stability ka matlab hai rocket ka nose wapas hawa ki taraf mudh jaata hai — lekin jab wapas mudta hai to overshoot karta hai, phir doosri taraf jhoolta hai, phir wapas. Yani **oscillation** shuru ho jaata hai. **Dynamic stability** yeh poochhti hai: kya yeh jhoolna dheere-dheere ruk jaayega (achha) ya badhta jaayega aur rocket tumble kar jaayega (bahut kharaab)? > > Jo cheez oscillation ko rokti hai woh hai **damping moment** — ek torque jo hamesha rotation *rate* ka virodh karta hai, bilkul shock absorber ki tarah. Iski strength ko $C_{m_q}$ (pitch) aur $C_{n_r}$ (yaw) kehte hain. Physics simple hai: jab rocket pitch karta hai rate $q$ se, to CG se $x$ door wali har surface transverse velocity $qx$ paati hai — yani ek extra local angle of attack. Yeh angle ek force banata hai jo lever arm $x$ par torque deta hai. Dono jagah $x$ aata hai, isliye damping **$x^2$** ke saath badhta hai. Matlab fins ko peeche shift karo, damping char guna! Yeh yaad rakhna: linear nahi, square. > > Ek important cheez: pura rocket ek **damped harmonic oscillator** ban jaata hai. Static term ($C_{m_\alpha}$) spring ki tarah kaam karta hai (frequency $\omega_n$), aur damping term ($C_{m_q}$) dashpot ki tarah (damping ratio $\zeta$). Safe hone ke liye **dono** chahiye — $C_{m_\alpha}<0$ aur $C_{m_q}<0$. Aur ek trap: jaise-jaise rocket upar jaata hai, air density $\rho$ kam hoti hai, to actual damping moment ($\propto \rho V$) girta hai. Isliye har altitude par $\zeta$ check karo, sirf launch par nahi. Bas itna samajh lo to yeh topic tumhara ho gaya. ![[audio/3.4.11-Dynamic-stability-—-pitch---yaw-damping-derivatives.mp3]]