Combustion instability — low-frequency (chugging), high-frequency (screaming)

WHAT is combustion instability?

WHY two names? Because there are two different physical reservoirs that can time-store the disturbance and hand it back in phase. Slow reservoir = the propellant feed lines. Fast reservoir = the standing sound waves inside the chamber.

The universal condition: Rayleigh's Criterion

Derivation from first principles (why this is the master rule): Take the linearized energy equation for a gas parcel. The acoustic energy $E$ changes because unsteady heat release does work on the acoustic field. For a perfect gas the rate of acoustic energy addition per unit volume is $\frac{dE}{dt} \;=\; \frac{\gamma-1}{\gamma\,\bar p}\; p'\,q',$ where $q'$ is the fluctuating heat-release rate per unit volume, $\bar p$ the mean pressure, $\gamma$ the specific-heat ratio.

• WHY the $p'q'$ product? Heat added to a gas raises its pressure ($p \propto$ heat at fixed volume). If you add heat while pressure is already high, you push pressure even higher — constructive. The overlap of the two signals is measured by their product.
• HOW we get a cycle criterion: Net energy gained per cycle is $\displaystyle \oint \frac{dE}{dt}\,dt \propto \oint p'q'\,dt$. If positive, each cycle leaves more energy than the last → growth, until nonlinear losses (nozzle radiation, viscous damping) cap the amplitude into a limit cycle.

So both chugging and screaming are the same Rayleigh condition — they differ only in what sets the time delay between a pressure blip and the resulting heat release.

Low-frequency: CHUGGING

HOW the frequency is set — derive the chug frequency.

Chamber mass balance: gas stored $\propto p_c$, so $\frac{V}{a^2}\frac{dp_c}{dt} = \dot m_{in} - \dot m_{out}.$

• $\dot m_{out}$ from a choked nozzle: $\dot m_{out} = \dfrac{p_c A_t}{c^*}$ (rises with $p_c$).
• $\dot m_{in}$ from injector: $\dot m_{in} \propto \sqrt{\Delta p_{inj}} = \sqrt{p_{feed}-p_c}$, but delayed by combustion lag $\tau$: it responds to $p_c(t-\tau)$.

Linearize ($p_c = \bar p_c + p'$) and you get a delay-differential equation: $\theta_c \frac{dp'}{dt} + p'(t) = -k\,p'(t-\tau),$ where $\theta_c = V/(a^2)\cdot(\text{gain terms})$ is the chamber time constant and $k$ the feed-coupling gain. Seeking $p'\sim e^{i\omega t}$, marginal instability occurs when the delayed feedback is in phase, roughly when $\boxed{\,\omega \tau \approx \tfrac{\pi}{2}\text{–}\pi \;\Rightarrow\; f_{chug}\sim \frac{1}{4\tau}\,}$

• WHY $1/\tau$ scaling? The only slow clock in the loop is the combustion/transit delay $\tau$ (milliseconds). $\tau \sim 2$ ms → $f \sim 125$ Hz. That's the chugging band.
• The fix (WHY it works): Increase injector $\Delta p_{inj}$. A stiff injector makes $\dot m_{in}$ nearly independent of $p_c$ — you break the feedback gain $k$. Rule of thumb: keep $\Delta p_{inj} \gtrsim 0.2\,p_c$.

High-frequency: SCREAMING (screech)

HOW the frequency is set — acoustic modes of the chamber. The chamber is a resonant cavity of speed of sound $a=\sqrt{\gamma R T_c}$. Its natural frequencies are set by the cavity geometry, e.g. for a cylinder the transverse (tangential/radial) and longitudinal modes: $\boxed{\,f_{mn\ell} = \frac{a}{2}\sqrt{\left(\frac{\alpha_{mn}}{\pi R}\right)^2 + \left(\frac{\ell}{L}\right)^2}\,}$ with $R$ chamber radius, $L$ length, $\ell$ longitudinal index, $\alpha_{mn}$ Bessel roots for the transverse pattern.

• WHY kHz instead of Hz? $a$ in hot gas is $\sim 1000$ m/s and $R\sim 0.1$ m, so $f\sim a/(2\pi R)\sim$ a few kHz. The clock here is the acoustic transit time across the chamber, microseconds — a thousand times faster than the feed lag → screaming.
• The most dangerous: the first tangential (1T) mode, because its pressure sloshes side to side, scrubbing hot gas against the wall.
• The fix (WHY it works): Acoustic damping — install baffles on the injector face (break up transverse modes) and Helmholtz-resonator acoustic cavities in the chamber liner (absorb energy at the target frequency). These make $\oint p'q'\,dt$ net-negative by adding loss.

Worked Examples

Common Mistakes

Flashcards

Connections

• Rayleigh Criterion — the master energy condition for all thermoacoustic growth.
• Injector Design & Pressure Drop — the chugging cure lives here.
• Acoustic Modes of a Cylindrical Cavity — sets screaming frequencies.
• Helmholtz Resonator and Injector Baffles — high-frequency damping.
• Choked Nozzle Flow — sets $\dot m_{out}\propto p_c$ used in the chug mass balance.
• Characteristic Velocity c* — combustion efficiency term in the chamber balance.
• Thermoacoustics — general field encompassing this topic.

Concept Map

Hinglish (regional understanding)