Injector design — impinging, coaxial, swirl injectors: Question bank

Before the questions, a one-line refresher on the symbols so nothing here is used unearned:

Recall Symbol refresher (open if any notation feels unfamiliar)

$\overset{m}{˙}$ ::: mass flow rate (kilograms of propellant per second through the orifice). $Δ p$ ::: injector pressure drop = manifold pressure minus chamber pressure, $p_{1} - p_{c}$ . $C_{d}$ ::: discharge coefficient, a fudge factor between 0 and 1 that folds in friction and jet contraction (vena contracta). $J$ ::: momentum flux ratio of the two coaxial streams, $ρ_{o} v_{o}^{2} / ρ_{i} v_{i}^{2}$ (outer over inner). $ϕ$ ::: swirl half-cone angle, with $tan ϕ = v_{t} / v_{x}$ (tangential spin speed over axial speed). $c^{*}$ ::: characteristic velocity, a scoreboard for how completely combustion happened — see Characteristic Velocity c-star.

True or false — justify

The naive orifice law

\overset{m}{˙} = C_{d} A 2 ρ Δ p

says doubling

Δ p

doubles the mass flow.

False —

\overset{m}{˙} \propto Δ p

, so doubling

Δ p

multiplies flow by only

2 \approx 1.41

. You would need to quadruple

Δ p

to double the flow.

A discharge coefficient

C_{d}

greater than 1 is possible for a very smooth, well-rounded orifice.

False —

C_{d}

accounts for losses (friction, vena contracta), so it can only reduce the ideal flow; the physical ceiling is

C_{d} = 1

, approached but never exceeded.

A swirl injector can atomize a single propellant on its own, with no second stream present.

True — swirl relies on centrifugal thinning of a self-supported hollow-cone sheet, so one propellant spun hard enough atomizes itself. See Atomization and the d-squared Law.

For a coaxial injector, if both streams move at the same velocity (

v_{o} = v_{i}

) the outer stream still atomizes the inner jet well.

False — coaxial atomization is driven by the relative velocity

Δ v = v_{o} - v_{i}

; with

v_{o} = v_{i}

there is no shear and the central jet barely breaks up.

A balanced (symmetric) unlike-doublet impinging pair, aimed at equal angles from the axis, sprays straight down the chamber axis.

True — the transverse momentum components cancel, so by momentum conservation the resultant sheet leaves along the axis (

α = 0

), which is exactly what avoids wall-scrubbing.

Raising the injector

Δ p

always improves combustion stability.

Mostly true up to a point — higher

Δ p

better decouples the feed system from chamber oscillations, but pushing it far higher wastes pump power for little extra stability, so designers stop near

Δ p ≳ 0.15 p_{c}

. See Combustion Instability.

The

Δ p

scaling comes from Bernoulli Equation, where pressure work is converted to kinetic energy

\frac{1}{2} ρ v^{2}

True — because kinetic energy goes as

v^{2}

, solving

Δ p = \frac{1}{2} ρ v^{2}

for

v

produces the square root, and

\overset{m}{˙} \propto v

inherits it.

Finer atomization (smaller droplets) reduces the required combustion chamber length.

True — evaporation time scales as

d^{2}

(the

d^{2}

-law), so smaller droplets burn faster and the propellant finishes reacting in a shorter chamber.

A wider swirl cone angle

ϕ

generally means a thinner sheet and finer atomization.

True — more tangential velocity widens the cone and stretches the sheet thinner, and thin sheets go unstable to ripples and shed droplets quickly.

Using many small orifices instead of one large one changes the total metered mass flow if

Δ p

is unchanged.

False — total flow depends on total open area; splitting the same area into many holes keeps

\overset{m}{˙}

fixed while improving mixing (more, finer jets).

Spot the error

"Since

\overset{m}{˙} = C_{d} A 2 ρ Δ p

, a denser propellant flows slower through the same orifice, so use it to reduce flow."

Error — density appears inside the mass-flow formula, and higher

ρ

raises

\overset{m}{˙}

(more mass per unit volume). The velocity

v = 2Δ p / ρ

drops with density, but mass flow, which is what's asked, increases.

"Coaxial and swirl injectors atomize identically; they just make differently-shaped clouds."

Error — coaxial uses shear from a second high-velocity stream (needs large

Δ v

); swirl uses centrifugal thinning of its own sheet (works at low

Δ p

, even solo). Different physics, different failure modes.

"Lowering

Δ p

is free efficiency because pump work drops and mixing is a separate concern."

Error — too-low

Δ p

lets chamber pressure oscillations push back through the orifice, coupling feed and chamber into Combustion Instability. Metering, stability, and

Δ p

are not separable.

"Large droplets are fine, the chamber is hot so they'll burn eventually."

Error — burn time

\propto d^{2}

, so doubling diameter quadruples burn time; if the droplet exits the finite chamber unburned,

c^{*}

efficiency drops. See Characteristic Velocity c-star.

"For the impinging spray angle

tan α = \frac{m ˙ _{1} v _{1} sin θ _{1} - m ˙ _{2} v _{2} sin θ _{2}}{m ˙ _{1} v _{1} cos θ _{1} + m ˙ _{2} v _{2} cos θ _{2}}

, the numerator sums both jets so a symmetric pair sprays sideways."

Error — the two jets sit on opposite sides of the axis, so their transverse (

sin

) terms carry opposite signs and subtract to zero; the numerator vanishes and

α = 0

, i.e. straight down.

"Since

tan ϕ = v_{t} / v_{x}

, a swirl injector with zero axial velocity produces a

0^{\circ}

(pencil) jet."

Error — with

v_{x} \to 0

the ratio

v_{t} / v_{x} \to \infty

, so

ϕ \to 9 0^{\circ}

: a flat radial disc, not a pencil. It is large

v_{x}

that narrows the cone.

Why questions

Why does the mass-flow law contain a square root rather than a linear dependence on

Δ p

Because pressure work converts to kinetic energy

\frac{1}{2} ρ v^{2}

, and energy scales with

v^{2}

; inverting that relationship to get

v

introduces the square root, which

\overset{m}{˙} \propto v

then carries.

Why do designers deliberately keep

Δ p

at roughly 15–25% of chamber pressure rather than minimising it?

A stiff pressure drop makes the flow through the orifice nearly insensitive to chamber pressure wobbles, decoupling the feed system from combustion oscillations and suppressing instability.

Why must the outer stream of a coaxial injector be fast precisely because it is light (e.g. gaseous H₂)?

Atomization is driven by momentum flux

ρ v^{2}

; low density

ρ

must be compensated by high

v

to build enough

J

to shear the dense central jet apart.

Why does a thin liquid sheet atomize better than a thick jet?

A thin sheet is easily destabilised by small surface ripples that grow, tear it into ligaments, and then into fine droplets — a thick jet resists this and stays coherent much longer.

Why does an unlike-impinging doublet mix and atomize at the same point, while like-on-like separates the two tasks?

In unlike impingement, fuel meets oxidizer at the collision, so the smashing sheet is already the mixed sheet; like-on-like collides same-with-same, atomizing first and relying on later diffusion for mixing.

Why do impinging designers "null the transverse momentum" of the resultant sheet?

So the spray goes straight down the axis instead of angling toward a wall; a sideways spray scrubs hot combustion against the chamber wall and risks burn-through. See Regenerative Cooling.

Why is the local mixture ratio at each injector element, not just the overall chamber

O / F

, what actually controls performance and wall heat?

Combustion happens locally in milliseconds, so a globally-correct chamber can still have fuel-rich and ox-rich streaks that hurt efficiency or overheat walls — see O/F Ratio and Mixture Ratio.

Edge cases

What happens to the jet velocity

v = 2Δ p / ρ

when

Δ p = 0

v = 0

: with no pressure difference there is no driving work, so nothing flows — the orifice is effectively closed regardless of its area.

What does the impinging formula give if only one jet fires (

\overset{m}{˙}_{2} v_{2} = 0

tan α = tan θ_{1}

, so the "resultant" is just the surviving jet at its own angle — the balance is lost and the single jet sprays off-axis, likely toward a wall.

For a swirl injector, what is the cone angle in the limit of pure axial flow,

v_{t} = 0

tan ϕ = 0/ v_{x} = 0

, so

ϕ = 0

: a straight pencil jet with no swirl — poor atomization, because the centrifugal thinning mechanism has switched off.

For a coaxial injector, what is

J

if the inner jet is momentarily stalled (

v_{i} = 0

J = ρ_{o} v_{o}^{2} /0 \to \infty

: the formula diverges, signalling maximal shear on a motionless core — practically the outer stream dominates entirely and violently shreds the inner jet.

What is the physical meaning of a discharge coefficient approaching

C_{d} \to 0

The orifice is passing almost no flow for its geometric area — extreme choking or blockage; nearly all the available pressure work is lost rather than becoming useful jet velocity.

What limits atomization as droplet diameter

d \to 0

in the

d^{2}

-law picture?

Burn time collapses toward zero, which is ideal for completeness, but real sprays cannot shrink

d

indefinitely — energy cost and sheet stability set a practical floor. See Atomization and the d-squared Law.