Visual walkthrough — Reaction control system — thruster selection, plume impingement limits

3.5.46 · D2 · Physics › Guidance, Navigation & Control (GNC) › Reaction control system — thruster selection, plume impingem

Yeh page Reaction control system — thruster selection, plume impingement limits ka picture-companion hai. Prerequisite ideas jin par hum rely karte hain woh Cross Product & Rigid-Body Torque, Attitude Dynamics — Euler's Equations, aur Control Allocation & Pseudo-inverse mein hain.

---

Step 1 — Ek thruster ek aisa arrow hai jo kisi jagah rahta hai

KYA HAI. Spacecraft ko ek solid brick samjho. Kahin uske beech mein ek khaas point hota hai: center of mass (CoM) — balance point, woh jagah jiske around poora body ghoomta hai jab tum use free space mein spin karo. Bahar ki taraf ek chhoti nozzle bolted hai. Jab yeh fire karti hai, hot gas ek taraf shoot karta hai aur ship ko doosri taraf push milti hai.

DO facts kyun, ek nahi. Push ko sirf "kitni zyada" se describe nahi kiya ja sakta. Yeh bhi matter karta hai ki kahan apply ki gayi aur kis taraf point karti hai. Toh har thruster $i$ do arrows carry karta hai:

• ${\mathbf{r}}_i$ — position arrow: CoM se start hota hai, nozzle par end hota hai. Yeh lever arm hai.
• ${\hat{\mathbf{u}}}_i$ — direction arrow: ek length-1 arrow jo woh direction batata hai jis taraf force act karta hai (chhoti hat $\hat{}$ ka matlab hai "length exactly 1", ek pure direction bina kisi size ke).

PICTURE. Brick, CoM dot, aur ek nozzle apne do arrows ke saath.

---

Step 2 — Force sirf "hardness times direction" hai

KYA HAI. Push-hardness $F_i$ ko direction arrow ${\hat{\mathbf{u}}}_i$ se multiply karo:

$\underset{\text{full force arrow}}{\underbrace{{\mathbf{F}}_i}}=\underset{\begin{array}{c}\text{how hard}\\ \ge 0\end{array}}{\underbrace{F_i}}\underset{\begin{array}{c}\text{which way}\\ \text{length }1\end{array}}{\underbrace{{\hat{\mathbf{u}}}_i}}.$

YEH SHAPE KYU. Ek unit arrow ki length 1 hoti hai; ise $F_i$ se scale karne par yeh length $F_i$ tak stretch ho jaata hai lekin direction same rehti hai. Toh ${\mathbf{F}}_i$ ek aisa arrow hai jo woh taraf point karta hai jis taraf thruster push karta hai aur utna lamba hota hai jitni zyada push hoti hai. Simple hai, lekin yeh woh atom hai jis se baaki sab kuch banta hai.

PICTURE. Unit arrow ${\hat{\mathbf{u}}}_i$ aur stretched arrow ${\mathbf{F}}_i$ side by side; same direction, zyada lamba.

---

Step 3 — Cross product kyun ek push ko twist mein convert karta hai

KYA HAI. Us force ka turning effect (torque) hai

${\mathbf{\tau }}_i=\underset{\text{lever arm}}{\underbrace{{\mathbf{r}}_i}}\times \underset{\text{force}}{\underbrace{{\mathbf{F}}_i}}=F_i({\mathbf{r}}_i\times {\hat{\mathbf{u}}}_i).$

$\times$ (cross product) kyun, ordinary multiplication kyun nahi. Asli sawaal poochho: yeh push body ko kitna spin karne ki koshish karta hai? Do cheezein decide karti hain. (a) CoM se door ki push CoM ke paas ki push se zyada twist karti hai — toh ${\mathbf{r}}_i$ ki length matter karti hai. (b) CoM ki taraf seedha aimed push kuch bhi spin nahi kar sakta; lever arm ke sideways aimed push sabse zyada spin karta hai — toh ${\mathbf{r}}_i$ aur ${\mathbf{F}}_i$ ke beech ka angle matter karta hai. Cross product precisely woh tool hai jo "lever length times force times kitna perpendicular hain" ka jawab deta hai, aur yeh spin axis ke saath ek arrow return karta hai. Koi doosra single operation yeh teeno facts package nahi karta. (Cross product ki full derivation Cross Product & Rigid-Body Torque mein hai.)

Result arrow padhna. ${\mathbf{\tau }}_i$ us axis ke saath point karta hai jiske around body spin karta hai; uski length twisting strength hai; apne right hand ki ungliyon ko ${\mathbf{r}}_i$ se ${\mathbf{F}}_i$ ki taraf point karo aur thumb arrow ki direction deta hai.

PICTURE. Lever arm, force, aur resulting torque arrow plane se bahar nikalta hua, swept angle shaded ke saath.

---

Step 4 — Do degenerate cases jo tumhe kabhi nahi bhuolne chahiye

KYA HAI. Extreme geometries dekho taaki baad mein koi case surprise na kare.

• Case A — push lever line ke saath aimed (${\mathbf{r}}_i$ collinear with ${\hat{\mathbf{u}}}_i$ — same way pointing ya exact opposite way): arrows ek line par hain, "kitna perpendicular" factor zero hai, toh ${\mathbf{\tau }}_i=0$ dono cases mein. Same-direction ka matlab force radially outward CoM line ke through point karti hai; opposite-direction ka matlab yeh radially inward same line ke saath point karti hai — dono CoM se pass karte hain, toh koi bhi body ko spin nahi kar sakta. Pure translation. Ship slide karta hai, kabhi spin nahi karta.
• Case B — push lever ke bilkul sideways (${\mathbf{r}}_i$ perpendicular to ${\hat{\mathbf{u}}}_i$): "kitna perpendicular" factor maximum hota hai, jo har newton par sabse bada twist deta hai, torque length $=F_i|{\mathbf{r}}_i|$.

YEH KYUN MATTER KARTA HAI. Real thrusters in dono ke beech kahin hote hain. Agli machine ka har column secretly "kitna shove karta hai" aur "kitna twist karta hai" ka blend hai, aur woh blend bilkul is angle se set hota hai. Dono ends jaanna tumhe kisi bhi thruster ko ek nazar mein sanity-check karne deta hai.

PICTURE. Do mini-diagrams side by side: aimed-along-the-lever (no spin, dono ways dikhaya gaya) vs. aimed-sideways (max spin).

---

Step 5 — Shove aur twist ko ek 6-number "wrench" mein stack karo

KYA HAI. Force 3 numbers mein rehti hai (x, y, z). Torque 3 numbers mein rehta hai (x, y, z ke around spin). Inhe ek tall stack of 6 mein glue karo:

$\mathbf{w}=\left[\begin{array}{c}\mathbf{F}\\ \mathbf{\tau }\end{array}\right]=\underset{\text{6 rows}}{\underbrace{\left[\begin{array}{c}{F}_x\\ F_y\\ F_z\\ {\tau }_x\\ {\tau }_y\\ {\tau }_z\end{array}\right]}}.$

INHE GLUE KYUN KARO. Controller jo bhi command issue karta hai woh hai "yeh net shove aur yeh net twist produce karo." Inhe ek object $\mathbf{w}$ (Attitude Dynamics — Euler's Equations se wrench) ke roop mein carry karna ek equation ko translation aur rotation saath mein handle karne deta hai. Top 3 rows = ship kahan jaata hai; bottom 3 rows = yeh kaise muda hai.

PICTURE. Ek single tall bracket, top half "force" se tinted, bottom half "torque" se tinted.

---

Step 6 — Thrusters ko columns ki tarah line up karo: allocation matrix $B$

KYA HAI. Har thruster apna chhota wrench contribute karta hai. Ek thruster ka unit wrench ($F_i=1$ par uska wrench) 6 numbers ke column ke roop mein likho:

${\mathbf{b}}_i=\left[\begin{array}{c}{\hat{\mathbf{u}}}_i\\ {\mathbf{r}}_i\times {\hat{\mathbf{u}}}_i\end{array}\right].$

Saare $N$ columns ko side by side ek grid $B$ mein khada karo (6 rows tall, $N$ thrusters wide). Push-hardnesses ko ek list $\mathbf{f}=[F_1,\dots ,F_N{]}^{\mathsf{T}}$ mein collect karo. Tab total wrench ek matrix–vector product hai:

$\boxed{\mathbf{w}=\underset{6\times N}{\underbrace{B}}\underset{N\times 1}{\underbrace{\mathbf{f}}},\mathbf{f}\ge 0.}$

YEH POORA IDEA KYUN HAI. Reading $B\mathbf{f}$ left to right: har thruster ka column ${\mathbf{b}}_i$ ko weight milta hai ki woh kitna hard fire kar raha hai, $F_i$, aur saare weighted columns add up hote hain. Woh sum exactly "yeh mix fire karo, yeh net wrench pao" hai. One-sided rule $\mathbf{f}\ge 0$ Step 1 se physics reminder hai: koi bhi jet pull nahi kar sakta. Yeh parent note ka central result hai — aur yeh Steps 2–5 ko har thruster ke liye ek baar likha aur stack kiya gaya hai.

PICTURE. Ek body par char thrusters, har ek ko phir se grid $B$ ko feed karne wale 6-tall column ke roop mein draw kiya gaya; list $\mathbf{f}$ unhe scale karti hai aur woh $\mathbf{w}$ mein sum hote hain.

---

Step 7 — Worked couple: zero drift ke saath twist

KYA HAI. Ek wheel of radius $r$ par do thrusters. Thruster 1 ${\mathbf{r}}_1=(0,+r,0)$ par baitha hai aur $+x$ mein push karta hai; thruster 2 ${\mathbf{r}}_2=(0,-r,0)$ par baitha hai aur $-x$ mein push karta hai. Dono same hardness $F$ se fire karte hain.

Step-by-step.

Net force (wrench ka top): $\mathbf{F}=F(1,0,0)+F(-1,0,0)=(0,0,0).$ Do shoves opposite taraf point karte hain aur cancel ho jaate hain — koi drift nahi, exactly wahi jo Step 4 Case B ne promise kiya tha ki hum ek thruster se nahi pa sakte.

Har torque (wrench ka bottom), $\mathbf{r}\times \mathbf{F}$ use karke: ${\mathbf{\tau }}_1=(0,r,0)\times (F,0,0)=(0,0,-rF),{\mathbf{\tau }}_2=(0,-r,0)\times (-F,0,0)=(0,0,-rF).$

Add karo: $\mathbf{\tau }=(0,0,-2rF).$

YEH CANCEL KYUN NAHI HOTE, ADD KYUN HOTE HAIN. Dono jets body ko $z$ ke around same taraf spin karne ki koshish karte hain (top jet ko ek taraf, bottom jet ko doosri taraf fire karo — jaise tumhare do haath steering wheel ghuma rahe ho). Opposite shoves, same twist. Is pairing ko couple kehte hain: pure torque, zero net force, length $2rF$.

PICTURE. Wheel, do opposing shoves cancel karte hue, do same-sense spins ek $z$-torque mein add hote hue.

---

Step 8 — Plume: kyun answer geometry ko bend karta hai

KYA HAI. Woh gas jisne force banai woh nozzle lip par vanish nahi hoti. Yeh near-vacuum mein ek widening cone ke roop mein fan out hoti hai. Yeh kisi surface ko distance $d$ par, cone ke centerline se angle $\theta$ par, jo strength deliver karta hai woh hai

$P_{\mathrm{imp}}(d,\theta )\approx \underset{\begin{array}{c}\text{reference}\\ \text{pressure}\end{array}}{\underbrace{P_0}}\underset{\begin{array}{c}\text{spreads over}\\ \text{a sphere}\end{array}}{\underbrace{{\left(\frac{d_0}{d}\right)}^2}}\underset{\begin{array}{c}\text{tightness of}\\ \text{the cone}\end{array}}{\underbrace{{\cos }^k\theta }}.$

HAR PIECE KYU.

• $P_0$ aur $d_0$ ek known calibration pair banaate hain: $d_0$ ek chosen reference distance hai (nozzle exit se ek fixed short distance jahan humne plume measure kiya ya compute kiya hai, e.g. $d_0=0.1$ m), aur $P_0$ woh pressure hai jo plume us reference distance par, on axis deliver karta hai. Baaki sab kuch is ek anchor point ke relative scale hota hai — tum "$d$ distance par pressure" nahi bol sakte bina pehle "kisi known distance par pressure" pin kiye. Toh $d_0$ koi naya physical effect nahi hai; yeh woh yardstick hai jo shape $1/d^2$ ko ek actual number mein convert karta hai.
• $(d_0/d{)}^2$: wahi gas travel karne par ever-larger spherical shells se pass hoti hai, toh uski intensity $1/d^2$ se dilute hoti hai. $d=d_0$ par yeh factor 1 ke barabar hai (tum $P_0$ recover karte ho), aur uske baad pressure fall karta hai — distance half karo aur pressure chaar guna ho jaata hai. Yahi reason hai ki close surfaces (deployed panels, ek docking target) danger mein hain, exactly wahi jo parent warn karta hai.
• ${\cos }^k\theta$: gas seedha aage sabse dense hoti hai ($\theta =0$, jahan ${\cos }^{k}0=1$) aur cone ke rim ki taraf thin hoti hai. Exponent $k$ plume ka focusing number hai: $k=0$ ka matlab hoga gas equally sab directions mein spread hoti hai (ek hemisphere, koi preferred direction nahi); small $k$ (roughly $1$–$2$) ek wide, gentle fan hai; large $k$ (roughly $4$–$40$ real space thrusters ke liye vacuum mein fire karte hue) ek tight, pencil-like jet hai jahan almost saari punch dead ahead hai aur rim almost empty hai. Bada $k$ = tighter, zyada focused plume. Yeh har nozzle ke liye test data ya Rarefied Gas Dynamics / Plume Modeling se fit kiya jaata hai.

Rule. Ek thruster allow tabhi hai jab fire kare, jab har sensitive surface $s$ ke liye, $P_{\mathrm{imp}}\le P_{\max ,s}$ ho. Yahan subscript $s$ label karta hai ki hum kaunsi surface ki baat kar rahe hain (solar panel, star tracker, docking target, ship belly, …); har ek kam ya zyada fragile hoti hai, toh har ek apna khud ka allowable pressure $P_{\max ,s}$ carry karta hai — woh sabse zyada impingement pressure jo surface $s$ safely tolerate kar sakti hai bina damage hue ya zyada push hue. Ek rugged radiator ka $P_{\max ,s}$ bada hota hai; ek delicate optic ka bahut chhota hota hai. Kyunki limit har surface per alag hai, hum har surface ko separately check karte hain. Practice mein yeh keep-out cone ban jaata hai: koi bhi delicate surface nozzle ke forward cone ke andar kisi range ke andar nahi honi chahiye.

Worked violation (parent se, re-derived): docking ke dauran ek forward thruster target ki taraf seedha point karta hai, $d=2$ m, $\theta =0$, $P_0=500$ Pa reference distance $d_0=0.1$ m par, $k=4$, panel limit $P_{\max ,s}=1$ Pa ke saath.

$P_{\mathrm{imp}}=500{\left(\frac{0.1}{2}\right)}^2{\cos }^{4}0=500\times (0.05{)}^2\times 1=500\times 0.0025=1.25\text{Pa}.$

Kyunki $1.25>1.0$, woh thruster inhibited hai — $F_i=0$ par locked hai. Step 6 ka selection ab $B\mathbf{f}={\mathbf{w}}_{\mathrm{cmd}}$ sirf remaining columns use karke solve karna padega, chahe woh thoda zyada propellant jalaye. Safety fuel se zyada zaruri hai.

PICTURE. Plume cone, $1/d^2$ shells, ${\cos }^k\theta$ falloff, aur ek panel keep-out cone ke andar baitha hua violation trigger karta hua.

---

Ek-picture summary

Ek flow: har thruster = (jagah ${\mathbf{r}}_i$ + direction ${\hat{\mathbf{u}}}_i$) → shove $F_i{\hat{\mathbf{u}}}_i$ aur twist $F_i({\mathbf{r}}_i\times {\hat{\mathbf{u}}}_i)$ → column ${\mathbf{b}}_i$ mein stack karo → saare columns $B$ mein line up karo → net command $\mathbf{w}=B\mathbf{f}$ hai $\mathbf{f}\ge 0$ ke saath → lekin kisi bhi firing thruster ka plume cone feasible set ko chhota kar deta hai (keep-out cones), toh final selection woh sabse sasta firing mix hai jo ${\mathbf{w}}_{\mathrm{cmd}}$ bhi hit kare aur kisi cheez par spray bhi na kare jo hurt ho sake.

Recall Feynman retelling — plain words mein bol ke sunao

Imagine karo tum space mein ek chair par float kar rahe ho jisme kuch chhote spray cans tape kiye hue hain, har ek ek fixed taraf point karta hai aur sirf spray kar sakta hai (kabhi suck nahi). Har can ek saath do kaam karta hai: yeh tumhe shove karta hai jis taraf point karta hai, aur agar yeh side mein hai toh yeh tumhe spin bhi karta hai. Yeh kitna spin karta hai woh is par depend karta hai ki yeh kitna bahar hai aur kitna sideways spray karta hai — woh "kitna far aur kitna sideways" combined cross product hai, aur yeh spin axis ke saath ek arrow deta hai. Main har can ka shove aur spin chhe numbers ke ek column ke roop mein likhta hoon, saare columns ek grid $B$ mein rakhta hoon, aur phir "kaunse cans, kitna hard" sirf ek list $\mathbf{f}$ of hardnesses hai (sab positive). Total effect $B\mathbf{f}$ hai. Agar main drift kiye bina muda chahta hoon, main do opposite cans fire karta hoon taaki unke shoves cancel ho jaayein lekin unke spins add ho jaayein — ek couple. Aakhir mein, spray khud ek cone hai jo distance ke saath weak hoti hai (flashlight beam ki tarah, $1/d^2$) aur side mein bhi weak hoti hai (${\cos }^k\theta$); agar woh cone meri solar panel ya kisi nearby ship ko us surface se zyada hit kare jo woh le sakti hai, main us can ko forbid kar deta hoon aur baaki ke saath re-solve karta hoon. Yeh poora reaction control system ek saansh mein hai.

Recall Quick self-test

Kaunsa operation force-at-a-place ko torque mein convert karta hai, aur kyun wahi ek? ::: Cross product ${\mathbf{r}}_i\times {\mathbf{F}}_i$ — yeh lever length, force size, aur kitne perpendicular hain package karta hai, aur spin-axis arrow return karta hai. $\mathbf{f}\ge 0$ kyun hona chahiye? ::: Ek jet sirf push (spray) kar sakta hai, kabhi pull nahi, toh har thrust magnitude zero ya positive hoti hai. Bina drift ke pure torque kaise paaoge? ::: Ek opposing pair (ek couple) fire karo: opposite shoves cancel ho jaate hain, same-sense spins $2rF$ mein add ho jaate hain. Nearby surfaces ke liye plume pressure itni fast kyun badhti hai? ::: Yeh $1/d^2$ se dilute hoti hai, toh distance half karne par pressure chaar guna ho jaata hai. Plume formula mein $d_0$ kya hai? ::: Ek chosen reference distance jahan $P_0$ known hai; $d=d_0$ par $1/d^2$ factor 1 hai aur $P_{\mathrm{imp}}=P_0$ on axis hota hai. $P_{\max ,s}$ mein subscript $s$ ka kya matlab hai? ::: Yeh specific sensitive surface ko label karta hai; har surface ka apna allowable pressure hota hai, toh har surface ko separately check kiya jaata hai. Docking example mein forward thruster kyun inhibited hai? ::: $P_{\mathrm{imp}}=1.25$ Pa panel limit $1.0$ Pa se zyada hai, toh yeh $F_i=0$ par locked hai.