2.1.23 · D3 · Physics › Analytical Mechanics › Torque-free rotation — Euler's equations, asymmetric top
Yeh page parent topic ka drill ground hai. Yahan hum Euler's equations dobara derive nahi karte — balki unhe har tarah ke input par exercise karte hain. Agar koi symbol unfamiliar lage, toh woh parent note mein build hua tha; neeche ke quick reminders unhe dobara anchor karte hain.
Recall Quick symbol re-anchor (padho agar koi notation naya lage)
I 1 , I 2 , I 3 — teen principal moments of inertia : body kitni mushkil se apne teen special body-glued axes ke baare mein spin resist karti hai. Inertia tensor and principal axes se build hua.
ω 1 , ω 2 , ω 3 — spin arrow ω ke teen components jo un body axes ke saath measure kiye jaate hain.
Ω (capital omega) — woh bada, near-constant spin rate jo us axis ke baare mein hai jise hum perturb kar rahe hain. Yeh kisi ek ω i ki ek specific value hai, koi naya object nahi.
ω ⊥ — transverse spin vector , do-component pair ( ω 1 , ω 2 ) ka shorthand jo axis 3 ke perpendicular plane mein rehta hai. Iska length hai ∣ ω ⊥ ∣ = ω 1 2 + ω 2 2 .
Ek dot matlab "rate of change per second": ω ˙ 1 = d ω 1 / d t .
T — rotational kinetic energy ; hum usually 2 T = I 1 ω 1 2 + I 2 ω 2 2 + I 3 ω 3 2 track karte hain. Koi external work na ho toh yeh conserved hai.
L = I ω — angular momentum . Iska length-squared L 2 = I 1 2 ω 1 2 + I 2 2 ω 2 2 + I 3 2 ω 3 2 torque-free motion mein conserved hai. Dekho Angular momentum in rotating frames .
k — dimensionless stability constant , Ex 3 mein defined — linearized wobble equation ko ω ¨ 1 = k Ω 2 ω 1 likh ke. Toh k ek pure number hai (isme Ω 2 nahi hai); Ω 2 har baar alag likha jaata hai. Iska sign stability decide karta hai: k < 0 ⇒ safe oscillation, k > 0 ⇒ runaway tumble.
Is topic ke har problem ko in cells mein se kisi ek mein daala ja sakta hai. Neeche ke examples har ek ko kam se kam ek baar hit karte hain. (Cells C–E mein jo stability constant k reference hai woh Example 3 mein defined hai — yahan use "woh pure-number sign jo stability decide karta hai" ka placeholder maano.)
Cell
Kya cheez ise woh case banati hai
Kyun iska alag treatment chahiye
Example
A. Saare I equal (sphere)
I 1 = I 2 = I 3
RHS → 0 : har derivative vanish, degenerate limit
Ex 1
B. Do I equal (symmetric top)
I 1 = I 2 = I 3
Ek equation khatam, doosre do steady precession dete hain
Ex 2
C. Saare I distinct, spin LARGEST axis ke paas
I 1 < I 2 < I 3 , spin ≈ axis 3
k < 0 ka sign ⇒ stable oscillation (k Ex 3 mein defined )
Ex 3
D. Saare I distinct, spin SMALLEST axis ke paas
spin ≈ axis 1
k < 0 ka sign ⇒ stable (woh doosra stable wala)
Ex 4
E. Saare I distinct, spin INTERMEDIATE axis ke paas
spin ≈ axis 2
k > 0 ka sign ⇒ exponential tumble
Ex 5
F. Conservation cross-check
koi bhi motion
Verify karo ki 2 T aur L 2 trajectory par actually constant hain
Ex 6
G. Zero / one-component spin (degenerate)
ω exactly ek axis ke saath
Pure principal-axis spin: RHS sab zero, wobble kabhi nahi
Ex 7
H. Real-world word problem
Earth ka Chandler wobble
Ek physics number ko ek real observable period mein convert karna
Ex 8
I. Exam twist (sign trap)
axis ordering scrambled
Check karta hai ki tum sign track karte ho, axis label nahi
Ex 9
Ek uniform solid sphere mein I 1 = I 2 = I 3 = I hai. Ise body frame mein ω = ( 2 , − 3 , 1 ) rad/s se spin diya gaya. Ek ghante baad ω kaisa hoga?
Forecast: Padhne se pehle andaza lagao — kya spin arrow drift karega, wobble karega, ya frozen rahega?
Teeno Euler equations mein plug karo.
I ω ˙ 1 = ( I − I ) ω 2 ω 3 = 0 , I ω ˙ 2 = 0 , I ω ˙ 3 = 0.
Yeh step kyun? Har coefficient do equal moments ka difference hai, isliye exactly 0 hai. Engine ke paas kuch bhi drive karne ki coupling nahi.
Integrate karo. ω ˙ 1 = ω ˙ 2 = ω ˙ 3 = 0 ⇒ ω = ( 2 , − 3 , 1 ) hamesha ke liye.
Yeh step kyun? Zero derivative matlab value kabhi nahi badlegi — body frame mein spin frozen hai.
Verify: Sphere ke liye I = I 1 , toh L = I ω ω ke parallel hai. Space mein L conserved hai (koi torque nahi), isliye ω bhi. Consistent. Ek ghante baad: abhi bhi ( 2 , − 3 , 1 ) rad/s.
Ek discus-jaisa symmetric top hai jisme I 1 = I 2 = I ⊥ = 2 , I 3 = 5 (kg·m²). Symmetry axis ke saath spin ω 3 = 4 rad/s hai, aur chota transverse start ω 1 ( 0 ) = 0.1 , ω 2 ( 0 ) = 0 hai. Body precession rate Ω b aur period nikalo.
Forecast: Kya transverse components decay karenge, blow up karenge, ya constant rate par circle karenge?
Pehle Euler 3. I 3 ω ˙ 3 = ( I 1 − I 2 ) ω 1 ω 2 = 0 , toh ω 3 = 4 constant rehta hai.
Yeh step kyun? Equal transverse moments woh coefficient zero kar dete hain, axial spin ko freeze kar dete hain — pehle lock down karne ki sabse clean quantity.
Euler 1 & 2 reduce karo. I 1 = I 2 = I ⊥ ke saath:
ω ˙ 1 = I ⊥ I ⊥ − I 3 ω 3 ω 2 , ω ˙ 2 = I ⊥ I 3 − I ⊥ ω 3 ω 1 .
Define karo Ω b = I ⊥ I 3 − I ⊥ ω 3 = 2 5 − 2 ⋅ 4 = 6 rad/s.
Yeh step kyun? Dono right sides opposite signs ke saath same magnitude Ω b share karte hain — yeh pair ( ω 1 , ω 2 ) ke pure rotation ki signature hai.
Circle solve karo. Re-anchor box se yaad karo ki ω ⊥ = ( ω 1 , ω 2 ) transverse spin vector hai. ω ˙ 1 = − Ω b ω 2 , ω ˙ 2 = Ω b ω 1 ke saath, ω ⊥ ki tip rate Ω b se rotate karti hai. Period T prec = 2 π / Ω b = 2 π /6 ≈ 1.047 s.
Yeh step kyun? Yeh equations ek point ke angular speed Ω b par circle mein jaane ki hain — exactly woh free precession cone hai.
Neeche ki figure yeh dikhati hai. Blue closed loop woh path hai jo transverse spin ω ⊥ = ( ω 1 , ω 2 ) trace karta hai; orange arrows aath instants par spin vector hain (sab same length — isliye loop perfect circle hai), aur red label constant turning rate Ω b = 6 rad/s mark karta hai. Ise aise padho: transverse spin ki tip steadily ek circle mein chalti hai aur kabhi spiral in ya out nahi karti.
Verify: ∣ ω ⊥ ∣ 2 = ω 1 2 + ω 2 2 ka = 0. 1 2 = 0.01 rehna chahiye. Kyunki d t d ( ω 1 2 + ω 2 2 ) = 2 ω 1 ω ˙ 1 + 2 ω 2 ω ˙ 2 = 2 ω 1 ( − Ω b ω 2 ) + 2 ω 2 ( Ω b ω 1 ) = 0 . Transverse magnitude constant hai — exactly circle, jaisa figure dikhata hai. Ω b = 6 rad/s, T prec ≈ 1.047 s. ✅
Asymmetric top I 1 = 1 , I 2 = 2 , I 3 = 3 . Axis 3 ke baare mein mainly spin Ω = 10 rad/s ke saath aur tiny ω 1 , ω 2 . Stability constant k aur wobble frequency nikalo.
Forecast: k ka sign: positive (tumble) ya negative (gentle wobble)?
Linearize karo. ω 3 ≈ Ω constant maano, ω 1 , ω 2 ko first-order rakho:
ω ˙ 1 = I 1 I 2 − I 3 Ω ω 2 , ω ˙ 2 = I 2 I 3 − I 1 Ω ω 1 .
Yeh step kyun? Do small quantities ka product negligible hai; yeh pure-spin state ke baare mein linearization hai.
Ek ODE mein combine karo aur k DEFINE karo. Pehle differentiate karo, doosra insert karo:
ω ¨ 1 = k ( pure number ) I 1 I 2 ( I 2 − I 3 ) ( I 3 − I 1 ) Ω 2 ω 1 .
Stability constant k exactly woh bracketed dimensionless coefficient hai — isme Ω 2 nahi hai, jo hum hamesha alag likhte hain:
k = I 1 I 2 ( I 2 − I 3 ) ( I 3 − I 1 ) = 1 ⋅ 2 ( 2 − 3 ) ( 3 − 1 ) = 2 ( − 1 ) ( 2 ) = − 1.
Yeh step kyun? k ko pure number rakhna (k = − 1 yahan) hamare liye stability sirf iske sign se padhna possible banata hai; physical growth/wobble rate phir k Ω 2 se aati hai.
Wobble read off karo. k = − 1 ke saath: ω ¨ 1 = k Ω 2 ω 1 = − Ω 2 ω 1 , simple harmonic. Frequency ω wob = ∣ k ∣ Ω = 1 ⋅ 10 = 10 rad/s.
Yeh step kyun? Negative k ek spring ki equation hai — bounded, safe oscillation. Axis 3 (sabse bada I ) stable hai. ✅
Verify: Dono factors: ( I 2 − I 3 ) = − 1 < 0 aur ( I 3 − I 1 ) = + 2 > 0 , product negative ⇒ k = − 1 < 0 . Wobble frequency ∣ k ∣ Ω = 10 rad/s, period ≈ 0.628 s. Stable, jaisa tennis-racket rule predict karta hai.
Same body I 1 = 1 , I 2 = 2 , I 3 = 3 , lekin ab mainly axis 1 ke baare mein spin Ω = 10 rad/s ke saath. Dikhao ki yeh stable hai aur frequency do.
Forecast: Do stable axes hote hain. Kya sabse chhhota wala sach mein stable hai, ya sirf sabse bada?
Stability constant ko cyclic-shift karo. Axis 1 ke baare mein bade spin ke liye, ω 2 ke liye linearized ODE hai
ω ¨ 2 = k Ω 2 ω 2 , k = I 2 I 3 ( I 3 − I 1 ) ( I 1 − I 2 ) ( pure number ) .
Yeh step kyun? "Jis axis ke baare mein spin kar rahe ho" ka role equations mein cycle karta hai; hum sirf ( 1 , 2 , 3 ) sahi se relabel karte hain aur k ko Ex 3 ki tarah dimensionless rakhte hain.
k compute karo. k = 2 ⋅ 3 ( 3 − 1 ) ( 1 − 2 ) = 6 ( 2 ) ( − 1 ) = − 3 1 ≈ − 0.333.
Yeh step kyun? Numbers plug karne se directly sign evaluate hota hai — k ke andar Ω 2 nahi.
Wobble read off karo. k < 0 ⇒ oscillation, frequency ∣ k ∣ Ω = 1/3 ⋅ 10 ≈ 5.77 rad/s. Stable. ✅
Verify: ( I 3 − I 1 ) = + 2 > 0 , ( I 1 − I 2 ) = − 1 < 0 ⇒ product negative ⇒ k = − 1/3 < 0 . Largest aur smallest dono axes k < 0 dete hain: exactly "BIG aur SMALL are SOLID" rule. Frequency ≈ 5.77 rad/s.
Same body I 1 = 1 , I 2 = 2 , I 3 = 3 , mainly axis 2 (intermediate) ke baare mein spin Ω = 10 rad/s ke saath. k nikalo aur disturbance ka e-folding time nikalo.
Forecast: Yeh Dzhanibekov / tennis-racket axis hai. Compute karne se pehle k ka sign guess karo.
Axis 2 ke liye stability constant.
ω ¨ 1 = k Ω 2 ω 1 , k = I 1 I 3 ( I 3 − I 2 ) ( I 2 − I 1 ) ( pure number ) .
Yeh step kyun? Same linearization machine, ab axis 2 "big spin" ke roop mein, aur k abhi bhi dimensionless number.
k compute karo. k = 1 ⋅ 3 ( 3 − 2 ) ( 2 − 1 ) = 3 ( 1 ) ( 1 ) = 3 1 ≈ 0.333.
Yeh step kyun? Dono factors yahan positive hain, toh unka product positive hai — yahi crucial sign flip hai.
Growth read off karo. k > 0 ⇒ ω ¨ 1 = + k Ω 2 ω 1 , solution ∝ e k Ω t . Growth rate k Ω = 1/3 ⋅ 10 ≈ 5.77 s − 1 ; e-folding time τ = 1/ ( k Ω ) ≈ 0.173 s.
Yeh step kyun? Positive k spring ko runaway bana deta hai — disturbance har 0.173 s mein e se multiply ho jaata hai. Tumble confirm. ❌
Neeche ki figure ise Example 3 se contrast karti hai. Red curve yeh case hai (axis 2): ek disturbance jo e k Ω t ki tarah unbounded grow karta hai; vertical dashed line e-folding time τ ≈ 0.173 s mark karti hai jahan yeh e se multiply ho chuka hai. Blue curve Example 3 ka axis-3 case hai: ek bounded, gently oscillating wobble jo kabhi nahi bhaagta. Ek picture, do fates — sirf k ke sign se decide.
Verify: ( I 3 − I 2 ) = + 1 > 0 aur ( I 2 − I 1 ) = + 1 > 0 ⇒ product positive ⇒ k = 1/3 > 0 . Ex 3–4 se contrast karo jahan ek factor negative tha. Growth rate k Ω ≈ 5.77 s − 1 , τ ≈ 0.173 s. Yeh "MIDDLE is a MUDDLE" cell hai — upar ki red curve se match karta hai.
Lo I 1 = 1 , I 2 = 2 , I 3 = 3 aur instantaneous spin ω = ( 3 , 1 , 2 ) rad/s. 2 T aur L 2 compute karo, phir Euler's equations use karke dikhao ki unke time derivatives zero hain. (Re-anchor box se yaad karo: T rotational kinetic energy hai, L = I ω angular momentum hai.)
Forecast: Bhaale har ω i time mein change ho, kya do scalars sach mein pinned rehte hain?
Constants evaluate karo.
2 T = I 1 ω 1 2 + I 2 ω 2 2 + I 3 ω 3 2 = 1 ( 9 ) + 2 ( 1 ) + 3 ( 4 ) = 9 + 2 + 12 = 23.
L 2 = I 1 2 ω 1 2 + I 2 2 ω 2 2 + I 3 2 ω 3 2 = 1 ( 9 ) + 4 ( 1 ) + 9 ( 4 ) = 9 + 4 + 36 = 49.
Yeh step kyun? Yeh do ellipsoids hain (2 T = const aur L 2 = const ω -space mein) jinke intersection par ω ride karta hai.
Energy differentiate karo aur Euler substitute karo.
d t d ( 2 T ) = 2 ( I 1 ω 1 ω ˙ 1 + I 2 ω 2 ω ˙ 2 + I 3 ω 3 ω ˙ 3 ) .
Ab I i ω ˙ i Euler ke right-hand sides hain, toh
d t d ( 2 T ) = 2 [ ω 1 ( I 2 − I 3 ) ω 2 ω 3 + ω 2 ( I 3 − I 1 ) ω 3 ω 1 + ω 3 ( I 1 − I 2 ) ω 1 ω 2 ]
= 2 ω 1 ω 2 ω 3 [ ( I 2 − I 3 ) + ( I 3 − I 1 ) + ( I 1 − I 2 ) ] = 0.
Yeh step kyun? Euler's equations se I i ω ˙ i substitute karne par common factor ω 1 ω 2 ω 3 nikalta hai, aur teen moment-differences telescope karke 0 ho jaate hain — energy ko koi external work nahi chahiye.
L 2 differentiate karo aur Euler substitute karo — poora dikhaya gaya.
d t d L 2 = 2 ( I 1 2 ω 1 ω ˙ 1 + I 2 2 ω 2 ω ˙ 2 + I 3 2 ω 3 ω ˙ 3 ) = 2 ( I 1 ω 1 ⋅ I 1 ω ˙ 1 + … ) .
Har I i ω ˙ i (Euler RHS) insert karo:
= 2 [ I 1 ω 1 ( I 2 − I 3 ) ω 2 ω 3 + I 2 ω 2 ( I 3 − I 1 ) ω 3 ω 1 + I 3 ω 3 ( I 1 − I 2 ) ω 1 ω 2 ] .
ω 1 ω 2 ω 3 factor out karo:
= 2 ω 1 ω 2 ω 3 [ I 1 ( I 2 − I 3 ) + I 2 ( I 3 − I 1 ) + I 3 ( I 1 − I 2 ) ] .
Bracket expand karo: I 1 I 2 − I 1 I 3 + I 2 I 3 − I 2 I 1 + I 3 I 1 − I 3 I 2 = 0 (har product ek baar + aur ek baar − ke saath aata hai). Isliye d t d L 2 = 0 .
Yeh step kyun? L ki length space mein fixed hai, toh L 2 constant hona chahiye — aur yahan hum kyun algebraically dekhte hain: har I a I b pair apne twin ko cancel karta hai.
Verify: 2 T = 23 (joules) aur L 2 = 49 (kg²·m⁴·s⁻²). d t d ( 2 T ) = 0 aur d t d L 2 = 0 dono VERIFY block mein symbolically confirm hain.
Asymmetric top I 1 = 1 , I 2 = 2 , I 3 = 3 . Exactly axis 2 ke baare mein spin: ω = ( 0 , Ω , 0 ) with Ω = 7 rad/s aur koi perturbation nahi. Yeh kya karta hai?
Forecast: Axis 2 unstable hai — lekin perfectly clean start ke saath, kya yeh phir bhi tumble karega?
Exact state plug in karo.
I 1 ω ˙ 1 = ( I 2 − I 3 ) ω 2 ω 3 = ( 2 − 3 ) ( Ω ) ( 0 ) = 0 ,
I 2 ω ˙ 2 = ( I 3 − I 1 ) ω 3 ω 1 = ( 3 − 1 ) ( 0 ) ( 0 ) = 0 , I 3 ω ˙ 3 = ( I 1 − I 2 ) ω 1 ω 2 = ( 1 − 2 ) ( 0 ) ( Ω ) = 0.
Yeh step kyun? Har right side mein ek zero component ka factor hai, toh saare derivatives vanish karte hain.
Conclude karo. ω = ( 0 , 7 , 0 ) ek fixed point hai — mathematically yeh hamesha axis 2 ke baare mein spin karta rahega.
Yeh step kyun? Pure principal-axis spin hamesha Euler's equations ka equilibrium hota hai, stability ke bawajood.
Instability se reconcile karo. Equilibrium unstable hai: koi bhi real ω 1 ya ω 3 , chahe kitna bhi tiny ho, e k Ω t ki tarah grow karta hai (Ex 5). Physics mein hamesha woh seed hota hai, toh real bodies tumble karti hain; idealized point nahi karti.
Yeh step kyun? Fixed point = stable — yahi distinction Ex 5 ne quantify ki.
Verify: Teeno ω ˙ i = 0 at ( 0 , 7 , 0 ) . Bhi 2 T = I 2 Ω 2 = 2 ( 49 ) = 98 aur L 2 = I 2 2 Ω 2 = 4 ( 49 ) = 196 ; dono trivially constant kyunki ω move nahi karta. Degenerate case handle ho gaya.
Earth almost ek symmetric top hai jisme I ⊥ I 3 − I ⊥ ≈ 305 1 ("dynamical ellipticity"). Yeh ek sidereal day mein ek baar spin karta hai, ω 3 = 2 π / ( 86164 s ) . Body frame mein free-precession (Euler/Chandler) period predict karo.
Forecast: Days? Ek mahina? Lagbhag ek saal? Order of magnitude guess karo.
Ex 2 se symmetric-top precession rate use karo:
Ω b = I ⊥ I 3 − I ⊥ ω 3 = 305 ω 3 .
Yeh step kyun? Earth ka tiny equatorial bulge I 3 = I ⊥ banata hai; woh difference wobble rate set karta hai.
Period mein convert karo. Wobble period hai
T wob = Ω b 2 π = 305 ⋅ ω 3 2 π = 305 × ( 1 sidereal day ) ≈ 305 days .
Yeh step kyun? 2 π / ω 3 ek din hai, toh wobble period sirf 305 din hai — direct, memorable result.
Reality se compare karo. Rigid prediction ~305 days hai; observed Chandler wobble ~433 days hai. Gap isliye hai kyunki Earth perfectly rigid nahi hai (elastic crust + fluid core period lengthen karte hain).
Yeh step kyun? Kyunki hamari derivation perfectly rigid body assume karti hai; model ko data se compare karna exactly woh assumption isolate karta hai jo fail hoti hai aur dikhata hai ki idealized model kya chhod deta hai — prediction ko measurement ke against check karne ki essential habit.
Verify: T wob = 305 sidereal days = 305 × 86164 s ≈ 2.628 × 1 0 7 s ≈ 304.2 solar days. Order-of-magnitude "lagbhag ek saal" forecast sahi hai; ~305 d rigid vs ~433 d observed.
Ek tricky exam deta hai I 1 = 5 , I 2 = 1 , I 3 = 3 (note: sorted nahi! ). Body mainly axis 3 ke baare mein spin kar rahi hai. Kya axis 3 stable hai? Label par trust mat karo — sign check karo.
Forecast: "Axis 3 sabse bada hai ⇒ stable" — kya yeh reasoning yahan safe hai?
Moments rank karo. Sorted: I 2 = 1 < I 3 = 3 < I 1 = 5 . Toh axis 3 intermediate moment hai, bhaale uska label "3" ho.
Yeh step kyun? Stability depend karta hai ki moment smallest/largest/middle hai ya nahi — value par, index par nahi.
Axis 3 ke baare mein spin ke liye k compute karo. Axis 3 ke baare mein bade spin ke liye, pure-number stability constant (Ex 3 ke pattern se) hai
k = I 1 I 2 ( I 2 − I 3 ) ( I 3 − I 1 ) = 5 ⋅ 1 ( 1 − 3 ) ( 3 − 5 ) = 5 ( − 2 ) ( − 2 ) = 5 4 = 0.8.
Yeh step kyun? Dono factors negative hain, toh unka product positive hai ⇒ k > 0 ⇒ instability, exactly Ex 5 ke intermediate case ki tarah.
Conclude karo. k = 0.8 > 0 : axis 3 yahan UNSTABLE hai. "Largest index" label ne tumhe fool kiya; intermediate value jeetti hai. Growth rate = k Ω = 0.8 Ω ≈ 0.894 Ω .
Yeh step kyun? Yahi trap hai parent note ke Mistake C mein — product ( I a − I b ) ( I b − I c ) ka sign decide karta hai, axis name kabhi nahi.
Verify: Generic Ω ke saath, growth rate = Ω 0.8 ≈ 0.894 Ω s − 1 . Positive k = 0.8 tumble confirm karta hai. Yahan sirf woh axes stable hain jo I = 1 (axis 2) aur I = 5 (axis 1) carry karte hain.
Recall Self-test — answers cover karo
I = ( 1 , 2 , 3 ) ka kaunsa axis unstable hai, aur kis sign rule se? ::: Intermediate (I 2 = 2 ); unstable kyunki ( I 3 − I 2 ) ( I 2 − I 1 ) > 0 se k > 0 milta hai.
Symmetric top I 1 = I 2 = 2 , I 3 = 5 , ω 3 = 4 ke liye, body precession rate kya hai? ::: Ω b = I ⊥ I 3 − I ⊥ ω 3 = 2 3 ⋅ 4 = 6 rad/s.
Kya stability constant k mein Ω 2 hai? ::: Nahi — k ek pure number hai; physical rate k Ω 2 hai, jisme Ω 2 alag likha jaata hai.
Exact single-axis spin mathematically kabhi tumble kyun nahi karta? ::: Har Euler RHS mein ek zero component ka factor hota hai, toh saare ω ˙ i = 0 hain — yeh ek fixed point hai (bhaale unstable ho).
Mnemonic One-line takeaway
I 's sort karo, phir sign test karo ( I a − I b ) ( I b − I c ) : negative = stable wobble, positive = runaway tumble. Axis label ek decoy hai.