2.2.23 · D1 · Physics › Fluid Mechanics › Boundary layer separation — adverse pressure gradient
Intuition Is poore topic ki ek core idea
Air (ya paani) jo surface ke saath slide kar raha hai, uski ek razor-thin skin hoti hai — slow fluid jo wall se friction ki wajah se chipki rehti hai. Agar us slow skin ke aage pressure badhne lage, toh skin ka push khatam ho jaata hai, ruk jaati hai, aur peeche ki taraf mudd jaati hai — toh smooth flow surface se peel off ho jaati hai. Ye page, bilkul zero se, har wo symbol build karta hai jo us sentence ko mathematics mein kehne ke liye chahiye.
Parent note padhne se pehle, tumhe har wo letter apna banana hoga jo woh tumhare saamne phenk ta hai. Hum unhe ek ek karke milenge, har ek ke saath ek picture aur ek reason ki woh exist kyun karta hai.
Is topic ki har cheez ek wall ke paas rehti hai. Humein pehle ek coordinate frame chahiye jo usi wall se chipka ho , room se nahi.
Definition Wall-hugging coordinates
x aur y
x = surface ke saath saath measure ki gayi doori, jis direction mein flow travel karta hai.
y = surface se seedha bahar measure ki gayi doori (uske perpendicular).
Toh wall khud line y = 0 hai, aur y badhta hai jab tum us se fluid mein door jaate ho.
Figure 1 (neeche). Ek curved wall white mein jisme flow-direction axis x (blue arrow) surface ko follow karte hue muda hua hai, aur height axis y (yellow arrow) seedha perpendicular nikal raha hai. Neeche shaded region solid hai; wall line y = 0 hai. Notice karo kaise x surface nahi chhodata chahe wall curve ho — is frame ka yahi to poora point hai.
Intuition Axes ko wall ke saath kyun modein?
Wall curved ho sakti hai (cylinder, wing). x ko surface follow karne dene se, "flow ke saath" ka matlab hamesha "increasing x " hota hai aur "wall se door" ka matlab hamesha "increasing y " — chahe surface kitna bhi curve kare. Isse baad ke saare equations simple rehte hain.
Ek fluid particle ek direction mein move karta hai. Hum us motion ko do arrows mein split karte hain, ek har axis ke saath.
Intuition Ek arrow ki jagah do letters kyun?
Motion ke equations ek direction ek time par likhna aasaan hota hai. u jawab deta hai "kitni tezi se aage slide ho raha hai?" aur v jawab deta hai "kya yeh wall ki taraf ya door drift kar raha hai?" Separation ek story hai u ki near-wall mein, toh u humara star hai.
Word streamline Bernoulli section mein aane wala hai, toh ise abhi build karte hain.
Ek streamline wo imaginary curve hai jo tumhe milti hai jab tum ek instant mein, har point par fluid ki velocity ka ek arrow draw karo aur phir un arrows ko tip-se-tail join karke ek smooth line banao. Construction se fluid velocity hamesha streamline ke tangent (parallel) hoti hai — streamline par sawaar ek particle kabhi use cross nahi karta; woh ussi ke saath behta hai, jaise taar par manka.
Intuition Kyun streamlines flow ke natural "roads" hain
Kyunki koi fluid streamline cross nahi karta, tum har streamline ko ek private lane maan sakte ho: ek particle ke paas lane ki shuruat mein jo bhi energy hai, woh usi lane mein le jaata hai. Yahi reason hai ki Bernoulli ka balance ek streamline ke saath bola jaata hai — yeh ek particle ke ek lane mein travel karne ki energy ka bookkeeping hai. y = δ ( x ) par edge streamline woh lane hai jo boundary layer ki top ko skim karti hai; yeh wahi lane hai jis par hum Bernoulli apply karenge.
"Layer ke bahar" ke baare mein baat karne se pehle, humein precisely bolna hoga ki layer kahan khatam hoti hai . Us edge ka ek naam aur ek symbol hai.
Definition Boundary layer
Boundary layer fluid ki woh thin sheet hai jo wall ke paas hoti hai jahan friction ne flow ko free-stream speed se neeche slow kar diya hai. Iske andar, u wall par 0 se chadh kar thodi door par bahar ki poori speed tak pahunchta hai.
δ ( x ) — boundary-layer thickness
Jaise tum y mein upar uthte ho, speed u outer value ke karib aati hai lekin kabhi exactly nahi pahunchti. Toh hum edge ek chosen cutoff par draw karte hain: δ ( x ) woh height hai jis par u outer speed ka 99% tak chadh jaata hai.
u ( x , y = δ ( x ) ) = 0.99 U ( x ) .
δ ke neeche = "layer ke andar" (friction important hai); δ ke upar = "free stream" (friction ignore kar sakte hain). Kyunki aage jaane par zyada fluid drag ho jaata hai, δ x ke saath badhta hai.
Figure 2 (neeche). White wall y = 0 par, red dashed edge curve y = δ ( x ) ke saath jo x badhne par moti hoti jaati hai. Teen stations par, ek blue velocity profile dikhata hai u wall par 0 se shuru hoke red dot tak edge par free-stream value tak chadh ta hai; green arrows along-wall direction mark karte hain aur yellow arrows layer ke upar uniform free stream U ( x ) dikhate hain. Red edge ko chadhte dekho — yahi δ ka x ke saath badhna hai.
U ( x ) — free-stream (outer edge) speed
U ( x ) edge y = δ ( x ) par aur usse aage flow speed hai — fast outer flow, position x par. Capital U = outer speed; small u = layer ke andar actual speed, wall par 0 se y = δ tak U tak.
δ ko U se pehle kyun define karna padta hai
"Layer ke bahar ki speed" ka koi matlab nahi jab tak tum fix na karo ki "bahar" kahan se shuru hota hai. δ ( x ) woh fence hai. Yahi woh reason bhi hai ki layer ko thin kyun kaha jaata hai : δ body ki length ke comparison mein tiny hai, aur yahi thinness woh assumption hai jis par baad ki har simplification depend karti hai.
edge speed U ( x ) ki kyun parwah karte hain
Thin layer itni kamzor hai ki apna khud ka pressure set kar sake; powerful outer flow usse dictate karta hai. Toh agar hum jaante hain ki U ( x ) kaise change hota hai, hum jaante hain pressure kaise change hota hai — aur pressure poori story ka villain hai.
Parent note quietly assume karta hai ki p sirf x par depend karta hai, height y par nahi. Yeh obvious nahi hai — aao actually ise derive karte hain.
y -momentum equation (simplify karne se pehle)
Newton ka second law ek steady, constant-density flow ke cross-stream (y ) direction ke liye likhne par:
ρ ( u ∂ x ∂ v + v ∂ y ∂ v ) = − ∂ y ∂ p + μ ( ∂ x 2 ∂ 2 v + ∂ y 2 ∂ 2 v ) .
Simple words mein: (mass-per-volume ρ ) × (sideways speed v ka acceleration) = (cross-stream pressure push − ∂ p / ∂ y ) + (v par kaam kar raha friction). Ab hum measure karte hain ki thin layer ke andar har piece kitna bada hai.
Intuition Thin-layer scaling argument — piece by piece
Layer ko x along length L aur tiny height δ do, jahan δ ≪ L . Free-stream speed U kaho. Do facts sab kuch drive karte hain:
Along-wall speed u , ∼ L doori mein ∼ U range karta hai, toh ∂ u / ∂ x ∼ U / L .
Mass conservation, ∂ u / ∂ x + ∂ v / ∂ y = 0 , phir force karta hai ∂ v / ∂ y ∼ U / L bhi. Kyunki v tiny height δ mein change hota hai, iska matlab v khud sirf ∼ U δ / L hai — chhota , kyunki δ / L chhota hai.
Ab y -equation mein har term ka wajan lagao:
Inertia ρ u ∂ v / ∂ x ∼ ρ U ⋅ ( U δ / L ) / L = ρ U 2 δ / L 2 .
Compare karo along -wall pressure term se (x -equation se), jo order ρ U 2 / L hai.
Unka ratio hai ( ρ U 2 δ / L 2 ) / ( ρ U 2 / L ) = δ / L ≪ 1 .
Toh har woh term jo pressure ko y -direction mein bend kar sake, main flow terms se δ / L factor se chhoti hai. Layer shrink karo (δ / L → 0 ) aur woh vanish ho jaate hain, sirf ye bachta hai:
∂ y ∂ p ≈ 0.
p = p ( x )
Kyunki ∂ p / ∂ y ≈ 0 , kisi bhi height par pressure seedha neeche wall ke pressure ke barabar hai. Toh outer flow apna pressure thin layer ke seedha andar stamp karta hai wall par. Yahi reason hai ki hum p ( x ) likh sakte hain (har station x par ek value) aur plain derivative d p / d x use kar sakte hain — koi y -dependence bachti hi nahi.
Intuition Yeh linchpin kyun hai
Isse inviscid outer flow pressure set karta hai aur viscous inner layer use bas inherit karti hai. Separation ki poori story inner layer ka us pressure ke against struggle hai jo usne choose nahi kiya — aur yeh assumption woh pressure unchanged hand down karti hai.
Definition No-slip condition
Solid surface ko touch karne wala real fluid usse chipak jaata hai: woh wall ki speed par move karta hai. Stationary wall ke liye iska matlab hai:
u = 0 aur v = 0 at y = 0.
Intuition "No-slip" poore topic ka seed kyun hai
Kyunki fluid right at the wall zero par frozen hai, uske theek upar wali layer bhi slow ho jaati hai. Yahi woh "thaka hua bachcha" hai jo chadhai par energy khatam kar leta hai. No-slip us energy-poor near-wall fluid ko create karta hai jo baad mein peeche push ho jaata hai.
u ko height y ke against sideways draw karo: woh curve velocity profile hai. Uski shape ke baare mein do cheezein sab kuch decide karti hain.
∂ y ∂ u — profile ka slope
Symbol ∂ (ek "curly d") ka matlab hai "doosre variables ko fixed rakh kar change ki rate." Toh ∂ y ∂ u padha jaata hai: "along-wall speed u kitni tezi se change hoti hai jab hum height y mein upar step lete hain?" — profile ki steepness .
d ki jagah partial derivative kyun?
u ek saath do cheezein par depend karta hai — kitna aage (x ) aur kitna upar (y ). Hum sirf height ke saath change chahte hain, x fixed rakh kar, toh hum partial ∂ u / ∂ y use karte hain ye kehne ke liye "sirf y vary karo." Plain derivative d u / d y maanegi ki u sirf y par depend karta hai, jo galat hai.
Figure 3 (neeche). Do velocity profiles (u horizontal, height y vertical). Left (green): ek favourable profile — wall par steep slope aur sirf ek taraf bend, koi inflection nahi, healthy aur attached. Right (red): ek adverse profile — wall slope almost zero tak flat ho gayi hai aur ek yellow dot inflection point mark karta hai jahan bend direction switch karta hai. Wall slopes compare karo: green wala steep hai (strong forward pull); red wala almost flat hai (reverse hone wala hai).
∂ y 2 ∂ 2 u — profile ki curvature
Height-slope operation ko do baar apply karo: yeh measure karta hai ki slope khud kaise change hoti hai jab tum upar jaate ho. Yeh profile ka bend (curvature) hai.
Positive ∂ 2 u / ∂ y 2 : profile is taraf bend karti hai ki slope height ke saath badhti hai (ek taraf curl karti hai).
Negative: slope height ke saath ghatti hai (doosri taraf curl karti hai).
Kisi height par zero curvature ⇒ ek inflection point , jahan bend direction flip karta hai.
Intuition Curvature itni important kyun hai
Ek profile jo sirf ek taraf bend karti hai (koi inflection nahi) stable hai aur attached rehti hai. Ek profile jise apna bend switch karne par majboor kiya jaaye — inflection lene par — ek wobble carry karti hai jo near-wall flow ko reverse mein flip kar sakti hai. Parent note dikhata hai ki rising pressure exactly wahi hai jo us inflection ko force karta hai.
ρ (rho) — density
ρ = fluid ke har unit volume mein packed mass (kilograms per cubic metre). Bhaari, denser fluid = bada ρ . Yeh batata hai ki kitni cheez ko accelerate karna hai.
μ (mu) — dynamic viscosity
μ = fluid ki "stickiness" ya internal friction. Bada μ (honey) shearing ka strongly resist karta hai; chhota μ (air) barely resist karta hai. Yeh velocity slope ko friction force mein convert karta hai.
ν (nu) — kinematic viscosity
ν = ρ μ . Yeh stickiness density par share out ki gayi hai — "is heaviness ke fluid ke liye friction kitni tezi se momentum diffuse karta hai." Tum ise viscous term ko multiply karte dekhoge.
Intuition Ek idea ke liye teen cousins kyun?
Woh alag sawaalon ke jawaab dete hain. μ jawab deta hai "given slope se friction force kitni badi hogi?" ρ jawab deta hai "push ke against kitna inertia resist karta hai?" aur ν = μ / ρ jawab deta hai "slow-down fluid mein upar kitni tezi se spread hota hai?" Parent equation ν use karta hai spreading term ke liye aur μ wall par force ke liye.
Wall shear formula koi definition nahi hai jo aasman se giri ho — yeh fluid ke baare mein ek physical assumption par tikhi hai.
Definition Newtonian-fluid assumption
Ek Newtonian fluid woh hai jiska internal friction (shear stress) directly proportional hota hai is baat se ki neighbouring layers kitni tezi se ek doosri ke past slide karti hain — woh sliding rate exactly velocity slope ∂ u / ∂ y hai. Air aur water ise khoobsurti se maante hain; honey, ketchup aur toothpaste nahi maante.
τ = μ ∂ y ∂ u .
Intuition Shear viscosity times
slope kyun hota hai
Heights y aur y + Δ y par fluid ki do neighbouring sheets imagine karo. Agar woh almost same speed par move kar rahi hain (tiny slope), woh barely rub karti hain — kam friction. Agar ek doosri ke past race kare (bada slope), woh hard rub karti hain — bada friction. Toh friction speed difference per unit height ke saath scale honi chahiye, jo precisely ∂ u / ∂ y hai. Proportionality constant jo "sliding rate" ko "force per area" mein turn karta hai woh stickiness μ hai — μ ka yahi sara content hai. Slope double karo toh drag double ho jaata hai: ek linear law, aur linear exactly wahi hai jo "Newtonian" ka matlab hai.
τ w kyun poora point hai
Wall par steep profile (bada slope) ka matlab hai fast-moving fluid surface ke paas baitha hai: flow aage push kar raha hai, healthy aur attached. Jab τ w zero tak shrink ho, wall par fluid ka koi forward slope nahi bachi — woh rokne wala hai. Us point ke theek baad slope negative ho jaata hai: near-wall fluid peeche slide karta hai. Woh zero-crossing, τ w = 0 , exactly separation point ki definition hai.
p ( x ) — wall ke saath pressure
p = push per unit area jo fluid feel karta hai, aur p ( x ) batata hai ki woh push surface ke saath move karne par kaise change hota hai. Jaise upar dikhaya, thin layer ke andar p height ke saath change nahi karta, toh yeh sirf x par depend karta hai.
d x d p — pressure gradient
Rate jis par pressure downstream move karne par change hota hai. Plain d / d x (partial nahi) yahan theek hai kyunki p sirf x par depend karta hai.
d x d p < 0 : pressure downstream fall ho raha hai — favourable , flow speed up hoti hai.
d x d p > 0 : pressure downstream rise ho raha hai — adverse , flow slow down hoti hai. Yeh woh chadhai hai jo tired near-wall fluid ko stall kar deti hai.
Intuition Rising pressure "adverse" kyun hai
Fluid naturally high pressure se low pressure ki taraf behta hai. Agar pressure uske aage badhta hai, toh fluid se kaha ja raha hai ki pressure ki badhti hui wall mein dhakka maar ke ghuse — yeh chadhai par cycle chalaana hai. Energy-poor near-wall fluid pehle yeh fight haar jaata hai.
Hum kehte rehte hain "U ( x ) ↓ forces karta hai p ( x ) ↑ ." Woh link Bernoulli's equation se aata hai — lekin sirf specific conditions mein, aur sirf ek specific form mein. Dono state karte hain. (Yaad karo upar se ki streamline ek aisi curve hai jis par fluid saath behta hai lekin kabhi cross nahi karta.)
Definition Bernoulli kab allowed hai
Ek single streamline ke saath, p + 2 1 ρ U 2 = const hold karta hai sirf tab jab flow:
steady ho (pattern time mein change nahi ho raha),
incompressible ho (ρ constant — air ke liye sound speed se kaafi neeche theek hai, aur paani ke liye),
inviscid ho (friction negligible) us streamline ke saath .
y = δ par outer edge streamline teeno satisfy karti hai: yeh friction-dominated layer ke bahar baithti hai, toh viscosity wahan ignore karne layak hai. Yahi precisely reason hai ki hum Bernoulli edge par apply kar sakte hain lekin layer ke andar kabhi nahi.
Intuition Yeh woh tool kyun hai jo sab kuch unlock karta hai
Yeh ek geometry fact ("channel wide ho raha hai, toh U girta hai") ko ek pressure fact ("toh p badhta hai, adverse") mein convert karta hai. Parent note mein har worked example secretly is ek equation se guzarta hai: ise do ki edge speed U ( x ) kaise change hoti hai, aur yeh d p / d x ka sign wapas deta hai — jo ek single switch hai jo decide karta hai ki boundary layer attached rahe ya separate ho.
Neeche diagram ek dependency map hai: har box is page ka ek idea hai, aur ek arrow "A → B " ka matlab hai B ka koi matlab nahi jab tak A apna na ho . Ise upar se neeche padho.
no-slip u equals 0 at wall
thin layer so p equals p of x
Bernoulli dp dx equals minus rho U dU dx
Separation tau w equals 0
Intuition Is map par kaise chalein
Bottom box "Separation τ w = 0 " mein jaane wala koi bhi path follow karo aur tum poori logic replay kar lete ho. Ek path: coordinates → velocity → profile → uska wall slope → wall shear τ w → separation. Doosra path: Bernoulli aur thin-layer assumption dono pressure gradient d p / d x ko feed karte hain, jo profile ki curvature ko shape karta hai, jo separation ko feed karta hai. Bottom box par land hone wale teen arrows woh teen ingredients hain jo parent note combine karta hai: ek wall slope (τ w ), ek curvature jo woh force karta hai, aur ek pressure gradient jo use drive karta hai. Agar koi bhi upstream box fuzzy hai, wahi tumhari agli cheez study karne ki hai.
Daayein side cover karo aur parent note padhne se pehle har ek recite karo.
Is topic mein x aur y kya measure karte hain? x surface ke saath chalta hai (flow direction); y perpendicular, wall se door, wall y = 0 par hai.
u aur U mein kya farq hai?u layer ke andar actual along-wall speed hai (wall par 0 se U tak); U ( x ) fast free-stream speed hai edge y = δ par aur usse aage.
Streamline kya hota hai? Ek curve jo har jagah velocity ke tangent ho; fluid ussi ke saath behta hai lekin use cross nahi karta, toh energy bookkeeping (Bernoulli) ek streamline ke saath ki jaati hai.
Boundary-layer thickness δ ( x ) kya hai? Woh height jahan u , U ka 99% reach kar le; uske neeche friction matter karta hai, uske upar free stream hai, aur δ x ke saath badhta hai.
Hum p = p ( x ) kyun likh sakte hain bina y -dependence ke? y -momentum equation mein har term main flow se δ / L ≪ 1 factor se chhoti hai, toh ∂ p / ∂ y ≈ 0 ; outer pressure seedha wall par stamp ho jaata hai.
No-slip condition kya kehti hai? Stationary wall ko touch karne wala fluid frozen hota hai: u = 0 aur v = 0 at y = 0 .
∂ u / ∂ y mein curly ∂ kyun likhte hain d ki jagah?u dono x aur y par depend karta hai; partial ka matlab hai "sirf height y ke saath change, x fixed rakh kar."
∂ y 2 ∂ 2 u physically kya represent karta hai?Velocity profile ki curvature (bend) — uski slope kaise change hoti hai jab tum upar jaate ho; sign change inflection point deta hai.
μ , ρ aur ν ko relate karo.ν = μ / ρ , matlab μ = ρ ν ; μ stickiness hai, ρ density hai, ν momentum diffusivity hai.
Newton's law of friction aur uske peeche assumption state karo. τ = μ ∂ u / ∂ y , Newtonian fluid ke liye valid jahan shear stress linearly proportional hoti hai velocity slope se (air, water).
Wall shear stress likho aur batao ki woh separation kab signal karta hai. τ w = μ ∂ u / ∂ y ∣ y = 0 ; separation wahan shuru hoti hai jahan τ w = 0 , theek downstream mein backflow ke saath.
Pressure gradient "adverse" kab hota hai? d p / d x > 0 — pressure downstream badhta hai, flow decelerate hoti hai (d U / d x < 0 ).
Bernoulli ki validity conditions aur uska differential form state karo. Steady, incompressible, inviscid along a streamline; differentiate karne par milta hai d p / d x = − ρ U d U / d x , toh U ↓⇒ p ↑ .
Related builds: Boundary layer theory · Reynolds number & transition · Drag — form vs skin friction · Flow over a cylinder and sphere · Diffusers and nozzles · Stall on an aerofoil .