Dimensional analysis — Buckingham π theorem: Question bank

2.2.26 · D5 · Physics › Fluid Mechanics › Dimensional analysis — Buckingham π theorem

True or false — justify

The theorem

p = n - k

predicts the exact formula of the physical law.

False. Yeh sirf

p

independent dimensionless knobs ki sankhya aur form

f (π_{1}, \dots, π_{p}) = 0

predict karta hai; actual function

Φ

aur constants jaise

2 π

experiment ya full theory se aate hain.

If two problems have the same

n

and

k

, they must have the same physics.

False. Same

n - k

ka matlab hai

π

groups ki same ginti, lekin kaun se variables aate hain aur groups kaise combine hote hain yeh bilkul alag hota hai — pendulum aur drag ek hi count share kar sakte hain phir bhi unka physics alag hoga.

Adding an irrelevant variable to your list always increases

p

by one.

Count mein True hai (

n

ek badh jaata hai, aur agar variable koi nayi dimension nahi laata toh

k

fixed rehta hai, isliye

p = n - k

badh jaata hai), lekin isse ek spurious

π

banta hai jis par real law depend hi nahi karta — garbage in, garbage out.

The number of fundamental dimensions

k

is always

3

because the universe has

M, L, T

False.

k

woh dimensions hain jo is problem mein actually present aur independent hain; ek purely geometric/kinematic problem mein

k = 2

(

L, T

) ho sakta hai, aur ek thermal problem mein

k = 4

(temperature

Θ

add karo).

k

equals the rank of the dimension matrix

D

, not just the number of distinct base symbols you wrote down.

True. Agar do "dimensions" hamesha ek fixed combination mein aate hain, toh woh ek independent row nahi add karte, isliye

rank (D)

symbols ki sankhya se kam ho sakta hai — hamesha rank use karo.

Every set of

k

variables can serve as repeating variables.

False. Unhe (a) saath milke saare

k

dimensions contain karne chahiye aur (b) dimensionally independent hone chahiye (unka

k \times k

sub-matrix invertible hona chahiye), warna woh secretly ek

π

bana lete hain aur tumhare groups collapse ho jaate hain.

The dimensionless groups

π_{1}, \dots, π_{p}

are unique.

False. Valid

π

groups ke powers ka koi bhi product (jaise

π_{1} / π_{2}

π_{1}^{2}

) ek aur valid

π

hai; sirf sankhya

p

aur span fixed hote hain, particular basis nahi jo tum likhte ho.

A quantity that is already dimensionless (like an angle or a pure number) counts as a variable that raises

n

Loosely true — yeh list mein enter ho sakta hai, lekin yeh apne aap mein ek

π

group hai aur

k

mein kuch nahi add karta, isliye yeh directly

π_{i}

mein appear ho jaata hai.

Spot the error

"For sphere drag I picked repeating variables

v

d

, and time

t = d / v

." — what's wrong?

t = d / v

dimensionally

v

aur

d

par dependent hai, isliye repeaters ka sub-matrix rank khota hai; woh condition (b) fail karte hain aur teen independent dimensions span nahi kar sakte.

"The pendulum period must depend on mass because heavy things feel different." — where's the flaw?

Mass

m

akela variable hai jo dimension

M

carry karta hai, isliye koi doosra variable isko ek dimensionless ratio mein cancel nahi kar sakta; isliye

m

single

π

group mein appear hi nahi kar sakta.

"I got

p = n - k = 5 - 3 = 2

, so drag

F = ρ v^{2} d^{2} \times R e

." — what's the mistake?

Theorem deta hai

F / (ρ v^{2} d^{2}) = Φ (R e)

ek unknown function

Φ

ke saath; tum dimensions se akele yeh assume nahi kar sakte ki

Φ (R e) = R e

(ya koi specific form) hai.

"My equation has

n = 4

variables and

k = 4

dimensions, so

p = 0

— dimensional analysis fails." — is this an error?

Yeh error nahi hai;

p = 0

legitimately matlab hai ki koi free dimensionless group nahi hai, isliye dimensions relation ko ek pure constant tak fix kar dete hain — yeh ek bahut strong (failed nahi) result hai.

"Density

ρ

and specific weight

γ = ρ g

are both in my list, giving two mass-bearing variables." — trap?

Haan agar

g

bhi listed hai:

γ

doosron ka ek function hai, isliye yeh redundant hai aur silently rank ko tumhari expectation se kam kar sakta hai — derived duplicates drop karo.

"I chose

F

(the drag force) as a repeating variable." — why is that risky?

Repeaters woh "building blocks" hain jo har

π

mein appear karte hain; target quantity

F

ko unme dalna wo cheez jo tum actually solve karna chahte ho use saare groups ke andar dafna deta hai — unknown ko repeaters se bahar rakho.

Why questions

Why does demanding "the law works in metres or miles" force the law into dimensionless groups?

Kyunki units rescale karna har term ko ek factor se multiply karta hai; sirf woh ratios jinke koi units nahi hain rescaling se untouched rehte hain, isliye ek rescale-invariant law sirf aisi ratios ke through expressible honi chahiye.

Why is

p = n - rank (D)

rather than

n

minus the number of dimension symbols?

p

D

ke null space ki dimension hai (dimensionless exponent vectors ka space), aur rank–nullity se yeh

n - rank (D)

ke barabar hota hai; rank, symbol count nahi, measure karta hai ki dimensions kitne independent constraints impose karte hain.

Why can dimensional analysis never hand you the constant

2 π

T = 2 π L / g

2 π

dimensionless hai, isliye yeh us unknown function/constant ke andar rehta hai jise dimensions pin down nahi kar sakti; sirf actual pendulum equation solve karne (ya measure karne) se yeh pata chalta hai.

Why do we combine each non-repeating variable with the repeaters one at a time?

Exactly

p = n - k

independent groups build karne ki guarantee ke liye — har naya non-repeater exactly ek naya

π

contribute karta hai, aur same repeater base use karna groups ko ek clean independent set rakhta hai.

Why does a bigger

p

make a problem experimentally harder?

Har

π

ek knob hai jise lab mein sweep karna padta hai;

p

groups ka matlab hai

p

-dimensional function

Φ

map karni hai, isliye experiments combinatorially badhte hain — drag with

p = 2

C_{D}

R e

ke curves chahiye, sirf ek number nahi.

Why does matching all

π

groups between a scale model and the real thing guarantee similar flow?

Agar har dimensionless group equal hai, toh dono systems ek hi dimensionless law

f (π_{1}, \dots) = 0

satisfy karte hain, isliye unka behaviour scale tak identical hota hai — yeh Model testing and similarity ka basis hai.

Edge cases

What does

n = k

(so

p = 0

) tell you physically?

Variables koi dimensionless ratio nahi bana sakte, isliye dimensions akele relationship ko ek single overall constant tak fix kar dete hain — yeh sabse strong possible dimensional prediction hai.

What if a variable's dimension is zero (already dimensionless, e.g. a strain or Mach number)?

Yeh

n

mein contribute karta hai lekin koi row/rank nahi add karta, isliye yeh apne aap ek

π

group ban jaata hai aur seedha final relation mein pass through kar jaata hai.

What happens if you forget a physically relevant variable?

Tumhara count

p

bahut chhota ho jaata hai aur jo "law" tum paate ho woh incomplete hota hai — missing quantity ek genuine

π

add karti jo real physics ko chahiye; dimensions tumhe warn nahi kar sakti ki kuch missing hai.

Can

rank (D)

ever be less than the number of dimensions that visibly appear?

Haan — agar woh dimensions sirf ek fixed combination mein appear hoti hain (jaise

L

aur

T

hamesha ratio

L T^{- 1}

ke roop mein), toh woh fewer independent rows mein collapse ho jaati hain, isliye

k = rank (D)

drop kar jaata hai.

What if two chosen repeating variables happen to have identical dimensions?

Unki rows parallel hain, sub-matrix singular hai, aur woh ek extra dimension span nahi kar sakte — tumhe ek ko dimensionally distinct variable se swap karna hoga.

R e

or its reciprocal the "correct" dimensionless group from the drag derivation?

Dono valid hain —

μ

-group

1/ R e

nikla; kyunki kisi

π

ka koi invertible power bhi ek

π

hai, convention bas ise

R e = ρ v d / μ

relabel karta hai. Dekho Reynolds number.

Connections

Reynolds number — archetypal $π$ group aur upar wala reciprocal trap
Drag force and drag coefficient — jahan $C_{D} = Φ (R e)$ aur "assume $Φ$ " error kaat ti hai
Dimensional homogeneity — yahan har "why" ke peeche unit-invariance principle
Model testing and similarity — saare $π$ groups match karne ka practical payoff
Navier–Stokes equations — unhe non-dimensionalise karna bina guess kiye $R e$ derive karta hai
Fundamental and derived units — $M, L, T, Θ$ ko sahi se count karna