Sound intensity — decibels (logarithmic scale): Visual walkthrough

1.6.23 · D2 · Physics › Oscillations & Waves › Sound intensity — decibels (logarithmic scale)

Step 1 — "Loudness" asal mein kya hai: energy ka ek area mein phailna

KYA. Kisi bhi formula se pehle, socho sound ek source se nikal rahi hai. Sound hawa ke pressure ki ek chalti hui ripple hai — dekho Sound waves — pressure & displacement — aur yeh energy carry karti hai (wahi energy tumhara eardrum push karti hai). Har second deliver hone wali energy power hai, jo watts (W) mein measure hoti hai.

KYUN. "Loud" sirf total power ke baare mein nahi hai. Ek loudspeaker jo $10$ watts output karta hai woh paas mein bahut tez lagta hai aur door mein bilkul faint — chahe watts kabhi change na ho. Jo cheez matter karti hai woh yeh hai ki woh power tumtak pahunche tab kitni concentrated hai — kitne watts tumhare eardrum ke har square metre par land karte hain.

PICTURE. Figure mein, same amber energy-arrows source se nikalte hain. Source ke paas woh ek chote patch se bheede hue guzarte hain; door jaake wahi arrows ek bade patch par spread ho jaate hain, toh har square metre kam pakad paata hai.

Step 2 — Distance ke saath intensity kaise kam hoti hai (sphere argument)

KYA. Ek tiny source equally sab directions mein radiate karta hai. Distance $r$ par, uski saari power $P$ radius $r$ ke ek sphere ki surface par spread ho jaati hai. Sphere ki surface area $A = 4 π r^{2}$ hai, toh

$I (r) = \frac{P}{sphere surface 4 π r ^{2}} .$

KYUN. Travel karte waqt koi energy destroy nahi hoti (free air mein, bina echoes ke). Woh bas ek hamesha bade hote balloon par paint hoti rehti hai. Bada balloon → patla paint → chota $I$ . Yeh Inverse-square law for radiation hai.

PICTURE. Do cyan spheres, radius $r$ aur $2 r$ . Bahar wale ki surface area 4× zyada hai, toh same power one-quarter intensity par dilute ho jaati hai. Dekho kaise amber "paint" patla hota hai.

Step 3 — Problem: numbers bahut absurd hain

KYA. Aao kaan jo intensities handle karta hai woh honestly list karte hain.

Sound	$I$ (W/m²)
Faintest jo sun sako	$0.000000000001$
Ek conversation	$0.000001$
Pain / damage	$1$

YEH KYUN DIKHATE HAIN? Taaki tumhe disaster feel ho. Faintest sound aur loudest sound ek trillion ( $1 0^{12}$ ) ke factor se alag hain. Inhe ek ordinary ruler par plot karo toh whisper zero ke paas invisibly baithta hai jabki painful sound ek mile door hai.

PICTURE. $0$ se $1$ W/m² tak ek linear axis. Almost har roz ka sound left edge ke paas ek sliver mein crush ho jaata hai — bilkul useless.

Step 4 — Ise fix karne ka tool: logarithm

KYA. Hum $lo g_{10}$ laate hain. $lo g_{10} (x)$ ko aise bolte hain: "main 10 ko kitni baar khud se multiply karun taaki $x$ mile?" Toh $lo g_{10} (1000) = 3$ kyunki $10 \times 10 \times 10 = 1000$ .

YEH TOOL KYUN, KUCH AUR KYUN NAHI? Hamare numbers sab powers of ten multiplied together hain, aur multiply karna jaldi ugly ho jaata hai. Earth par woh ek function jiska poora kaam multiplication ko addition mein badalna hai woh logarithm hai: $lo g_{10} (a \times b) = lo g_{10} a + lo g_{10} b .$ Yahi woh "×10 becomes +constant" behaviour hai jo Step 3 ne demand kiya tha. Koi aur elementary tool yeh nahi karta — isliye log hai aur kuch nahi. (Full toolkit: Logarithms and exponentials.)

PICTURE. Step 3 ki same intensities, ab unke $lo g_{10}$ ke hisaab se placed. Trillion-fold crush evenly spaced tick marks ban jaata hai. Log ne scale ko unbend kar diya.

Step 5 — Zero point choose karna: reference intensity $I_{0}$

KYA. Humein ek starting line chahiye — woh value jo "0" de. Hum ek healthy ear ki sabse quiet detectable sound choose karte hain: $I_{0} = 1 0^{- 12} W/m^{2} (1 kHz par ==threshold of hearing==) .$

KYUN. Is choice se, threshold par exactly baithne wali sound $I / I_{0} = 1$ deti hai, aur $lo g_{10} (1) = 0$ . Toh silence-threshold 0 read karta hai — ek natural, tidy floor. Sab audible sounds "sabse quiet whisper se 10 ke kitne factors upar" ka ek positive count hain.

PICTURE. Ek number line jiska origin $I_{0}$ par nailed hai. Right ki taraf har step = ek aur ×10. Amber marker $I = I_{0}$ ke liye 0 par baitha hai.

Step 6 — Fine resolution: decibel

KYA. Ek bel ek poora factor of ten hai — bahut bada jump. "Conversation" aur "shouting" ke beech tum sirf ek fraction of a bel move karoge, jo bolna awkward hai. Toh hum har bel ko ten equal parts mein chop karte hain (deci = one-tenth) 10 se multiply karke:

$β = deci: tenths of a bel 10 count factors of ten lo g_{10} (\frac{I your sound}{the zero-line I _{0}}) dB$

EXTRA 10 KYUN? Purely ek comfortable scale ke liye: yeh useful audible range ko "0 to 12 bels" se "0 to 120 dB" mein stretch karta hai, toh ordinary conversation decimals mein buried nahi hoti. Physics untouched rehti hai — ek 10× cosmetic rescaling.

PICTURE. Do aligned rulers: upar bels mein ( $0 \dots 12$ ), neeche decibels mein ( $0 \dots 120$ ). Same landmarks, bas neeche ten times zyada tick marks.

Step 7 — Payoff: intensity mein har ×10 exactly 10 dB add karta hai

KYA. Intensity ko 10 se multiply karo aur level dekho: $β^{'} = 10 lo g_{10} \frac{10 I}{I _{0}} = 10 [= 1 lo g_{10} 10 + lo g_{10} \frac{I}{I _{0}}] = 10 + β .$ Beech wala split tabhi possible hai jab log "×10" ko "+1" banata hai (Step 4). Do levels ki intensities multiply karo aur subtract karo: $Δ β = β_{2} - β_{1} = 10 lo g_{10} \frac{I _{2}}{I _{1}} .$ $I_{0}$ cancel ho jaata hai — changes ke liye tumhe kabhi reference ki zaroorat nahi.

YEH KYUN MATTER KARTA HAI. Yahi woh rule hai jo tum actually use karte ho: tum rarely absolute intensities jaante ho, lekin tumhe aksar ek ratio pata hota hai (do baar zyada singers, chaar baar distance). Ratio seedha daalo.

PICTURE. Ek staircase: har tread intensity mein ek ×10, har riser ek fixed +10 dB. Amber landmarks: ×2 ≈ +3 dB, ×100 = +20 dB.

Step 8 — Edge & degenerate cases (kabhi gap mat chhodna)

KYA. Har scale ki boundaries hoti hain. Aao har ek walk karte hain taaki koi scenario surprise na kare.

$I = I_{0}$ (exactly threshold): $β = 10 lo g_{10} (1) = 0 dB$ . Floor, by design.
$I < I_{0}$ (threshold se neeche): $I / I_{0} < 1$ , aur 1 se neeche ke number ka $lo g_{10}$ negative hota hai. Toh $β$ negative jaata hai — e.g. $I = 1 0^{- 13}$ deta hai $β = - 10$ dB. Bilkul legal; iska matlab hai "faintest audible sound se bhi quiet". Anechoic chambers kareeb $- 9$ dB tak pahunche hain.
$I \to 0$ (true silence): $lo g_{10} (0) \to - \infty$ . Formula $β$ ko minus infinity bhejta hai. Yeh maths honestly bata raha hai "bilkul no sound ka ratio scale par koi jagah nahi" — tum kabhi wahan pahunch nahi sakte, sirf approach kar sakte ho.
Do equal sources jodna: intensities add hoti hain, toh $I \to 2 I$ , deta hai $+ 3$ dB — dB number double nahi hota. (Step 7 ka ×2 rule.)
Distance ke saath combine karna: Step 2 se, $r \times 4 \Rightarrow I \div 16 \Rightarrow Δ β = 10 lo g_{10} \frac{1}{16} \approx - 12$ dB.

YEH KYUN DIKHATE HAIN. Negative aur $- \infty$ cases wahi hain jahan beginners panic karte hain ("kya dB negative ho sakta hai?!"). Haan — scale ek comparison hai, aur comparisons dono taraf ja sakte hain.

PICTURE. Full number line: amber $I_{0}$ at 0 dB, cyan region right mein (audible, positive), ek hatched region left mein (threshold se neeche, negative) true silence par $- \infty$ ki taraf trail off karte hue.

Ek-picture summary

Neeche, poori derivation ek single frame mein hai: raw intensities linear ruler par crush hui (top), log unhe bel ladder mein unbend karta hai (middle), aur ×10 = +10 dB staircase sab landmarks ke saath (bottom). Ise left→right, top→bottom trace karo aur tum Steps 1–8 re-walk kar chuke ho.

Recall Feynman retelling — poora walkthrough plain words mein

Sound ek source se ud rahi energy hai. Woh energy bade area par spread karo toh har patch ko thoda milta hai (quiet); choti jagah squeeze karo toh har patch ko bahut milta hai (loud) — yahi intensity hai, watts per square metre. Jaise tum door jaate ho, energy ek bade aur bade balloon par coat hoti hai, toh yeh distance-squared ke inverse par thin hoti hai.

Problem: jo faintest sound tum sun sakte ho aur ek painful sound mein trillion times ka fark hai. Normally likho toh quiet wala loud ke paas invisible hai. Toh hum raw numbers ki jagah zeros count karte hain — yahi logarithm karta hai: "times ten" ko "plus one" banata hai. Hum har sound ko faintest audible sound $I_{0} = 1 0^{- 12}$ se measure karte hain, toh whisper-threshold tidy 0 score karta hai. Phir, kyunki ten ke pore factors baat karne ke liye bahut coarse hain, hum har ek ko ten mein slice karte hain aur slice ko decibel bolte hain — yahi woh akela "10" saamne hai.

Jo magic niklata hai: intensity mein har ×10 ek fixed +10 dB step hai, toh loudness 0 se 120 tak ek friendly staircase ban jaati hai. Do equal sources? Sirf +3 dB. Chaar baar door? −12 dB. Threshold se quieter? Negative dB — bilkul theek. Absolute silence? Formula minus infinity ki taraf bhaag jaata hai, apna polite tarika yeh kehne ka "woh point ek comparison scale par nahi hai." Yahi decibel hai, ek single idea se bana: zeros count karo.

Recall

$lo g_{10}$ , koi aur function nahi, yahaan sahi tool kyun hai? ::: Hamare numbers ten ke factors se multiply hote hain; sirf logarithm multiplication ko addition mein badalta hai, "×10 louder" ko ek fixed step banata hai. Difference rule $Δ β = 10 lo g_{10} (I_{2} / I_{1})$ mein reference $I_{0}$ kyun gayab ho jaata hai? ::: Do level formulas subtract karne par, $10 lo g_{10} (I_{2} / I_{0}) - 10 lo g_{10} (I_{1} / I_{0})$ , $lo g_{10} I_{0}$ terms cancel ho jaate hain. Negative dB value ka kya matlab hai? ::: Sound threshold reference $I_{0}$ se quieter hai, kyunki 1 se neeche ke ratio ka $lo g_{10}$ negative hota hai. Formula mein akela factor of 10 kahaan se aata hai? ::: "Deci" se — hum har coarse bel ko ten decibels mein split karte hain ek comfortable 0–120 scale ke liye.

Parent: Sound intensity — decibels (logarithmic scale) · See also perceived loudness · Doppler effect