Structural design process — load cases, FOS (factor of safety): Visual walkthrough

3.6.2 · D2 · Physics › Spacecraft Structures & Systems Engineering › Structural design process — load cases, FOS (factor of safet

Yeh deep dive Structural design process — load cases, FOS (factor of safety) ke central result ko expand karta hai. Agar koi word yahan heavy lage, toh parent note mein reference table hai.

Step 1 — Ek part, aur usse lagta ek dhakka

KYA HAI. Ek real spacecraft ka piece socho: ek metal strut. Kuch cheez usse push ya pull karti hai. Woh push ek force hai — hum uski size ko $P$ kahenge, jo newtons (N) mein measure hota hai, push ki unit (roughly ek chote apple ka weight $1 N$ hota hai).

YE SE KYU SHURU KAREIN. Kisi bhi formula se pehle, humein ek honest physical cheez chahiye: ek load. Baaki sab kuch sirf is single arrow ke aas-paas ki bookkeeping hai.

PICTURE. Strut gray rod hai. Magenta arrow force $P$ hai jo usse stretch karne ki koshish kar raha hai.

Step 2 — Hum $P$ ko exactly kabhi nahi jaante: limit load

KYA HAI. Real life mein hum launch push ko fly karne se pehle measure nahi kar sakte — hum usse forecast karte hain. Part ko kabhi bhi lagne wale sabse bade push ka hamara best forecast limit load kehlata hai, likha jaata hai $P_{limit}$ .

EK SPECIAL NAAM KYU. "Limit" ka matlab hai expected world ki edge — sabhi launch load cases mein se worst case. Lekin nature mein scatter hota hai: asli push hamare forecast se thoda upar bhi ja sakta hai. Yeh baat yaad rakho — yahi poora reason hai ki factor of safety exist karta hai.

PICTURE. Possible real loads ka ek fuzzy cloud. Hamara forecast $P_{limit}$ (violet line) us cheez ke top par baitha hai jo hum expect karte hain, lekin cloud ka ek tail uske aage bhi nikla hua hai.

Step 3 — Darr demand ko inflate karta hai: $F O S$ se multiply karo

KYA HAI. Scatter ko (aur apni analysis errors, kamzor welds, temperature effects ko) cover karne ke liye, hum deliberately ek bade load ke against design karte hain jitna hum forecast karte hain. Hum multiply karte hain: $load we size for P_{design} = a number > 1 F O S \times our forecast P_{limit}$

$F O S$ (factor of safety) ek pure number hai jiska koi unit nahi, jaise $1.25$ .
Multiply karna demand arrow ko stretch karta hai. Yeh material ko nahi touch karta — yeh us push ko inflate karta hai jise hum pretend karte hain ki feel ho raha hai.

LOAD KO MULTIPLY KYU KAREIN, METAL KO NAHI. Uncertainty load mein rehti hai (humne push guess ki thi). Toh hum us cheez ko pad karte hain jiske baare mein hum unsure hain. Material strength ek measured fact hai; use hum alone chhodte hain.

PICTURE. Violet limit arrow $F O S$ dwara stretch hokar longer orange design arrow ban jaata hai. Same direction, zyada length.

Step 4 — Part actually kya le sakta hai: allowable

KYA HAI. Har material ki ek ceiling hoti hai — ek load (ya stress) jiske baad woh misbehave karta hai. Us ceiling ko hum allowable kehte hain, $S_{allow}$ . Stress and Strain — Yield vs Ultimate Strength se do ceilings hain:

Yield ceiling — iske past jao toh part hamesha ke liye bent reh jaata hai (permanent deformation).
Ultimate ceiling — iske past jao toh part toot jaata hai (rupture).

DO CEILINGS KYU. Ek bent bracket phir bhi hold kar sakta hai; ek snapped wala kabhi nahi. Toh humein do alag limits milte hain aur dono ko clear karna zaroori hai.

PICTURE. Material ke liye ek vertical strength bar, jisme do marks hain: lower yield line aur higher ultimate line.

Step 5 — Aamnaa-saamna: design load vs allowable

KYA HAI. Ab dono ko ek hi axis par rakho aur compare karo. Hum demand karte hain: $what it can take S_{allow} \geq what we make it face F O S \times S_{limit}$

EK HI AXIS KYU. Force ko strength se compare karna tabhi sense deta hai jab dono ek hi tarah ki quantity hon — dono loads, ya dono stresses. Hum inflated demand bar ko allowable bar ke saamne line up karte hain aur bas dekhte hain kaun sa taller hai.

PICTURE. Ek saath do bars: orange = inflated demand $F O S \times S_{limit}$ , violet = allowable $S_{allow}$ . Agar strength bar taller hai, toh hum survive karte hain.

Step 6 — "Taller" ko ek number mein badlo: Margin of Safety

KYA HAI. "Kitna taller?" Hum strength bar ko inflated demand bar se divide karte hain, phir $1$ subtract karte hain: $M S = \frac{S _{allow}}{F O S \times S _{limit}} - 1$

Term by term padho:

$\frac{S _{allow}}{F O S \times S _{limit}}$ = strength ka inflated demand se ratio. Agar strength exactly equal hai, toh yeh $1$ hai.
$- 1$ "just barely OK" point ko zero par shift kar deta hai, toh number fractional spare ke roop mein padhta hai.

Toh $M S = 0.20$ ka matlab hai " $20%$ strength factored load ke baad bhi spare hai".

$1$ SUBTRACT KYU KAREIN. Log " $0$ = exactly enough, positive = spare, negative = fails" ko " $1$ = exactly enough" se kahin zyada easily eyeball kar paate hain. $- 1$ ruler ko pass/fail line par recenter kar deta hai.

PICTURE. Step 5 ka bar comparison, ab top par extra strip $M S$ label ke saath — woh fraction jitna strength bar demand bar ke upar stick up karta hai.

Parent ka Example 2 picture par try karo: $S_{allow} = 180$ , $S_{limit} = 120$ , $F O S = 1.4$ . Inflated demand $= 1.4 \times 120 = 168$ . Phir $M S = 180/168 - 1 = 0.071$ — spare ka ek patla sa sliver, aur picture exactly dikhata hai kitna patla.

Step 7 — Edge cases: teen tarike jismein picture degenerate hoti hai

KYA HAI. Ek rule tabhi trustworthy hai jab hum jaanein ki extremes par woh kya karta hai. Teen special situations:

$M S = 0$ exactly — strength bar demand bar ke equal hai. Pass hai, lekin zero spare. Ek kharab weld aur yeh fail ho jaata hai.
$M S < 0$ — demand bar strength bar se taller hai. Fail. Redesign karo: thicken karo, rib add karo, ya material badlo (dekho Finite Element Analysis of Spacecraft Structures ki kahan fail hota hai).
$F O S = 1$ — koi inflation nahi. Demand bar = raw limit load. Tab $M S$ strength ko hamare unpadded forecast se compare karta hai — forecast ke upar koi bhi real load scatter part ko toda deta hai. Isliye real spacecraft par kabhi $F O S = 1$ use nahi hota; qualification testing (dekho Qualification vs Acceptance Testing) ultimate se upar ke factors use karta hai.

TEENO KYU DIKHAO. Taaki tum kabhi koi bar chart na milo jo tum padh na sako. Positive, zero, negative, aur degenerate no-cushion case sirf yehi outcomes hain — aur yahan sab hain.

PICTURE. Teen mini bar-pairs ek saath: pass with margin (green tick), $M S = 0$ knife-edge (amber), aur fail (red cross).

Ek-picture summary

KYA HAI. Poori chain, left se right: load forecast karo → $F O S$ se inflate karo → material allowable ke saamne khada karo → $M S$ padho.

Recall Feynman: poora walkthrough kisi dost ko batao

Mere paas ek rocket part hai aur kuch cheez usse dhakka deti hai. Mein us dhakke ko fly karne se pehle measure nahi kar sakta, toh mein worst dhakke ka best guess karta hoon — woh hai limit load. Lekin mera guess kam bhi ho sakta hai, toh mein part ko ek dhakka face karwaata hoon jo, say, $1.25 \times$ bada ho — woh multiplier hai factor of safety; yeh push ko badhata hai, metal ko kabhi nahi. Ab mein dekhta hoon part actually kya hold kar sakta hai — uska allowable — aur do answers hain: "hamesha ke liye bend mat karo" wala answer (yield) aur "snap mat karo" wala answer (ultimate). Mein inflated push ko strength ke saamne ek ruler par khada karta hoon. Agar strength taller hai toh mein safe hoon; kitna taller, fraction ke roop mein, woh hai margin of safety $M S$ . Mein $1$ subtract karta hoon taaki $0$ ka matlab ho "exactly enough", positive ka matlab "spare", negative ka matlab "redesign". Zero margin ek knife-edge hai, aur $1$ ka factor matlab hai koi cushion nahi — dono hi tarike hain jisse launch day par parts khatam hote hain. Bas itna hai: push guess karo, pad karo, strength se compare karo, margin padho.

Recall

$F O S$ kaun sa arrow stretch karta hai — load ya strength? ::: Load (demand). Yeh material strength ko untouched chhod deta hai. $M S = S_{allow} / (F O S \cdot S_{limit}) - 1$ mein, ratio kya equal hota hai jab strength exactly inflated demand se milti hai? ::: Exactly $1$ , toh $M S = 0$ — knife-edge, just barely OK. Real spacecraft par $F O S = 1$ kabhi kyun nahi aata? ::: Yeh koi cushion nahi deta; forecast ke upar koi bhi real load scatter part ko toda deta hai. Do allowable ceilings kya hain aur dono ko pass karna kya prevent karta hai? ::: Yield (koi permanent bend nahi) aur ultimate (koi rupture nahi); part ko dono clear karne chahiye.