Setup (WHY this coordinate choice): put the foci symmetrically on the x-axis at F1(−c,0) and F2(c,0) so the algebra is symmetric and the centre sits at the origin.
Let P=(x,y). Then
(x+c)2+y2−(x−c)2+y2=±2a.
Step 1 — isolate one root.Why? One square root at a time is manageable.
(x+c)2+y2=±2a+(x−c)2+y2.
Step 2 — square both sides.Why? Kills the isolated radical.
(x+c)2+y2=4a2±4a(x−c)2+y2+(x−c)2+y2.
Expand and cancel x2,c2,y2:
4cx=4a2±4a(x−c)2+y2⇒cx−a2=±a(x−c)2+y2.
Step 3 — square again.Why? Remove the last radical.
c2x2−2a2cx+a4=a2[(x−c)2+y2]=a2x2−2a2cx+a2c2+a2y2.
Cancel −2a2cx, regroup:
(c2−a2)x2−a2y2=a2(c2−a2).
Step 4 — name the leftover. Since c>a, the quantity c2−a2>0. Define
b2=c2−a2.Why define b? It makes the equation clean AND turns out to be the vertical semi-extent that fixes the asymptote slopes. Divide by a2b2:
The vertical version just swaps roles:
a2y2−b2x2=1(foci on the y-axis,(0,±c)).
Directrix form: each focus pairs with a directrix line x=±a/e, and for every point P:
(distance to that directrix)PF=e.
This is the single unifying definition of all conics.
Derivation. Solve for y:
y2=b2(a2x2−1)⇒y=±abx1−x2a2.
As ∣x∣→∞, 1−a2/x2→1, so
y→±abx.
Gap → 0 proof (Why they never cross): the vertical gap between curve and line is
abx−abx1−x2a2=abx(1−1−x2a2)x→∞0,
because the bracket shrinks like 2x2a2 and x only grows linearly. Product →0: the curve hugs but never meets the line.
far out, the "1" is negligible so curve → x2/a2−y2/b2=0
Recall Feynman: explain to a 12-year-old
Imagine two thumbtacks (the foci) and you walk so that one tack is always exactly, say, 4 steps closer than the other. Not the total — the difference. If you keep that difference fixed and trace where you can stand, you get a curve that swoops toward one tack, then a mirror copy swoops toward the other: two separate arcs. Way out at the edges, these arcs straighten out and run almost perfectly alongside two crossing straight rulers (the asymptotes) — getting closer and closer forever but never quite touching, like a car racing beside a wall it never scrapes. How "wide open" the arcs are is one number, e; for a hyperbola e is always bigger than 1.
Dekho, hyperbola samajhne ka sabse aasan tareeka yeh hai: ellipse mein hum do foci se distances ko add karte the aur constant rakhte the. Hyperbola mein bas ek operation badal jaata hai — ab hum distances ka difference constant rakhte hain, yaani ∣PF1−PF2∣=2a. Yeh chhota sa minus sign hi wajah hai ki curve do alag-alag branches mein tut jaata hai jo infinity ki taraf bhaagti hain, band nahi hoti.
Standard equation a2x2−b2y2=1 hai. Yaad rakho sabse important cheez: hyperbola mein foci vertices ke bahar hote hain, isliye c>a aur relation c2=a2+b2 (plus!) — ellipse waala minus yahan mat lagana, yeh sabse common galti hai. Eccentricity e=c/a hamesha 1 se bada hota hai. Aur ek shortcut: e=1+b2/a2 — isse tumhe a,b alag se nahi chahiye.
Asymptotes woh do straight lines hain y=±abx jinke paas curve door jaakar chipak jaata hai par kabhi touch nahi karta. Kaise nikaale? Equation mein jab x bahut bada ho, toh "1" ignore ho jaata hai, aur bacha a2x2−b2y2=0 — yeh do lines hi hain. Slope b/a hota hai (na ki a/b), yeh central box ke corner (a,b) se check kar lo.
Yeh cheez real life mein bhi kaam aati hai — jaise navigation systems mein jahan do stations se signal ka time-difference constant hota hai, wahan tumhari position ek hyperbola pe aati hai. Toh concept clear rakho: difference constant, plus sign, e>1, slope b/a. Bas!