PDA

View Full Version : If a + b + c = abc, prove that at least one of (a,b,c) is < or = sqrt(3)


maverick280857
Jun5-04, 08:04 AM
Hi

Here's a question....

Let a, b, c be three non-zero real numbers such that

a + b + c = abc

Prove that at least one of these three numbers (a, b or c) is less than or equal to the square root of 3.

Can you prove this without trigonometry? The trigonometric solution follows...

Solution (using Trigonometry)

Let a = tan(A), b = tan(B), c = tan(C) (for some nonzero angles A, B, C which are real) so that the given constraint becomes

tan(A) + tan(B) + tan(C) = tan(A)tan(B)tan(C)

which can be true iff A + B + C = n*pie (n is an integer)

If n = 1, then A, B, C are the angles of a triangle (as the constraint is true for angles of a triangle). The result follows by considering cases: of an equilateral triangle where A = B = C = pie/3 radians so that each of a, b and c is equal to sqrt(3); next consider the case of a scalene triangle where A, B and C are all distinct. If A > pie/3, then B+C = pie-A = pie-(qty less than pie/3) and so either B or C is less than pie/3.

--------------------------------------------------------------------------

Cheers
Vivek

arildno
Jun5-04, 08:42 AM
i) a=b=c=\sqrt{3} is a solution of the equation
ii) Setting a=\sqrt{3}+\hat{a} and similarly for b and c

The proof follows readily from this
(i.e., at least one of the hatted numbers must be non-positive.)

maverick280857
Jun6-04, 11:06 AM
Hi

Thanks. Your method is interesting...we get a new constraint on the hatted numbers now...am I right? (I haven't completed the solution using your substitutions..)

Cheers
Vivek

matt grime
Jun6-04, 11:44 AM
proof without using trig or substitution: use the arithmetic mean/geometric mean inequality

Gokul43201
Jun6-04, 12:00 PM
Hi

Thanks. Your method is interesting...we get a new constraint on the hatted numbers now...am I right? (I haven't completed the solution using your substitutions..)

Cheers
Vivek

Yes, you end up with an equation which can be written like

f(\hat{a},\hat{b},\hat{c}) = -g(\hat{a},\hat{b},\hat{c})

And you'll find that for positive values of \hat{a},\hat{b},\hat{c}, the functions f and g must give positive numbers. So you have LHS = positive number and RHS = negative number...a contradiction ! So, one of \hat{a},\hat{b},\hat{c} must not be positive.

maverick280857
Jun6-04, 12:11 PM
Thanks Gokul43201

I get it now :-D

Cheers
Vivek

StonedPanda
Jun6-04, 07:38 PM
I don't =[