Mathwords logoMathwords

Bessel Functions — Definition, Formula & Example

Bessel functions are a family of oscillating functions that come from problems with circular or cylindrical symmetry, such as a vibrating drumhead or heat flowing in a pipe. They look like cosine or sine waves that slowly shrink as x grows.

For a constant α\alpha, Bessel's differential equation is x2y′′+xy′+(x2−α2) y=0x^2 y'' + x y' + (x^2 - \alpha^2)\,y = 0. The Bessel function of the first kind, Jα(x)J_\alpha(x), is the solution that is finite at x=0x = 0 when α≥0\alpha \ge 0, normalized by the power series below. For integer order α=n\alpha = n, the general solution is y=c1Jn(x)+c2Yn(x)y = c_1 J_n(x) + c_2 Y_n(x). Here YnY_n, the Bessel function of the second kind, is unbounded as x→0+x \to 0^+.

Key Formula

Jα(x)=∑m=0∞(−1)mm! Γ(m+α+1)(x2)2m+αJ_\alpha(x) = \sum_{m=0}^{\infty} \frac{(-1)^m}{m!\,\Gamma(m+\alpha+1)}\left(\frac{x}{2}\right)^{2m+\alpha}
Where:
  • Jα(x)J_\alpha(x) = Bessel function of the first kind of order α\alpha
  • α\alpha = the order, a constant from the differential equation (often a nonnegative integer nn)
  • mm = summation index, m=0,1,2,…m = 0, 1, 2, \dots
  • Γ\Gamma = the Gamma function, which satisfies Γ(k+1)=k!\Gamma(k+1) = k! for whole numbers kk

Worked Example

Problem: Use the first four terms of the series to estimate J0(1)J_0(1).
Step 1: Set α=0\alpha = 0. Then Γ(m+1)=m!\Gamma(m+1) = m!, so the series simplifies.
J0(x)=∑m=0∞(−1)m(m!)2(x2)2mJ_0(x) = \sum_{m=0}^{\infty} \frac{(-1)^m}{(m!)^2}\left(\frac{x}{2}\right)^{2m}
Step 2: Substitute x=1x = 1 and compute the terms for m=0,1,2,3m = 0, 1, 2, 3.
1,−(1/2)21=−14,(1/2)44=164,−(1/2)636=−123041,\quad -\frac{(1/2)^2}{1} = -\frac{1}{4},\quad \frac{(1/2)^4}{4} = \frac{1}{64},\quad -\frac{(1/2)^6}{36} = -\frac{1}{2304}
Step 3: Add the terms over the common denominator 2304. The series alternates and its terms shrink, so the error is smaller than the next term, (1/2)8(4!)2=1147456≈0.0000068\frac{(1/2)^8}{(4!)^2} = \frac{1}{147456} \approx 0.0000068.
1−14+164−12304=2304−576+36−12304=17632304≈0.765191 - \frac{1}{4} + \frac{1}{64} - \frac{1}{2304} = \frac{2304 - 576 + 36 - 1}{2304} = \frac{1763}{2304} \approx 0.76519
Answer: J0(1)≈17632304≈0.76519J_0(1) \approx \dfrac{1763}{2304} \approx 0.76519, which agrees with the true value J0(1)≈0.7652J_0(1) \approx 0.7652 to four decimal places.

Why It Matters

When you solve the wave or heat equation in polar or cylindrical coordinates by separation of variables (a standard topic in a PDEs or mathematical physics course), the radial part is a Bessel function. The radial part of each vibration mode of a circular drumhead is Jn(kr)J_n(kr), and the modes' frequencies are set by the zeros of JnJ_n. Engineers also use Jn(β)J_n(\beta) to find the sideband strengths of an FM radio signal.

Common Mistakes

Mistake: Treating J0J_0 like a cosine curve, with constant amplitude and evenly spaced zeros.
Correction: Jn(x)J_n(x) is not periodic. For large xx, its amplitude shrinks roughly like 2/(πx)\sqrt{2/(\pi x)}. The gaps between consecutive zeros are uneven at first and only approach π\pi as xx grows. Look up the zeros in a table or compute them numerically. Do not assume they fall at multiples of π\pi.

Related Terms