If you could keep just one thing from calculus for physics, it might be this. Nearly any smooth function, however complicated, looks like a simple polynomial when you zoom in. Physicists use this constantly to turn impossible equations into easy ones.
- Start with \(\sin x\) and raise the highest power one at a time. Watch the amber curve hug more and more of the wave.
- Look at where it's close at power 1: for small \(x\), \(\sin x \approx x\).
- Try \(e^x\), then \(\ln(1 + x)\). The log stops improving past \(x = 1\) however many terms you add.
The idea
Choose a polynomial that matches your function at \(x = 0\): the same height, the same slope, the same curvature, and so on. Each extra term copies one more derivative.
Working it out for \(e^x\)
Every derivative of \(e^x\) is \(e^x\), and at \(x = 0\) that's 1.
Put those into the formula.
The ones you'll use forever
A pendulum's true equation has \(\sin\theta\) in it and has no simple solution. For small swings, \(\sin\theta \approx \theta\), and it becomes the easy spring equation. And Einstein's energy \(mc^2\left(1 - v^2/c^2\right)^{-1/2}\) gives \(mc^2 + \frac{1}{2}mv^2\) for small \(v\), so Newton's kinetic energy falls straight out of relativity.
Problems
Use the series to estimate \(e^{0.1}\) to three terms.
Answer
\(1 + 0.1 + 0.005 = 1.105\). The true value is 1.10517.
Estimate \(\cos 0.2\).
Answer
\(1 - \frac{0.04}{2} = 0.98\). The true value is 0.98007.
Estimate \(\sqrt{1.02}\).
Answer
\((1 + 0.02)^{1/2} \approx 1 + \frac{1}{2}(0.02) = 1.01\). The true value is 1.00995.
Differentiate the series for \(\sin x\) term by term. What do you get?
Answer
\(\frac{d}{dx}\left(x - \frac{x^3}{6} + \dots\right) = 1 - \frac{x^2}{2} + \dots\), which is the series for \(\cos x\).