2.2.2FoundationsStretch

Using Logarithms in Physics

Not every physical relationship gives you a nice straight-line graph. Radioactive decay curves downward, a pendulum’s period grows with the square root of its length, and neither is easy to analyse accurately by eye. Take the logarithm of the right variable, though, and both problems collapse into a straight line — one of the most quietly powerful tricks in the whole course.

35 min Video by A Level Physics HQ Module 2.2 Making measurements and analysing data
Analyse log graphs in physics — A-Level Physics How ToCurated by Kepler, not produced by us. Watch it before the notes below.
Open on YouTube

What you'll be able to do

  • Explain why taking logarithms can turn a curved graph into a straight line
  • Linearise a power-law relationship y=Axny=Ax^n using logarithms
  • Linearise an exponential relationship y=Aekxy=Ae^{kx} using natural logarithms
  • Find unknown constants (A, n or k) from the gradient and intercept of a log graph
  • Identify situations elsewhere in the course where this technique applies
1

Why logarithms linearise a graph

A straight-line graph is far easier to analyse than a curve — you can read off a gradient and intercept with confidence, and easily judge how well data fits a proposed relationship. Many physical relationships are not\textbf{not} naturally linear, but two very common non-linear forms — power laws and exponentials — can both be turned into straight lines by taking an appropriate logarithm of one or both variables.

Tip — The general habit: if a relationship has an unknown POWER, take logs of both x and y. If it has an unknown quantity in an EXPONENT, take the natural log of y only (keeping x as it is).

2

Linearising a power law

A relationship of the form y=Axny=Ax^n (where AA and nn are unknown constants) becomes linear when you take logarithms of both sides: log⁡y=log⁡A+nlog⁡x\log y = \log A + n\log x. Plotting log⁡y\log y against log⁡x\log x gives a straight line whose gradient is the power nn, and whose yy-intercept is log⁡A\log A.

log⁡y=nlog⁡x+log⁡A\log y = n\log x + \log A
Linearised power law; gradient = n, intercept = log A.
Example 1A graph of log(T) against log(l) for a simple pendulum (period T, length l) gives a straight line of gradient 0.50 and y-intercept 0.10. Find the power relationship between T and l, and the value of the constant A.
1Gradient =n=0.50=n=0.50, so T∝l0.5T \propto l^{0.5} — i.e. T∝lT \propto \sqrt{l}, consistent with T=2πl/gT=2\pi\sqrt{l/g}.
2Intercept =log⁡A=0.10=\log A = 0.10, so A=100.10A = 10^{0.10}.
3A≈1.26A \approx 1.26.
Answer
n = 0.50 (T ∝ √l); A ≈ 1.26.

Tip — A gradient of exactly 0.5 on a log-log graph is a strong clue you’re looking at a square-root relationship — very common for period-vs-length type formulae in this course.

3

Linearising an exponential relationship

A relationship of the form y=Aekxy=Ae^{kx} becomes linear when you take the natural\textbf{natural} logarithm of both sides: ln⁡y=ln⁡A+kx\ln y = \ln A + kx. Plotting ln⁡y\ln y against xx (not ln⁡x\ln x — the independent variable here stays as it is) gives a straight line with gradient kk and yy-intercept ln⁡A\ln A.

ln⁡y=kx+ln⁡A\ln y = kx + \ln A
Linearised exponential relationship; gradient = k, intercept = ln A.
Example 1A graph of ln(current) against time, for a discharging capacitor, has gradient −0.50-0.50 s⁻¹ and y-intercept 2.30. Find the initial current and the time constant of the circuit.
1The discharge equation is I=I0e−t/RCI=I_0e^{-t/RC}, so ln⁡I=ln⁡I0−tRC\ln I = \ln I_0 - \dfrac{t}{RC}; comparing, gradient =−1RC=-\dfrac{1}{RC} and intercept =ln⁡I0=\ln I_0.
2Intercept: ln⁡I0=2.30\ln I_0 = 2.30, so I0=e2.30≈9.97I_0 = e^{2.30} \approx 9.97 A.
3Gradient: −1RC=−0.50-\dfrac{1}{RC}=-0.50, so RC=10.50=2.0RC = \dfrac{1}{0.50} = 2.0 s.
Answer
I0≈10.0I_0 \approx 10.0 A; time constant RC = 2.0 s.

Tip — This is exactly the method used later in the course for both radioactive decay (ln N or ln A against t, gradient = −λ) and capacitor discharge (ln Q, ln V or ln I against t, gradient = −1/RC) — learn it once here and reuse it everywhere.

4

Where this is used elsewhere in the course

This single technique reappears across several later modules: finding a radioactive decay constant from a ln(activity)–time graph, finding a capacitor’s time constant from a ln(charge)–time graph, and testing whether an experimental relationship (for example, between a pendulum’s period and its length, or a resistor’s resistance and temperature) genuinely follows a proposed power law or exponential model, rather than just assuming it does.

Equation recap

log⁡y=nlog⁡x+log⁡A\log y = n\log x + \log A
Linearised power law y = Ax^n (gradient = n).
ln⁡y=kx+ln⁡A\ln y = kx + \ln A
Linearised exponential y = Ae^(kx) (gradient = k).

Common mistakes to avoid

Taking the log of both x and y when linearising an exponential relationship.
For y = Ae^(kx), only take the (natural) log of y — x stays as it is on the horizontal axis.
Taking the log of only y when linearising a power law relationship.
For y = Ax^n, BOTH x and y must be logged (log y against log x) — logging only one variable will not give a straight line.
Reading the constant A directly off the graph’s intercept value without "undoing" the logarithm.
The intercept gives log A (or ln A), not A itself — you must raise 10 (or e) to the power of the intercept to recover A.

Key takeaways

  • Taking logarithms turns awkward non-linear relationships into straight-line graphs that are far easier to analyse.
  • Power law y = Ax^n: plot log y against log x — gradient = n, intercept = log A.
  • Exponential y = Ae^(kx): plot ln y against x — gradient = k, intercept = ln A.
  • This technique underlies later work on radioactive decay constants and capacitor time constants.

Test yourself

Answer a few questions, then review anything you missed.

Prefer paper? Open the written self-check

Answer each question before opening its answer. This self-check does not record a quiz score.

  1. To linearise a power law relationship y = Ax^n, you should plot:

    • log y against log x
    • ln y against x
    • y against log x
    • log y against x
    Show answer and explanation

    Answer: log y against log x

    Both variables must be logged to linearise a power law.

  2. To linearise an exponential relationship y = Ae^(kx), you should plot:

    • ln y against x
    • log y against log x
    • y against ln x
    • ln y against ln x
    Show answer and explanation

    Answer: ln y against x

    Only y is logged (naturally); x is left unchanged on the horizontal axis.

  3. On a graph of log y against log x, the gradient represents:

    • The power n in y = Ax^n
    • The constant A directly
    • The value of y at x=0
    • The value of x at y=0
    Show answer and explanation

    Answer: The power n in y = Ax^n

    log y = n log x + log A, so gradient = n.

  4. On a graph of ln y against x, the y-intercept represents:

    • ln A
    • A itself
    • k, the exponential rate constant
    • The gradient
    Show answer and explanation

    Answer: ln A

    ln y = kx + ln A, so the intercept is ln A, not A itself.

  5. A log-log graph has gradient 2.0. The underlying relationship between y and x is:

    • y ∝ x²
    • y ∝ x
    • y ∝ √x
    • y ∝ 2x
    Show answer and explanation

    Answer: y ∝ x²

    Gradient = power n, so n=2 means y ∝ x².

  6. A ln y vs x graph has intercept 1.61. The value of A is closest to:

    • 5.0
    • 1.61
    • 0.48
    • 16.1
    Show answer and explanation

    Answer: 5.0

    A = e^(intercept) = e^1.61 ≈ 5.0.

  7. A radioactive sample’s activity is analysed with a ln(activity)–time graph, gradient −0.20 s⁻¹. The decay constant λ is:

    • 0.20 s⁻¹
    • −0.20 s⁻¹
    • 5.0 s⁻¹
    • 0.693 s⁻¹
    Show answer and explanation

    Answer: 0.20 s⁻¹

    ln A = ln A₀ − λt, so gradient = −λ; λ = 0.20 s⁻¹ (a positive physical quantity).

  8. Why is a log-log graph particularly useful for testing a proposed power-law relationship in an experiment?

    • A genuine power law produces a straight line, whose gradient directly gives the unknown power
    • It removes the need for any units
    • It only works if the power is exactly 1
    • It converts the data into a bar chart
    Show answer and explanation

    Answer: A genuine power law produces a straight line, whose gradient directly gives the unknown power

    If the data truly follows y=Ax^n, a log-log plot will be linear, and any curvature reveals the model does not fit.

Finished this lesson?

Loading your progress…

Next step

Keep practising

Start with the quiz above to check the key ideas. You can also choose another topic from the practice library.

Browse practice topics

Related lessons

All 4 lessons