SHM acceleration treated as constant: the one motion in the course where every kinematic formula is illegal

01The mistake

Students use $v^2 = v_0^2 + 2a\Delta x$ and its siblings on an oscillator, taking $a$ as the maximum acceleration or as some average. The answers are wrong in a way that is hard to see, because they have the right units, the right rough size, and the right dependence on amplitude.

The conceptual signature is the equilibrium point. Ask where the acceleration is greatest and where it is zero. A constant-acceleration thinker often says acceleration is greatest at the centre, reasoning from the fact that the mass moves fastest there. Speed and acceleration have been fused into a single idea of “how much motion,” and in SHM those two quantities are exactly out of phase — the acceleration is zero precisely where the speed peaks.

A related surface form: students say that because the mass is momentarily at rest at the turning point, the acceleration there must be zero too. That is the velocity-acceleration conflation from Unit 1 reappearing in a new context, and in SHM it is maximally wrong, since the turning point is where the acceleration is largest.

Watch also for $a = -\omega^2 x$ being memorised without its consequence. Students can recite it and still answer that acceleration is constant during the motion, because they have not read the equation as saying that $a$ depends on $x$ and $x$ is changing. The formula is stored as a fact rather than as a description.

02Why it makes sense to the student

Everything before this unit had constant acceleration. Free fall, inclines, the entire dynamics unit: $a$ was a number you found once and used throughout. That is not a bad habit, it is the correct habit for every problem the course has posed so far, and SHM is the first place it fails.

The motion looks smooth and regular, and “regular” slides easily into “uniform.” A pendulum swinging steadily gives a strong impression of something unchanging, and students transfer that impression to the wrong quantity. The period is constant; the acceleration is not.

The restoring force is introduced as $F = -kx$, which is a compact statement that the force depends on position. But students meet Hooke's law long before they meet oscillation, usually in static contexts where a spring is stretched and held. In that setting $x$ is fixed, so $F$ is a number, and the position dependence never has to be confronted.

And $a = -\omega^2 x$ has the same shape as formulas that are constants times constants. Students pattern-match it to an expression that yields a fixed value, rather than reading it as a function evaluated at a moving point.

03The correction

Start from the defining relation and read it aloud as a sentence about change:

$$F_x = -kx \quad\Longrightarrow\quad a = -\frac{k}{m}x = -\omega^2 x$$

The acceleration is proportional to the displacement and opposite in direction. Since $x$ changes continuously through the cycle, $a$ does too. That is not a side effect of SHM; it is the definition of it, and it is why none of the constant-acceleration equations can be used anywhere in this unit.

Make the phase relationship explicit, because it is the thing students get backwards. At the turning points, $|x|$ is maximum, so $|a|$ is maximum and $v = 0$. At equilibrium, $x = 0$, so $a = 0$ and $|v|$ is maximum. Speed and acceleration peak at opposite ends of the motion. Have students mark both on a single diagram of the oscillator, at five positions across a full cycle.

The demonstration that settles it: a mass on a spring, moving slowly enough to watch. Ask the class to point at the moment the mass is being pushed hardest. It is at the extremes, where it is momentarily not moving at all — the one place their intuition says nothing is happening. Then ask when it is being pushed least, and the answer is the instant it is flying through the middle.

Then close the loop with the calculus, since this is the course where you can. $a = d^2x/dt^2 = -\omega^2 x$ is a differential equation whose solution is $x(t) = A\cos(\omega t + \phi)$; differentiate twice and the $-\omega^2$ falls out. Students who see the sinusoid produced by the restoring condition stop treating SHM as a separate topic with its own formula sheet, which is also the repair for C-U7-PH3.

04A sample question

05What each wrong answer reveals

• A Speed and acceleration fused. The student has reasoned that the fastest point must be the most forced point, which treats acceleration as a measure of how much motion there is rather than of how motion is changing. This is the Unit 1 velocity-acceleration conflation arriving in a context where it produces the exact opposite of the truth. It is the most common wrong answer here and it needs the phase relationship drawn, not restated.
• B Correct. $a = -\omega^2 x$, so $|a|$ is greatest where $|x| = A$. At the turning points the block is momentarily at rest, so the speed there is zero. Acceleration and speed peak at opposite ends of the motion.
• C Half right, and the wrong half is diagnostic. The student has located the acceleration correctly — they are reading $a = -\omega^2 x$ properly — and then attached the maximum speed to the same place, because in their model the extremes of the motion are where everything is extreme. They need only the velocity half of the picture, which makes this a much quicker repair than A despite scoring identically.
• D Constant acceleration imported wholesale. The reasoning cited is revealing: the spring constant does not change, therefore the acceleration does not. The student has read $a = -kx/m$ and taken the constants to fix the value, without registering that $x$ is the variable. This is the misconception in its purest form and it needs the equation re-read as a function before anything else will land.

D and A are different failures despite both scoring zero. D has not noticed that $a$ depends on anything that moves; A has noticed but has attached the dependence to speed. C is nearly there and needs a single correction. The fastest way to tell them apart is the sketch: ask for $x(t)$, $v(t)$ and $a(t)$ stacked on shared axes, and the three wrong models produce three visibly different sets of curves.

06Try it in Mistake Master