Non-Inertial Frames & Pseudo-Forces

Every launch you have watched runs pseudo-force engineering. Astronauts train in long-arm centrifuges where the rotating-frame centrifugal force stands in for launch loads of $3g$ to $6g$ — flight surgeons speak fluently in "eyeballs-in" $g_{\text{eff}}$, exactly your $\sqrt{g^2 + a_0^2}$ with bigger numbers. The same bookkeeping runs the other way in orbit: a space station is a frame in permanent free fall, $g_{\text{eff}} = 0$, which is why "weightlessness" is frame language, not the absence of gravity — Earth's pull at the ISS's altitude is still 89% of the surface value. And the largest non-inertial laboratory is the planet itself: Earth's rotation shaves about $0.3\%$ off effective gravity at the equator and steers every cyclone and ocean gyre through the Coriolis pseudo-force — the rotating-frame deep dive waiting for you at UG level.

Run the test that defines everything. Place a puck on the frictionless floor of a stationary train and release it: it sits still. Newton's first law, verified. Now let the train accelerate forward at $a_0$ and release the puck again. The ground observer sees nothing strange — the puck, force-free, stays put while the train floor slides forward beneath it. But the passenger sees the puck accelerate backward at $a_0$ with no force in sight. Her frame has failed the first-law test. That is the operational definition, and it needs no philosophy: a frame is non-inertial precisely when free objects refuse to keep constant velocity in it.

The passenger's repair is beautifully cheap. Every unexplained acceleration she observes is the same for all objects: $-a_0$, the negative of her frame's acceleration, regardless of mass. Heavy toolbox, light phone — both drift backward at $a_0$ when released. So she can blame a single universal "force" that scales with mass: charge every object $m$ a pseudo-force $ma_0$ pointing backward, and Newton's second law works perfectly again in her coordinates. The pseudo-force is not an interaction — nothing exerts it, it has no Newton's-third-law partner, and it vanishes the moment you step back to the ground frame. It is a bookkeeping correction: the frame's own acceleration, moved to the force side of the ledger and dressed in newtons. That mass-proportionality is the deepest thing in this article. Only one genuine force in nature scales with the mass of the object it acts on: gravity. A pseudo-force is therefore indistinguishable, locally, from extra gravity — which is why a braking bus feels like the floor tilting forward, and why an accelerating lift feels like gravity strengthening. Einstein promoted this coincidence into the equivalence principle, the seed of general relativity. At exam level it gives you the single most useful concept of the chapter: effective gravity. Inside a frame accelerating at $a_0$, everything behaves as if gravity were the vector combination of $g$ downward and $a_0$ reversed:

where: $\theta$ — the angle effective gravity makes with the true vertical; a plumb line, a hanging pendulum, the surface of water in a glass — all align themselves with $g_{\text{eff}}$, not with $g$. For a vertically accelerating frame (a lift) the vectors are parallel, and the combination collapses to the scalar $g_{\text{eff}} = g \pm a_0$ — the entire lift-and-apparent-weight archetype in one substitution.

The rule costs one line to state. If your frame accelerates at $a_0$ relative to the ground, then for every mass $m$ you analyse inside that frame:

where: $a_0$ — the frame's acceleration (not the object's); $m$ — the mass of the object being analysed, so a 60 kg passenger in a bus braking at $3\ \mathrm{m/s^2}$ is charged $180\ \mathrm{N}$ forward. Add this to the real forces, then solve $F = ma$ using accelerations measured inside the frame. An object at rest in the frame is in equilibrium in the frame — statics, with one extra force on every FBD. Where does the rule come from? From nothing deeper than adding velocities. Positions measured from the ground and from the moving frame differ by the frame's own position. Differentiate that statement conceptually — rates of change of position are velocities, rates of change of velocity are accelerations — and the accelerations obey the same addition:

where: $a_{\text{rel}}$ — the object's acceleration as measured inside the frame; $a_0$ — the frame's acceleration. This is the same relative-motion bookkeeping you used for river-boat problems, applied to accelerations instead of velocities. It holds whenever the frame translates without rotating.

One warning before the classic application: the derivation used only translation. A rotating frame is also non-inertial, but rotation contaminates the bookkeeping in a richer way and generates velocity-dependent corrections — that is the Coriolis story, a UG deep dive. The piece of it you need now is the simplest rotating-frame pseudo-force, previewed below.

Centrifugal force — the rotating-frame preview

Ride a merry-go-round turning at angular speed $\omega$ and stand at radius $r$. In the ground frame you are accelerating inward at $\omega^2 r$ — some real force (friction, a railing) must supply the centripetal demand $m\omega^2 r$. In your rotating frame you are at rest, so the repair kit must charge you a pseudo-force to balance the books:

where: $r$ — your distance from the rotation axis. This outward pseudo-force is what your body reports as "being flung outward." It is legal currency only inside the rotating frame: in a ground-frame FBD there is no outward force on you — there is an inward net force, and calling anything "centrifugal" there is the single most-penalized error of the circular-motion chapter. The full rotating-frame toolkit (including the velocity-dependent Coriolis pseudo-force that steers cyclones and Foucault pendulums) is UG territory — flagged below as a deep dive.

Which route answers which question

1 · Board level — the plumb line in an accelerating train (full method)

A small bob hangs by a light string from the roof of a train accelerating uniformly at $a_0 = 2.5\ \mathrm{m/s^2}$ on level track. Find the steady angle the string makes with the vertical and the tension if the bob's mass is $0.40$ kg. Take $g = 10\ \mathrm{m/s^2}$.

Step 1. Choose the train frame and declare the toll: pseudo-force $ma_0$ backward on the bob, plus real forces $T$ and $mg$. The bob hangs steady in this frame ⇒ equilibrium.

Step 2. Resolve along horizontal and vertical: $T\sin\theta = ma_0$ and $T\cos\theta = mg$.

Step 3. Divide: $\tan\theta = a_0/g = 0.25 \Rightarrow \theta = \boxed{14°}$ (to two significant figures).

Step 4. Square-and-add for the tension: $T = m\sqrt{g^2 + a_0^2} = 0.40\sqrt{100 + 6.25} = 0.40 \times 10.31 = \boxed{4.1\ \mathrm{N}}$. Sanity: $T > mg = 4.0$ N — the string must both hold the bob up and drag it forward, so tension exceeds the weight. The string leans backward (away from the acceleration), which matches the lurch you feel.

2 · Board level — reading a bathroom scale in a decelerating lift

A 50 kg student stands on a scale in a lift. The lift, moving upward, decelerates at $2.0\ \mathrm{m/s^2}$ as it approaches her floor. What does the scale read? ($g = 10\ \mathrm{m/s^2}$.)

Step 1. Moving up + slowing down ⇒ acceleration points downward, magnitude $2.0\ \mathrm{m/s^2}$.

Step 2. In the lift frame the pseudo-force is opposite the frame's acceleration: $ma_0 = 100$ N upward. Effective gravity: $g_{\text{eff}} = g - a_0 = 8.0\ \mathrm{m/s^2}$.

Step 3. In the lift frame the student is in equilibrium: $N = mg_{\text{eff}} = 50 \times 8.0 = \boxed{400\ \mathrm{N}}$.

Step 4. Cross-check from the ground frame: $mg - N = ma_0 \Rightarrow N = 50(10 - 2) = 400$ N ✓. The scale reads less than her true weight (500 N) — arrival at an upper floor always feels light. The full family of these problems lives in the Lifts & Apparent Weight archetype.

3 · Competitive level — block on a smooth accelerating wedge (JEE-Main pattern, ~2.5 minutes)

A block rests on the smooth $30°$ incline of a wedge. With what horizontal acceleration $a_0$ must the wedge be pushed so that the block does not slide relative to the wedge? ($g = 10\ \mathrm{m/s^2}$.)

Step 1. Ride the wedge. In the wedge frame the block must sit in equilibrium under three forces: weight $mg$ down, normal $N$ perpendicular to the incline, pseudo-force $ma_0$ horizontal (backward, i.e. into the slope).

Step 2. Resolve along the incline (the direction the block could slide). Down-slope component of gravity: $mg\sin 30°$. Up-slope component of the pseudo-force: $ma_0\cos 30°$. Equilibrium along the slope:

Step 3. Solve: $a_0 = g\tan 30° = 10/\sqrt{3} = \boxed{5.8\ \mathrm{m/s^2}}$.

Step 4. Bonus mark: the normal force. Perpendicular to the incline: $N = mg\cos 30° + ma_0\sin 30° = m(10)(0.866) + m(5.77)(0.5) = 11.5m$ — larger than $mg$, because the wedge is shoving the block as well as supporting it. Timing note: with the pseudo-force this is a 90-second, two-line problem — $a_0 = g\tan\theta$ is worth memorizing with its logic. The ground-frame route (block and wedge share horizontal acceleration; resolve $N$; two equations) gives the same answer in roughly triple the time. Budget 2.5 min including the $N$ part.

4 · Competitive level — JEE-Advanced stretch: the tilted water surface (~4 minutes)

A closed rectangular tank half-full of water accelerates horizontally at $a_0 = 7.5\ \mathrm{m/s^2}$. (a) At what angle to the horizontal does the water surface settle? (b) A cork is released from rest at the bottom of the tank (fully submerged, then floating up). In the tank frame, in which direction does the cork initially accelerate? ($g = 10\ \mathrm{m/s^2}$.)

Step 1. (a) In the tank frame every parcel of water feels effective gravity: $g$ down combined with $a_0$ backward. The free surface always sets itself perpendicular to $g_{\text{eff}}$ — that is what "level" means to the fluid.

Step 2. The surface therefore tilts from the horizontal by $\theta$ where $\tan\theta = a_0/g = 0.75 \Rightarrow \theta = \boxed{37°}$, rising toward the rear of the tank (effective gravity leans backward, so the surface banks up at the back).

Step 3. (b) Buoyancy always points opposite effective gravity — it is "up" as the fluid defines up. A cork is less dense than water, so in the tank frame the net of (effective weight $mg_{\text{eff}}$ along $g_{\text{eff}}$) and (buoyancy $> mg_{\text{eff}}$ against it) points along $-g_{\text{eff}}$: the cork accelerates perpendicular to the tilted surface — i.e. forward and up, not straight up.

Step 4. Direction check with numbers: $g_{\text{eff}}$ points $37°$ backward from vertical; the cork sets off $37°$ forward of vertical. Counter-intuitive and classic: light objects lean into the acceleration (the same reason a helium balloon in an accelerating car swings forward). Timing note: part (a) is the one-liner $\tan\theta = a_0/g$ — the same triangle as the pendulum, because every "which way is down?" question is the same question. Part (b) is the discriminator: write "buoyancy antiparallel to $g_{\text{eff}}$" and the answer falls out with zero computation. Budget 4 min; if you catch yourself computing pressure fields, restart with $g_{\text{eff}}$.

Pseudo-forcing the wrong frame. The toll is charged only when you work inside the accelerating frame. If you write ground-frame accelerations ($a_{\text{ground}}$ on the right-hand side) and add $ma_0$ to the forces, you have paid the toll twice; answers come out shifted by exactly $ma_0$. First line of every solution: "Working in the ___ frame." Then stay there.

Wrong direction. The pseudo-force opposes the frame's acceleration — not the frame's velocity, not the object's motion. A bus braking while moving forward has $a_0$ backward, so the pseudo-force is forward (you lurch toward the windshield). Speed is irrelevant; only the acceleration vector matters.

Wrong mass. Each object is charged its own $m a_0$. In multi-body wedge problems, the block pays $m a_0$ and the wedge pays $M a_0$ — a merged $(M+m)a_0$ applied to one body is a wrong FBD.

Centrifugal force in the ground frame. In an inertial frame, circular motion needs an inward net force and there is no outward one — see the pitfall box. JEE-Advanced multiple-choice options are built to catch this: any ground-frame statement containing "the outward centrifugal force on the body" is bait.

Third-law partner hunting. Pseudo-forces have no reaction pair — nothing exerts them. If an exam asks you to list interaction pairs, pseudo-forces never appear; conversely if your FBD "reaction" to $ma_0$ acts on the frame itself, delete it.

Forgetting they vanish. The moment a question shifts to the ground observer ("what does a person on the platform see?"), all pseudo-forces evaporate. The physics they explained is re-explained by real forces and honest accelerations.

$g = 9.8$ vs $10$. These are JEE-Main/Advanced problems: the paper almost always declares $g = 10\ \mathrm{m/s^2}$ — use exactly what is stated, and keep $g_{\text{eff}}$ consistent with it throughout.

Texts. HC Verma, Concepts of Physics Vol. 1, Ch. 5 (frame of reference discussion) · DC Pandey, Mechanics 1 (accelerating-frame problems) · Kleppner & Kolenkow, An Introduction to Mechanics, Ch. 9 (the honest UG treatment).

Provenance. All results are standard. A verifier should recompute: the pendulum $\tan\theta = a_0/g$ both routes; Example 1's $\theta = 14°$, $T = 4.1$ N; Example 2's $N = 400$ N both frames; Example 3's $a_0 = g\tan 30° = 5.8\ \mathrm{m/s^2}$ and $N = 11.5m$; Example 4's $37°$ tilt from $\tan\theta = 0.75$.

Prerequisites. Force, Inertia & Newton's First Law · Newton's Second Law · Free-Body Diagrams. Leads to. Block on an Accelerating Wedge · Lifts & Apparent Weight · Pendulum in Lift, Cart & on Incline (oscillations). Deep dives (UG). Non-Inertial Reference Frames · the Coriolis pseudo-force story · Centrifugal Force.