Torque vs Work: Same Units, Different Quantities

The difference is one word of geometry

Two quantities, both defined as a force times a length, both reducing to $\text{kg}\cdot\text{m}^2/\text{s}^2$. One is energy and the other is not. The reason is not deep and it is not a convention: they resolve the force against different reference directions.

Work takes the component along the displacement. AP Physics 1 EK 3.2.A.3.i: only the component of the force exerted on a system that is parallel to the displacement of the point of application of the force will change the system's total energy. Hence the cosine:

Torque takes the component perpendicular to the position vector. EK 5.3.A.1: torque results only from the force component perpendicular to the position vector from the axis of rotation to the point of application of the force. Hence the sine:

Both lines are printed on the AP Physics 1 equation sheet, in the two columns of the mechanics block, and the two $\theta$ symbols do not mean the same thing. In the work equation $\theta$ is measured from the displacement. In the torque equation it is measured from the position vector running out of the axis.

So: parallel to a displacement, or perpendicular to a radius. Everything on this page unfolds from that one line. The parallel component moves energy; the perpendicular component turns things. Those are different jobs, done by different parts of the same push, which is why the two products cannot be the same quantity even when the arithmetic returns the same number.

Torque vs work, side by side

Read the two base-unit rows together. They are the same, and that is not a coincidence to be explained away: both quantities really are a force times a length. What differs is which length, measured from where, and against which direction the force was resolved. Dimensional analysis is blind to all three, which is exactly why the units cannot tell these apart and you have to.

The "does anything have to move" row is the most useful line in the table. A torque exists on a stationary object. Work does not. Stand on a spanner clamped to a rusted bolt and you are exerting a real torque, measurable, large, and doing exactly zero work, because nothing has moved. Worked example one prices that at $45\ \text{N}\cdot\text{m}$ and $0$ J.

The "can be added to an energy" row is why the joule is withheld. Work goes into an energy equation and adds to kinetic and potential terms. A torque cannot be added to any of them; the sum would be meaningless even though the units match. Reserving the joule for energy is what stops a torque from ever wandering into an energy total by looking like it belongs there.

Why torque is never written in joules

The units genuinely reduce to the same thing. $1\ \text{N}\cdot\text{m}$ and $1\ \text{J}$ are dimensionally identical, and no algebra will separate them. The separation is done by convention, and the convention is universal enough to be treated as a rule.

Write torque in $\text{N}\cdot\text{m}$. Write energy and work in $\text{J}$. Never swap them.

The evidence that this is how the AP courses operate is on the sheet rather than in a written rule. The Table of Information's unit symbols table lists the joule as $\text{J}$, alongside the newton, the watt and the rest, and the symbol keys beside the equations list $W$ as work and $\tau$ as torque as separate entries. No AP source writes a torque in joules anywhere, and the word joule appears in all four CEDs only in that unit symbols table.

The reason the convention is worth keeping is practical rather than cosmetic. Energies are additive: you can put kinetic energy, potential energy and work into one equation and the sum means something. Torque is not a term in any energy equation. If both were labelled $\text{J}$, nothing in a written line would stop a torque being carried into an energy total, and the result would be a number with the right units and no meaning.

A parallel case makes the point without the arithmetic getting in the way. Frequency and angular frequency are both $\text{s}^{-1}$, and they are written $\text{Hz}$ and $\text{rad/s}$ so that nobody adds them. Same trick, same reason.

Practical rule for a free-response answer: if the quantity is going into a conservation-of-energy statement, it is in joules. If it is going into $\sum \tau$ or into $\alpha = \tau/I$, it is in newton metres. Nothing goes into both.

The bridge: a torque becomes work when the thing turns

The two are not sealed off from each other. There is one equation joining them, and it is printed on the sheet.

EK 6.2.A.1 states the condition and it is worth quoting exactly: a torque can transfer energy into or out of an object or rigid system if the torque is exerted over an angular displacement. The word doing the work in that sentence is if. EK 6.2.A.2 then gives the amount: the work done on a rigid system by a torque is related to the magnitude of that torque and the angular displacement through which the rigid system rotates during the interval in which that torque is exerted.

Read $W = \tau\Delta\theta$ as a conversion with a price. A torque is not energy; a torque multiplied by an angle is. Set $\Delta\theta = 0$ and the work is zero however large the torque, which is the stuck bolt again, now as a line of algebra.

The reason no unit conversion appears in that equation is worth a sentence, because it is the piece that makes the whole thing hang together. $\Delta\theta$ must be in radians, and the radian is a ratio of two lengths, an arc length over a radius, so it carries no dimensions at all. The AP sheet prints $s = r\theta$ in its geometry table and $\Delta x_{\text{cm}} = r\Delta\theta$ in the mechanics block, both of which are only true in radians. Multiply $\text{N}\cdot\text{m}$ by a dimensionless number and you get $\text{kg}\cdot\text{m}^2/\text{s}^2$, which you then label $\text{J}$ because it is now energy.

Substitute degrees and there is no unit mismatch to warn you. A torque of $6.72\ \text{N}\cdot\text{m}$ through $115^\circ$ is $13.5$ J. Multiply by $115$ instead of by $2.007$ and you get $773$, a factor of $57.3$ too large, in a line that looks perfectly well formed. This is the arithmetic error most likely to survive a check, because the units come out right either way.

AP Physics C: Mechanics prints the integral form, $W = \int \tau \cdot d\theta$, for the case where the torque varies with angle. EK 6.2.A.3, in both courses, gives the graphical route: the work done on a rigid system by a given torque can be found from the area under the curve of a graph of torque as a function of angular position. Same reading as force against displacement, one axis relabelled.

The case that separates them: the same push, computed twice

For a rotating body, the deepest version of the distinction is not that the two use different components. It is that they use the same component, and multiply it by different lengths.

Here is why that is not a contradiction. Work resolves the force against the displacement. Torque resolves it against the position vector from the axis. For a body turning about a fixed axis, the point of application moves along a circular arc, and that arc is perpendicular to the position vector. So the component of the force perpendicular to $\vec{r}$ is the component parallel to the displacement. They are the same physical part of the force, described relative to two reference directions that happen to sit at right angles to each other.

Work then multiplies it by the arc length $s = r\Delta\theta$, and torque multiplies it by $r$. The ratio of the two answers is $\Delta\theta$, which is $W = \tau\Delta\theta$ read backwards.

Worked example two runs a single $24$ N force pushed at $53^\circ$ to the radius of a $0.35$ m wheel, and gets both numbers from the same resolved component:

• Tangential part: $(24)(0.8) = 19.2\ \text{N}$. Radial part: $(24)(0.6) = 14.4\ \text{N}$.
• Torque: $(0.35)(19.2) = 6.72\ \text{N}\cdot\text{m}$.
• Work over $2.0$ radians: the point of application travels $s = (0.35)(2.0) = 0.70\ \text{m}$, so $W = (19.2)(0.70) = 13.44\ \text{J}$.
• Check: $\tau\Delta\theta = (6.72)(2.0) = 13.44\ \text{J}$. The two routes agree exactly, as they must.

The $14.4$ N radial part does neither. It is parallel to $\vec{r}$, so it contributes no torque by EK 5.3.A.1. It is perpendicular to the displacement, so it does no work by EK 3.2.A.3.i. One component of one force, and two separate zeros for two separate reasons. That is the cleanest demonstration available that the two definitions are genuinely different tests, even when they agree about which part of the force is doing something.

EK 3.2.A.3.ii is the general statement of the second zero: the component of the force exerted on a system perpendicular to the direction of the displacement of the system's centre of mass can change the direction of the system's motion without changing the system's kinetic energy.

Where the confusion costs a mark

• Writing a torque in joules. The units match; the quantities do not. Torque goes in $\text{N}\cdot\text{m}$ and stays there.
• Putting a torque into an energy equation. $\frac{1}{2}I\omega^2 = \tau$ is dimensionally wrong by a factor of an angle and physically meaningless. The torque has to be multiplied by $\Delta\theta$ first.
• Using degrees in $W = \tau\Delta\theta$. Off by $180/\pi$, that is $57.3$, with no unit clue that anything went wrong. Radians, always.
• Claiming a torque does work when nothing has turned. EK 6.2.A.1 requires the torque to be exerted over an angular displacement. Holding a spanner against a seized bolt is real effort, and it is zero work in the physics sense.
• Using $\sin$ where the work equation wants $\cos$, or the reverse. The two equations resolve against different reference directions. In $W = Fd\cos\theta$, $\theta$ is between the force and the displacement. In $\tau = rF\sin\theta$, $\theta$ is between the force and the position vector from the axis. Getting them the wrong way round is a version of this confusion that survives checking, because both give plausible-looking numbers.
• Quoting a work with an axis, or a torque without one. Work needs no axis. Torque is meaningless without one, as torque vs force sets out.
• Assuming a big torque means a lot of energy delivered. It says nothing about energy until an angle is supplied. Worked example three needs four full revolutions before $32\ \text{N}\cdot\text{m}$ amounts to $804$ J.
• Forgetting that torque can do negative work. A bearing dragging on an axle exerts a torque opposing the rotation, so $\tau$ and $\Delta\theta$ carry opposite signs and energy leaves the system. Work is a signed scalar by EK 3.2.A.2, and the rotational version is signed the same way.
• Treating $W = \tau\Delta\theta$ as an extra equation rather than the same one. It is $W = F_{\parallel}d$ with $F_{\parallel}$ traded for $\tau/r$ and $d$ traded for $r\Delta\theta$. The $r$ cancels.

The procedure for the torque side is in how to calculate torque, the energy accounting in the work-energy theorem, and neither is repeated here.

When the numbers coincide, and why that is a trap

There are three ways these two land on the same figure, and each of them teaches a wrong lesson if it is the only case you have met.

A torque turned through exactly one radian. $\Delta\theta = 1$ makes $W = \tau$ numerically. A torque of $8.0\ \text{N}\cdot\text{m}$ through one radian does $8.0$ J of work. Nothing has become energy that was not; a dimensionless factor of $1$ has been applied, and the label changed because the quantity did.

A one-newton force and a one-metre length. Push $1$ N through $1$ m and you do $1$ J of work. Apply $1$ N at $1$ m from an axis and you exert $1\ \text{N}\cdot\text{m}$ of torque. Same two numbers, same multiplication, two quantities that cannot be added to each other.

Any problem where the arithmetic is done before the labelling. This is the practical one. Both calculations are a multiplication of a force by a length, so the working looks the same right up to the last line, and the last line is where the physics is.

The distinction bites the moment a question does any of these:

• Asks what happens with no rotation. Torque survives, work goes to zero.
• Asks for a change in kinetic energy. Only work can supply it, via the work-energy theorem at EK 3.2.A.4 and its rotational counterpart.
• Asks for angular acceleration. Only torque can supply it, via $\alpha_{\text{sys}} = \frac{\sum \tau}{I_{\text{sys}}}$ at EK 5.6.A.2.
• Changes the axis. The torque changes, and the work done by that force does not, since work does not refer to an axis at all.

That last one is worth sitting with. Move the axis in a problem and every torque in it takes a new value, while every work stays exactly as it was. Two quantities that responded identically to a change of axis could plausibly be the same thing. These do not.

What the CED asks, and how the exam frames it

The two sit in different units of AP Physics 1, and one of them has a topic named after the pairing.

Work: Unit 3, Work, Energy, and Power, weighted at 18 to 23 percent of the multiple-choice section. Topic 3.2, Work, has one learning objective, LO 3.2.A: describe the work done on an object or system by a given force or collection of forces. EK 3.2.A.1 defines work as the amount of energy transferred into or out of a system by a force exerted on that system over a distance, with five sub-points covering path independence for conservative forces, path dependence for nonconservative ones and the fact that potential energies are associated only with conservative forces. EK 3.2.A.2 makes work a scalar that may be positive, negative or zero. EK 3.2.A.3 and its two sub-points give the component rule and the equation $W = F_{\parallel}d = Fd\cos\theta$. EK 3.2.A.4 gives the work-energy theorem and EK 3.2.A.5 the graphical area. Suggested skills for Topic 3.2 are 1.B, 2.B, 2.D, 3.A and 3.B.

Torque: Unit 5, Torque and Rotational Dynamics, weighted at 10 to 15 percent. Topic 5.3 carries LO 5.3.A and 5.3.B, EK 5.3.A.1 on the perpendicular component, EK 5.3.A.2 on the lever arm, and EK 5.3.B.2 with $\tau = rF_{\perp} = rF\sin\theta$. Its boundary statement puts the direction of torque beyond the scope of AP Physics 1.

The bridge: Unit 6, Energy and Momentum of Rotating Systems, weighted at 5 to 8 percent of the multiple-choice section over about 8 to 14 class periods. Topic 6.2 is titled Torque and Work, which is as clear a signal as the CED gives that the pairing is examinable. LO 6.2.A asks you to describe the work done on a rigid system by a given torque or collection of torques, and carries EK 6.2.A.1 on the angular-displacement condition, EK 6.2.A.2 with $W = \tau\Delta\theta$, and EK 6.2.A.3 on the area under a torque against angular position graph. Suggested skills for Topic 6.2 are 1.B, 2.A, 2.C, 2.D and 3.A. The topic carries no boundary statement.

On the equation sheet, the AP Physics 1 and AP Physics 2 Tables of Information print $W = F_{\parallel}d = Fd\cos\theta$ in the translational column and both $\tau = r_{\perp}F = rF\sin\theta$ and $W = \tau\Delta\theta$ in the rotational column. Both AP Physics C sheets replace the two work lines with integrals, $W = \int_a^b \vec{F} \cdot d\vec{r}$ and $W = \int \tau \cdot d\theta$, and replace the torque line with the cross product $\vec{\tau} = \vec{r} \times \vec{F}$. The symbol keys keep $W$ as work and $\tau$ as torque throughout, in every course.

The CED framing is at Topic 3.2, Topic 5.3 and Topic 6.2, with the calculus treatment at C: Mechanics Topic 6.2. Problems are in the torque and rotational motion set and the work, energy and power set.