Work–Energy Lab · Guide · glidepathscience.org
GuideWork–Energy Lab
Contents
This Guide explains the ideas and equations behind every Walkthrough step, Concept Challenge question and Predict round, in the order a textbook would take them. Part 1 prepares the Intro track and Part 2 the Advanced track; each takes about fifteen minutes to read. Each section ends with a worked example that is not one of the sim's own problems.
Part 1 · The essentials
Work: a force along a displacement
Forces are how energy moves from one object to another. A push that speeds up a cart gives it energy; a brake that slows it takes energy away. The measure of that transfer is work, and learning to calculate it, sign and all, is the first step toward the energy bookkeeping that runs through the whole sim.
Push a cart with a steady force F in the direction it moves, through a distance d, and the work you do on it is the force times the distance. When the force is tilted at an angle θ to the motion, only its component along the displacement, $F\cos\theta$, does work:
$W = Fd$ $W = Fd\cos\theta$
Here F is the size of the force in newtons, d the distance moved in meters, and θ the angle between the force and the displacement. The unit, the newton-meter, is the joule (J). Work is a scalar with a sign, and the cosine sets it. A force along the motion does positive work and adds energy; a force against the motion (θ = 180°) does negative work and takes energy away; a force at right angles to the motion does no work at all, however large it is. The normal force from a level track is one; the upward part of a tilted pull is another.
The usual mistake is to multiply the whole force by the distance. The upward part of the pull does something, since it lifts some of the sled's weight off the snow, but it moves nothing along the snow and adds no energy to the motion. A second mistake is to think a strong force must do positive work: a hand pushing back on a cart rolling forward does negative work, however hard it pushes.
Positive work adds energy and negative work removes it. The next question is where that energy shows up.
Work changes kinetic energy
The first place work goes is into motion. The kinetic energy of an object of mass m moving at speed v is
$K = \frac{1}{2}mv^2$
It is never negative, and it grows with the square of the speed: twice as fast means four times the kinetic energy, and four times the work to get there. The work–energy theorem says that the net work done by all the forces on an object equals its change in kinetic energy:
$W_{net} = \Delta K = \frac{1}{2}mv^2 - \frac{1}{2}mv_0^2$
where $v_0$ is the starting speed and v the final one. For a steady force along a line it follows from constant acceleration: with $a = F/m$, $v^2 = v_0^2 + 2ad$, and multiplying by $\frac{1}{2}m$ gives $\frac{1}{2}mv^2 = \frac{1}{2}mv_0^2 + Fd$. Part 2 shows that it holds for any force and any path.
The word net matters: if friction, gravity or a second push also act, their work counts too. On the sim's level, frictionless track the robot's work is the net work; on a ramp or a rough floor it is not. The other trap is arithmetic. Solving $\frac{1}{2}mv^2 = W$ gives $v = \sqrt{2W/m}$. Dropping the ½ makes the speed too small by a factor of √2, and forgetting the square root leaves $v^2$, which is not a speed.
A steady force makes the work easy: force times distance. Most forces change as they act, and then we need a picture.
A force that changes: work as an area
Real forces rarely stay constant: a spring pushes harder the more it is squeezed, and a hand's push grows and fades. To find the work, plot the force along the motion against position. For a steady force this F(x) graph is a flat line, and $Fd$ is the rectangle beneath it. For any force the rule is the same:
$W = \text{area under}\;F(x)$
Area above the axis, where the force points along the motion, is positive work; area below the axis is negative work. Break awkward shapes into rectangles and triangles: a force that grows steadily from zero to $F_{max}$ over a distance d encloses $\frac{1}{2}F_{max}d$, half of its rectangle.
Because net work is the change in kinetic energy, the running area tells you the kinetic energy at every point. A cart that starts from rest stops only where the running area returns to zero. That answers a question many students get wrong: a cart does not stop when the push ends, or when the push turns into a pull. The pull only starts to slow it, and the cart coasts on until the negative area has used up every joule the positive area put in. If the pull is still on, it then pulls the cart back.
One force that changes along the way deserves a section of its own: the spring.
A spring stores the triangle
Springs store energy and give it back, in launchers, toys, car suspensions and running shoes. An ideal spring pushes back in proportion to its compression, $F = kx$, where x is the compression in meters and k the spring constant in N/m. (Why springs behave this way is the subject of Springs & Elastic Potential Energy.) As you compress it, the force grows from zero to $kx$, so the work you do is a triangle's area, $\frac{1}{2}(x)(kx)$, stored as elastic potential energy:
$U_s = \frac{1}{2}kx^2$
Released against a block on a level, frictionless track, the spring hands all of it back: when the block leaves the spring, $\frac{1}{2}kx^2 = \frac{1}{2}mv^2$, so it launches at $v = x\sqrt{k/m}$. Twice the compression stores four times the energy and launches twice as fast. Two slips are common. Writing $\frac{1}{2}kx$ gives half the final force, not an energy (the unit is N, not J). And with k in N/m the compression must be in meters: a 4.0 cm squeeze is 0.040 m, or the energy comes out ten thousand times too large.
The spring returns all the energy it stores. Gravity does the same, with an even simpler rule.
Gravity's work depends only on height
Gravity does the work on every roller coaster, slide and ski run. Near the Earth it is a steady force, mg, straight down, with g = 9.81 m/s². Along a slope only its component along the motion does work, and that component times the distance moved always comes out to the weight times the change in height. So every path between the same two heights gives the same result:
$W_{grav} = -mg\,\Delta h$
where Δh is the rise. Going up, gravity's work is negative; coming down, positive. On a steep ramp gravity pulls hard along a short path, on a gentle one weakly along a long path, and the product is the same. That is what lets us give gravity a stored energy, the gravitational potential energy $U_g = mgh$, measured from any height we choose as zero. Released from rest at a height h on a frictionless track, a cart reaches the bottom with $\frac{1}{2}mv^2 = mgh$:
$v = \sqrt{2gh}$
Neither the mass nor the slope appears. Carts released from the same height down steep and gentle frictionless ramps arrive at the same speed; the steep one gets there first, but it is not faster. Expecting more speed from the steeper ramp is the misconception here, along with losing the 2 under the root.
Springs and gravity give back what they store. Friction is different: what it takes, it keeps.
Friction turns kinetic energy into thermal energy
Brakes, skids and sliding boxes all stop because of friction, and the surfaces get warmer. Kinetic friction acts against sliding with a force $f = \mu_k N$, where N is the normal force pressing the surfaces together (mg on a level floor) and $\mu_k$ the coefficient of kinetic friction for the two materials. Over a slide of length d, friction turns kinetic energy into thermal energy in the block and the floor together:
$\Delta E_{th} = \mu_k m g\,d$
A block that slides to a stop turns all its kinetic energy into thermal energy, $\frac{1}{2}mv^2 = \mu_k mg\,d$, so it stops in $d = v^2/(2\mu_k g)$. The mass cancels, since a heavier block has more kinetic energy but feels proportionally more friction. The speed enters squared: twice the speed, four times the distance. Friction does not destroy energy, as students often say. The energy is still there, as the random motion of molecules in two warm surfaces, but nothing will turn it back into motion. The other mistake is the wrong coefficient. The static $\mu_s$ says how hard you must push to start something sliding; once it slides, only the smaller $\mu_k$ applies.
With kinetic, elastic, gravitational and thermal energy all in play, we need a way to keep the books.
Energy bar charts and the system
An energy bar chart keeps track of every joule, like a bank statement. First choose a system: the objects inside a boundary you imagine around them. That choice decides what each interaction looks like. A force from an object outside does work across the boundary, a W bar, positive or negative. A conservative interaction between two objects inside (gravity with the Earth inside, a spring that is inside) is a potential-energy bar, $U_g$ or $U_s$. Friction between two surfaces that are both inside is a thermal bar, $\Delta E_{th}$. Then the stores before, plus the work across the boundary, equal the stores after:
$K_i + U_i + W = K_f + U_f + \Delta E_{th}$
The classic error is to count an interaction twice, with a $U_g$ bar and a work-by-gravity bar in one chart. With the Earth inside, gravity is $U_g$ and nothing else; with the Earth outside, gravity does work and there is no $U_g$. Friction works the same way: floor inside, $\Delta E_{th}$; floor outside, negative work on the block. That bar is friction’s center-of-mass term, $-\mu_k N d$: exact for the block’s kinetic energy, though part of the heat stays in the block (Part 2, friction’s fine print). Different systems draw different charts of the same event, and each one balances.
The loop-the-loop puts kinetic and gravitational energy together with a condition from circular motion, and shows that enough energy is not always enough.
The loop-the-loop
To get around a loop of radius R, a frictionless cart must reach the top, 2R above the bottom. But reaching that height is not enough. At the top the track is above the cart and can only push, never pull, so the cart must be moving fast enough that turning it around the circle takes at least its whole weight. That means a speed of at least $\sqrt{gR}$ at the top (Circular Motion & Banked Turns shows why):
$v_{top}^2 \ge gR$
Energy then gives the lowest release height. Released from rest at h, the cart arrives at the top with $mgh = mg(2R) + \frac{1}{2}mv_{top}^2$, and with the smallest allowed speed, $v_{top}^2 = gR$, the masses cancel:
$h_{min} = 2.5R$
The extra half radius pays for the speed at the top. Released from 2R, the cart has the energy to reach the top's height but would arrive with no speed, so it never gets there: the normal force falls to zero at a height of 5R/3, and the cart leaves the track and falls inside the loop. At the bottom of a loop entered from $h_{min}$, the track pushes up with six times the cart's weight, though that normal force does no work.
So far we have asked how much energy moves. The last idea of Part 1 asks how fast.
Power: the rate of doing work
Lifting a load slowly or quickly takes the same work, but not the same engine. The difference is power, the rate of doing work, in watts (W): joules per second. A horsepower is about 746 W. Over a time Δt the average power is the work divided by the time, and at any instant it is the force along the motion times the speed, since in a short time the force does $F\,\Delta x$ of work and $\Delta x/\Delta t$ is v. They agree only at steady power: a 1.5 kg cart pushed from rest by 3.0 N for 2.0 s averages 6.0 W, but ends at 4.0 m/s receiving $Fv$ = 12 W.
$P_{avg} = W/\Delta t \qquad P = F v$
A winch raising a load at a steady speed pulls with the load's weight, so its power is $mgv$; a person climbing stairs does the same for their own weight. Students most often confuse the two quantities: work is the total, in joules, and power the rate, in watts. The same job in half the time takes the same work at twice the power. And the force in $Fv$ is a force, mg, not the mass.
Part 2 returns to these ideas with calculus: work as an integral, potential energy from force, and forces that do no work at all.
Part 2 · Going further
Work as an integral, and why W = ΔK
The area under F(x) is easy when its edges are straight. For a smooth curve, cut the motion into short pieces Δx, short enough that the force barely changes across each, and add up $F\,\Delta x$. As the pieces shrink the sum becomes an integral:
$W = \int F\,dx$
In two or three dimensions only the part of the force along each step counts, $W = \int \vec F\cdot d\vec r$: the idea of $Fd\cos\theta$ applied piece by piece along a curved path. The work–energy theorem now follows for any force. Newton's second law says $F = m\,dv/dt$, and in a time dt the body moves $dx = v\,dt$, so $F\,dx = mv\,dv$ and
$\int F\,dx = \int m v\,dv = \Delta K$
Look at what this needed. F must be the net force, because Newton's law is about the net force, and the body must move as a single particle, every part sharing the same dx. It did not need a constant force, a straight path or the absence of friction. The particle condition is the one that fails for a skater and a sliding block, as later sections show. For a changing force, the tempting shortcut of the starting force times the distance overestimates the work whenever the force falls off.
The integral gives a force's work along a path. Some forces, remarkably, do not care which path.
Conservative forces and potential energy
Gravity's work depends only on the starting and ending heights. Such a force is conservative, and the test is that its work around every closed path, one ending where it began, is zero:
$∮ \vec{F}\cdot d\vec{r} = 0$
Gravity and ideal springs pass. Friction fails, because it opposes the sliding on every leg of a trip. Around a closed path on a level table it does
$W = -\mu_k m g \times \text{length}$
which is never zero: ending back at the start does not cancel friction's work, the most common misconception about closed paths. For a conservative force we can define a potential energy U, depending only on position, whose decrease equals the force's work. In one dimension:
$F = -\frac{dU}{dx}, \quad U(x) = -\int F\,dx$
The force is minus the slope of U, not U itself. On a track $U = mgh(s)$, so a graph of U against the distance along the track s is a picture of the track's height along it, and the force along it is $F_s = -dU/ds = -mg\sin\alpha$, with α the local slope. The force is largest where the track is steepest and zero at every crest and valley, even though U is largest at a crest. Confusing a large U with a large force is the mistake this picture prevents.
A graph of U against position does more than give the force. It shows where an object can go.
Energy diagrams and turning points
When only conservative forces do work, the total mechanical energy $E = K + U$ stays constant. Draw it as a level line over the graph of U(x). At every position the gap between the line and the curve is the kinetic energy, and since K cannot be negative, the object can only be where the curve lies at or below the line:
$E = U(x), \quad K = E - U \ge 0$
Where the line meets the curve, K is zero and the object stops for an instant and turns back: a turning point. A cart released from rest at h on a frictionless track therefore clears any crest lower than h, as long as it stays on the track, and turns back on the slope of any crest higher than h. A starting speed adds $v_0^2/2g$ to the height it can reach. Two misconceptions show up here: that the object must sit where U is lowest (any point under the line is possible), and forgetting the starting kinetic energy.
In one dimension, every force that depends only on position has a potential energy. In two dimensions that is no longer true.
Position-dependent, yet not conservative
It is tempting to think that any force depending only on position must be conservative. In one dimension it is: the only way from a to b along a line is back and forth along it, so the work is $\int F\,dx$ whatever the route, and $U(x) = -\int F\,dx$ always exists. In a plane the route matters. The swirl $\vec F = c(-y, x)$ has size $cr$ at a distance r from the origin and always points around it counterclockwise. Carried once counterclockwise around a circle of radius R, an object feels a force cR along its motion the whole way and gains $cR \times 2\pi R$. Around any closed loop the swirl does work proportional to the enclosed area:
$∮ \vec{F}\cdot d\vec{r} = 2c \times \text{area}$
positive counterclockwise and negative clockwise. The deciding test is the curl, $\partial F_y/\partial x - \partial F_x/\partial y$ in two dimensions, which for the swirl is $c - (-c) = 2c$ everywhere. A force is conservative when its curl is zero everywhere (in a region without holes). The swirl's curl is the same everywhere, so the work around any loop is the curl times the area it encloses, wherever the loop sits.
Conservative or not, every force must land somewhere in the energy books. The general rule is the energy equation for a system.
The energy equation for a system, and friction's fine print
The bar-chart rules of Part 1 are one equation. For any system, the work done by forces from outside equals the changes inside:
$W_{ext} = \Delta K + \Delta U + \Delta E_{th}$
ΔU includes every potential energy whose two partners are inside, and $\Delta E_{th}$ the thermal energy of surfaces that are both inside. A source inside, such as a battery, fuel or muscles, adds the change in its own store. A conservative force between objects inside never appears as work, only as −ΔU; move one partner outside and the same interaction becomes work across the boundary.
Friction has fine print. For a sliding block alone, Newton's law for its center of mass gives $F_{net}\,\Delta x_{cm} = \Delta K$, the center-of-mass equation (Sherwood's "pseudowork"). It always gets the kinetic energy right, so $-\mu_k N d = \Delta K$ correctly gives a sliding block's change in speed. But it is not a statement about energy transfer. The thermal energy $\mu_k N d$ appears at the rubbing surfaces and is shared between block and floor in a split that depends on the microscopic contact, so the real work friction does on the block is not $-\mu_k N d$. Put block and floor in one system and the books are exact.
The center-of-mass equation matters most for forces that push on an object without doing any work on it at all.
Forces that do no work, and energy from inside
A skater who pushes off a wall speeds up, but not because the wall did work. Work needs the point where a force acts to move, and the wall pushes on the skater's hands, which stay against it while it pushes. The same holds for the road on a car's drive tires, whose contact patches are momentarily at rest, and the floor on a jumper's feet. These forces still change the motion, and for the center of mass
$F_{net}\,\Delta x_{cm} = \Delta K$
gives the right change in kinetic energy. The energy itself comes from inside: the skater's or jumper's chemical energy, released by muscles, or a car's fuel, through its engine. A real body is far from perfectly efficient, so it spends more chemical energy than it gains and warms up. Two misconceptions meet here: that the wall must have done the work, and its mirror image, that with no work done ΔK must be zero.
A car's engine converts fuel energy at some rate. Holding that rate fixed gives a motion that is not uniformly accelerated at all.
Driving at constant power
An engine delivering a steady power P adds kinetic energy at a steady rate, so over a time t, with no drag on a level road,
$Pt = \Delta K$
and $v = \sqrt{v_0^2 + 2Pt/m}$: from rest, the speed grows only as √t. The time to go from $v_0$ to v is
$t = \frac{m(v^2 - v_0^2)}{2P}$
Because $P = Fv$, the driving force is $P/v$, large at low speed and falling as the car speeds up, which is why treating the motion with a constant force or acceleration fails here, the usual mistake. At very low speed $P/v$ has no limit, and no car can deliver that: the tires slip first. With all its weight on driven wheels, a car can be pushed by at most $\mu_s mg$, so below $v^* = P/(\mu_s mg)$ the tires, not the engine, set the limit.
Real cars never reach the speeds this formula predicts, because the air pushes back harder and harder.
Air drag and the limits of the model
At everyday speeds air resists a car or a cyclist with a drag force growing as the square of the speed, $F_{drag} = \frac{1}{2}\rho C_d A v^2$, where ρ is the air's density, about 1.2 kg/m³, and the product of the drag coefficient $C_d$ and the frontal area A is the drag area. The power to push through the air is the force times the speed, one more factor of v:
$P = \frac{1}{2}\rho C_d A v^3$
The cube makes speed expensive: twice the speed takes eight times the power. A vehicle at constant power speeds up until the air takes all of it, at its top speed
$v_{max} = \left(\frac{2P}{\rho C_d A}\right)^{1/3}$
and doubling the power raises the top speed only by $2^{1/3}$, about 26 %. Forgetting the extra factor of v, as if the power grew like the force, is the common error.
The sim's particle model leaves things out, and a good modeler knows which: the wheels' rotational energy (rolling belongs to Inclined Planes & Friction), rolling resistance, air drag outside the Lift scene, friction on curved track, how thermal energy splits between surfaces, the spring's own mass, and every frame but the ground's (Frame Shift shows how work depends on who is watching). The Field Manual lists the rest.
Reference
Common mistakes
- Using the whole force. Only the part along the motion does work: $Fd\cos\theta$.
- Losing the ½ or the square root. $\frac{1}{2}mv^2 = W$ gives $v = \sqrt{2W/m}$.
- Stopping when the push stops. A cart stops where the running area under F(x) returns to zero.
- Writing ½kx, or x in centimeters. $U_s = \frac{1}{2}kx^2$, with x in meters.
- Expecting a steeper ramp to give more speed. Gravity's work depends only on height: $v = \sqrt{2gh}$.
- Saying friction destroys energy, or using $\mu_s$ for a slide. It becomes thermal energy, through $\mu_k$.
- Counting an interaction twice. Earth inside: $U_g$; Earth outside: work by gravity. Never both.
- Releasing from 2R. The loop needs speed at the top: $h_{min} = 2.5R$.
- Mixing up work and power, or a force with a mass: $P_{avg} = W/\Delta t$ and $P = Fv$ at an instant, the force in newtons. The two are equal only at steady power.
- Assuming a position-only force is conservative, or that a closed path cancels friction.
- Crediting the wall. A force whose point of contact does not move does no work; the energy comes from inside.
Equation sheet
| Part 1 | |
|---|---|
| work, steady force | $W = F d\cos\theta$ |
| work, any force | $W = \text{area under}\;F(x)$ (+ above the axis, − below) |
| work–energy theorem | $W_{net} = \Delta K = \frac{1}{2}mv^2 - \frac{1}{2}mv_0^2$ |
| spring | $U_s = \frac{1}{2}kx^2$ |
| gravity, any path | $W_{grav} = -mg\,\Delta h$; from rest, $v = \sqrt{2gh}$ |
| friction, level floor | $\Delta E_{th} = \mu_k m g\,d$ |
| loop | $v_{top}^2 \ge gR$; $h_{min} = 2.5R$ |
| power | $P_{avg} = W/\Delta t, \quad P = F v$ |
| Part 2 | |
|---|---|
| work as an integral | $W = \int F\,dx$ |
| potential energy | $F = -\frac{dU}{dx}, \quad U(x) = -\int F\,dx$ |
| turning points | $E = U(x), \quad K = E - U \ge 0$ |
| conservative; friction (level) | $∮ \vec{F}\cdot d\vec{r} = 0$; $W = -\mu_k m g \times \text{length}$ |
| swirl $F = c(-y, x)$ | $∮ \vec{F}\cdot d\vec{r} = 2c \times \text{area}$ (counterclockwise +) |
| a system | $W_{ext} = \Delta K + \Delta U + \Delta E_{th}$ |
| center of mass | $F_{net}\,\Delta x_{cm} = \Delta K$ |
| constant power | $Pt = \Delta K$; $t = \frac{m(v^2 - v_0^2)}{2P}$ |
| drag | $P = \frac{1}{2}\rho C_d A v^3$; $v_{max} = \left(\frac{2P}{\rho C_d A}\right)^{1/3}$ |
Symbols: W work (J) · F force (N) · d, x distance or position (m) · θ angle between force and motion · m mass (kg) · v, $v_0$ final and starting speeds (m/s) · K kinetic energy · k spring constant (N/m) · $U_s$, $U_g$ elastic and gravitational potential energy · g = 9.81 m/s² · h, Δh height and rise · $\mu_k$, $\mu_s$ kinetic and static coefficients · N normal force · $\Delta E_{th}$ thermal energy · R loop radius · P power (W) · E mechanical energy · c swirl strength (N/m) · $\Delta x_{cm}$ center-of-mass displacement · ρ air density (kg/m³) · $C_d A$ drag area (m²).
Further reading
OpenStax, University Physics Volume 1 (CC BY-NC-SA 4.0), chapters 7 (work and kinetic energy) and 8 (potential energy and conservation of energy). A. Van Heuvelen and X. Zou, "Multiple representations of work–energy processes," Am. J. Phys. 69, 184 (2001), for energy bar charts; B. A. Sherwood, "Pseudowork and real work," Am. J. Phys. 51, 597 (1983), for the center-of-mass equation. The Field Manual lists every source and the model's limits.
Work–Energy Lab v1.5 · © 2026 Dr. David Meier · CC BY-NC-SA 4.0 · glidepathscience.org