This simulation needs a browser with HTML canvas support.
↗ Electric field
⊙ ⊗ Magnetic field
Tap or click to place a charge.
t = 0.00
5 units
100%

The physics behind the field

This is a two-dimensional slice through the three-dimensional vacuum fields of one positive point charge. You can drag the charge in the plane or choose automatic motion along the vertical axis. This is not a two-dimensional version of Coulomb’s law, and it does not integrate the charge’s response to its own radiation.

1. What does a distant point know?

The fields use the charge’s position, velocity and acceleration at the retarded time, the earlier time at which its influence left the source. Every sample solves the implicit equation below. This is why a new pulse does not reach the whole screen at once.

tr + R(tr)/c = t
R(tr) = |x − rq(tr)|
n = [x − rq(tr)] / R(tr)

Here x is the observation point, rq is the source position, R is the retarded distance, c is the speed of light, and n is the unit direction from the retarded source to the observer. The numerical solver uses safeguarded Newton iterations with a bracket guaranteed by the source speed bound. If D is the distance from the current source and V is its maximum speed, the triangle inequality gives D/(c + V) ≤ t − tr ≤ D/(c − V). This remains valid for a freely moved charge. Its target equation residual is 10−10 normalized time units.

2. Two contributions, one electric field

The simulation evaluates the Liénard–Wiechert field formula, as given in Michael Fowler’s University of Virginia lecture, “The Lienard-Weichert Potentials and Fields.” It is used because it separates the velocity contribution from the acceleration contribution without a slow-motion approximation.

E = Ev + Er
Ev = K (1 − |β|²)(n − β) / [κ³ R²]
Er = K n × [(n − β) × a] / [c² κ³ R]
B = n × E / c
K
q/(4πε₀), where q is charge and ε₀ is vacuum permittivity.
β
v/c, the source velocity as a fraction of the speed of light.
κ
1 − n · β, the retarded directional factor.
a
The source acceleration, the derivative of its velocity.
Subscripts v, r
Velocity field and radiation field. Every source quantity in these expressions is evaluated at tr.

In this plane, E has x and y components. The cross product gives Bz = (nxEy − nyEx)/c. A positive value points out of the screen (⊙), and a negative value points into it (⊗). This relationship is applied separately to each selectable contribution.

Two important distinctions. The velocity term is Coulomb-like, but is not generally radial from the retarded source. Its direction is n − β. The acceleration term is already transverse at every distance, not only in the far field.

3. Why does the far field look transverse?

The cross product of a vector with n is perpendicular to n. With u = (n − β) × a, the formula therefore gives

n · Er = K [n · (n × u)] / [c² κ³ R] = 0.

For fixed emission conditions and direction, the velocity term decreases as 1/R² while radiation decreases as 1/R. Their magnitude ratio thus grows in proportion to R whenever the radiation coefficient is nonzero. The total field approaches a transverse field as the velocity term becomes negligible. It is not necessarily exactly transverse at any finite distance.

There are exceptions to radiation dominance. For the automatic vertical motion, the radiation field is zero on the vertical line through the motion center, and it is also zero at phases with zero acceleration. An angle is undefined for a zero field. The snapshot records an undefined angle there.

4. Motion, units and honest visualization

Continuous wave. Around your chosen center (x₀, y₀), the source follows x(t) = x₀ and y(t) = y₀ + A sin(2πft), where A is amplitude and f is frequency. Its derivatives are vy(t) = 2πfA cos(2πft) and ay(t) = −(2πf)²A sin(2πft). This steady oscillation exists for all earlier times, including negative times. Changing source parameters restarts this steady-state experiment rather than simulating a causal change of the drive.

Single pulse. The charge is stationary before t = 0 and after t = T, where T = 1/f. During the pulse, set u = πt/T and use x(t) = x₀ and y(t) = y₀ + A sin⁴u. Differentiation gives vy = 4A(π/T)sin³u cos u and ay = 4A(π/T)²(3sin²u cos²u − sin⁴u). Position, velocity and acceleration join continuously to the stationary portions. “Emit another pulse” restarts this experiment.

Normalized units set c = 1 and q/(4πε₀) = 1. Distances and times are not labeled as meters or seconds. Source speed is limited to 0.65c. Automatic parameters are clamped when necessary. Mouse motion follows a speed-limited target controller. A small disk of radius 0.14 units around the current charge is excluded because the point-charge field is singular there. A larger visual source marker can cover nearby samples at distant zoom levels.

Vector scale. Directions are unmodified. Let s = ln(1 + g|E|), where g is 250 times the visibility setting. The arrow length is L = Lmaxs/(1 + s). This chosen display mapping is monotonic but is not proportional to physical field strength. Magnetic symbols similarly encode c|Bz|. Snapshot readings use the unscaled numerical field.

Electric arrows and magnetic glyphs use interleaved grids, each evaluated at its own actual sample position. Propagation guides are circles of constant emission time centered on the source position at that time. They are not electric field lines or a claim of equal radiation in every direction. Dashed gray rays join the retarded source position to the probe. The probe’s angle uses the selected electric component and this retarded direction, not the direction from the current charge.

Mouse motion. There is no pre-placed charge. Your first click chooses a stationary initial condition: the charge is assumed to have existed at that location for all earlier times. This is not a charge suddenly created in empty space. Restarting in Mouse mode establishes a new stationary initial condition and clears the old history.

The cursor specifies a target, not an instantaneous teleport. The source follows it with bounded speed and acceleration. Very fast drags show an outlined target while the real charge catches up. Mouse mode defaults to 24 normalized time units per real second at 1× playback. The Mouse time rate control changes that display clock, not the speed of light in the field equations. Dragging a paused charge explicitly resumes playback.

The recorded path uses steps no longer than 0.05 normalized time units. Let Δ = tk+1 − tk and u = t − tk within one step. The implementation chooses bounded endpoint velocities and defines

ak = (vk+1 − vk) / Δ
r(t) = rk + vku + ½aku²
dr/dt = vk + aku = v(t)
dv/dt = ak

These expressions follow by integrating constant acceleration over a step. Position and velocity remain continuous. Acceleration is piecewise constant and may jump at step boundaries. Because velocity interpolates between two vectors whose magnitudes do not exceed 0.65c, the triangle inequality keeps every intermediate velocity below the same bound. The full stored trajectory is used for retarded fields, including after you release the mouse. Only decorative guide circles are limited in number.

Switching from an automatic trajectory to Mouse control preserves its earlier analytic history. Switching to an automatic motion starts a new experiment centered at the current charge position. The charge is prescribed externally, with no self-force, radiation reaction, or response to the noninteracting probe.

Arrow bases are spatial sample points. They do not represent particles moving through space.

Keyboard and pointer controls

Charge / probe / pan toolsQ / P / G
Play or pauseSPACE
One time step.
Total / velocity / radiation1 / 2 / 3
Restart timeR
Zoom in / out+ / −
Recenter overviewH
Move focused canvas probeARROWS
Close these notesESC

Click or tap to place your first charge. Drag the charge or the probe directly. The Charge tool can also set a new target anywhere. The Probe tool places or drags the probe anywhere. Use the Pan tool, Shift-drag, or the middle mouse button to pan. Scroll or pinch to zoom. Q selects Charge, P selects Probe, and G selects Pan. Arrow keys move the probe when the canvas is focused. Hold Shift for a larger keyboard step. Use Reset view or H to recenter the camera on the charge. A snapshot exports the current canvas with its field mode, time and numerical probe readings. The page starts paused with no charge. Placing or dragging the charge is an explicit action that starts playback.

Formula references

Michael Fowler, University of Virginia. The Lienard-Weichert Potentials and Fields.
Electric-field decomposition, retarded geometry, and the transverse radiation field.

The Feynman Lectures on Physics, Volume II, Chapter 21. Solutions of Maxwell’s Equations with Currents and Charges.
Retarded potentials and fields of a moving point charge.

The analytic trajectory, visual mapping and numerical solver are implemented locally. There are no libraries, fonts, trackers or remote assets required to run this file.