Quantum Eraser: Visualisation · Signals from the Periphery

← Intro

Configuration

Open

Which-path marker

off

Eraser φ

—

Screen

INTERFERENCE

QUANTUM ERASER · THE HIDDEN FRINGE
one photon · two slits · a tag you can erase
mark the path · the fringes die
erase the tag · the screen stays flat
sort the ± channels · the fringes return
no signal goes back · only a correlation

Ledger

exact ⇄ staged · live

Marking the path flattens the fringes, unread

distinguishability, not observation, is what does it

ESTABLISHED

P±(y) ∝ env·[1 ± sin2φ·cos δ]

the fringe and the anti-fringe, exactly out of step

ESTABLISHED

Combined screen stays flat once a path was tagged

P₊ + P₋ = 2·env · erasing shows nothing on its own

ESTABLISHED

V = |sin 2φ| · complementarity V² + D² ≤ 1

Englert 1996 · the trade the analyzer rides

ESTABLISHED

Delayed choice changes nothing in the raw screen

Kim 2000 · Ma–Zeilinger 2013 · fringes live in the coincidence

ESTABLISHED

The ± channels stand in for coincidence counting

a lab eraser uses entangled signal/idler pairs

STYLISED

One photon source · a single clean angle φ

beam speed and slit sizes drawn for watchability

STYLISED

Does erasing rewrite the past? It does not.

no information travels back · the sorting reveals a correlation

READING

View

Configuration · what's on the bench?

ESTABLISHED · both paths open, self-interference

▶ Fringes only ever appear in the sorted ± subset. The combined screen is flat the moment a path is tagged.

Build it yourself

Which-path marker (H/V tags)

45° eraser analyzer

Analyzer angle φ45°

0° reads the path (no erasing, sorted V = 0); 45° erases it fully (V = 1). The slider rides the exact V = sin 2φ trade.

Delayed choice (idler, later)

Sorted channels

Show + channel (fringes)

Show − channel (anti-fringes)

Geometry

Slit separation d…

wider slits pack the fringes closer: Δy = λL/d. It sets the stripe pitch, not whether they appear.

Slit width a…

the single-slit envelope under the stripes. Narrower slits spread the whole pattern wider.

Slit → screen L…

throw distance to the detection screen. Farther screen, coarser fringes.

Beam rate120 / s

how fast photons arrive. A real single-photon source is far slower; dialled up so the pattern builds in seconds.

Overlays

Theory curve(s)

Apparatus & labels

Live formula

Readout · exact from your dials

sorted V = |sin 2φ|

1.00

fringe Δy = λL/d

…

+ channel hits

− channel hits

screen window

…

detected

What is exact vs staged

Established: the marked pattern going flat, the P±(y) fringe/anti-fringe split, the always-flat total, V = sin 2φ and the delayed choice. Stylised: a single photon source stands in for entangled pairs, the ± channels for coincidence counting, and the beam runs far faster than any real source. Reading: whether "erasure" rewrites the past. It does not; this instrument takes no other side.

The whole trick, honestly

Erasing the which-path tag does not restore fringes in the data you can see. The combined screen is always flat once the paths were ever tagged. Fringes reappear only when you sort the same hits by the eraser's two outputs: one half is a fringe, the other its anti-fringe, and they sum back to flat. The delayed choice is harmless for the same reason: no signal travels back, only a correlation is revealed.

…1 / 5