Day 0from start date
0 d48121620242832 d
Moon orbit: 27.32 d sidereal · 29.53 d new moon to new moonStrip under the slider: gold = Moon on Earth's day side of the terminator, blue = night side
18:00local time
00:0003:0006:0009:0012:0015:0018:0021:0024:00

Solar system, seen from above the orbital plane

Zoom
Terminator (day/night boundary) Sight line to Venus Sight line to Moon Direction of Sun
Sun–Earth–Venus angle–
Sun–Earth–Moon angle–
Observer past terminator–
Venus distance–

Sky from the observer's position

Altitude × azimuth · west centred
BodyAltitudeAzimuthFrom SunVisible
Moon → Venus separation–
Venus relative to Moon–
Moon illuminated–
Venus illuminated–

What the geometry actually says

  1. "Night side" and "what you can see" are different planes. The terminator is a plane through Earth's centre, face-on to the Sun. Your horizon is a plane tangent to the ground under your feet. Just after sunset you are only a few degrees past the terminator, and your western horizon points almost exactly at the Sun. You can see everything on the sky side of that tangent plane, out to 90° from straight up. The red horizon line in the sky panel is that plane. Watch which side of it Venus and the Moon sit on as the evening goes by.
  2. Venus is never "beside the Sun" in the way the premise assumes. Its orbit is 0.72 AU, so from Earth it can stand up to asin(0.72) ≈ 46° away from the Sun's direction. On 15 August 2026 it was at that maximum. A body 46° from the Sun is above your horizon for a while after the Sun goes below it. That is the whole story: Venus is an evening or morning object, never a midnight one. Drag the slider to 00:00 and the model puts Venus well below the horizon.
  3. Earth's size plays no part. Venus is roughly 16,000 Earth radii away on this date. Whether you are standing on the "front" or "back" of a ball 1 unit across does not change your sight line to something 16,000 units off. Only angles matter, and the angles are shown on both panels.
  4. Why the Moon is above Venus. The Moon was 4 days past new and about 50° east of the Sun along the ecliptic; Venus was at 46°. Both lie near the same line in the sky. From the Southern Hemisphere in August the evening ecliptic stands steeply, so "further from the Sun along the ecliptic" means "higher", which puts Venus a few degrees below the Moon. The dashed ecliptic line in the sky panel makes the alignment visible.
  5. What would break the model. Venus more than about 47° from the Sun on any date. Venus above the horizon at local midnight. Venus with a full phase while near greatest elongation. None of these occurred on 16 August. Any of them, if ever observed, would falsify the orbit shown on the left.

The same observation on a flat plane

You asked for the flat-earth model as the base. It has no published equations that place Venus on a given date, so there is nothing to simulate against. It does make one hard geometric claim we can test with this very evening: bodies circle at some height above the plane and never pass below it. For an observer at horizontal distance D from a body at height H, the altitude is atan(H/D). That number is positive for every finite D.

Sun height HDistance D for 0° altitudeAltitude at D = 20,000 km

On 16 August, from Johannesburg, the Sun reached −44° by 21:00, Venus set at – and the Moon at –. A negative altitude is unreachable on a plane. The globe model reproduces the setting times from orbital arithmetic; the flat model has no mechanism for a body to set at all. The premise of "a night side facing away from the Sun" is itself a description of a sphere.

Method: JPL approximate Keplerian elements (Standish, 1800–2050) for Earth and Venus; Meeus low-precision lunar series (22 longitude terms, 12 latitude terms); mean obliquity and mean sidereal time. Expected error under 0.3°. No atmospheric refraction is applied, so real setting times run about 3–5 minutes later than shown. Cross-checks: greatest eastern elongation 15 Aug 2026 (published 46°, model 45.9°); Venus 48% lit (model 49%).