Repository navigation
Viewer: the dipole moment arrow is too long (and its origin and convention are not stated) #266
Description
9.0.0-beta.1. For water (μ ≈ 2 D), Properties > Dipole Moment draws an arrow about three times as long as the O–H bonds. It dominates the picture.
Proposed
- Length: a smaller default scale, e.g. about 0.5 Å per Debye, so a 2 D dipole is roughly an O–H bond long. Add a length control in the Dipole Moment panel, and show the value (|μ| in D, components) next to it.
- Origin: draw the arrow from the centre of mass, or the centre of nuclear charge, not from an atom (here it starts at O).
- Convention: state which one is used. Here the arrow points from the hydrogens past the oxygen, towards the negative end (chemists' convention); physics points from − to +. Show it in the panel ("arrow points to the negative end"), or offer both.
- Arrow head: the head and shaft should scale with the length (thickness relative to the bonds, Viewer: bonds poorly visible on a white background #227).
Check against the stored dipole components for a few molecules (water, ammonia, HF), and against the code's own output, so the direction is verified, not assumed.
Activity
ohlincha commentedon Oct 10, 2026
Checked 2026-10-10: the Dipole Moment panel's Scale is already 0.50 by default (1 Å per Debye × 0.5) in 9.0.0-beta.1, and the arrow starts at the centre of nuclear charge (the default "Center"; for water close to O), which is legitimate. So the long arrow in the report does not come from the default scale. To find out next: a remembered Scale from an earlier session, or how VRVector turns the scaled vector into a length (head and shaft added?). Change of AddDipoleMomentCmd's preferred scale reverted: it had no effect.
Viewer: the dipole arrow is 0.5 A per Debye and the panel says what i…
ohlincha commentedon Oct 11, 2026
e8004ed: the arrow is 0.5 Å per Debye by default (adjustable 0.05–5) with fixed shaft and head thickness; the panel gives |μ| in Debye and states the origin (centre of nuclear charge) and the direction (as the code reports μ, negative to positive end). Checked for NWChem, ORCA, Gaussian 16 and MOPAC water. Open: the Quadrupole and Stress panels inherit and draw a meaningless arrow.
Viewer: arrow only on the Dipole panel, centre of mass is mass-weight…
ohlincha commentedon Oct 11, 2026
1d7b8d1: the Quadrupole, Polarizability, Hyperpolarizability and Stress panels no longer draw an arrow or show the Tensor Color / Scale row; only Dipole and MP2 Dipole do. The "mass" centre is now mass-weighted (no menu offers it yet). To be checked on screen.
ohlincha commentedon Oct 12, 2026
Checked on screen (Linux): the arrow length and the panel text look right. Closing.