I' m using Meeus' algorythms : OK
For sundials and equation of time I'm using Denis Savoye algorithms (some mistakes in the first edition, but corrected in the last edition).
I wrote some tens of programs.
Implementation in QBasic of Meeus helped me to discover a bug in QBasic when I was computing the Julian day.
Implementation on a pocket Casio 25+ helps me to verify the sundials when I'm visiting some of them and discovered many sundials with bad realisation and important drifts. I'm working with a target of 3minutes or less precision for my own sundials, which is very low. Only a little fraction of gnomonists are able to reach that precision. Remind that every angle error of 1° may lead to a 4 minutes drift.
Formulas may be found too in the old ephemeris books perfectly verified.
Debug perfectly your softwares.
Much pleasure during nights with the well named "Calculs astronomics"
