Qimena Open Data: Solar-Term Comparison, Ju Table and Whole-Chart Comparison (CC0, Free Download)

Published

How Qimena casts a chart, and how it was checked, is written up on the calculation rules page. This page puts the data behind it into files you can download, for researchers, developers and other software authors who want to check it, cite it, or test their own calculator against it.

Everything is released under CC0 1.0 (public domain): no sign-up, no email, free to copy, modify and use commercially, no attribution required.

The three datasets

FileContentsRows
solar-terms-vs-hko-2026-2028.csvThe 72 solar terms of 2026–2028: the Hong Kong Observatory’s time, the time our engine computes, and the gap in minutes72
ju-number-table.csvThe Chai Bu ju table: 24 solar terms by the upper, middle and lower yuan, marked Yang Dun or Yin Dun24
board-comparison-china95.csv56 Beijing-time moments compared with china95.net (rotating plate + Chai Bu, no run-month): four pillars, ju, Chief Star and Chief Door, and the stems, star, door and deity of every palace56
qimen-open-data.jsonAll three together, with the sources152

The CSV files are UTF-8 with a byte-order mark, so Excel opens the Chinese correctly.

How to read each one

Solar terms. hko_time_hkt and qimena_time_hkt are both Hong Kong Time (UTC+8, the same as Beijing time). A positive diff_minutes_qimena_minus_hko means our engine runs later than the Observatory. All 72 terms across three years are within 1 minute (the Observatory publishes to the minute). Note that qimena_time_hkt comes from the precomputed solar-term table our engine uses (generated with tyme4ts), so matching the Observatory is expected; the independent check is a comparison of the table’s 4,824 terms for 1900–2100 with a second astronomy library, astronomy-engine: largest gap 1.0 minute, average 0.2. The Observatory’s times come from data of HM Nautical Almanac Office (UK) and the US Naval Observatory, and were read on 2026-09-29.

Ju table. The first row is the Winter Solstice and the last is Major Snow; Winter Solstice to Grain in Ear is Yang Dun and Summer Solstice to Major Snow is Yin Dun. The yuan comes from the branch of the day’s Fu Tou: Zi, Wu, Mao and You are the upper yuan, Yin, Shen, Si and Hai the middle, Chen, Xu, Chou and Wei the lower.

Whole-chart comparison. A result of identical means the four pillars, ju, Chief Star and Chief Door, and the heaven stem, earth stem, star, door and deity of all eight outer palaces match the other site exactly. different rows carry first_differences (what each side shows) and a note with the cause. Of the 56 moments, 16 are solar-term boundaries and special cases and 40 are random odd hours. The result: 49 identical and 7 different, and all 7 differences trace back to the other site’s solar-term times or month-pillar rules (details on the calculation rules page).

Before you use it

  • The whole-chart comparison uses the plain Beijing clock time, with no True Solar Time correction. That differs from the default on qimena.com, which corrects for the longitude of the casting place, so cast with the setting described above when you compare.
  • Only one other site was compared. Software using other schools (Zhi Run, flying plate and so on) lays out different charts, which does not mean either is wrong; see choosing a school.
  • The other site was read in October 2026. It may change its rules later; the table records what it returned at the time.

Citing and reproducing

To cite: Qimena open data (qimena.com/en/wiki/qimen-open-data/), 2026-10-13, version 1.0. To reproduce the whole-chart comparison, enter any moment from the table into your program and into china95.net’s rotating-plate Qi Men (Chai Bu, no run-month) and compare: one moment, one chart, a few minutes each.

If you find an error, or get a different result from other software, email [email protected] with the software name, the time and the difference. To see how this data is used in a chart, try the free AI chart.

FAQ

Can I use this data freely?

Yes. All three datasets are released under CC0 1.0, which waives copyright: copy, modify and use them commercially, with no attribution and no need to contact us. If you would like to credit the source, "Qimena open data" is enough.

How do I test my own Qi Men calculator with it?

Cast the same 56 Beijing-time moments in your program and compare the four pillars and the ju with the whole-chart table. Use the solar-term table to test your solar-term algorithm and the ju table to check the Chai Bu lookup. Use the plain Beijing clock time, with no True Solar Time correction, to match how the comparison was run.

Where does the data come from, and could it be wrong?

The solar-term instants are compared with the times the Hong Kong Observatory publishes, the ju table is the traditional rhyme table, and the whole-chart comparison casts the same moment on china95.net and compares item by item. The other site's results are what we read at the time and may change. If you find a problem, email [email protected]; we will check, correct it and add it to the corrections log.