A panchang answers five questions about a day, and the calculator above answers them for whichever date and city you pick. The tithi is the lunar day, set by how far the Moon has pulled ahead of the Sun. The nakshatra is the segment of sky the Moon is currently crossing. The yoga and the karana are two further quantities derived from the same two positions. The vara is the weekday, which begins at sunrise rather than at midnight.
Those five limbs are what the word itself means: pancha anga, five parts. Everything else on the page hangs off them: the Rahu Kaal window, the Abhijit muhurta, the moment your tithi ends and the next begins.
One thing this page does that most do not: it shows you which conventions produced the numbers. If your family panchang disagrees, the reason is almost always in that block, and you will be able to point at it.
The five limbs, and what each one measures
Tithi. The Moon gains on the Sun by roughly twelve degrees a day. Each twelve degrees of that gap is one tithi, so there are thirty in a lunar month, fifteen while the Moon waxes and fifteen while it wanes. Because the Moon's speed varies, a tithi is not a fixed length. It runs anywhere from about nineteen to about twenty-six hours, which is why tithis begin and end at odd times rather than at dawn.
Nakshatra. The ecliptic is cut into twenty-seven equal segments of 13°20′ each. The nakshatra is whichever one the Moon occupies. Each divides again into four padas of 3°20′, and each carries a ruling planet that later matters for dasha calculation.
Yoga. Add the Sun's longitude to the Moon's, and divide that sum into twenty-seven parts. That is the yoga. It has no obvious physical meaning, being a derived quantity the tradition tracks rather than an observable one. But note that the arithmetic uses a sum. That has a consequence we will come back to.
Karana. Half a tithi. Six degrees of elongation instead of twelve, so sixty in a lunar month, drawn from eleven names: seven that cycle through the month and four that appear once each.
Vara. The weekday. The Hindu day runs sunrise to sunrise, so the vara changes at dawn, not at midnight.
The day that does not begin at midnight
This is the single most useful thing to understand about a panchang, and it explains most of the confusion people have with them.
Every element above is reported as it stands at sunrise. When the page says today's tithi is Dwadashi, it means Dwadashi was running at sunrise this morning, even if it ended at eight o'clock and Trayodashi has been running since.
Two things follow. First, a moment after midnight but before sunrise still belongs to yesterday's vara. A festival "on Monday" may fall on what your phone insists is Tuesday. Second, if a tithi begins after sunrise and ends before the next sunrise, it never holds a sunrise at all. That is a kshaya tithi, skipped in the count.
This is also why panchangs printed for different cities disagree. Sunrise in Kolkata is roughly forty minutes earlier than in Mumbai, so a tithi ending in that window belongs to different days in the two cities. Neither is wrong.
The day's windows: Rahu Kaal, Abhijit and the rest
Take the time between sunrise and sunset and cut it into eight equal parts. Rahu Kaal is one of those parts, and which one depends only on the weekday. Yamaganda and Gulika are two more parts, assigned by their own weekday tables.
Here is the part almost nobody is told, and it is worth stating plainly.
Rahu Kaal has nothing to do with where Rahu actually is.
It is an ordinal: the second part of the day on Monday, the seventh on Tuesday, and so on. The calculation never consults Rahu's position. You can verify this: change the ayanamsa on the calculator above, which moves every planetary position, and the Rahu Kaal timing does not budge.
Because the window is one eighth of daylight, its length changes with the season. In Delhi it runs about 77 minutes at the December solstice and about 105 minutes at the June solstice. In Chennai, nearer the equator, the swing is much smaller, roughly 85 to 97 minutes. At the equinoxes everywhere gets about 91 minutes.
Abhijit muhurta is built the same way but with fifteen divisions instead of eight: it is the eighth of fifteen equal muhurtas, so it straddles local noon. Note that many traditions do not observe Abhijit on Wednesdays.
Brahma muhurta is the period roughly 96 to 48 minutes before sunrise.
How this panchang is computed
The chain runs like this.
Positions of the Sun and Moon come from astronomy-engine: the Sun and planets from VSOP87, and the Moon from the Nautical Almanac Office's Improved Lunar Ephemeris of 1954, which derives from E. W. Brown's lunar theory. Those are apparent geocentric positions referred to the true equinox of date.
Vedic calculation needs sidereal longitudes, measured against the fixed stars rather than the moving equinox. Converting means subtracting two things: the nutation in longitude, which brings the frame to the mean equinox, and then the ayanamsa. The default here is Lahiri, the Government of India standard fixed by the Calendar Reform Committee in the 1950s. You can switch to Raman or Krishnamurti above and watch what changes.
Our Lahiri implementation has been checked against Swiss Ephemeris 2.10.03 at eight epochs from 1900 to 2050, agreeing within 0.05 arcseconds, and against the mean value published in the Indian Astronomical Ephemeris for J2000.
Sunrise and sunset use the upper limb of the Sun with atmospheric refraction, the common civil definition. You can switch to the geometric centre of the disc, which some Vakya panchangs use; the two differ by a few minutes. Each city carries its own elevation, which depresses the horizon slightly and moves sunrise a little earlier.
Time zones are resolved for the date in question, never assumed. India used Madras Mean Time until 1906 and wartime daylight saving from 1942 to 1945. Bombay kept its own local time, UTC+04:51, until 1955, and Calcutta kept UTC+05:53:20 until 1948. Neither of those is expressible in the standard time-zone database, so both are handled explicitly here. Most birth-chart sites apply a flat +05:30 to every Indian birth ever recorded, which is 39 minutes wrong for a pre-1955 Bombay birth and enough to move an ascendant by about ten degrees.
Above the Arctic and Antarctic circles the Sun may not rise at all. Rather than invent a sunrise, the page falls back to local noon as the reference instant.
Why your panchang differs from another one
Modern ephemerides agree on where the Sun and Moon are to a fraction of an arcsecond. So when two panchangs disagree, the astronomy is not the culprit. It is nearly always one of four things.
A different ayanamsa. Lahiri and Raman sit about 1.45° apart. A nakshatra is 13°20′ wide, so that gap is roughly a tenth of a nakshatra. About one birth in ten therefore lands on a different star depending on which system is used.
A different sunrise definition. A few minutes, which only matters when a boundary falls inside them. It often does.
Drik versus Vakya. Drik systems compute actual positions. Vakya systems follow older tabular methods that can differ by a full day on the tithi. Both remain in use, particularly in the south.
A different city. Panchangs are location-specific, and a printed one is right only for the place it was computed for.
Now the part that settles a surprising number of arguments:
Tithi and karana cannot differ because of the ayanamsa.
Both are computed from the difference between the Moon's longitude and the Sun's. Whatever offset you subtract, you subtract it from both, so it cancels exactly. The same is true of the nutation correction.
Nakshatra and rashi come from an absolute longitude, so they shift with the ayanamsa. Yoga is built from a sum, so it shifts too, and by twice as much.
The practical upshot: if two panchangs disagree on tithi, the ayanamsa is not the reason. Look at the sunrise rule or the location instead. If they disagree on nakshatra but agree on tithi, the ayanamsa almost certainly is the reason.
Where the five limbs come from
The five-part scheme is old and stable. The classical siddhantic astronomy texts, the Surya Siddhanta foremost among them, set out the computation of tithi from the elongation, the nakshatra from the Moon's position, and the derived yoga and karana, in essentially the form used today. What has changed is the accuracy of the underlying positions, not the definitions built on them.
Regional almanac traditions carried these forward with local variation. The differences that survive are mostly about which convention to apply, not about what a tithi is.
What this panchang cannot tell you
It cannot tell you whether your day will go well.
It carries no claim about your health, your fertility, your finances, your relationships or your lifespan, and it never will. Those pages exist elsewhere on the internet and they are usually attached to something for sale.
What it does is arithmetic: here is where the Sun and Moon are, here is what follows from that under the conventions named above, and here is how to check the working. Nothing here is a prediction about a person.
Reading a panchang across regions
The month name you see depends on whether your tradition ends the month at the new moon (amanta, common in the west and south) or at the full moon (purnimanta, common in the north). The same day can carry two different month names, both correct.
The year number depends on the era: Vikram Samvat runs about 57 years ahead of the Common Era, Shaka Samvat about 78 behind.
And nakshatra names vary in transliteration far more than in substance. Krittika and Kārttikā are the same star group.
Using it day to day
Most people arrive at a panchang for one of three reasons, and each wants a different part of the page.
Checking a fast or a festival. Look at the tithi and, crucially, at when it ends. Observance rules attach to the tithi running at a particular moment, and that moment is not always sunrise. Ekadashi fasting, for instance, is tested at sunrise in the smarta reckoning but at arunodaya, roughly the last four ghatis before sunrise, in the vaishnava one. That difference is why two households on the same street can fast on different days and both be following their tradition correctly.
Avoiding an inauspicious window. That is the Rahu Kaal block. Read the caveat above it first: the window is an ordinal into the daylight, not an astronomical event, and its length is a function of how long the day happens to be.
Finding a good moment. Abhijit is the general-purpose auspicious window in most northern traditions, and it always straddles local noon because of how it is constructed. Anything more specific than that belongs to muhurta proper, which weighs several factors at once.
Two habits worth forming
First, note the city. A panchang computed for Delhi is not a panchang for Chennai, and around a boundary the two genuinely differ. Sunrise separates them by more than half an hour at some times of year.
Second, when an element ends late in the day, check the following day too. A tithi that ends at 23:40 has a successor that owns almost none of the day it is named for, and if you are planning around the successor you want its actual window rather than its label.
The honest limits of a calculator
Everything above is arithmetic performed on planetary positions. It is reproducible, it is checkable, and the working is shown. What it is not is a statement about you.
The tradition itself is more careful here than its modern commercial packaging suggests. A panchang is an almanac. It records the shape of a day in a system of reckoning that many people find meaningful, and that reckoning is genuinely old, genuinely mathematical and genuinely worth understanding.
None of that makes it a forecast, and this page will not pretend otherwise.
This page is a calculator with its working shown, not a source of predictions. Where traditions genuinely disagree, it shows the disagreement rather than picking a winner.