Tarabala & Chandrabala

How a day's Moon stands relative to your birth Moon, counted two ways. With panchak and bhadra for the same day, and how often each of the four actually occurs.

2026-10-14 · Budhavara · Moon in Anuradha, Vrischika

Add a birth date to get tarabala and chandrabala, which are both counted from your own birth Moon. Panchak and bhadra below need no birth details, since they are properties of the day itself.

Panchak

Not running. The Moon is in Vrischika (Anuradha).

Panchak is the Moon in Kumbha or Meena. That is the same interval as "the last five nakshatras from Dhanishtha's second half": Dhanishtha's third pada begins at exactly 300°, which is exactly where Kumbha begins, and Revati ends where Meena does. Two signs of twelve, so it runs about a sixth of the time — roughly five days in every lunar month.

Bhadra (Vishti karana)

Not running at sunrise. The karana is Vanija, until 12:17.

Bhadra is nothing more than the Vishti karana being current. A karana is half a tithi, and Vishti occupies eight of the sixty karanas in a lunar month, so it runs about thirteen per cent of the time. It is reported here at sunrise; it can begin or end at any hour, so check the window rather than the day.

How this was calculated — and why another panchang may differ
Ayanamsa
Lahiri (Chitrapaksha) — Government of India standard
mean 24°13′52.44″ · true 24°14′00.43″

Positions are computed from the mean value; printed panchangs usually quote the true value, which adds nutation. Both are shown so the number you see elsewhere is recognisable.

Sunrise
Upper limb, with refraction (−0°50′) — the common civil definition

Every element below is read at sunrise, so this choice moves the boundaries. The two common conventions differ by three to five minutes, which matters on any day a tithi ends inside that gap.

Ephemeris
astronomy-engine 2.1.19 — Moon from the Improved Lunar Ephemeris (1954), after E. W. Brown
Time zone
UTC+05:30

Resolved for this date, not assumed. India used Madras Mean Time until 1906 and wartime daylight saving from 1942 to 1945.

Julian Day (UT)
2461327.270833
Algorithm
v1.0.0
Place data
GeoNames, used under CC BY 4.0

Every populated place in India plus every city worldwide over 15,000, with the IANA zone each one actually sits in. Coordinates are the gazetteer's; elevation is only held for the curated cities, so a looked-up place is treated as being at sea level, which moves sunrise by under a minute for anywhere below about 1.5 km.

Why another panchang may not agree

Disagreements are almost never about astronomy. Modern ephemerides agree on where the Sun and Moon are to within a fraction of an arcsecond. They are about convention:

  • A different ayanamsa. Lahiri and Raman sit about 1.2° apart, which is roughly a tenth of a nakshatra — so about one chart in ten lands on a different birth star.
  • A different sunrise definition. Upper limb with refraction, or the centre of the disc geometrically.
  • Drik versus Vakya. Drik computes real positions; Vakya follows older tabular methods that can differ by a day on the tithi.
  • A different place. Panchangs are location-specific. A panchang printed for one city is wrong for another.

One property is worth knowing because it settles many arguments: tithi and karana do not depend on the ayanamsa at all. They are computed from the angle between the Sun and Moon, so the offset cancels exactly. If two panchangs disagree on tithi, the cause is the sunrise rule or the location — never the ayanamsa. Nakshatra, rashi and yoga do shift with it.

Tarabala depends on both nakshatras, so it shifts with the ayanamsa at both ends; Lahiri and Raman are about a tenth of a nakshatra apart. Sources also differ on whether Janma tara is favourable, on whether to count from the Moon's nakshatra or the birth nakshatra of the person versus the day's ruling star, and on how many of the nine taras are treated as difficult. The counting convention itself is the largest single source of disagreement: an exclusive count shifts every result by one tara.

Four separate tests get applied to a day before it is called good or bad. Tarabala counts nakshatras from your birth star. Chandrabala counts signs from your birth moon sign. Panchak asks where the Moon is. Bhadra asks which karana is running.

The calculator above runs all four for any date. It also states how often each fires, because those four numbers together answer a question the individual tests cannot: how often is a day clear on all counts at once?

Tarabala

Count from your birth nakshatra to the day's nakshatra. That count, taken modulo nine, names one of nine taras: Janma, Sampat, Vipat, Kshema, Pratyari, Sadhaka, Vadha, Mitra and Ati-Mitra.

Three of the nine are treated as unfavourable: Vipat, Pratyari and Vadha, the third, fifth and seventh. Janma, your own birth star, is disputed: some sources count it as difficult, others do not, so this page marks it mixed rather than resolving it.

The count is inclusive, and this is where implementations most often go wrong. Counting from a nakshatra to itself gives 1, not 0. Ashwini to Ashwini is Janma; Ashwini to Bharani is Sampat. An exclusive count shifts every result by one tara and produces an answer that looks entirely plausible.

Because 27 nakshatras divide evenly by nine, each tara comes up exactly three times in the cycle, and each is therefore exactly one ninth of days. That makes the frequency exact rather than estimated: three taras of nine are unfavourable, so a third of all days have a bad tarabala for you.

Chandrabala

The same idea one level coarser. Count from your birth moon sign to the day's moon sign, inclusive again, giving a number from one to twelve. The 4th, 8th and 12th are treated as unfavourable.

Three signs of twelve is exactly a quarter. A quarter of all days have a bad chandrabala for you.

Chandrabala is more forgiving of a missing birth time than tarabala is, and for a reason worth knowing. The Moon holds a sign for about two and a quarter days but a nakshatra for only about one, so a birth date with no time settles your moon sign far more often than it settles your birth star. If you have only a date, the chandrabala figure is the more trustworthy of the two.

Panchak

Panchak is the Moon in Kumbha or Meena.

You will more often see it defined as the last five nakshatras, from the second half of Dhanishtha through Revati. Those two definitions describe exactly the same interval, which is worth proving rather than assuming.

Dhanishtha spans 293°20′ to 306°40′, and its four padas are 3°20′ each, so its third pada begins at 293°20′ + 6°40′ = exactly 300°00′. Kumbha also begins at exactly 300°00′. From there, Dhanishtha's second half plus the four following nakshatras run 6°40′ + 13°20′ × 4 = exactly 60°, ending at 360°, which is exactly where Meena ends.

One identical 60° arc, described two ways. Two signs of twelve, so panchak runs about one sixth of the time, which is roughly five days in every lunar month. Hence the name: panchak means a group of five.

The type is named by weekday: Roga, Raja, Agni, Chora or Mrityu panchak depending on which day it begins.

Bhadra

Bhadra is the Vishti karana being current. That is the whole definition.

A karana is half a tithi. Seven karanas repeat cyclically through the month, and Vishti is the seventh of them. Of the sixty karanas in a lunar month, fifty-six are these movable ones and four are fixed, which puts Vishti in eight of sixty. About thirteen per cent of the time, which the tool measures at 12.8% over four hundred sampled days.

Two practical points follow from bhadra being a karana rather than a day. It can begin and end at any hour, so "today is bhadra" is imprecise, and what you want is the window. And because a karana is roughly twelve hours, bhadra typically runs less than a full day, unlike panchak which runs five.

Where the nine taras come from

The structure is worth a moment, because it explains why the count is modulo nine rather than any other number.

Twenty-seven nakshatras divide by nine to give exactly three. So the taras run through their cycle three times as you count round the zodiac, and the tool reports which of the three cycles your count falls in: 1 to 9 is the first, 10 to 18 the second, 19 to 27 the third.

That the division is exact is not a coincidence. Twenty-seven was chosen partly because it factors so conveniently: 27 = 9 × 3 = 27 × 1, and it also divides into four padas each to give 108. A system built on 28 nakshatras, which some early lists used, would not support a clean nine-tara cycle at all, and that is one of several reasons the twenty-eighth was dropped.

Some sources weight the three cycles differently, treating a bad tara in the first cycle as more consequential than the same tara in the third. The tool reports the cycle so that reading is available, without endorsing it.

The four together

Here is the arithmetic nobody assembles, and it is the reason to have all four on one page.

TestFires on
Tarabala unfavourable1 day in 3
Chandrabala unfavourable1 day in 4
Panchak running1 day in 6
Bhadra runningabout 1 day in 8

If those were independent, the chance of a day being clear on all four would be ⅔ × ¾ × ⅚ × ⅞, which is about 36%.

They are not fully independent, since tarabala and chandrabala both track the Moon and panchak is a subset of positions tarabala also sees, so the true figure is somewhat higher. But the order of magnitude holds: on any given day, a fairly large share of the tests will flag something, and finding a date clear on all four takes looking.

That is not an argument that the tests are meaningless. It is the context that makes any single one interpretable. Being told "there is bad tarabala today" sounds like a warning about that day. It describes a condition that holds one day in three, for you specifically, forever.

Why the Moon and not something slower

There is a structural reason these day-quality tests are built on the Moon rather than on any other body, and it is practical rather than doctrinal.

A test is only useful for choosing a day if its answer changes from day to day. The Moon moves about 13° a day and changes nakshatra roughly daily, so tarabala gives a different answer nearly every day, which is exactly what a day-selection tool needs.

Try the same construction with Saturn and it collapses. Saturn stays in a sign for two and a half years, so a Saturn-based day test would return the same verdict for nine hundred consecutive days and be useless for picking between them. That is precisely why Saturn's slow transits are read as life periods, like Sade Sati, rather than as day quality.

The choice of body follows the timescale of the question. Day selection needs the Moon; life periods need the outer planets; the hour-scale windows like Rahu Kaal and Choghadiya need nothing astronomical at all, because they are subdivisions of the day itself. Seen that way, the various systems are not competing accounts of the same thing. They are tools cut to different grain sizes.

Why another site gives a different tarabala

The counting convention. Inclusive against exclusive, worth exactly one tara on every single result. This is the largest single source of disagreement and it is almost never stated.

Whether Janma is favourable. Sources genuinely split. A site treating it as bad reports four bad taras of nine rather than three, taking the unfavourable rate from a third to nearly a half.

The ayanamsa, which enters at both ends here. Tarabala compares two nakshatras, and both shift with the ayanamsa. Lahiri and Raman are about a tenth of a nakshatra apart, so about one case in ten moves.

Which Moon the day's nakshatra is read at. Sunrise is the classical convention, but the Moon can change nakshatra during a day, and a site reading it at midnight or at the current moment will sometimes give a different star.

A missing birth time, which affects tarabala much more than chandrabala, as above.

Using this practically

If you are looking for a date rather than checking one, the frequencies suggest an order of operations.

Panchak and bhadra are properties of the day itself and are the same for everybody, so they can be checked once and shared. Panchak runs about five days a month in a single block, which makes it easy to plan around: find the block, avoid it, and you have cleared a sixth of the calendar in one step.

Tarabala and chandrabala are personal and differ for every individual, which is why a date good for one person is not automatically good for another. For any event involving two people, both sets have to be checked, and the chance of a day clearing both people's tarabala is roughly four ninths rather than two thirds.

That compounding is the practical reason muhurta selection is difficult. Each additional person and each additional test multiplies the constraint, and with enough of them the set of acceptable dates becomes very small. Traditional practice knows this, which is why the classical literature also carries rules about which tests may be waived for which purposes.

A note on what these tests are made of

None of the four looks at a planet. That is worth saying because it separates them from most of what gets called astrology.

Tarabala and chandrabala compare two positions of one body, the Moon, at two different times. Panchak asks where that same Moon is now. Bhadra asks about the karana, which is built from the angular distance between the Moon and the Sun.

So all four are functions of the Moon, and three of the four are functions of the Moon alone. There is no Mars, no Saturn, no lagna, no house. That makes them cheap to compute and easy to verify, and it also means they are wholly determined by the lunar cycle, which is why their frequencies come out as clean fractions rather than as things that need measuring.

It also means the four are correlated with each other in ways the independence estimate above does not capture. A day in panchak has the Moon in Kumbha or Meena, which fixes its nakshatra to one of five, which in turn constrains what your tarabala can be. The tests are not four separate looks at a day; they are four readings of one position.

What this page will not do

It will not tell you whether to travel on Tuesday.

What it computes is a count between two nakshatras, a count between two signs, whether the Moon lies in a 60° arc, and which karana is running. Every one is a small piece of arithmetic on positions computed elsewhere on this site and checkable against any panchang.

The reason to see all four together, with their frequencies, is that they are usually met one at a time. A page that tells you bhadra is running and stops has given you a fact without a denominator. Thirteen per cent of days have bhadra. A third have bad tarabala. Whatever weight you give them, that is the scale they operate at, and knowing it costs nothing.

Common questions

What is tarabala?

A count from your birth nakshatra to the day's nakshatra, taken modulo nine, naming one of nine taras: Janma, Sampat, Vipat, Kshema, Pratyari, Sadhaka, Vadha, Mitra and Ati-Mitra. Three of the nine are treated as unfavourable, so a third of all days have a bad tarabala for you.

Is the tarabala count inclusive or exclusive?

Inclusive, and this is where implementations most often go wrong. Counting from a nakshatra to itself gives 1, not 0, so Ashwini to Ashwini is Janma and Ashwini to Bharani is Sampat. An exclusive count shifts every result by one tara and produces an answer that looks entirely plausible.

What is chandrabala?

The same idea one level coarser: a count from your birth moon sign to the day's moon sign, inclusive, giving one to twelve. The 4th, 8th and 12th are treated as unfavourable, so exactly a quarter of days. It is more forgiving of a missing birth time than tarabala, since the Moon holds a sign for about two and a quarter days but a nakshatra for only about one.

What is panchak?

The Moon in Kumbha or Meena. That is exactly the same interval as the more commonly quoted definition of the last five nakshatras from Dhanishtha's second half through Revati: Dhanishtha's third pada begins at exactly 300 degrees, which is exactly where Kumbha begins, and Revati ends where Meena does. Two signs of twelve, so about a sixth of the time, roughly five days a month.

What is bhadra?

The Vishti karana being current, and nothing more. A karana is half a tithi, and Vishti occupies eight of the sixty karanas in a lunar month, so it runs about thirteen per cent of the time. It can begin and end at any hour, so the window matters rather than the day.

How often is a day clear on all these tests at once?

If the four were independent it would be about 36%: two thirds clear of tarabala, three quarters of chandrabala, five sixths of panchak, seven eighths of bhadra. They are correlated, since all four are functions of the Moon, so the true figure is somewhat higher. Either way, a fairly large share of days will flag something.

Why does another site give a different tarabala?

The counting convention is the biggest cause, worth exactly one tara on every result and almost never stated. After that: whether Janma tara counts as favourable, which sources genuinely split on; the ayanamsa, which enters at both ends; and whether the day's nakshatra is read at sunrise or at some other moment.

Do these tests use the planets?

No, and that is unusual. All four are functions of the Moon, and three of them are functions of the Moon alone. There is no Mars, no Saturn, no lagna and no house. That is also why a day test is built on the Moon: it moves about 13 degrees a day, so its answer changes daily, while a Saturn-based test would return the same verdict for nine hundred days running.