The navamsa cuts each sign into nine parts and reads each part as a sign of its own. That gives a second twelve-house chart built from the same planetary positions, conventionally read alongside the rashi chart rather than instead of it.
It also multiplies your birth-time problem by nine, and that is the part this page is built around.
The rule, which is shorter than the textbooks make it
The usual presentation gives three cases. Movable signs count their navamsas from themselves. Fixed signs count from the ninth sign along. Dual signs count from the fifth.
All three are the same statement wearing a disguise. Cut the entire zodiac into 108 equal parts of 3°20′, and read them off cyclically as the twelve signs:
navamsa sign = floor(sidereal longitude ÷ 3°20′) mod 12
Check it against the cases. Mesha begins at part 0, which is Mesha, so a movable sign starts from itself. Vrishabha begins at longitude 30°, part 9, which is Makara, and Makara is the ninth sign from Vrishabha. Mithuna begins at 60°, part 18, which is Tula, the fifth from Mithuna. The three-case table is a consequence, not an extra rule.
It works because the cyclic reading starts sign s at 9s mod 12, and that expression happens to reproduce the movable/fixed/dual pattern exactly. The implementation here is checked against the classical table at all 108 navamsas.
The shortcut is specific to the navamsa and does not generalise to other divisional charts. Applied to the drekkana it matches the classical rule in only 4 cases out of 36; to the dwadashamsa, 12 out of 144.
Those vargas start each division at the sign itself and step by a different amount. Any calculator that runs every divisional chart through one shared "divide and read cyclically" formula is producing a correct D9 and wrong everything else.
Working one out by hand
The formula is short enough to check without a computer, which is worth doing once so you know what the tool is doing.
Take Saturn at 0°45′ of Makara on the example chart. Makara is the tenth sign, so its start is 9 × 30 = 270°, and Saturn's sidereal longitude is 270.75°.
Divide by the navamsa width: 270.75 ÷ 3.3333 = 81.2. Take the whole part, 81. Then 81 mod 12 = 9, and sign 9 counting from zero is Makara.
So Saturn's navamsa sign is Makara, the same sign it occupies in the rashi chart, which makes it vargottama. And within Makara it sits in the first navamsa, since 0°45′ is inside the first 3°20′ band. The table on the page shows exactly that: Makara, 1/9, vargottama.
One more, to show the general case. The Sun at 22°27′ of Vrishabha has longitude 30 + 22.45 = 52.45°. Divided by 3.3333 that is 15.7, whole part 15, and 15 mod 12 = 3, which is Karka. Within its own sign, 22.45 ÷ 3.3333 is 6.7, so it is in the seventh navamsa. The table reads Karka, 7/9, and no vargottama because Karka is not Vrishabha.
Why the D9 needs a better birth time than the D1
A sign is 30° wide and the ascendant crosses it in roughly two hours. A navamsa is 3°20′ wide, so the D9 ascendant changes about every thirteen minutes.
That is the whole difficulty in one comparison. A birth time recorded to the nearest fifteen minutes, which is a perfectly ordinary level of precision, is usually good enough to fix your rashi lagna and usually not good enough to fix your navamsa lagna.
The default chart on this page demonstrates it. For a 10:00 birth in Delhi on 7 June 1990, the rashi lagna is Karka and stays Karka across a window of ±15 minutes. The navamsa lagna over that same window runs from Dhanu through Makara to Kumbha: three different signs, and therefore three different D9 charts, from a birth time nobody would consider especially vague.
The page reports this directly. It computes the D9 lagna at your stated time and at both ends of your stated uncertainty, and if they disagree it says so and names the range. It also points out when your rashi lagna is stable over the same window, because that is the situation people misread: the D1 looks settled, so the D9 is assumed to be settled too.
None of that says the D9 is wrong. It says that a D9 cast from a rounded birth time is a guess with a chart drawn around it, and you are entitled to know which of the two you are holding.
The numbers, laid out
It is easier to see the sensitivity as a table than as an argument.
| Rashi (D1) | Navamsa (D9) | |
|---|---|---|
| Width of one division | 30° | 3°20′ |
| Ascendant crosses it in | about 2 hours | about 13 minutes |
| Divisions in the zodiac | 12 | 108 |
| A 15-minute error moves the lagna by | a fraction of a sign | roughly one whole navamsa |
| A 1.45° ayanamsa change moves it by | about a tenth of a sign | about half a navamsa |
Nothing in that table is a claim about astrology. It is the consequence of dividing by nine, and it applies whatever you think the chart means.
Vargottama
A graha is vargottama when it occupies the same sign in the D1 and the D9.
The arithmetic makes the frequency exact. Each sign contains precisely one navamsa that maps back to itself, so vargottama covers exactly one ninth of the zodiac. Which navamsa it is depends on the sign's type: the first navamsa for movable signs, the fifth for fixed, the ninth for dual.
Two consequences follow, and they pull in opposite directions.
Any given graha has about an eleven per cent chance of being vargottama, so in a nine-graha chart you would expect about one on average. Finding one is unremarkable. Finding none, or finding four, is the mildly interesting case.
And because the vargottama band is only 3°20′ wide, whether a graha qualifies is sensitive to the ayanamsa. Lahiri and Raman sit about 1.45° apart, which is nearly half a navamsa, so a graha near the edge of its vargottama band can gain or lose the label purely by switching systems. The selector above lets you check your own case.
Reading the two charts together
The D9 is not a replacement for the rashi chart and does not answer the same questions. It is conventionally read as a second view of the same positions, which is why this page draws both side by side rather than making you open two tabs.
A few structural facts worth having straight, all of which are arithmetic rather than interpretation.
Houses in the D9 are counted from the D9 lagna, not from the rashi lagna. That is what makes it a chart rather than an annotation. Some sites instead overlay D9 signs onto D1 house positions, which produces a different result from identical positions, and is worth checking if two navamsas disagree.
A graha's D9 house is generally not its D1 house. On the example chart above, not one of the nine grahas keeps its house number between the two. That is normal and is the point: if the D9 reproduced D1 house numbers it would carry no additional information at all.
The D9 moves nine times faster. Over any stretch of zodiac, the navamsa sign changes nine times as often as the rashi does. Everything that makes the D1 sensitive makes the D9 nine times more so, including the ayanamsa, the birth time, and the time zone applied.
Why nine, and what 108 is doing here
Two questions people reasonably ask about the scheme itself.
Why nine divisions? The tradition gives the ninth division special standing among the vargas, and it is the one chart cast alongside the D1 as a matter of routine rather than on request. What can be said without interpretation is that nine divides the sign into parts of 3°20′, which is exactly the width of a nakshatra pada. So the navamsa boundaries and the pada boundaries are the same boundaries. A graha's navamsa and its nakshatra pada are two labels for one position, which is why the two are often quoted together and why moving one moves the other.
And 108? Twelve signs of nine divisions gives 108 parts, and 108 recurs throughout Indian tradition in ways that have nothing to do with astrology. Here it is simply 12 × 9, arrived at by arithmetic rather than chosen for its own sake. The pleasing number is a consequence of the division, not a reason for it.
The pada connection has a practical use. If you know a graha's nakshatra and pada, you already know its navamsa without any further calculation, because the twenty-seven nakshatras of four padas each are the same 108 divisions counted a different way. The tool reports both on the rashi chart page for that reason.
Why another site gives you a different navamsa
Every reason a rashi chart can differ applies here, amplified.
The birth time, as above, and this dominates everything else. Thirteen minutes is a whole navamsa.
The ayanamsa. A 1.45° difference is about a tenth of a rashi but nearly half a navamsa, so grahas comfortably inside their sign in the D1 can still land in different navamsas.
The time zone applied. For Indian births before 1955 this can be nearly an hour, which is four or five navamsas of ascendant.
Where the houses are counted from, as described above.
Mean against true lunar node, worth up to about 1.5° for Rahu and Ketu, which again is nearly half a navamsa.
The pattern is consistent: differences that are marginal in the D1 are decisive in the D9. If two rashi charts agree and two navamsas do not, that is expected behaviour rather than evidence that one site is broken.
Checking a navamsa you have been handed
Three quick tests, in order, before reading anything into a D9.
Confirm the birth time is known to better than a quarter of an hour. If it is not, the D9 lagna is unsettled and the house numbers are provisional. Everything else can wait until that is resolved.
Check Rahu and Ketu are six houses apart. They are always exactly opposite, in every chart and every divisional chart, because they are opposite points by definition. If the D9 shows them anywhere other than six houses apart, the chart is broken and nothing else about it is worth checking.
Check the vargottama count is plausible. About one graha in nine is vargottama, so a nine-graha chart averages one. Zero is common, two is unremarkable. A chart claiming six or seven vargottama grahas is far more likely to be miscomputed than remarkable, and is worth recomputing elsewhere before anyone reads anything into it.
What this page will not do
It will not tell you what your navamsa says about your marriage.
The D9 is strongly associated with marriage and with the spouse in popular practice, and a great deal is sold on that association. This page computes which of the 108 divisions each graha occupies, which sign that division maps to, and which house that sign holds relative to the D9 lagna. Every one of those steps is arithmetic you can check by hand with a calculator and the formula above.
What the tradition builds on top of it is a separate question. The one thing worth saying is structural rather than interpretive: because a navamsa is thirteen minutes of ascendant wide, any reading that rests on the D9 lagna rests on a birth time known to better than about a quarter of an hour. If yours is not known that well, the honest position is that the chart is provisional, and that is what the page will tell you.