Every few months a family finds that two calendars, or two relatives, name different days for the same Ekadashi, and sometimes one printed panchang lists it against two dates. Nobody has made a mistake. Ekadashi is a tithi, a lunar day, and a lunar day does not begin at midnight or at sunrise. A fast, though, is kept on a calendar day, so someone has to decide which calendar day the tithi belongs to. This page shows why the two do not line up, the rule most calendars use, where Smarta and Vaishnava calendars differ, and how to check the day for your own city. It is one of the life questions this site answers by showing the working, not by deciding for you.
A tithi is measured by the Moon, not the clock
A tithi is defined by the angle between the Moon and the Sun. This app's engine takes the Moon's sidereal longitude, subtracts the Sun's, and divides the result by 12 degrees: the first 12 degrees are tithi 1, the next 12 are tithi 2, and so on to 30. Tithis 1 to 15 make the bright half, Shukla Paksha, and 16 to 30 the dark half, Krishna Paksha. Ekadashi, the eleventh, is tithi 11 in the bright half and tithi 26 in the dark half, so there are two in each lunar month.
Because the Moon does not move at an even speed, the time it takes to gain 12 degrees on the Sun changes from tithi to tithi. In this app's computation for New Delhi, the Devutthana Ekadashi of November 2026 began at 7:18 AM on 20 November and ended at 6:34 AM on 21 November, a span of 23 hours 16 minutes. The Vijaya Ekadashi of March 2027 began at 4:47 AM on 3 March and ended at 7:27 AM on 4 March, a span of 26 hours 40 minutes. Neither begins or ends at sunrise, and neither is exactly a day long.
The sunrise rule, and the three cases it cannot settle alone
A printed panchang names, for each calendar day, the tithi that is current at local sunrise. Most Ekadashi lists start from the same idea: the fast falls on the day whose sunrise falls inside Ekadashi. This site's Ekadashi date pages use exactly that rule for New Delhi. Most of the time it gives one clear answer, because a tithi of 23 to 27 hours usually contains one sunrise. Three cases do not:
| Case | What happens | Example (New Delhi, this app) | What the date pages show |
|---|---|---|---|
| Two sunrises | Ekadashi starts just before one sunrise and ends just after the next | Vijaya Ekadashi: 4:47 AM on 3 Mar 2027 to 7:27 AM on 4 Mar 2027; sunrises 6:45 AM and 6:44 AM | Both days, with a note to follow your own calendar |
| No sunrise | Ekadashi starts after one sunrise and ends before the next | Devutthana Ekadashi: 7:18 AM on 20 Nov 2026 to 6:34 AM on 21 Nov 2026; sunrise 6:48 AM | The day it begins, 20 Nov, with the Vaishnava day, 21 Nov, named |
| Short Dwadashi | The next tithi, Dwadashi, holds no sunrise | Pausha Putrada Ekadashi: listed on 19 Jan 2027; Dwadashi ends 4:43 AM on 20 Jan 2027 | 19 Jan, with the day earlier, 18 Jan, named |
The first case is what most people mean by "Ekadashi on two days": the same tithi really is current at both sunrises, so the sunrise rule alone names both. The second case is the reverse: the tithi is squeezed between two sunrises and owns neither. The third concerns the day after, because the fast is broken in Dwadashi, and a Dwadashi that holds no sunrise leaves no Dwadashi morning to break it in; calendars that break the fast in Dwadashi move the Ekadashi a day earlier. In the 2026 and 2027 New Delhi lists, these flags fall on Padmini (26 or 27 May 2026), Yogini (10 July 2026, Vaishnava 11 July), Devutthana (20 November 2026, Vaishnava 21 November), Pausha Putrada (19 January 2027, or 18 January), Vijaya (3 or 4 March 2027) and Kamika (29 July 2027, Vaishnava 30 July).
There is a fourth, smaller reason: arithmetic. Two almanacs that compute the Moon a few minutes apart can disagree on a day when the tithi changes very close to sunrise. This site's date pages mark any day where the tithi changes within 10 minutes of the deciding sunrise, and its notes record that this engine's tithi times run 2 to 3 minutes later than one widely used published panchang. The Yogini Ekadashi of July 2026, which ends at 5:25 AM against a 5:31 AM sunrise, carries that mark.
What the classical texts say about Ekadashi
This site's classical sources are BPHS, Saravali, the Bhrigu Sutras, Daivajna Vallabha, Jyotisha Siddhanta Sara and Advanced Predictive Astrology Vol. II. None of them states an Ekadashi fasting rule, the sunrise rule, or a Smarta or Vaishnava rule. A search of all of them for "Ekadasi", "Smarta", "Vaishnava", "Dashami", "vrat" and "fast" finds no verse about which day to fast. These rules belong to the vrat literature and to living calendar traditions, which this site does not hold, so this page reports them only as calendars apply them.
The texts do count tithis. Saravali Ch. 5 Sl. 16-17 reports a view of "the elders among Yavanas" that grades the Moon's strength by lunar day: medium for the ten days from the first of the bright half, "exceedingly strong" from "the 11th day, or Ekadasi of Bright Lunar Half" through the fifth of the dark half, and of little strength from the sixth of the dark half to Amavasya. BPHS Ch. 92 Sl. 2 names the first two ghatikas of the Nanda tithis (the 1st, 6th and 11th) and the last two of the Purna tithis (the 5th, 10th and 15th) as Tithi Gandanta, a junction point, in a chapter about births. So the texts treat the 11th tithi as a span with a beginning and an end, and the meeting of the 10th and 11th as a named point; neither verse is about fasting.
The only other appearance of the word is Saravali Ch. 20 Sl. 35, which says Venus rules "the mendicants of Saivite and Vaishnavite sects", in a chapter on renunciation. It says nothing about calendars.
Smarta and Vaishnava calendars
On most Ekadashis every calendar agrees. Where they part, this site's date pages state the difference as follows: Smarta and Vaishnava calendars apply different rules where Dashami, the tenth tithi, touches the Ekadashi, and Vaishnava calendars, ISKCON among them, can then keep the next day. In the no-sunrise case above, the Ekadashi of 20 November 2026 begins at 7:18 AM, after a sunrise that was still in Dashami, so the date pages list 20 November and name 21 November as the day Vaishnava calendars keep.
This app does not compute the Vaishnava test itself; it computes the tithi at sunrise and flags the days where the traditions are known to part. If you follow a Vaishnava calendar, its own published date is the one to keep. If you follow your family's Smarta panchang, its date is. Neither is a correction of the other: they are two rules applied to the same sky.
How to check this in your own panchang
- Open the Panchang calculator, enter the date and the city where you will keep the fast. With a place, it reads the tithi at that city's sunrise and shows the sunrise time.
- Enter the day before and the day after as well. If the tithi at sunrise reads 11 (or 26) on two days in a row, Ekadashi holds two sunrises in your city. If it jumps from 10 to 12 (or 25 to 27), Ekadashi held no sunrise there.
- Read the paksha beside the tithi: Shukla for 11, Krishna for 26. The name of the Ekadashi depends on the lunar month, which the date pages give.
- For the year's list, open this site's Ekadashi date pages for New Delhi from the vrat dates section; each flagged day carries a note naming the other day calendars use.
- If you are planning an event around the fast, the Muhurat calculator scores time windows by tithi, nakshatra, paksha and Rahu Kaal for an activity. It does not pick fast days, and Ekadashi is not on its list of tithis it marks down.
For more on reading a day's entry, see how to read a panchang, and for the general date question, when is Ekadashi and how is the date decided. The muhurat feature explains how this app uses the panchang for timing.
How to choose between the two days
- Follow your tradition's calendar. If your family or temple follows a Vaishnava calendar, keep its date; if it follows a Smarta panchang, keep that one. This is the answer the date pages themselves give for the two-sunrise case.
- Use the calendar made for your city. A list for New Delhi can be a day off for Chennai, and more often for a city abroad with its own clock.
- Check that your calendar gives the parana (fast-breaking) time for the next morning. This site gives the fast day and when Ekadashi begins and ends, but not the parana window.
- When the two days are flagged as close, a few minutes of difference between almanacs decided it; either choice follows a real rule.
Common misreadings
- "One calendar is wrong." Usually both are right by their own rule: they share the tithi times and differ on how to place them.
- "Ekadashi is the 11th day after the new moon." It is the 11th tithi, which can start at any hour; counting calendar days from Amavasya can land on the wrong date.
- "Two days means two fasts." The two-day listing names one Ekadashi that touched two sunrises. Calendars keep the fast on one of them.
- "The date is the same everywhere." The tithi changes at the same instant everywhere, but sunrise does not, so the day can differ by city.
What astrology cannot tell you about the Ekadashi day
The computation can tell you when the tithi begins and ends and whether it holds a sunrise in your city. It cannot tell you which tradition to follow; that is a matter of family, community and faith, not of astronomy. It cannot tell you that one day earns more than the other, and this site's sources make no such claim. It does not compute the Vaishnava test or the parana time, so for those, your tradition's published calendar or your temple is the place to ask. And it is not health guidance: whether you should fast at all, and how, is a question for your doctor, especially with a health condition, pregnancy or medicines. See can I keep the Ekadashi fast with a health condition. How this app computes the Moon and the Sun is set out in its methodology, and the books it draws on in its sources. A related question on lunar days is why Amavasya is considered inauspicious.
Questions people ask about Ekadashi on two days
- Which day should I fast when Ekadashi is listed on two days?
- The one your tradition's calendar names. When Ekadashi is current at both sunrises, calendars keep the fast on one of the two days by their own rule. This site lists both and asks you to check the calendar you follow.
- Why do ISKCON or Vaishnava calendars sometimes fast a day later?
- Because Vaishnava calendars do not fast on an Ekadashi that Dashami touches. On a day like 20 November 2026 in New Delhi, where Ekadashi begins after a sunrise still in Dashami, they keep the next day, 21 November.
- Can Ekadashi have no sunrise at all?
- Yes. A tithi shorter than the gap between two sunrises can fit between them. The Devutthana Ekadashi of November 2026 ran from 7:18 AM on 20 November to 6:34 AM on 21 November in this app's computation, missing both New Delhi sunrises.
- Why does my city have a different Ekadashi date from Delhi?
- Because sunrise differs by place. On 21 November 2026 Chennai's sunrise falls inside Ekadashi and Delhi's does not. Enter your own city in the Panchang calculator to see the tithi at your sunrise.
- Do BPHS or Saravali say which day to fast?
- No. They count tithis, and Saravali names "Ekadasi of Bright Lunar Half" in a verse on the Moon's strength, but no text in this site's corpus gives an Ekadashi fasting rule or a Smarta or Vaishnava rule.
- Does this site give the parana time?
- No. It gives the fast day and when Ekadashi begins and ends. The parana window on the next morning is not computed here; take it from your own calendar.
Daily Panchang
Tithi, paksha, the Moon's phase, choghadiya, hora and Rahu Kaal for any calendar date.
This date is sent only to compute the answer. Nothing is stored, no account is created, and no email is asked for.