Downloaded calendar shows different candle lighting and Havdalah times than online version

Tzipora 3 years ago updated by Steven DuBois 2 years ago 16

I use Hebcal each year to set up the calendar for different schools I work with. This year, I am noticing that when I download the PDF of the calendar and check it against the calendar online, the candle lighting and Havdalah times start off the same in January, but begin to show some 1-minute discrepancies in April and May, increasing throughout June and July, reaching 3-minute discrepancies in August and September, then decreasing again from October through December. It seems like the PDF is being generated using a different set of parameters than the online version. 

Has anyone else noticed this? Is this an existing issue with a known workaround or a new issue?



Hi Nathan, thanks for using Hebcal and thanks for your message!

Differences of 2 minutes are fully expected and are not considered a bug. Please read this detailed post for more details:


You're right! My mistake

Hi Steven, thanks for sending the screen shots. It looks like you're using different Havdalah options for Skokie, IL 60076.

In the web browser screenshot, you've picked the Havdalah tzeit hakocahvim option (when the sun is 8.5 degrees below the horizon). You can see this because it says "Havdalah" but no fixed number of minutes in parentheses after.

In the PDF view, you've selected the Havdalah fixed 42 minutes option. You can see this because it says "Havdalah (42 min)".

If you choose the same Havdalah setting for both, you will see consistent times.


I am noticing that there is still sometimes a ~1 minute difference between the havdalah times displayed on the website and in a PDF, at least when using a fixed number of minutes. See these screenshots as an example for the recent zmanim for 60076. 





Hi Nathan, thanks for using Hebcal and thanks for your message!

Differences of 2 minutes are fully expected and are not considered a bug. Please read this detailed post for more details:


Not sure if this should be a new topic, but the issue still seems to exist with havdalah times.  For example, this coming Shabbat (Beshallach 5781) with the ZIP code 10023, Hebcal shows a havdalah time of 5:52PM.  Myzmanim shows 5:55pm (really, 5:54:17).  This sun-position-calculator https://www.sunearthtools.com/dp/tools/pos_sun.php?lang=en shows that at 5:52 the sun is only at 8.1 degrees at 5:52pm, and is at 8.5 degrees between 5:54 and 5:55pm.  The Python zmanim library (https://pypi.org/project/zmanim/) has it at about 5:55 too (5:54:48, later than myzmanim by about 30 seconds, which could just be slightly different locations for 10023 vs my lat/lon).

Thank you so much for the update! I have been monitoring the situation and really appreciate that you took the time to let us know. And thank you for the service you provide!

Hi Tzipora and Yaakov,

Thanks again for your patience and feedback.

Another update: after some additional research, we've come to the conclusion that the sunset algorithm we were using in the new version of Hebcal wasn't accurate enough. As of yesterday, we have swapped out the sunset engine for a version that matches what we previously published on Hebcal.com and most other sources (including OU).

Some of the Hebcal event feeds are cached up to 1 week, so subscriptions on Google Calendar, iPhone, etc might not refresh until next Sunday August 23.

Thank you so much for the update, Michael!

I see that the times across the website and downloads are now consistent with each other, and seem to match the "new" set of parameters, where the times slowly start to get later and later, up to 3 minutes in August and September, and then the gap  between the new candle-lighting times and the original candle-lighting times narrows back down. 

When I compare the Hebcal times to the OU Zmanim, the new ones are the same or 1 minute different in January, but 2-3 minutes different in September. In comparison, the Hebcal times that were generated a year ago were consistently the same or only 1 minute different than the OU Zmanim.

What is the reason for the new seasonal difference?


Hi, thanks for your patience!

We've finished our website upgrades and now the candle-lighting times are consistent across the website and the downloads/subscriptions in Google Calendar and other calendar apps that consume hebcal.com feeds. Calendar apps like Google or the iPhone tend to refresh calendar subscriptions approximately once a week, and it's now been long enough since we updated our servers that all calendar subscription feeds should be fully refreshed.

Thanks again for using Hebcal!


The reason why I noticed it this past week is because there was a discrepancy of 6 minutes! 


Depending on the week, it varies by more than 1-2 minutes

I downloaded the calendar directly from Hebcal into Google Calendar for my zip code (11210), so in theory, they should be exactly the same.

July 3: Hebal is 8:12 and Google calendar is 8:13, so it's late by 1 min
July 10: Hebal is 8:09 and Google calendar is 8:11, so it's late by 2 mins
July 17: Hebal is 8:05 and Google calendar is 8:11, so it's late by 6 mins

July 24: Hebal is 8:00 and Google calendar is 8:03, so it's late by 3 mins
July 24: Hebal is 7:53 and Google calendar is 7:55, so it's late by 2 mins

Under review

Thanks for using Hebcal. We have recently begun upgrading our website and we've introduced a new, more accurate candle-lighting times engine. We're working hard to complete the upgrade, which will remove any remaining inconsistencies between what's displayed on the website and the PDF/downloads.

In the meantime, please note that differences of 1 or 2 minutes are fully expected. Candle-lighting and Havdalah times can only be approximated and are never 100% precise. To determine candle-lighting times, Hebcal uses latitude and longitude, which is in derived from the city's geographic and time-zone information.

If you ever have any doubts about Hebcal's times, consult your local halachic authority. 

Thank you so much for this resource, and for responding and updating us as to the status of the upgrades taking place! 
Meanwhile, which times are accurate? Are the times on the website correct? Or are the times in the PDF correct? Or even the subscribed calendars, as shown below?

While differences of 1 or 2 minutes may be expected, it's unclear which set of times is being generated by the updated/most accurate algorithm - or if none of them is currently reliable. If you can let us know which set of times to use right now, it would be extremely helpful. 

Thank you so much!

Yes! I am having this problem too! Please let me know if there is a solution - kind of a pain.