my RTC RV-8803 showing strange dates

Development Boards ESP32
1

Hi, I have been working on a project using this RTC.

https://www.sparkfun.com/products/16281

It has been working great up until just a few days ago when it has started printing its human readable time with things like this:

RTC Time: 25/45/2021 33:78:11PM

An example of an epoch timestamp that is coming from it shows that corresponds to a date in the future:

2304696526 → Monday, January 12, 2043 5:28:46 PM

I ran an example sketch that uses NTP to set the RTC to correct time. This worked for a while, but a few hours later it had started reading the invalid date again. Once again I reset it using the NTP code and it has since kept accurate time. I have no idea what would have caused it to leap years into the future like that.

Any chance this RTC board has some kind of bug in its firmware or something? Anyone else reporting similar problems?

Thanks,

-Matt

2

A few months ago I had the same bad experience with a simple DS1307 module. As far as I can remember there was a low voltage issue. Have you checked the voltage of the coin cell battery? Is it giving accurately 3V?

3

I have not checked the coin cell voltage. However, I just bought this board a couple months ago. Also, since resetting it via NTP the other day it has continued to work normally.

Other than that jump into the future the board seems to still accurately measure the passage of time. I just can’t seem to figure out what would cause it to shift like that.

4

As a followup, here is some interesting output from a recent test. I have the Thing Plus C sitting on my desk being powered by a 3.7V LiPo. My code sleeps for an hour, checks the battery voltage, and if it is >10% goes back to sleep for another hour. (I wanted to see how many days this would run and the answer is ~12 days.)

I was keeping timestamps in a logfile, with the timestamp coming from my RV-8803. It looks like it was keeping perfect time with the exception of two anomolies:

12/30/2021 07:16:40AM: Target time is still in the future.

12/30/2021 07:16:41AM: sleep time is: 3600

12/30/2021 08:16:18AM: NO emergency

165/30/20165 08:16:18AM: Voltage: 3.91 V

12/30/2021 08:16:18AM: Percentage: 65.57 %

12/30/2021 08:16:18AM: Alert: 1

12/30/2021 08:16:18AM: The nap time in microseconds is: 3600000000

12/30/2021 08:16:18AM: Started sleep : 3577 seconds ago

12/30/2021 08:16:18AM: Target time: 1642573782

12/30/2021 08:16:18AM: Time stamp: 1640852178

12/30/2021 08:16:18AM: Time left: 1721604

12/30/2021 08:16:18AM: Target time is still in the future.

12/30/2021 08:16:18AM: sleep time is: 3600

12/30/2021 09:16:05AM: NO emergency

165/165/20165 153:165:05PM: Voltage: 3.91 V

12/30/2021 09:16:05AM: Percentage: 65.28 %

12/30/2021 09:16:05AM: Alert: 1

12/30/2021 09:16:05AM: The nap time in microseconds is: 3600000000

12/30/2021 09:16:05AM: Started sleep : 3587 seconds ago

The date/timestamps all look correct except for those two above which start with 165. This seems to not have had any negative impact on how my project behaved, but its a little troubling when the clock is not reporting sane values. And once it has gone nuts like this, why would it go back to being correct?

I’m running one more test on this setup and then I will check the voltage of my RTC battery and replace it if necessary.

Any idea what’s going on here?

Thanks.