What exactly will trigger an RTC reset ?

[Zingemneire]

Hi,

In my first tests it appeared to be behaving strangely but in the end it appears to work well, this is what I think happens:

1) Each time the unit is powered up from a state of being totally without power that seems to trigger a complete RTC reset: the seconds and microseconds counters start from 0. Nothing wrong with that and it would be entirely logical but I would like to be sure that is the case.

2) I have also found that , while keeping the ESP32-WROVER module under permanent power, if I flash new software which ends with the "Hard resetting via RTS pin..." the RTC counters appear to be reset as well. There is nothing wrong with that but I have been unable to find any documentation that states it does so. Hence the question: is the observed behaviour exactly as it should behave?

3) Other than that I am wondering if there any other reasons that may trigger an RTC reset, assuming the module remains permanently battery powered, as it will be in our application. I assume there are not but again I would like to be sure.

Any help/info much appreciated.

[Zingemneire]

Correction: further tests show that a hard reset sets the clock to value 0xFC3D9900 and not to 0 as I assumed.

That causes my software to detect the clock's EPOCH value is higher than 0x80000000 so it sets the EPOCH value to 01/01/2018 at 00:00:00.

The value of 0xFC3D9900 does not appear to be a random value as I can reproduce it each and every time.

Edit:

Yep, it is not a random value it is -63072000 which is used in my software to represent a reference EPOCH time of 1/1/2000 at 00:00:00 as if it were a Unix/Linux based epoch time of 01/01/1972 at 00:00:00 so the internal RTC counters are effectively reset to 0 whenever there is a reset.

In the mean time I have found all of the relevant information in section 3 of the "esp32_technical_reference_manuel_en.pdf file" which describes the complete unit's reset behaviour in great detail.