Do you have a way to measure the current in the DS3231's Vcc line? Nominally, it's about 100uA for the processor itself, and then if you're using a module of some kind, whatever the pullups might draw and the power indicator LED if present. It's just hard to see why such a large cap would be needed to make it work right. The traditional ZS-042 module has two caps on that line - 0.1uF and 1uF, but it also has an EEPROM chip.
Could you try powering the RTC's Vcc and pullups from a GPIO pin? Then you could intialize I2C before bringing up Vcc. I have an AVR project that only brings up Vcc when communicating with the RTC, which otherwise runs on Vbat. Seems to work fine. And The GPIO is active low when not high, so the pullups are actually pulldowns, and it still works when everything goes back high.
It just seems like there must be large currents flowing somewhere that shouldn't be.
I think this may be the case as I'm seeing this behavior, though less prevalent even without the coin-cell battery present. If I measure the current on start-up I see a spike of 100mA after which it settles to around 24mA and if I remove the cap the spike drops to around 70mA, likely from the other caps I have on the PCB that charge during power on.
I believe I need to implement a soft-start circuit to limit this spike, correct? Any recommendations on the soft-start circuit design/components? (my circuit is 5V powered and will draw 1.4 Amp max when under load)
Home
Help
Search
Login
Register
