The arduino is only "gobbling power" during measurements.
Which would be at worst for 2 seconds every 10 minutes (in my case).
The 4.8µA (measured) quiescent current is with Arduino, DS3231 RTC, SD Card module and sensors connected.
This is a very advanced topic. There is a lot of erroneous posts about it. A few are correct. With my 50+ yrs general experience and several years of interest in low power I have a good but not perfect understanding.
The ZS-042 is NOT a real RTC, it has many problems. Look at either the bare DS3231 chip or the Chronodot board. The latest version even has optional EEPROM.
Good luck.
Also look for Cave Pearl.
@sonofcy
Please enlighten me.
The ZS-042 uses the same DS3231SN chip from Analog Devices / Maxim Integrated.
How is it that a ZS-042 module is not a RTC but the bare DS3231 chip and the Chronodot board are?
P.S.
I know the ZS-042 has issues with the "faulty" charge circuit for the LIR2032.
But it can be "fixed" by removing one resistor. (Disabling the charge circuit)
The thing is (here locally) a naked DS3231SN costs THREE TIMES more than a whole ZS-042 board with battery! https://www.segor.de/#Q=DS3231SN https://www.segor.de/#Q=RealTimeClock-Modul
It has to be modified by cutting traces. MOST users do NOT know that. There are many articles written about it. One of the worst things is it comes with a charge circuit BUT a non-rechargeable battery. I don't think it will explode, but it tells you how badly it's engineered.
Much better to use the Chronodot or bare chip.
I have even seen some engineering purists first cut the I2C traces, then solder on 'better' resistors.
I just feel my time is better spent building rather than re-engineering a faulty board. I bought 8 Chronodots at $1.90 each, hardly a cost consideration.
I spent a lot of time searching. The information is out there. The resistor packs have to be cut. Just get a Chronodot, so much easier AND it works correctly. Good luck.
That doesn't make sense. Just desolder the upper SMD 4x4k7/1206 resistor network. And add external pullups. Some people seem to be a little bit too eager with the scalpel.
You might be right, remember I am 83, so sometimes I have a hard time remembering details from a couple of years ago. If the extra work is ok and the timer interrupt works, so you have true RTC functionality, then go for it. For me, there is also a size issue. SEEED boards are tiny, and my best small design requires a physical design of approx 25mm x 25mm up to 50mm, depending on LiPo size and final implementation current needs.
Sorry, I'm only 53.
The timer interrupt works, even without VCC.
You just have to remove the upper resistor network and pull up SQW separately from VCC.
Then supply VCC on demand if you want to communicate with the RTC.
No fancy circuit reworks like in DS3231 RTC on ZS-042 module - if low power is important
needed
@ sonofcy
Well...
I did do my research (for weeks).
I did do my own tests.
I find it quite insulting to be told that I'm "someone who has not spent many hours tracking down all the information
So please stay on topic or stay away from the discussion.
