Update 29.07.18 - current consumption without SD
This won't surprise most of you, but the SD card makes up for like 99% of the power-down current consumption...
Current draw is <2µA in sleep mode when no SD is present. So I'm going to add the high-side switch again and do my tests/code rewrites till I manage to turn the SD off. On/off itself is just one line of code, but re-initializing the SD after power-off isn't exactly supported by the library. Seems to work for others, so I'll give it a shot.
Hardware-side of things is done by soldering the framed components and moving the solder bridge from # to *:
The new board layout is a success. I know that current measurements aren't the exact strength of the UT61E multimeter, but that's all I have at the moment. And in comparison to the old version I cut that current consumption (at least) by half!
The SD card used is a Transcend one with 16Gb (fairly new model) which is not switched on/off whatsoever.
The stated 73µA are in powerDown mode with BOD and ADC disabled. All peripherals are disabled as well (BLE module, DS3231).
So my next tasks are to modify the V1.1 program for the use of the:
- 512kBit EEPROM
- BLE module
What really annoys me is the use of Digital Pin 10 (hardware pin 14, SS on the Atmega328P-Au chip) when using SPI. It has to be switched high (and that's done by the SDfat lib) to work properly, even if you use another pin for your SD select function.
One of the LEDs is connected to that pin and is always on. I had to desolder it for the whole thing to work. Bummer!
Now I'll just use one LED or just occupy one of the unused digital ports for it.
Anyway, with this setup I'll get much more than one year out of one 18650 li ion cell: NICE!