Ed van den Enden

@going @djurny @IBV Reason why I need an working hardware clock: Because my access point need to work without internet connection, but with always the right date and time. Following made the battery operated hardware clock ds3132 to work: OS: Armbian Bookworm from 24-04-2025 apt-get update (No upgrade! ) Edit /boot/armbianEnv.txt add following lines: overlays=i2c0 dtoverlay=i2c-rtc,ds1307 Edit /etc/modules add following lines i2c-dev aml_i2c Edit /etc/rc.local add following lines echo ds1307 0x68 > /sys/class/i2c-adapter/i2c-0/new_device hwclock -s Reboot Check date and eventual correct date and time. Check if the i2c has the right status with: i2cdetect -y 0 ==> address 068 should show UU Check hwclock is communicating with: hwclock --verbose Save the right system time to the hwclock with: hwclock -w Check if you can read hwclock with: hwclock -r

@djurny I believe the first code you show me is in "C" and is a ds3231 driver. The missing driver in the today kernel. (Kernel: 6.12.23-current-sunxi) This driver would work for me but I have no clue how to compile. The rtc-sync program, makes sense to me but how can I make it to run?

For my complete log and Banana pi M2 Zero data, see: https://paste.armbian.com/yusewapuzi Kernel: 6.12.23-current-sunxi RTC module ds3231, connected to pin 3 and 5 (i2c0) 3.3V from 1 and 9 overlays=sun8i-h3-i2c0 dtoverlay=i2c-rtc,ds1307

I will leave it for what it is for now. I will move back to RPi zero 2W, solder an antenna, and I'm good to go. Would have loved to go to the BPI m2z and later to the BPI m4z, but for now not an option. By the way, with the latest Armbian upgrade an hour ago, my USB camera is also not working anymore. Thank you very much for the support. Kind regards, Ed👏

Nop, that didn't work had also both i2c0 and i2c1 activated root@clearview:~# i2cdetect -y 0 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: -- -- -- -- -- -- -- -- 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 30: 30 -- 32 33 34 35 36 -- -- -- -- -- -- -- -- -- 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 50: 50 51 52 53 54 55 56 57 58 59 5a 5b 5c 5d 5e 5f 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 70: -- -- -- -- -- -- -- -- root@clearview:~# i2cdetect -y 1 Error: Could not open file `/dev/i2c-1' or `/dev/i2c/1': No such file or directory root@clearview:~# ls /dev/i2c-* /dev/i2c-0

Indeed, the rtc_ds1307 is compatible with the ds3231 and removing pps-gpio w1-gpio did the errors disappear. The RTC module still doesn't work. 😭

I notice that there is a linux driver available at GITHUB for the ds3231, but I have no clue how to compile/install this driver. https://github.com/AxelElRojo/ds3231-linux

The below output with the RTC module taken off the BPI M2Z Further below I saw that there were some pin errors. (most probably nothing to do with this topic) https://paste.armbian.com/axoduwanuq [ 1.020506] usbcore: registered new interface driver usb-storage [ 1.022284] sun6i-rtc 1f00000.rtc: registered as rtc0 [ 1.022371] sun6i-rtc 1f00000.rtc: setting system clock to 1970-01-01T00:00:04 UTC (4) [ 1.023147] i2c_dev: i2c /dev entries driver [ 1.025025] sunxi-wdt 1c20ca0.watchdog: Watchdog enabled (timeout=16 sec, nowayout=0) [ 7.721526] sun8i-h3-pinctrl 1c20800.pinctrl: pin PD14 already requested by onewire@0; cannot claim for pps@0 [ 7.721577] sun8i-h3-pinctrl 1c20800.pinctrl: pin-110 (pps@0) status -22 [ 7.721597] sun8i-h3-pinctrl 1c20800.pinctrl: could not request pin 110 (PD14) from group PD14 on device 1c20800.pinctrl [ 7.721613] pps-gpio pps@0: Error applying setting, reverse things back

Another remark/question: The latest Armbian does not show a driver RTC-ds3231, just the RTC-ds1307 driver. Therefore I used fo my ds3231 RTC module the available driver RTC-ds1307. Could that be the reason for the malfunction of the hwclock?

To my understanding and the documentation I found, by investigating the RTC address, with the outcome 068, it indicates that the RTC driver is not loaded, otherwise the address request would show UU as outcome.

When connected to WiFi root@clearview:~# timedatectl status Local time: Wed 2025-04-23 09:56:59 CEST Universal time: Wed 2025-04-23 07:56:59 UTC RTC time: Wed 2025-04-23 07:56:58 Time zone: Europe/Brussels (CEST, +0200) System clock synchronized: no NTP service: active RTC in local TZ: no After power down and disconnected from WiFi root@clearview:~# timedatectl status Local time: Wed 2025-04-23 10:04:43 CEST Universal time: Wed 2025-04-23 08:04:43 UTC RTC time: Wed 1970-01-01 00:02:45 Time zone: Europe/Brussels (CEST, +0200) System clock synchronized: no NTP service: active RTC in local TZ: no

Just measured again the battery on the battery pins of the RTC-ds3231, and it shows me 3.6 volt (not the BPI 3.3 volt pins (1 + 9)

See link https://paste.armbian.com/alehuretiw Now connected to WiFi

Indeed, the hardware clock has been set and checked before power off. 1) date -s "April 22, 2024 11:54:00" 2) hwclock -w 3) hwclock -r 4) battery hwclock 3.6V 5) works flawless with RPi Zero 2W 6) i2cdetect -y 0 shows address 068 Why did you modify the hwclock-set script ? Suggested by many, but without modifying this script, it does not change mallfunction of the hwclock. When connected to WiFi hwclock --verbose tells me following: hwclock from util-linux 2.38.1 System Time: 1745389217.903091 Trying to open: /dev/rtc0 Using the rtc interface to the clock. Last drift adjustment done at 1745317708 seconds after 1969 Last calibration done at 1745317708 seconds after 1969 Hardware clock is on UTC time Assuming hardware clock is kept in UTC time. Waiting for clock tick... ...got clock tick Time read from Hardware Clock: 2025/04/23 06:20:20 Hw clock time : 2025/04/23 06:20:20 = 1745389220 seconds since 1969 Time since last adjustment is 71512 seconds Calculated Hardware Clock drift is 0.000000 seconds 2025-04-23 08:20:18.895813+02:00 After power down and When disconnected from WiFi (my application and so the reason why I need a hwclock to work) hwclock --verbose tells me following: hwclock from util-linux 2.38.1 System Time: 1745389772.399976 Trying to open: /dev/rtc0 Using the rtc interface to the clock. Last drift adjustment done at 1745317708 seconds after 1969 Last calibration done at 1745317708 seconds after 1969 Hardware clock is on UTC time Assuming hardware clock is kept in UTC time. Waiting for clock tick... ...got clock tick Time read from Hardware Clock: 1970/01/01 00:04:02 Hw clock time : 1970/01/01 00:04:02 = 242 seconds since 1969 Time since last adjustment is 1745317466 seconds Calculated Hardware Clock drift is 0.000000 seconds 1970/01/01 01:04:00.760027+01:00

I'm unable to make an RTC module to work: RTC module ds3231, BPI M2Z and ARMBIAN Bookworm i2c pins 3 and 5 Driver RTC-ds1307 loaded, because RTC-ds3231 is not available. Why do I need a RTC-module? Often in the field I'm not online but I do need the right time on my observation camera's. My application does not allow me for having a monitor. My system does the recordings by motion detection, and the recordings do need the right time on it. armbian-config i2c0 set Tools installed: apt install python3-smbus i2c-tools i2cdetect -y 0 Detects address 068 and not UU as proof of driver been loaded File "hwclock-set" adapted #if [ -e /run/systemd/system ] ; then # exit 0 #fi Fake hwclock removed sudo apt -y remove fake-hwclock sudo update-rc.d -f fake-hwclock remove armbianEnv.txt overlay and dtoverlay correctly set modules.conf added: rtc-ds1307 Still, after a power down, the hardware clock shows 1970 as the date The same module tested on my RPi Zero 2W, and all is working perfect. If somebody can help me, that would be wonderful.