Hello,
I'm using a Witty Pi 4 Mini paired with a Raspberry Pi Zero 2W for an automated image acquisition project.
My program works as follows: the system turns on every 20 minutes every day between 6 a.m. and 8 p.m. The system remains completely inactive between 8 p.m. and 6 a.m. The system is powered by a battery and a solar panel. The battery level is maintained between 80% and 100%. I currently have 25 prototypes of this system. I'm having two types of problems with my systems:
1- I have a recurring issue with one particular prototype: I’ve noticed that data is acquired correctly for a variable period (between 3 and 4 days), and then the system becomes out of sync. My program allows me to detect battery power failures. In the Witty Pi log, I specifically look for “power supply is newly connected” as an indicator that the battery has shut down and then reconnected. This detection is recorded in a system incident report. For this prototype, the incident report never mentions any battery failures, and yet the system is no longer set to the correct time (the images captured from 11 a.m. to 8 p.m. are completely black, which tells me they were actually taken at night). What could be the causes of desynchronization? I am currently conducting tests using a wall outlet to see if this problem persists without possible power interruptions.
2- For some other prototypes (in a somewhat random manner and without repetition) I noticed that data collection can stop during the transition from Sunday to Monday (the data from Sunday evening at 8 p.m. was recorded, and then the Raspberry Pi never woke up again). However, when I manually check my prototypes, they are on time (so I think there is no battery issue for those ones). Has this happened to other users before? What could be the reasons why the WittyPi stops waking up the Raspberry Pi even though the program is running and the batteries are working?
Thank you for your responses.
Cheers.
For Witty Pi 4 (all models), the logs are stored on the Raspberry Pi side, so they are not guaranteed to contain the exact moment or cause of every failure — especially if the Pi did not start up or shut down normally.
However, checking the logs is always the recommended first step. Without the logs, there is unfortunately not enough information to analyze what happened, and any explanation would mostly be speculation.
Please check and provide the relevant wittyPi.log and schedule.log, preferably covering the period from the last normal operation to the first abnormal one. The schedule script you are using would also be useful.
With those logs, we can then see whether the RTC/system time changed unexpectedly, whether the next startup was scheduled correctly, and whether there were any other clues around the failure.