Hi everybody!
Our organization recently deployed 21 AudioMoths for recording birds, bats and grasshoppers. They were set to record 3 times 2 hours per day during one week, and encased in the waterproof case. Batteries are no older than one year and were fully charged a couple of days before deployment. The AMs sat in the fields for maybe 3-4 days before recording periods started.
After getting the AMs back from the field, I noticed that about only a third of the devices recorded correctly. 14 AMs either recorded no files at all (3 devices), stopped recording after a few days (2) or recorded increasingly short periods during the week (6).
I updated the AMs to the newest firmware before deployment (version 1.7.1). We recorded using the same settings last year and got much longer recorded periods.
Is it possible the batteries are two weak? I inspected the .wav metadata of the files recorded on one device using exiftool, and battery voltage seemed to have been fine.
Temperatures got to around 0°C and at the beginning of the week there was some strong rain. I don't think this should pose a problem?
However, the decreasing file size over time suggests some kind of voltage problem.
What else can I do to find the cause of these problems and prevent them from happening in the future?
Thank you kindly for your suggestions! 😊
PS: I attached the config file as .txt.
Update:
I did a full charge cycle (charging, then discharging) with the batteries. Charched them again, tested the AudioMoth with the same settings, and all of the devices (except for 1 out of 21) worked normal again.
The voltage is continuously measured internally during recording and the recording will abort if it drops too low. The battery voltage reported at the start is before the recording started when the battery is under no load. It is indicative of the battery condition, but often batteries will exhibit a large voltage drop when under load, especially when cold. AudioMoth 1.2 devices operate at a slightly lower voltage that AudoMoth 1.1 and AudioMoth 1.1 to make the maximum use of 1.2V NiMH batteries, but these in general 1.2V NIMH batteries should work fine on all models if initially charged well. The LIFEPLAN project has deployed many hundreds of AudioMoth using 1.2V NiMH batteries. There is a lot of variation in the power requirements of different SD cards from different manufacturers - https://www.openacousticdevices.info/sd-card-guide. These large cards might exhibit write currents that are higher than other cards. We normally use Sandisk Extreme cards. I'd suspect that these batteries weren't fully charged or don't hold their charge very well anymore.
If the files are shorter than expected, but are readable, then the WAV file comments will always report the reason for stopping early. Normally this is a report that the voltage was too low. What batteries and what SD cards were you using. However, in this case, the config file attached here also shows that you were using an amplitude threshold recording. This will only make recordings when the amplitude of the signal is above the set threshold, and the resulting recordings will be shorter than the total recording time. Was this intentional? You can use the expansion tools in the configuration app to expand the T.WAV files to their original length, with the 'unrecorded' sections replaced with silent periods.