This is a copy of an unanswered question from the FB Audiomoth User's Group from November, plus a few more details. This might not be within most people's experience, but perhaps someone has had the same issue. I use an external home-made 2-capsule microphone for use in a parabolic dish, attached to an AM version 1.2 + adapter and firmware V 1.6. The specs of this mic are the same as those used by others connecting via a 3.5mm Jack (and at least one contact has even used 3 capsules without significant extra battery drain). Rather randomly, while out actively recording (birds and other audible wildlife sounds), the AM at some point after successfully recording a or some files, fails to record at all AND/OR RECONFIGURES ITSELF roughly* to the default setting (01011970) if I then switch to Default (wherein it records OK again). If I don't notice this reconfiguration while outdoors (despite getting the solid red and flashing green light), as it's in my pocket inside a double plastic bag, it just creates a blank file then nothing after. (from November) So, today's example was interesting, since it recorded 3 half-hour sessions correctly, then it failed on the fourth, reconfiguring itself to the default settings and unsuccessfully tried to create a file in the process dated 01/01/1970. However, when I noticed afterwards and actively switched the unit across to the Default recording setting, it worked again and created a file for 01/01/1970 (even though this file creation occurred on 01/01/2098!). This is probably easier to see in the details of a few files listed below. NEW: 6th Feb '22 - it's occurred again today after another series of random occurrences between November and now, leaving me with a completely empty file for the second hour period of recording (in this case as that's the time standard I now use) and nothing at all thereafter. On checking through the files generated, I then came across that some of files from the interim period in Default mode had actually been dated as 10011970 as well s the standard 01011970! (I'm certain however, that I didn't leave it for 10 days without use after a self-disconfiguration event before using it in Default mode, which would explain that occurrence). I also do not think that it's to do with the microSD card and buffer, since I now have the AM configured to record for 3595s plus 5s sleep (but I haven't tried recording without sleep between the recording blocks to see what loss there is; it seems to be in the order of 3s looking at the Nov files below) and the high speed card has no issues and has been reformatted successfully at least once. I also now have a 1m break in the continuous recording configuration over 24h, so that it's not one continuous block (in case this could somehow lead to disconfiguration). I also do not think that it's to do with the microSD card and buffer, since I now have the AM configured to record for 3595s plus 5s sleep (but I haven't tried recording without sleep between the recording blocks to see what loss there is) and the high speed card has no issues and has been reformatted successfully at least once. I also now have a 1m break in the continuous recording configuration over 24h, so that it's not one continuous block (in case this could somehow lead to disconfiguration). So, could it just be that something moves inside the unit due to having it loose (albeit within two small plastic bags) in a fleece or jacket pocket (not subject to hard knocks or bangs, I hasten to add)? Could a loose Jack connection do this, or would it simply stop the current file and then restart once the connection went back into place (as in the video and see at the bottom), or at least just record with the internal mic? That said, I now use a small rubber band to exert gentle pressure on the connection to ensure that it doesn't come loose (and it doesn't). Perhaps there's a tiny bit of play in the SD card connection? Could it be a battery connection? Is there anything else that it could be inside the unit that could move and cause this disconfiguration? Here are the data for the file series in November: 20211116_101847 16/11/2021 10:30 63.095 KB 20211116_103002 16/11/2021 10:59 168.563 KB 20211116_110001 16/11/2021 11:29 168.563 KB 20211116_113001 [ BLANK ] 0 KB 19700101_000003 01/01/2098 00:31 178.752 KB* 19700101_000016 [ BLANK ] 0 KB CONFIG Device ID : 242A26075FDF**** Firmware : AudioMoth-Firmware-Basic (1.6.0) Time zone : UTC Sample rate (Hz) : 48000 Gain : Low Sleep duration (s) : 0 Recording duration (s) : 1800 Active recording periods : 1 Recording period 1 : 00:00 - 23:59 (UTC) Earliest recording time : ---------- --:--:-- Latest recording time : ---------- --:--:-- Filter : - Amplitude threshold : - Minimum trigger duration (s) : - Enable LED : Yes Enable low-voltage cut-off : Yes Enable battery level indication : Yes Always require acoustic chime : No Disable 48Hz DC blocking filter : No Enable energy saver mode : No Finally, I came back today after the latest disconfiguration, reconfigured the unit and have had it running continuously and successfully on the parabolic outside since for almost the past 5 hours, i.e., with the AM completely inmobile in the same plastic bags, with the first short file being that from the internal mic, then switching automatically on detecting the external mic from the second file onwards and then running as expected until I switched it off. Any thoughts very welcome!