Audiomoth stops recording

My solution to this problem was to use a longer sleep time. I think I just went to 60 seconds, and have not had any problems since.

Hi John, Flashing red and green at the same time always indicates an error. When writing to the SD card during a file this might be due to the battery being too low to continue recording or an SD card write error. If it is the former then the file will be closed correctly and you will have a correct WAV file which is shorter than expected but is still readable. The comments in the WAV file header will report why the file stopped earlier than scheduled. In the latter case, the AudioMoth won't be able to close the file correctly so you will see the filename in the file listing but the file will have zero length.

Can you confirm what batteries you are using? You mention Energizer Ultimate Lithium but say that you are recharging them? The Energizer Ultimate Lithium batteries aren't rechargeable as far as I know.

Alex

OK. I've tried a couple of tests using two short record / stop cycles with two different microSD cards, having brought the AM back indoors after supposedly recording last night (which failed completely; all blank files from start to finish).

However, I switched it to default just before taking the card out and it recorded successfully for c. 10s (I didn't try longer).

I tried the first test with 55s recording + 5s rest (empty files) over 5 cycles, with all files appearing and all empty. This despite it swtiching on OK and apparently recording with a red flashing light, before switching to dual red-green flashes. I tried a quick test immediately afterwards by switching to Default, which failed.

I then tried another test with the same record/sleep cycle on a second card and it failed again. This time it barely came up with flashig red light to indicate recording, rather remaining almsot constantly on the dual red and green simultaneously flashing lights).

So I'm assuming that it just doesn't have enough power.

Having seen that they were recommended, I bought a pack of Energizer Ultimate Lithium batteries at the beginning, which I assume are genuine, off Amazon, though they come with no mAh rating either on them or on the external packaging (odd, since I've read -goodness knows where on the web- that they're rated either as 2600 mAh or 2000 mAh. And they worked perfectly at the start, with the AM saying they were at >4,9V. After a very few recharging cycles though, it could be that they're become useless for the job, AND/OR my recharger isn't doing it's job either.

Would it make sense that it saves file names and is triggered at the correct times, but leaves the files completely empty due to insufficient power?

This is very frustrating!

Many thanks in advance for any further suggestions / comments.

John

I have changed the sd-cards, e.g. with the slower and also cheaper SanDisk Ultra class 10 or other similar cards I had regularly this recording problems. With the faster SanDisk Extreme XC1 V30 A1 U3 I never had this problems.

The problem is 0 seconds between files. Change that back to 5 sec. and you will be fine. Also 10 or 15 minute files allow you to see slightly more detail in the spectrogram in your editor.

Hello.

I'm new to Audiomoth use but have experienced what appears to be the same problem after just a couple of weeks of use.

With brand new batteries and unformatted microSD cards at the start, I had no problems, but with the recharger now leaving the (supposedly very good) batteries at say 4.6V starting value (rather than >4.9V), I'm now getting very varied results for the files. They start recording fine, but at some point simply stop, so I either get the full recording period saved (FYI, it's set at 1800s with 0s between from around dusk until just after dawn and low battery cut-off is disabled), or file stamps for the right cycle times (every half hour), but which contain no data. This is in contrast to the very beginning when I used some old rechargeable batteries for practice and found that they would record part of the files until the power dropped to low and then start again (for progressively recording time for each recording cycle) until the overall power got to low. So this is different (as it's all or nothing within each file).

I have also reformatted the cards on Windows 10 as well though, and wonder if somehow it could have had an effect?

If you discovered what the problem was, I'd be very interested to hear it, and thanks in advance!

John

Update: With two devices I had the problem, that 133 out of 6143 files were empty (0kb). This problem occured identicly with two devices, which had the same settings. Suprisingly, I cannot generate a new order or file on the SD-card, if I try, the explorer crashes. Anyone an idea, what could be the problem?

Thanks Michael. My test ran smoothly as well.

Update on test: The flashing of the red LED for two days indicated that the three units were in recording mode.On the third day, a green LED on one unit was blinking continuously, the others were still correctly red. I checked the SD card and found that two devices had recorded absolut correctly, but the third SD card could not be read and needed to be formatted. I couldn't find out if the error was caused by the device or if it had something to do with the SD card, over two days the device also showed that it was recording. I will remove this SD card at least for use with Audiomoth.

Thanks for keeping me in the loop. I will run a three day test myself using a similar configuration.

After Device ID : 248D9B045CC5C964

Firmware : AudioMoth-Firmware-Basic (1.4.4)

Time zone : UTC

Sample rate (Hz) : 48000

Gain : High

Sleep duration (s) : 5

Recording duration (s) : 55

Active recording periods : 2

Recording period 1 : 00:00 - 08:00 (UTC)

Recording period 2 : 20:00 - 24:00 (UTC)

Earliest recording time : 2020-09-04 00:00:00 (UTC)

Latest recording time : 2020-09-13 00:00:00 (UTC)

Filter : -

Amplitude threshold : -

Enable LED : true

Enable low-voltage cutoff : false

Enable battery level indication : true

-------------------------------------------------------------------------------------

I deposited the device on 03.09. in the afternoon. The first recording was on 04.09. at 07:10 UTC. Time of the CONFIG file is 07:11 UTC.

I don't have the other config file anymore, the start was correct on the other device but after 2 days the recording ended. Settings were the same.

Currently I have a test with three devices running at home over several days. I will report on the results.

Hi Michael, Are you still seeing problems? It would be great if you could share your CONFIG.txt file to help debug this. Thanks Andy

Michael, my issue is resolved by changing the sleep schedule to 5 seconds, but I am concerned about yours as well since I leave mine in the field for multiple nights. I can't help but think yours is a configuration issue as well. Do you mind reporting your problem on the configuration forum. I got an answer fairly quickly from one of the administrators. I will follow your progress there.

I run with the version 1.3.6 and firmware version 1.4.4

Michael What version of the configuration app are you currently running? I was running 1.3.0. I went in and found the latest is 1.3.6 and downloaded that one. I am going to try again and see it the bug we are experiencing has been fixed

New batteries made no difference. Once again every third file did not record. This is a problem I need to solve. I am beginning the second year of a project. Michael, I don't see my original post to the configuration support forum. I will try there again, and perhaps someone has an answer. I see that the administrators have been active on that page.

I went into the Configuration Support forum and asked this same question. We cannot be the only ones with the problem. I do remember - vaguely - that there is a way to go back and use a previous configuration app, pre-upgrade. As I mentioned above, I installed new Duracell alkaline batteries to see if additional voltage makes a difference. That test should finish in about 2-1/2 hours. I will post.

The test at home with the settings recording time 00:00 - 24:00 (UTC) and recording duration (s) : 360; (sleep duration 5 s) resulted in the first recording at 5:40 UTC although I started the device already yesterday evening. After interrupting the recording to check the SD card the recording starts immediately.

I can't figure it out but the device is no longer reliable. Is there a way to flash back the firmware to the previous version?

I ran two more tests and both failed. I recorded from 1000 to 0600 last night and the recorder skipped every third session, 10 files??? It makes no sense. I have never had to replace batteries that clearly had sufficient charge, but I will try that before bugging the administrators.

Hi Michael, I let my test finish. I forgot to reconfigure for each recording segment and had the default 55 seconds and 5 secs paused. I had all 180 files in sequence. I an running it again the rest of the day to verify. As to your problem of the recorder stopping altogether, I'm no help here. Is it possible you have something on the Advanced Settings configuration page checked such as amplitude threshold?

Thanks. I did not experience a complete stop. Mine might have been okay for the better part of the night and then stopped toward dawn for 30 or 45 minutes, with one or more final segments recorded. I was just writing a friend. I will wait to hear back from you.

I have now a test running at home, we will see if it stops again.

When I first encountered the problem, I removed exactly this checkmark when I reconfigured, but the problem still exists. Recording devices one week in the field, but only recorded something for 1-2 days. My batteries were still at 4.3 and 4.4 volts.

I think the answer is on the first page of the configuration app. At the bottom uncheck the box for "Enable Low-voltage Cutoff. In my first response to this issue I suggested new batteries. I did not make that change. I have noticed when I have come in from the field that when I measure the remaining voltage in the recorders with my own testing equipment, I can get one reading, and some minutes later it is higher. That is consistent across all my units. Last night I noticed in the configuration app that my batteries were at 4.0 volts. During recording a segment they apparently dropped below that level and the next 15 minute segment was not recorded. Then the unit recovered, and recording proceeded. I typically let them run down to 3.5 volts before discarding the batteries unless I expect them to be deployed for 5-7 nights. This action is not well described in the start-up of the configuration guide nor is there any indication I have found what low voltage level triggers the cut-off. I is pretty obviusly 4.0 volts.Perhaps I missed these points somewhere. In any event I made these changes to my test unit. It is about half-way through the exercise and it has recorded all 6 - 15 minute segments flawlessly. Hope this helps.

Looking for an answer to the same problem. I have been recording in 15 minute segments (900) with '0' sleep (still 1 second) for about a year. I did not experience this problem until the first configuration upgrade as I recall. Last night the same problem again with one unit I was playing with. I just deployed that one again in the garden with batteries reading 4.0v. I have easily recorded all night long in the past with less than 4.0 volts. This could be a serious issue. I hope the administrators are monitoring this thread. I will put in fresh batteries and reconfigure to see if that makes a difference. I will report back. This is a bummer!

While I don't have an answer on how to fix this issue, I am encountering a similar problem with a batch of AudioMoths I just pulled out of the field. I programmed these AudioMoths to begin recording on July 15th, and record for 2 4-hour periods per day until battery failure. (~32 days)

I programmed a batch of AudioMoths all at the same time, deployed them throughout June and early July and I am just starting to pick them up now. Of the 7 I have pulled from the field so far, only 2 recorded the expected amount of data, 1 recorded just 3 files (2 hours and 3 minutes worth of data) on July 15th, and the remaining 4 AudioMoths did not record anything at all. The only thing on each of those memory cards is the CONFIG text file.

I am at a total loss on how this could have occurred, as the batteries of the ones that failed to record anything or only recorded for 2 hours are all still apparently full (>4.9V). When I plugged an AudioMoth (the one that recorded for 2 hours and then stopped) into the computer I did notice something funky going on with the clock though. The date appears to be stuck on July 15th and the time is moving extremely slowly. I have attached a 30 second video clip of what the Configuration App looks like when I plug one of these AudioMoths in. For reference the local time this clip was recorded should read (16:14:02 11/9/2020 UTC-7).