To test this feature, visit your live site.

Bad recordings

I tried to use Audiomoth 1.2 for long hours for the first time. For about 22 hrs continuous. Recording duration = 5 mins and sleep duration = 5 sec. From about 260 files it generates, 50+ files got bad recording.

Issue starts for a file like this >>>

And later entire next files looks lake this >>>

And next 2 or 3 files gets corrupted like this. And later it starts getting normal. And after few normal files, issue starts again.

Note: Frequency > 5K is OK as these are Cicadas and / or other insects. It is present with my other recorder as well.

Issue started appearing first after about 2 hours of use. And kept on occurring for next few hours (bad recoding for about 5 to 6 files continuously, then no issues for about 20 files, then again the issues).

Then normal for few hours.

And then again issue for few hours.

I am using the memory card supplied with the device: SanDisk Extreme 64GB.

Sample rate: 48 kHz

Gain: between Medium & High

Filtering: Band - 0 KHz to 16 kHz

Schedule: 00:00 to 24:00

Custom mode.

Battery: Duracell Ultra Alkaline

Max temperature: Aprox 35 Deg C

Min temperature: Aprox 25 Deg C

Humidity: Aprox 80%

Using the casing supplied with the device.

For reference, normal (OK) file from the same lot (without cicadas) looks like this >>>

Can someone please help with this issue.

Is this device issue, memory card issue, environmental factors (temperature / humidity), mobile tower signal, high power electricity line, battery?

Regards,

Nishant.

6 comments

6 Comments

Nishant Bhagwat

Mar 26

Thank you Alex for your time and the analysis of banding🙏.

Taking clue from you, searched and found that most of the Cicadas sings at day time and likes heat as well. And perhaps that's why I got clean recordings after midnight. And issue started again just after few mins of Sunrise. Hence this also rules out heat issue as it's not hot at the Sunrise.

Yes, mostly it could be a Cicada very close to the recorder.

Regards,

Nishant

Alex Rogers

Mar 26

Replying to

You can turn the gain down and that will probably help. There is also an 'Enabled low-gain range' option in the 'Advanced' tab of the Config App which substantially reduces the gain. This was introduced for noisy underwater settings but might be useful in this case.

Nishant Bhagwat

Mar 27

Replying to

Thank you. I will take trial with low gain as well as enabled low gain range, with Cicadas in background. This may eliminate the issue of saturation of input but need to check how much it affects the sounds of interest (bird sounds).

Alex Rogers

Mar 26

Hi, Can you some of the example WAV files somewhere so we can look at them. Alex

Nishant Bhagwat

Mar 26

Replying to

Hello Alex, thank you for the reply.

I have kept 4 files (1 OK and 3 not OK) in google drive.

Link: https://drive.google.com/drive/folders/1hNQrKGYh2bdJ74sEhztNZXS4g43Y0BJQ?usp=sharing

Thank you.

Regards,

Nishant.

Alex Rogers

Mar 26

Replying to

Could these be cicadas that are very close to the AudioMoth and are saturating the input? It doesn't look like a hardware issue, and you can see the recording returning to normal in this section.

The banding here is 0.2s apart while the SD card writes will be occurring 0.37s apart, so this looks more like real environmental noise, rather than device noise.

Does the device record okay when deployed on a shelf?

Edited