To test this feature, visit your live site.

Hydromoths malfunctioning 2 days after deployment

We recently deployed 6 hydromoths at nearshore coral reef sites in Indonesia. All units were programmed to record with a sleep/record cycle of 50s recording and 10s sleep. After 3 days, we checked the units underwater and noticed that the units were not recording correctly. The red and green LEDs were flashing together. Their flashes matched this error: "Short flashes (10ms) of both LEDs between recordings mean an earlier scheduled recording has been cut short due to a recording failure." We noticed that after ~two days of successful recording, the recorders began saving shorter and shorter files, until they stopped saving files entirely. This happened on all six units.

Reading posts in this forum, it seemed that bad batteries was the most likely culprit. So, we did some tests using different battery types, including some that were recommended by colleagues who do PAM in the region. I have attached the results of this test here. We originally deployed with ABC batteries, and it is easy to see why there was an error after two days. In our test, these batteries started out with a combined voltage of ~3.7 V. But for the other two battery types, AA Alkaline Krisbow and AA Alkaline Energizer Max Plus, the voltage starts out at 4.9 and then drops to < 3.6 V in just two days. The hydromoths were updated to the latest firmware for this test.

Why are these batteries draining much faster than expected? What advice do you have for our team so that we can attain the expected recording duration (~7 days) for these hydromoth units using alkaline batteries. We still would like to record continuously and keep this sample rate.

Our team from Indonesia to Ithaca sincerely appreciates your help! Thank you.

6 comments

6 Comments

blg85

Jul 29

Hi Alex, Just wanted to follow up. We did a test with a new SD card (Sandisk Extreme Pro 256 GB) and it fixed the issue. The unit recorded as programmed until the end of Day 7 when the batteries ran out of juice (we were using a set of gently used batteries for this test). Many thanks for your help.

Alex Rogers

Jul 29

Replying to

Great. Glad that fixed it. We’ve found SanDisk Extreme (red and gold) cards to be the best. They are a bit quieter than the SanDisk Extreme Pro cards.

Aguirre Joe

Jul 01

Looks like you’ve done a lot of troubleshooting already. Battery drainage can be a pain, especially with such specific recording needs. Have you tried using lithium batteries? They usually last longer. Also, double-check your settings and firmware – sometimes little tweaks can make a big difference. Good luck! hill climb racing

Alex Rogers

Jul 01

Hi, Can you use the 'Summarise AudioMoth Files' option under the 'Process' menu option in the Configuration App to generate a summary of one of the SD cards so that we can see what files were generated? The file will show the battery voltage at the start of each recording and also show the WAV comment that will report the reason for the shortened files. The most common reason for faster than expected depletion of batteries is the choice of SD cards. Can you confirm which SD card make and model you use? Can you also confirm the sample rate - you don't specify in the description above. SD cards are very variable in their energy consumption. We recommend SanDisk Extreme (red and gold) cards which are reliable and have low energy consumption. The Configuration App presents the energy consumption based on measurements of this card. https://www.openacousticdevices.info/sd-card-guide Other cards can use between x2 and x5 as much energy over a deployment. Alex

Edited

blg85

Jul 04

Replying to

Hi Alex,

Thanks very much for this helpful information. We now have the answers to your questions. We are using the SanDisk Ultra SD cards (photo attached). We have tried using the Summarise AudioMoth Files, but this did not work for us. We have already converted all the WAV files to FLAC using Raven Compass. Potentially because of this step, no information was written in the summary excel file (summarise steps attached).

Could SanDisk Ultra be draining much more than SanDisk Extreme? We will try to do another test, keep the wav files, and do the summary to get more information on voltage drainage.

Thanks for your help,

The Sangihe Dugong Team

Alex Rogers

Jul 04

Replying to

Hi, This is almost definitely the batteries going flat due to an SD card that uses more energy than necessary and low-capacity batteries. The 'Summarise AudioMoth Files' tool reads the comments and metadata inside the WAV files. This describes the battery voltage and the reason for any short files, so it's useful to do this to each SD card before converting the files.

The SanDisk Ultra card uses more energy, is slower, and more noisy than the SanDisk Extreme (red and gold) SD card. We only recommend using SanDisk Extreme (red and gold) SD cards. The SanDisk Ultra card will probably use about twice the energy of the SanDisk Extreme card: https://www.openacousticdevices.info/sd-card-guide You don't say what sample rate you were using but it sounds like this is quite an energy-intensive recording schedule using a high sample rate. In this case, be sure to use good-quality batteries. A good quality alkaline battery has a capacity of about 2600mAh, good quality NiMH rechargeable batteries will have a capacity of about 2850mAh, and Energiser Ultimate Lithium batteries will have a capacity of about 3600mAh. https://www.openacousticdevices.info/batteries The ABC and Krisbow batteries do not make datasheets available and should be avoided. Only use batteries that have published datasheets describing their discharge curves (e.g. https://data.energizer.com/pdfs/e91.pdf or https://data.energizer.com/pdfs/l91.pdf). The panel at the bottom of the Config App will tell you the energy used each day. This is calculated for a 32GB SanDisk Extreme SD card so your actual capacity will be a little less. Always test the proposed recording schedule with the actual SD cards and batteries to be used prior to deployment to see the actual consumption.

From the SUMMARY.CSV file you will be able to plot the voltage over time. The two examples below are for a specific set of recording settings. AudioMoth will start to cut recording short when the batteries are below about 3.6V as at this point they begin to struggle to supply the necessary current to write to the SD card successfully. The limit is determined by the AudioMoth monitoring the supply voltage during SD card writes and stopping if it drops below 2.8V. The battery voltage is generally higher than this but you can see the rapid drop in battery voltage at the end of their capacity. NiMH and Lithium batteries tend to look similar, and alkaline batteries have a flatter discharge profile.

Alex