The purpose of this post is to describe an error we’re experiencing with two of our Anabat Swifts. The third Swift in our stable works perfectly well, we think. In this post, I’ll summarize our problem. Then I will give a detailed report on our last encounter with the problem. If we solve the problem, I will edit this post to include an explanation of the source of the problem (as I understand it) and the solution to the problem.
We are getting “error” and “format error” messages from our Anabat Swift passive bat detectors/recorders. To isolate the cause of the error, we have purchased and used new 128gb SanDisk Ultra Plus SD cards from Amazon.com. We have verified that all the cards we use mount on our laptops, are formatted, and are empty of any data.
Our SD cards are numbered so we can keep track of which cards are in which units and have which data. This report will refer to them as SD18 and SD21, for example.
Notes on Errors from First Unit
We have two Anabat Swift units deployed in the field for testing. We put them out yesterday afternoon, and this morning we returned to them to see if they recorded any bat activity. I took meticulous notes of our experience replacing the SD cards with fresh SD cards.
At Site #1, our Swift (P03) is attached to a post. We verify that the detector, mic, and cord look unmolested.
Strap off. Detach mic from case. Power down. Remove 2 SD cards: SD11 from slot 1, SD10 from slot 2. The former opens with 12 files. The later is empty. Both are put in our SD card portfolio to take back to basecamp.
We put SD07 in slot 1 and SD09 in slot 2. Attach mic. Power on.
Slot 2 displays “error”. Reformat SD09 in slot 2 using the Swift. Main screen, see “error” on slot 2. Waited. Dialog box opens, “Do you want to reformat slot 1?” Answer ‘no’. Swift asks again. Answer ‘no’. Back to main screen. Now see “error” on slot 1, too. (That is, “error” on both slots.).
Wait. Now slot 1 “error” goes away and shows that slot 1 SD card is good, but there’s still an “error” on slot 2. Power down. Wait a long 20 seconds. Power up.
Main screen shows “error” on both slots. Pause. Main screen changes to show an error only on slot 2; slot 1 looks fine, like it is working condition. Reformat SD09 in slot 2 using the Swift. Oddly, the device reformats slot 1 (according to the message on the display). Then the main display shows an “error” for both slots; slot 1 reads “error, 0 GB” and slot 2 reals “error”.
Then reformatted both cards with the device. Main screen shows a “format error” for slot 1 and a “reformat error” on slot 2.
Power down. Try new cards: SD15 in slot 1 and SD19 in slot 2. Power up. Slot 1 looks like it is working normally, but slot 2 shows “error”. Reformat SD19 in slot 2 using the Swift Get “format error” on display. Decide to replace Swift with our spare, P02.
Power down. Detach mic. Remove battery pack from P03 and put it into P02. Put SD20 into slot 1 and SD21 in slot 2 of replacement Swift. Attach mic. Power on.
Both cards look good. Waiting to acquire GPS. Signal acquired. Closed unit. Strapped to post. Check that there is not play in the attachment. Verify set-up and activity with magnet. Light blinks. Good to go.
Notes on Errors with Second Unit
Detector, mic, and cord look unmolested. Unchanged from when they were placed yesterday. Strap off. Detach Mic from case. Power down. Because of “error”s thrown on slot 1 when we were placing the Swift in the field, the unit has only one SD card installed, and it’s in slot 2.
Remove SD08 from slot 2. This card SD08 mounts in a laptop to show 134 files recorded, 1gb storage used.
Place SD13 in slot 1 and SD14 in slot 2. Attach mic. Power on.
Get “error” in slot 1. Reformat SD13 in slot 1 using the Swift. Main screen then shows “format error” in slot 1 which soon changes to “error” in slot 1.
Power down. Remove SD13 from slot 1. Replace with SD18. Power on.
Main screen shows that both appear good. Then screen closes on its own. When it reopens, it shows “error” in slot 1.
Reformat SD18 in slot 1 using the Swift. Get “format error” for slot 1 on main screen.
Power off. Wait 20 long seconds. Power on. See “error” for slot 1. Power off. Take SD 18 out of slot 1 and put SD21 into slot 1. Power on.
See “error” in slot 1. Power off. Remove SD21. Decide to replace the unit with only one SD card. Verify that slot 2 is still good. Close unit. Verify that GPS is acquired. Strap to post. Test with magnet – doesn’t blink.
Update 1: For what it’s worth…
Back at basecamp, just for kicks, I opened up the Swift to check the firmware version. Saw that it was version 1.0. Read in the manual how to update the firmware, and I did that. But something else remarkable happened when I powered up the Swift. BOTH SD CARDS READ PERFECTLY!! No errors.
My summer research group is trying to discover which species of bat are inhabiting Santa Rosa Island (in the Channel Islands National Park). We are searching for bats using acoustic methods. Our tool of choice is the Anabat ultrasonic bat detector, and we are hoping to get permission from the Channel Islands National Park to place some detectors for long-term monitoring of their bat populations. We’ll use the recordings, hopefully, to identify the species of bat that make Santa Rosa Island home.
The tool that we will use for our passive bat detection efforts is the Anabat swift (Titley Scientific, LLC; Columbia, MO). This post describes this detector.
The detector specifications (below) show that the unit is small, light, weather hardened, and battery powered.
Pictured here with a 12 ounce can of soda (to give you a sense of scale), you can see the detector is about the size of a paperback book. What is pictured is actually the hard, weather-proof case that contains the detector. Sticking up out of the case is a removable omnidirectional microphone.
The detector itself is inside the case. Its internal clock is powered by a button battery. It can be made nocturnal. That is, it can be configured to use GPS information to start recording shortly before dusk and stops recording shortly after dawn, going into sleep mode during the day to conserve power.
It can be powered by four or eight AA batteries. Below it is pictured with a carriage for four batteries. When placed in the field we will use a carriage for eight AA batteries. Also pictured below is a one (1) meter long cable used to place the omnidirectional microphone at a distance from the detector, a configuration that can be important when recording (i.e., improves the quality of recordings). Also pictured is a simple black webbing strap that can be used to temporarily and non-destructively secure the detector to a post, tree trunk, tree limb, or other stable object.
When we use a cable for the microphone, we can use zip-ties or recyclable variations to secure the omnidirectional microphone so that its cone of sensitivity is aimed at spacial volumes where we expect bats to be hunting for food.
We’ve been very pleased with the Swift and the way it has allowed us to record calls over long periods of time. It has slots for two SD cards, allowing for long-term deployment. Our longest deployments have been overnight on the University campus. Depending on the activity on Santa Rosa Island, we hope to be able to deploy for 3-4 weeks before running out of power and space for recordings.
If you’re a member of the CI faculty, and you’re interested in work being done at one (or more) of the commands at Naval Base Ventura County, then you should consider applying to participate in the ONR’s Summer Faculty Research Program. The application is not onerous, and you don’t even have to line up a collaborator, first. (Though I imagine that would help.)
In your application, you can identify two of the labs that you’re interested in working with. Our labs are easy to identify in the list. When you submit the application, your CV will be sent to a manager at the lab, and the CV will be circulated among scientists and engineers to see if anyone has an interest in working with you. (This is where identifying a collaborator ahead of time might be a good idea.)
Note that the funds to support you do not come from the ONR. They come from the local lab which sends the funds to ONR who administers all the program details. According to my sources, the local lab involvement in the program varies from having hosted no faculty recently, to having hosted about three faculty from around the country per summer. To my knowlege, no CI faculty have particiapted in the program in recent memory. Let’s change that.
The ONR also has a sabbatical leave program for faculty that appears to align nicely with the CSU difference-in-pay program. Check that program out, too.