Bat detector

I'm not sure if you are referring to the 0.015mm offset here.

center.jpg

The misalignment of the display stuck to the red PCB probably is bigger than this 0.015 mm or maybe there is an error in the library file. I did my best to get it centered, there was a small error in the first boards.

Or am I not getting your point here?
 
No sorry, I mean the display is of-center on the pcb, the left border about 3mm narrower than right border.
I'm sure you explained in past the reason why and difficulty to change that.
But maybe when changing the dimming circuit it gives a bit of space to slide the display a bit to the center.
 
Must be an older (2018) version pcb 2019 or later have the display board 15/1000 of a mm off center a distance I think nobody would notice. The offset on the 2018 boards was just a beginners mistake.
 
Ah ok, so that is solved already. I've built an 16-01-2020 version and that one still was asymetrical. And the picture you just posted with 3.2" display was asymetrical too.
But when looking at Thierry's 3D printed housing, left and right border looks the same width so I assume newer boards indeed have the display in center.
 
I used black screws, it looks a lot better too. In the new circuit board you could also activate the second operational amplifier in order to be able to record low-frequency signals with larger capacitors. At the moment I use an external microphone with a very low-noise single transistor amplifier (BC849C) like the one built into the "Franzis" bat detector.
 
Hi,

I just found this cool project and decided to build it.Already got the Teensy 3.6 and the audio shield. Is the file for this printed circuit available for download, so that I can order it?

Diane
 
Hi Diane,

There are no gerber files publicly available. I do share boards + microphone PCB's via a big US based international marketplace. Or you could try contacting me directly. I know you will get 5 pcb's from china for that price but you have to wait much longer and there is no microphone included.

I did put up a website www.teensybat.com to try and keep all relevant information together.

Kind regards, Edwin
 
TEENSYBAT V1.2 release pending (coming days)

Hi all,

We (Edwin, Adrian, Thierry and I) have been very busy since the release of the V1.1 "development" update last month and we are pleased to annouce that we are planning to release a new version this weekend (22/23 may).

We had several important goals for this release and the most prominent change is that we finally can step away from the limitation of short filenames. This was restricting and pushing us for instance towards using "HEX" filenames to code date_time. But we now have switched to TeensyDuino 1.154 and thus can use the new SDfat library which supports long filenames. This new release will therefore save all files with a full date_time stamp (YYYYMMDDTHHMMSS.wav) and to make things more organized saving will be done in folders (YYYYMMDD).

Another important change is that the replay of files is not done using pre-set playback framerates but you will be able to replay using DIRECT (speed=samplerate) and anywhere from 1/5 to
1/20 of the original samplerate. Having a mix of files with different samplerates thus will allow you to hear all of them with the same replay_speed and the TimeExpansion sound
during replay will therefore be comparable.

When selecting files for replay you will see the filename, the length of the file in milliseconds and the samplerate. And if you have recorded the file using the new V1.2 release you will also get to see the overall powerspectrum for the complete recording before playing the file. You will thus be able to find that file with an "odd" spectrum easier than before, especially if you have a lot of files recorded using AutoRecord we think this will come in handy.

When actively listening to bats or replaying a recording you can now choose from 10 different colourschemes for the waterfall graphs.

And loads more small but often important changes have been made, some visible, some not. During the past week this version has been tested (as far as the weather was cooperating) but we cant guarantee its "bugfree" :)

But we do hope this will be usefull for the coming "bat" season,

greetings also on behalf of Edwin, Adrian and Thierry,

Cor
 
Teensybat V1.2 released 20210522

Hi all,

I have just finished copying both the libraries/code and HEX files onto the github repository:
https://github.com/CorBer/teensy_batdetector/tree/master/version1_2

The HEX files will probably be used by most people so here's a bit more information:

firmwareV1_2_20210522_080508_default.hex
this is the default release. Debug is OFF and PWM is ON. Also if you have not altered the TFT to use PWM to control the backlight you can still use this version.

firmwareV1_2_20210522_080550_debug.hex
this is the debug version (PWM is ON). This version can be of use if you encounter problems. The debug will be using the serialport at a 19200 baud 8N1 setting.

firmwareV1_2_20210522_nopwm.hex

this is a version where PWM is not actively used. A few things will work slightly different and some menu options will possible be missing.

If you plan to compile using the source/libraries be aware that this project is maintained using VisualCode/PlatformIO on a linux system (Linux Mint 19.3) and thus my settings
and OS are not "default". All code was compiled using the 1.154.0-beta7 version of TeensyDuino and will NOT work in any lower version of TeensyDuino.
If you are using windows as an OS than you will need to "comment out" the "#ifdef UNIXBASE" in the file "bat_defines.h" before compiling !

If you encounter problems, share them with us using this forum or via a PM.

kind regards, also on behalf of the rest of the testing team (Edwin, Adrian, Thierry),

Cor
 
Hi Helge,

If you longpress the left pushbutton (longer than 2 seconds) it will create a screen-recording. During that time you will see the word "SDUMP" on the topleft of your screen.
See it as a late "easter" egg ;)

cheers
Cor
 
A hidden Easter egg, how does the reinitialization of the display work? I disconnect this from the battery when recording. So I have much less interference at 23KHz and their harmonics.
 
When you use PWM (Edwin has described this before) to control the backlight of the TFT it can be brought to sleep and we also set the backlight OFF. That way the TFT does consume the bare minimum and can be revived instantly. IF you disconnect the TFT completely you will have trouble starting the communication, the only thing I can think of is to restart the initialisation of the TFT over SPI. Not something our project is planning to use.
 
Last edited:
@Helge: Alternatively, when you pickup your detector after switching off the TFT and using AREC to have it somewhere outside waiting/recording the best procedure is to switch off the detector. Then reconnect your TFT and restart the detector, that way you will be having a working fully working detector with TFT.
 
I do not experience display noise, the micropohone is more sensitive on 23kHz and I know I do see that. Specially in (near) silent recordings the spectrum preview upon playing shows all the collected noise, there is not much we can do about that.

Also using the uit with high gain setting will show more signal on 23khz but most is due to higher sensitivity of the microphone an higher gain settings. For the perfect recording you should be close to the bats using almost no gain.

There are a few things that could help improve sound quality.

- Check for a good seal of the microphone to the enclosure. (using a thin layer of foam-tape)
- Keep the opamp close to the PCB (do not bring it closer to the display using an ic-socket)
- (tip by another builder) Use some sound absorbing foam in the enclosure, if any noise is generated inside the enclosure the foam can capture that noise.
 
@Helge: whats the problem with that ... we had the choice to NOT use the left pushbutton (since you cannot record with it) or to allow it to have its previous functionality when no SD is present.
 
I was just wondering, it's so okay. Great, now also supports sdFAT. Accidentally plugged in the wrong SD from the AudioMoth and everything worked.
 
sdFat is the library that supports long filenames and different formats. In our setup we allow SDcards to be in FAT16, FAT32 and exFAT. Thats all possible due to the great library by bill Greiman.
Good to hear things are working as planned !
 
Sometimes with long 384k recordings on exFAT, only part of it can be played. The file size is correct. If an hour is recorded, only 23 minutes, even 46 minutes, can be played. Sometimes it's right too.
 
Hi Helge,

The code was never intended to play recordings with that length, but if the recordings are fine (you hopefully have checked that with other software) than this is not an issue. What size are these files ?

cheers
Cor
 
The file size is well below the WAV file limit of 4 GB. For playback I use CoolEditPro and Audacity. The error does not always occur.
 
Hi,

Although I will not change the code - as there is no clear problem with the recording setup - be aware that the recording system we use is not based on sdFAT but on uSDFS. If that is less reliable above 2Gb ... good you found it. But I dont think this is the functionality this system was designed for.

What kind of errors do you get from Audacity ?

Cor
 
Back
Top