Bat-Duino
Andrew Thomas
I emailed about a pcb yesterday, but haven't heard anything (probably an old link or on holiday). However, I had a look on the internet this morning and found some useful links on bat detectors that I've put below, in case they're any use.
I bought some 4024's on ebay and looking at ballpark figures I think the components (exc. case, earphone, socket, battery) would be less that £5. The third link uses an opamp on the output, which wouldn't cost much more and would allow use of normal headphones or a speaker.
The Pico detector gave me an interesting idea. Arduino A2D is too slow for bat detecting, but digital input reading ?should? be fast enough. So I wondered whether pre-amplifying a transducer and feeding it through a switching transistor might be a good way of detecting bats with one of Nikki's diyduinos? Theoretically it might be possible even to get some sound level info by feeding the pre-amp into two or more switching transistors. Might need a high-pass filter though, to reject audio frequencies.
If so, it should be relatively easy to make a walk around bat-duino, with a simple audio output. However, it could be more exciting, as with a cheap RTC circuit it could be put outside overnight to detect when (and how many) bats are nearby, and it would be easy to graph the data later. In fact, the full 4024 circuit could be combined with an arduino circuit to get good listening and detection/logging.
So, I wondered what you both (and anyone else) think of that as a project, or even if it would work? If it worked, and Gary does his eagle-tutorial-a-thon, then maybe it could even be built on a pcb for putting in a nice case? It's just a thought, but if anyone's interested in an open-source bat-duino project then it could be fun.
A page with lots of bat detector circuit links (some dead links):
http://bertrik.sikken.nl/bat/index.html
PicoBat, a simple PIC bat detector:
http://www.micro-examples.com/public/microex-navig/doc/077-picobat.html
Another nice 4024 based design:
http://bertrik.sikken.nl/bat/my_div.htm
Best wishes,
Andrew.
Adrian Beaumont
Anyway, this is all way over my head technically but in laymans terms it seems we are looking at seperate projects:
1. a simple bare-bones bat detector on a PCB as a kit of parts for classes (sort your own case and battery out) for a bat detector themed evening, goal price around a fiver.
2. an enhanced version of the above with an opamp for those interested.
3. An open source Batduino project based on a barebones arduino with possible future enhancements such as identification/logging etc. It would be nice to then sell kits on line to raise funds for fizzpop whilst still keeping it open-source, of course.
Have I understood that correctly?
If so then I would be happy to work offline with you, Andrew to get 1 & 2 sorted, bearing in mind my limited knowledge. 3 would stay open for wider discussion. It would be great to do 1. in October and possilby even put a piece in the local paper. You know the type of thing "Local club runs batty course to make a spooky project" etc. Could raise the profile of fizzpop and get some new members in..?
Thanks, Ade
--
You received this message because you are subscribed to the Google Groups "Birmingham Hack Space" group.
To post to this group, send email to birmingham...@googlegroups.com.
To unsubscribe from this group, send email to birmingham-hack-...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/birmingham-hack-space?hl=en.
Andrew Thomas
I think that's it in a nutshell (I like the batty bit :) Using your numbering:
1) Basically I've bought some chips and when they arrive I'll try making the simple detector circuit. If that works I can put together some instructions for your batty-building evening if you like.
2) When it's working, I can then try adding an opamp for the output. Maybe change the circuit to use a dual opamp for the input stage, instead of 2 lm386's, just to make it easy for beginner batty builders.
3) If that works, I could try hooking the arduino into parts of the circuit, to make a logger/detector. Then if it works maybe do a pcb for a combined listener/logger (maybe even as a shield). My guess is around £20 max for a kit.
I'll let you know when I've got the parts and tried Step 1. If anybody else wants to get involved we could talk about it at fp?
Best wishes,
Andrew.
Thanks, Ade
> components (exc. case, earphone, socket, battery) would be less that £5 The
> birmingham-hack-...@googlegroups.com<birmingham-hack-space%2Bunsu...@googlegroups.com>
> .
> For more options, visit this group at
> http://groups.google.com/group/birmingham-hack-space?hl=en.
>
>
--
genzaichi
they're very pleased with for less than £1 in parts.
I'm not sure if that's the right sort of amp you're looking for, but
it'll be one of the kits they're doing at Geek's World and is probably
worth a butchers anyway.
nikki
G Bulmer
I think all the 4024 counter-based circuits amplify the input (from
the transducer) until it clips, turning it into a binary signal (0V
and +V) which is used to drive the clock on the 4024 counter. All its
doing is dividing the input signal to make it audible.
So the same digital signal could be feed into an Arduino, no A2D
involved. If the signal is fed into T0 or T1 (digital pin 4 or 5),
then that will drive timer0 or timer1, so the job is done mostly in
hardware.
By loading, say, 16 into the timer, it will divide the frequency in
hardware, like a 4024. Wiggle an output pin connected to a headphone
socket, and it should be done. I think the output side of the timer
still works when driven by an external clock, so I think the whole of
the 4024 could be moved into a timer, and the output pin is a PWM pin
from the timer.
By using one of the other timers, it should be practical to get a good
estimate of the frequency of the sound, and hence try to identify some
of the species.
As this only needs two timers and a couple of pins, it should be
feasible to make one with a cheaper AVR than an Arduino, e.g. an
ATtiny25, which is about £1.15 1-off (£0.99 10-off). Use the internal
8MHz oscillator to save pins and cost. Then there is a base to build
on to do a little bit of species identification (4 spare pins if there
is no RESET) for the cost of LEDs and resistors. (Irritatingly, the
clock input clashes with SPI, or it would have been worth upgrading to
an ATtiny45, add an SD memory card socket, and record everything,
which would be very cute).
It might make a nice little PCB if we used a little ATtiny. Probably
smaller than the 4024-based one, as an ATtiny25 is smaller.
If they could be made on 25mm x 40mm (say) PCB, 16 would fit on one of
those Olimex 100mm x 160mm PCB's.
HTH - GB
G Bulmer
The instructions at http://bertrik.sikken.nl/bat/my_div.htm suggest
using a comparator instead of an opamp to drive the 4024. All it needs
to do is drive the signal full on or off, and opamps are a bit smarter
than that :-)
If it just used a 'naked' ATmega or ATtiny, it shouldn't add much to
the cost.
GB
G Bulmer
That audio amp is likely ideal. It's probably just an IC and a few
parts.
Can you get one or two to look at?
GB
Noel Welsh
Been on holiday and just catching up...
On Fri, Aug 20, 2010 at 12:46 PM, Adrian Beaumont <ajbea...@gmail.com> wrote:
> 3. An open source Batduino project based on a barebones arduino with
> possible future enhancements such as identification/logging etc. It would be
> nice to then sell kits on line to raise funds for fizzpop whilst still
> keeping it open-source, of course.
Our company (www.untyped.com) has done some work for Dr Kate Jones
(http://www.zsl.org/science/ioz-staff-students/jones,1087,AR.html) in
the past. One project we discussed is building a better bat detector.
You can see their current kit here:
http://www.ibats.org.uk/page.aspx?tabid=264 It is very expensive, and
processing the recordings is a very time consuming and error prone
task for them. They really need to do bat identification and GPS
logging on the bat detector device, so this is a bit beyond what you
can squeeze out of an Arduino. If anyone is interested in taking this
project beyond the hobbyist stage I think Kate would be interested and
might be able to find some money. Let me know and I'll liaise with
Kate.
N.
Andrew Thomas
Excellent insights :) Would the clock be needed if it's a problem with spi? Wouldn't it be just as easy, and still pretty fast, to just input pulses into an interrupt pin? The interrupt handler could then count pulses and the main routine (or a timed interrupt) could then decide whether or not to toggle the audio out.
Presumably that would leave time to save to storage without worrying about pin clashes? My knowledge of these things isn't up to yours though, so I still plan on building the 4024 circuit to play with and do some learning about bat detection :)
Best wishes,
Andrew.
Andrew
HTH - GB
On Aug 20, 11:56 am, Andrew Thomas <andrewmarktho...@yahoo.com> wrote:
> Hi Adrian and Gary,
>
> I emailed about a pcb yesterday, but haven't heard anything (probably an old link or on holiday). However, I had a look on the internet this morning and found some useful links on bat detectors that I've put below, in case they're any use.
>
> I bought some 4024's on ebay and looking at ballpark figures I think the components (exc. case, earphone, socket, battery) would be less that £5. The third link uses an opamp on the output, which wouldn't cost much more and would allow use of normal headphones or a speaker.
G Bulmer
Agreed, the bat-detection could be done mostly in software+interrupts.
The only down side of that is more code, and the possibility of
clashing with another interrupt.
The pretty part of a timer-hardware based solution is, I think, that
it would "Just Work" (TM) without any software intervention, after the
initial setup(). So writing to SD memory, or doing "chirrup-print"
analysis could be asynchronous, and not get in the way of producing
the audio.
The three SPI-pins would be used to read/write to SD memory. That's
why it's irritating that the pins clash on an ATtiny25/45. If they
didn't clash, and bat-input, human-speed-bat-output, and SD memory all
worked harmoniously, it would be a super-nifty little project. A
simple bat detector, bat-analyser (I think that was quite a huge gizmo
in the TV series) and bat-recorder for under £10.
An ATtiny25/45 even has a built in analogue-comparator, so for a
software+interrupts solution, it might not need much of an amplifier
on the input, making it even smaller.
I agree, with software+interrupts, there are no pin clashes, so the
hardware is unchanged. I'm not sure what happens if an SPI interrupt
and pin-change interrupt happen at "the same time".
The old creative juices are warming up to this :-)
GB-)
Andrew Thomas
I agree your way is much better. It kind of reminds me of how the prop chip does A2D. In fact a prop chip might be ideal for bat detecting as it wouldn't need interrupts and can handle a mic input without an amplifier (something for me to think about when I have time).
I asked mostly because you mentioned the interrup clashes, but from what you say it sounds like it wouldn't be a problem on a 168 or 328? By the way, wouldn't the hardware pin interrupt get priority during spi? I suppose that could cause spi timing problems? However, would the biggest problem be that if you use an interrupt for spi it might delay the hardware pin interrupt routine causing it to become re-entrant?
Sorry for all the questions. However, you only have yourself to blame as you mentioned Batman: Robin and the Commisioner are always asking silly questions ;-)
Best wishes,
Andrew.
Andrew
> > I bought some 4024's on ebay and looking at ballpark figures I think the components (exc. case, earphone, socket, battery) would be less that £5 The third link uses an opamp on the output, which wouldn't cost much more and would allow use of normal headphones or a speaker.
G Bulmer
On Aug 24, 9:59 am, Andrew Thomas <andrewmarktho...@yahoo.com> wrote:
> Hi Garry,
>
> I agree your way is much better. It kind of reminds me of how the prop chip does A2D. In fact a prop chip might be ideal for bat detecting as it wouldn't need interrupts and can handle a mic input without an amplifier (something for me to think about when I have time).
>
so sensitive.
> I asked mostly because you mentioned the interrup clashes, but from what you say it sounds like it wouldn't be a problem on a 168 or 328? By the way, wouldn't the hardware pin interrupt get priority during spi?
I suppose that could cause spi timing problems? However, would the
biggest problem be that if you use an interrupt for spi it might delay
the hardware pin interrupt routine causing it to become re-entrant?
>
Yes, external interrupts (INT0/INT1) and pin change are both higher
priority than SPI.
I usually disable interrupts for a little bit when I am manipulating
shared variables, like buffer pointers, so a simplistic implementation
(which I sometimes write :-) might lose an interrupt occasionally.
>
Robin, but many of the people talking may be silly to think they know
how to answer" (or then again, he might not).
> Best wishes,
>
> Andrew.
Good thinking, keep it coming!
I've looked at the ibats link, and there is a lot of 'food for
thought'. It reads more like 'Indiana Jones, and the ...', than 'Five
go batty in Dorset'.
GB
Andrew Thomas
I've got a prop chip soldered into a circuit board I'm not using, so in a few weeks I might rescue it and give it a go. The prop dev board schematic just shows a little condensor mic and some filter components, yet works quite well.
Although you'd expect the A2D to be fairly low quality, it's up to 16 bit which is similar to a 10 bit avr chip with preamp. Certainly I got quite clean A2D results when playing with the parallax LC measuring circuits.
I'm thinking maybe a cog (their term for a core) for monitoring a transducer, one to synthesise audio out, maybe one to write to storage, and one maybe to provide a serial interface (leaving a few spare).
My prop dip chip is fairly large (40 pin), but the smd ones are pretty small and cheap enough for a bat detector. However, I'm assuming its tokenised code would be fast enough, but they have pretty fast clock rates so hopefully ok.
It might make an interesting project for me on cold winter evenings :)
Anyway, must dash as the bat beacon is glowing (also better go as I didn't know it's important to know what ones talking about when replying ;)
Best wishes,
Robin.
-----Original Message-----
From: G Bulmer <gbu...@gmail.com>
Andrew
> > To post to this group, send email to birmingham-hack-space@googlegroupscom
G Bulmer
I found an article (via oe of Andrews links) about bat-microphones (I
guess they need them to make the batphone).
Anyway, highly recommended is the Panasonic WM-61A. They look
identical to something Rapid sell, but apparently they are not
comparable, the Panasonic is way better.
They recommended eBay (but found none) and Digikey.
Digikey have a secret uk lair (I believe VAT is paid, so no hassle
from the Joker's at the Royal Mail).
So this is the part:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=P9925-ND
It is claimed to have a good response upto 60KHz, which isn't quite
upto all bat-species, but, it's only £1.33 each, or £1.18 each 10-
off !!
This compares very favourably to these chaps
http://www.ultrasoundgate.com/microphones.htm
where their cheapest ultrasonic microphone is 420 euro's.
Why so expensive? Only the Riddler can say!
So I plan on ordering some. At this price, they are comparable to a
cheap piezo ultrasonic receiver, but appear to have a much better
frequency response.
Digikeys postage was quite steep, so does anyone want to order some
parts from Digikey, and split the postage?
GB
>
> read more »
G Bulmer
WM-61A Electret Condenser microphones 'catridges') if I order some?
Their spec shows a reasonable response from 10KHz to 60KHz, so it
should be good for listening to squirrels, mice and other rodents,
who, like bats, also vocalise above human hearing.
Also other animals, mainly insects but also spiders and a few others,
use stridulation to create ultrasonic sound.
http://en.wikipedia.org/wiki/Stridulation
On Aug 25, 4:22 pm, G Bulmer <gbul...@gmail.com> wrote:
> More bat-tastic discoveries.
>
> I found an article (via oe of Andrews links) about bat-microphones (I
> guess they need them to make the batphone).
> Anyway, highly recommended is the Panasonic WM-61A. They look
> identical to something Rapid sell, but apparently they are not
> comparable, the Panasonic is way better.
>
> They recommended eBay (but found none) and Digikey.
> Digikey have a secret uk lair (I believe VAT is paid, so no hassle
> from the Joker's at the Royal Mail).
>
> It is claimed to have a good response upto 60KHz, which isn't quite
> upto all bat-species, but, it's only £1.33 each, or £1.18 each 10-
> off !!
>
>
> read more »
genzaichi
Really busy, so have only been scanning posts and not sure where this
fits in on the 1-3 of projects, but if you think it's within my
intellectual and financial means to make the rest of it then put me
down for one!
nikki
On Aug 25, 10:54 pm, G Bulmer <gbul...@gmail.com> wrote:
> I should ask, does anyone want any Bat-microphones (really Panasonic
> WM-61A Electret Condenser microphones 'catridges') if I order some?
>
> Their spec shows a reasonable response from 10KHz to 60KHz, so it
> should be good for listening to squirrels, mice and other rodents,
> who, like bats, also vocalise above human hearing.
>
> Also other animals, mainly insects but also spiders and a few others,
Andrew Thomas
Can I have three? Any chance you have an estimate of how much that would be with the steep postage costs?
G Bulmer
Yes I can add three to the order for you.
I am hoping to get a big enough order that shipping will be free.
If not, the shipping is £12, I am grateful to anyone willing to share,
so I'll ensure shipping will be either 24p/£ (£12/£50), or if over
£50, £0
i.e. if the parts came to, 3 x £1.33 + 17.5% VAT = £4.69,
then shipping will be no more than £1.13 (and maybe £0.0).
The parts will actually be less than £1.33, because I'll get enough to
get the 10-off discount, where the price £1.18 each, which also cuts
your share of the P&P to a either £1.00, or £0
Does that seem reasonable?
GB
On Aug 25, 11:47 pm, Andrew Thomas <andrewmarktho...@yahoo.com> wrote:
> Garry,
>
> Can I have three? Any chance you have an estimate of how much that would be with the steep postage costs?
>
> Best wishes,
>
> Andrew.
>
>
>
> -----Original Message-----
> From: G Bulmer <gbul...@gmail.com>
> Sent: 25 August 2010 22:54
> To: Birmingham Hack Space <birmingham...@googlegroups.com>
> Subject: [birmingham-hack-space] Re: Bat-Duino
>
> I should ask, does anyone want any Bat-microphones (really Panasonic
> WM-61A Electret Condenser microphones 'catridges') if I order some?
>
> Their spec shows a reasonable response from 10KHz to 60KHz, so it
> should be good for listening to squirrels, mice and other rodents,
> who, like bats, also vocalise above human hearing.
>
> Also other animals, mainly insects but also spiders and a few others,
Andrew Thomas
I'm happy just to call it a pound :)
Best wishes,
Andrew.
Andrew
Does that seem reasonable?
GB
--
You received this message because you are subscribed to the Google Groups "Birmingham Hack Space" group.
G Bulmer
pound of flesh from around the heart. Is that correct?-)
I did the calculation in case anyone else wanted a few components
added to the order, and was wondering how much postage might be.
Garry
Andrew Thomas
-----Original Message-----
From: G Bulmer <gbu...@gmail.com>
Sent: 26 August 2010 21:41
To: Birmingham Hack Space <birmingham...@googlegroups.com>
Subject: [birmingham-hack-space] Re: Bat-Duino
Okay, a pound it is. I seem to remember it is traditional to take a
pound of flesh from around the heart. Is that correct?-)
I did the calculation in case anyone else wanted a few components
added to the order, and was wondering how much postage might be.
Garry
On Aug 26, 10:42 am, Andrew Thomas <andrewmarktho...@yahoo.com> wrote:
> Hi Garry,
>
> I'm happy just to call it a pound :)
>
[The entire original message is not included]
genzaichi
Just looking at my sonar goggles for The Bloop (Shameless plug - this
Sunday http://barg.org.uk/make/programme_for_enter_the_arena/) and
idly wondering if it'd be possible to build counterpart moth ears with
enough directional information to enable eared players to avoid the be-
goggled ones?
nikki
G Bulmer
several?
Do you care if the 'eared' players are helplessly confused if there
are several be-goggled coming from different directions?
On Aug 27, 6:35 pm, genzaichi <genzai...@gmail.com> wrote:
> Just looking at my sonar goggles for The Bloop (Shameless plug - this
genzaichi
stage. I suspect it might be 1:1 though.
I tend towards thinking the general almost-but-not-quite-helplessly-
confused edge of things is where it starts to get interesting game-
wise. :)
nikki
G Bulmer
thinking and skill may come into action ;-)
Well, there are several phenomena that could be used.
1. As the frequency increases sound becomes more directional, and
bends less well around objects, like, er, a head!
So, if the sound is coming from one side, two receivers one on either
side of the head, should get very different signal strengths.
So using two simple frequency division bat detector circuits, with a
little bit of "sample and hold" (like audio level meters) to control
the volume should work. 'Run away from the louder sound'.
2. Sound travels at approximately 340m/second, so the wavelength of
40KHz sound is 340/40,000 = 0.0085m = 8.5mm
If the sonar goggles emitted bursts of sonar, with gaps bigger than
the width of a head, then detecting which direction the sound comes
from should be straightforward; time the different arrival times of
the sonar. One area of possible confusion is when the sonar is
straight in front, or straight behind, but it shouldn't take moths
long to figure out how the effect works, and how to use the ears
better.
This is all made much more complex by sound reflection, and will need
a bit of tweaking, but in principle is reasonably doable with an
Arduino. (Of course Orone should be really good at this, as he may be
as quick as, about, 80x faster than an Arduino A2D, and certainly 4x
faster than the ultrasound :-)
HTH - GB