1. Home
  2. Electronics Down Under
  3. How to get 1.5Volts from a microphone.

How to get 1.5Volts from a microphone.

I have cobbled together several devices for triggering remote cameras, based on a Passive Infrared Receiver. Briefly, the PIR, which is designed to sense a passer-by and turn on six white LEDs for a few seconds to illuminate a stairway or whatever, has one of the LEDs replaced with a connection to an analogue-to-digital input of a Picaxe. The Picaxe program waits until the passer-by (in this case a wild lyrebird singing and displaying on his performance mound ) triggers the PIR and takes photographs every five seconds for one minute. After the minute, the Picaxe program tries again, the assumption being that if the bird is still performing, the PIR will pick him up again. Not so. What happens is that the bird stands stock still for up to twenty minutes and doesn't trigger the PIR again until he leaves. Our only photographic record of his memorabloe performance is the first minute after he arrives, and a brief shot of his retreating backside as he leaves. But all the time he is there he is singing, and deafeningly. So what I need is a signal from a microphone. I have programmed the Picaxe to recognise a level, however fleeting, of 1.5Volts from the PIR, and I'd like to add to the existing setup a similar, ~1.5Volts from a microphone. Just a voltage. No attempt to reproduce the actual sound. I would need some kind of level control, in order to adjust the output to respond to the target bird's close-up song and not respond to his rivals who are some distance away and nowhere near as loud. I have looked at a kit for a voice-controlled microphone which would probably work but needs 12Volts worth of batteries. I need at least four of these devices and possibly eight, so they must be cheap. I feel that there must be a simple way to achieve this. On past experience, if there is a simple answer, this is the newsgroup where I will find it. Can anyone help?

Reply to
L.A.T.
Loading thread data ...

Why not just have to camera trigger for 20 minutes instead of 1 minute then? Storage is cheap...

You'll have to experiment a bit and figure out what levels you get out of your electret mic, and what the background levels are. That will require some form of data logging capture of the actual audio levels (or waveforms) so you'll have some reference levels to work from for your final solution. That starts getting a bit messy and potentially unreliable I suspect. This solution doesn't give good vibes... You might get better background sound discrimination with a directional shotgun mic.

I'd just shoot images for longer using your existing setup and sort through the images later.

Silly question, but if you want to capture the bird "performing", wouldn't a video camera be a better option? i.e. if the bird is just sitting their "stock still" as you say (singing?), what's the point getting that on still camera?

Dave.

--
---------------------------------------------
Check out my Electronics Engineering Video Blog & Podcast:
http://www.alternatezone.com/eevblog/
Reply to
David L. Jones

................. He can come for a couple of minutes or for up to twenty. ..................The warmth of the sun reflecting off the clay, interrupted by moving branches, gives two or three false triggers each session. ..................The camera batteries will last for up to seven hours while waiting, but will run out after less than an hour and a half of shooting.

.................The cameras' CF cards are limited to 2Gig.

...........................The Mics will be directional.

..........................On any given day, we will have six to eight mounds staked out. ...........................On most days he won't perform where we want him.

.......................We have eight Canon G2 cameras. Most were bought on eBay for about $150 each. Video cameras are too expensive. (YET. Maybe in a year or so).

................. Sorry, a bit of hyperbole there. He is displaying his amazing tail feathers in a most photogenic series of poses. ..................His warm, infrared-emitting body is not moving enough to affect the PIR.

Reply to
L.A.T.

You could always a microwave or ultrasonic doppler sensor instead of a PIR, alternatively the better quality PIR's had different Fresnel lenses for the front for different applications: longer range/better gain.

It may well be that a really cheap PIR is better suited as it won't tune out the lyrebird as background noise.

Reply to
Mark Harriss

Using sound detection here is an interesting problem that I've dealt with several times in other scenarios. Just measuring level is likely to result in a spurious trigger rate that's far too high if the setup is sensitive enough to detect anything. I'd be considering some processing of the sound input, as far as I know, Lyre bird song can contain single frequency sounds, it's easy to pick this with a small microprocessor these days, with fast fourier processing. It would require someone with micro programming skills, however there's tons of FFT code googleable. Bird song whistle will show as a clear spike in a data array that will stand out from typical background noise in a bush environment, which will be mostly white noise or close to it. You wouldn't need to know the frequency, just that there was one or a few isolated frequencies present which are significantly above the background.

I'd get a recording of the sort of thing you expect, and use that to test the algorithm.

Sorry if that all sounds high-tech, but I believe it could be a viable solution.

Reply to
Bruce Varley

Have you tried motion dection using using the camera?

formatting link

Dave.

--
================================================
Check out my Electronics Engineering Video Blog & Podcast:
http://www.alternatezone.com/eevblog/
Reply to
David L. Jones

....................Alas, this so0ftware works on Canon G7 and G9. ....................Our cameras are Canon G2, which are seven? years old. .....................When we were capturing images to laptops. we used ZoneAlarm, with limited success.

Reply to
L.A.T.

No matter what quality of lens the PIR only triggers if there is movement - then bird standing still and singing won't qualify.

Reply to
David Eather

How about a notch filter tuned to the bird's frequency? - simple, cheap and plenty of info on how available on the net.

Reply to
David Eather

How about one of the rare tap dancing birds? ;-)

--
You can\'t have a sense of humor, if you have no sense!
Reply to
Michael A. Terrell

A PIR Fresnel lens assembly is comprised of smaller lenses, it's possible to buy different types that trade off range with the amount of movement detected: If you fitted a lens with many more smaller lenses in it, the range suffers but it will now trigger with only slight sideways movements as the IR light from the bird is focussed onto the sensor by different parts of the array.

With the right lens the bird would have to be totally stock still to avoid detection.

Reply to
Mark Harriss

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.