sound-controlled flash trigger

[sylvie369]

I've been working on a project for my brother, the photographer

http://www.somajones.com/

to build a sound-based trigger for a flash unit. I've put together this circuit with no luck so far:

http://diyphotography.net/universal_..._flash_trigger

I'm fairly sure that my problem has to do with the amplifier part of the circuit coming from the microphone. I have a Propscope (thanks, Parallax!), and have been using it for troubleshooting, and I'm very pleased that it has been so useful. So far I've verified:

- that when I apply power, the 1 uF capacitor properly charges up.
- that when the input to the MOC 3020 optoisolator is triggered, the flash fires reliably.

I have a small electret microphone from Sparkfun, and I had hoped to find a test point in the circuit where I could verify (using the Propscope) that the signal from the mic is being properly amplified before going into that second BC 547 transistor. I put the scope ground to circuit ground and then the probe to the base of that transistor, and made sound into the microphone, but saw nothing on the screen. It seems to me that if things are working properly in the amplifier part of the circuit, I should get a voltage there that changes with the sound level into the microphone.

I also have that same microphone mounted onto a small amplifier board (also from Sparkfun). When I put the scope on the output of that board, I do see waveforms as expected. I hooked that output to the base of that second BC 547 in the circuit. When I turn the circuit power on, the flash triggers after a few seconds, sound or no sound. Here's what I believe is happening: When I just hook up the output of the mic/amp board to the scope, I see waveforms around a roughly 2V offset. I suspect that the 2V that's always there is itself enough to trigger the transistor, and therefore when the 1 uF capacitor builds up enough charge to fire the circuit, it fires right away, regardless of sound level.

That would be good news, I think, as it would suggest that everything in the original circuit except the amplification of the microphone is working fine. It also suggests that I can replace the amplification part of that circuit with the amplified breakout board mic, if I can get rid of that 2V offset.

How am I doing so far?

So it occurs to me to hook up both channels of the Propscope, one to a variable voltage (using a potentiometer/voltage divider circuit) input to the base of the second transistor (the one that triggers the circuit) and the other channel to the capacitor input to the MOC 3020, which builds up as the capacitor charges and then drops off to zero when the circuit fires. That way I can see what the voltage threshold for firing that transistor is, and use the potentiometer/voltage divider to bring the output of the microphone/amp to a point below that but close enough that sounds into the mic trigger the circuit.

Make sense? Or am I hopelessly confused?

[Leon]

My Nikon D80 has an IR sensor and I built a remote control for it based on a PIC circuit with software that I found for an older Nikon camera. If I wanted a sound-actuated circuit for it I'd simply add a microphone and amplifier to the IR remote control. The camera can trigger the flash, of course.

[Phil Pilgrim (PhiPi)]

The biasing on the first transistor looks a bit high to me. Try adding bias resistors of various values from the base of the first transistor to ground. I was wrong. I tried that part of the circuit as presented, and it works fine.

-Phil

[Zoot]

Leon -- triggering the shutter is not the issue. The method he wants to use involves taking a 2 second or so exposure in utter darkness. A separate flash is triggered by an event (sound, IR beam, whatever) so that the only exposure is during the brief flash when the event actually happens. There is no delay for the "shutter" because it's already open and exposing.

[sylvie369]

Zoot is correct (though Leon's point is a good one for another project I'll like to build for my brother).

Phil - it's very possible I've got a wire in the wrong place or something like that. That's why I was trying to test it. Should there be an amplified sound signal (a voltage that varies with sound intensity) at that point where the wire goes to the base of the second transistor? I'm not getting one. Or do I have to measure it somewhere earlier?

If there is supposed to be a signal there, I'll tear apart the first part of the circuit and rebuild it on a separate board and see if I can get it to work there.

Thanks for the help.

[Phil Pilgrim (PhiPi)]

Should there be an amplified sound signal (a voltage that varies with sound intensity) at that point where the wire goes to the base of the second transistor?

Yes. Also on the collector of the first transistor. The circuit responds well to loud impulses, such as a hand clap. Did you try that?

-Phil

[sylvie369]

Yes. Also on the collector of the first transistor. The circuit responds well to loud impulses, such as a hand clap. Did you try that?

-Phil

I'll try again tomorrow, but yes, I used hand claps first. No luck.
I probably have it wired wrong. I'll tear it apart and start over.

[BR]

Is there some reason a sound impact sensor wouldn't work?

[sylvie369]

Is there some reason a sound impact sensor wouldn't work?

Nope. And I can't explain how I failed to know that Parallax sells one. I just ordered two of them - thanks for pointing them out. One of those with a MOC 3020 and I uF capacitor should be all that's necessary to build a sound-based flash trigger. I'll post here about how that goes.

I'm still going to try to figure out what's wrong with the do-it-myself version, though.

[sylvie369]

I connected up the sound impact sensor to the flash trigger. It works, but the timing is off.



The normal solution is to move the microphone closer to the source of the sound, but in these pictures (I have four, all like this one) the mic is only inches away. I can't get much closer. I wonder if it's a matter of the flash itself taking longer to trigger than it should.

It does work reliably, which is pretty slick, and for somewhat longer events would be a pretty good way to capture things. I'm going to put it aside for a bit and think about it - suggestions welcome.