Simple sound trigger for cameras and flashes
Experiments with the external trigger function in my EOS 400D inspired me to develop a simple do-it-yourself sound trigger. The circuit detects sound pulses that exceed a certain pressure level. The sensitivity of the trigger is limited, but high enough to detect popping champagne bottles, bursting balloons or eggs hitting the floor. It can trigger cameras as well as flashes, as both are activated by the same mechanism of short-circuiting two wires.
The electret microphone is connected in the usual way with R4 and C3. I used the MCE-2000, but any other electret type will do. If you need more sensitivity look for a high V/Pa ratio mic. This value is usually specified in dB. C3 forms a high-pass filter and blocks the DC component in the mic signal. The voltage divider R1-R6 sets the polarisation voltage of the transistor slightly below the conduction threshold which is about 0.7V. Now, if the mic generates a positive signal, is adds up to the polarisation voltage and opens the transistor. The collector voltage is now pulled down from VCC to ground. The falling edge triggers the 555 for one pulse (mono-stable operation). The pulse duration is set by R3 and C2 according to the formula
Higher values for C2 or R3 result in longer pulses. The output pin of the 555 can be connected to an opto-coupler or an NPN transistor. The fist option provides some additional safety for the camera or flash unit.
In older flash units the trigger pin is connected directly to high voltage. In such cases an opto-triac or opto-thyristor should be used. Special care must be taken to avoid electrical shock!
The values of most components can vary between about -20% and +50%. The transistor is an NPN small signal type, e.g. BC547-550. If using a modern CMOS version of the 555 timer (e.g. LMC555, TLC555) the supply voltage can be reduced to less than 3V allowing operation from two 1.5V batteries.
The schematics of the circuit for Eagle can be downloaded here.
Before use the polarisation voltage has to be adjusted with R1 to obtain maximum sensitivity. Turn the voltage up slowly until the timer has triggered. Now go back a little to get below the threshold - ready. Turning the voltage further down reduces the trigger sensitivity. This might be useful in some situations.
Now you have to adjust the trigger pulse duration with R3 according to your needs. For high-speed photography it's a good idea to block the trigger for a few seconds after detection by selecting a long pulse duration. This helps to avoid secondary triggering due to ricocheting champagne corks, falling glass fragments and so on. Longer trigger pulses also allow to use the serial image mode of the camera to shoot multiple images after detection.
The pics above were shot by triggering the camera directly. The shutter lag is clearly visible in the right picture. If you need more information on shutter lag issues, have a look here or here. For real high-speed shots the flash unit must be triggered as in the pics below.
It is important to limit the flash duration by reducing the flash power. Values of 1/32 and 1/64 gave good results on my Canon Speedlite 430EX. At higher flash outputs the fast moving parts are blurred. By moving the mic away from the balloon the captured phase of the burst can varied. I used a small tube as a blowgun to puncture the balloons with a screw.
In the image above left the sensitivity of the sound trigger was too high, so a "parasitic" sound pulse was captured.
Shutter lag measurements for the Canon EOS 400D
Comparison of Shutter Lag & Startup Delay
Remote trigger pinout of the Canon EOS 400D
Supply for popping cork experiments
Note, that the potentiometer R1 needs to adjusted first. Turn it up until the circuit triggers without any sound and then turn it back a bit.
The 5V connection is connected to the battery positive terminal and the ground (GND) is the negative. This is almost always the case EXCEPT with split rail supplies, which this isn't.
As you cannot get a 5V battery easily, here are a few thoughts:
- I am confident 6V will not damage the circuit. The NE555 chip will operate at over 15V (from memory) so although things like timings may vary very slightly, the adjustments in the circuit will compensate. Note that I have not tried this but commenting based on experience.
- A battery pack of 4 x 1.5V cells will deliver 6V which will be completely satisfactory.
- That same holder can be used with 4 x rechargeable Nicads or nickel-metal-hydride batteries which at 1.2V each will give you 4.8V - again, completely satisfactory. Using these battery chemistries reduces the long term cost of powering the circuit.
- The circuit will almost certainly run from 4.5V, meaning a 3 cell battery holder with 3 x standard AA cells will operate adequately.
- Larger cells, such as C's will work and live longer.
- If you have a supply and really want to "lose" some voltage, simply wire a diode in series with the positive connection and you will get rid of about .6-.7V per diode. Use silicon diodes for this, a 1N4001 is a cheap and commmon device costing only a few cents. The more diodes you wire in series, the bigger the drop. The 1N4004 is another common diode/rectifiier which will work.
Hope this helps. My email is email@example.com should there be other simple queries...
you are absolutely right, RESET should be connected to VCC for better stability. 10nF at pin 5 is also recommended, but I achieved good results without the cap. It depends on the application.
When in 'Start system', the LED blinks 3 times, the camera shutter is opened (must be in BULB mode), tere is a 0.5 sec delay and then the PIC waits for the sound trigger signal. When it gets this, it waits for the set time and then triggers the flash, waits 0.5 secs again and then closes the shutter.
The delay time will be done with a 16 position hex BCD switch, which represents a time delay (in tens on milli secs). After the sound trigger is activated, the circuit waits for the set time before firing the flash.
I haven't got the components yet, so haven't yet tested it. Also I'm am not sure if having the hex BCD switch setting the delay in 'tens of milli secs' will work, maybe it should be 'milli secs'
I need a sound trigger to set off a flash unit for track and field athletics. Obviously the sound is a pistol report, unfortunately
a .38 calibre so the report is quite loud. How can your project
be adapted for the large inital sound produced. Any help would be gratefully received.
I was wondering if you could create an optical trigger so I can use the flash from my Canon EOS 350D camera to trigger a slave flash.
A friend and I tried to create the "Lullaby DIY Universal Flash" but we can't make it work. (http://lullaby.homepage.dk/diy-camera/usoft.html)
Of course you can trigger the cam only and use the cameras sync contact to fire the flash.
you can delay the trigger slightly by moving the mic away. One meter distance makes delays about 3 ms (speed of sound / distance). I'll think about a delay stage for the circuit.
Looking forward to playing with this now and getting some great shots. Would love to have the ability to delay the trigger by small amounts (in the milliseconds range) to capture bursting balloons or eggs at various stages.
Thanks and well done!