I’d like to create a sound trigger for doing some high-speed photography. This would cause a sound event picked up by a microphone to trigger an off-camera flash. I’ve found a good electrical schematic that seems to do what I want, on this page (link). I’m interested in the top one with “double-fire protection”. The reason I like this one is that pulses an LED to activate my flash’s optical slave function. This is easier than other trigger circuits I’ve found, which involve a cable to the flash. Such a setup would involve me rigging up a hot shoe adapter
If I were just following those directions, this would be simple enough. But I’d like to create an adjustable delay between the sound event and the triggering of the flash. The author mentions adding a delay circuit, but only links to another page (link) with a delay schematic. I want to put this circuit into the first one.
Here’s the catch: I’m not an electronics expert…or even close to it. I can build curcuits of this level of complexity given a schematic, but I don’t really understand how to “read” circuits like this. I tend to get confused trying to trace my way through ICs and transistors. So I need some help in understanding what the result needs to look like when you add the two circuits above.
Do I put the timer between the amplifier and the light generator halves of the main circuit, thereby replacing the output of the amplifier with a pulse from the 556 chip?
Or do I replace the simple LED in the original circuit with the timer in its place (LED attached directly to the timer output)?
In either case, it would seem that I want to eliminate the SCRs from the timer circuit. If I understand their function, they use the timer pulse to close a “switch”, pulling the output to ground for the duration of the pulse. That makes sense for a cable-operated flash, where I’m just trying to bridge the terminals on the flash to fire it. But in this case, I don’t want to close a switch…I want to power an LED for a short pulse.
So if you electronics buffs were going to add this delay circuit to the trigger circuit, how would you do it?
If I were doing it, I’d skip the cumbersome analog timing circuitry altogether an build it with a microcontroller, such as the inexpensive PIC 16F84 or 16F88. These include internal timers and a main advantage of these chips is that if you wish to make changes to the behavior of your device, it’s simply a matter of programming. I’d be more than happy to code the software and program the chip for you, if you decide to go this route–while the PIC chips are cheap enough, the initial outlay for a programmer and the time necessary to learn how to program them might be more than is worth to you.
I considered the microcontroller approach. I like the flexibility it provides, but not already having the equipment to use one kept me from going that way. I know they could be very useful for other projects, so it’s something I may consider. But for now, I’m going to try to make it the harder, cheaper way. =)
This is an extremely simple project for anyone who knows and I think using a PIC is overkill.
I’m willing to lend a hand but we need some specs: How loud a sound? Like a sudden clap or something like that? And an adjustable delay between what and what? (I mean amount of time)
If you need a very exact delay, repeatable every time with great accuracy, then maybe a PIC is the way to go. How much accuracy do you need?
I didn’t reply to this thread when I first saw it because I was going to say basically everything that QED said. If it were me, I’d use a PIC. They are simple and cheap, and for me, programming one is easy.
It doesn’t take a lot of money to get into PICs. The software to program it is free from Microchip. A dirt simple programmer can be constructed out of very simple and cheap parts.
555 timers and such are great if you want to do things the way we did back in the 70’s. With a little creativity you can do a lot with them. But, as long as you can handle the programming, I’d recommend getting into microcontrollers.
I disagree. They’re highly flexible, will drastically reduce one’s part count and are dirt cheap (16F84s can be had for under a dollar each), especially if someone else is letting you use their equipment and software to program one. That part of my post appears to have been skipped over.
It wasn’t skipped over by me. Your offer is greatly appreciated…but relying on others for this type of support isn’t a long-term solution for a DIY type like myself. So I’ll either invest in the means to do microprocessor programming myself, or I’ll do it another way.
Again, I appreciate your help. But I can’t take you up on it, because most of my reason for doing this is the learning experience.
I’ll look into making something like engineer_comp_geek posted. Thanks to both of you for pointing me in that direction.
Indeed. If you didn’t want to go through the trouble of designing your own and scraping up the parts to make it, I’ve seen programmer kits for around $20–some soldering required. And they normally include some interface software to load the HEX file your compiler/assembler generates into the chip. Some even include a freeware compiler.
One note about MPLAB–the programming and debugging environment from Microchip: while it is free, it doesn’t include a C compiler, only an assembler. Now, that’s fine for people, like me, who understand assembly language but C is a lot easier for neophytes to deal with, IMO. Enter Hi Tech C. They offer a lite version of their PICC compiler which supports a number of low-end and midrange PIC chips, including the 16F84. This has the further advantage of being able to integrate with MPLAB and the whole is pretty easy to learn and use. Plus, you’ve got at least two of us who’d be willing to help you get started and answer questions.
Incidentally, this is the programmer I use. I highly recommend getting the optional ZIF socket–it makes it much easier inserting and removing chips. It comes with the PICC Lite software I mentioned above.
I studied the circuit. It’s rather clever, but it does not easily lend itself for implementing a delay.
The easiest way to implement an analog delay is to insert an RC time constant into the circuit somewhere, where a single-pole RC circuit is arranged like an LP filter. But in order to implement such a delay the drive circuit must have a fairly low source impedance.
The left-hand circuit is a common-emitter amplifier with capacitively-coupled input and output. Because you’re dealing with AC pulses (in the small signal model), there’s not much you can do in terms of adding an RC delay. At least not easily.
The right-hand part of the circuit has a fairly high impedance when a DC analysis is performed. This is due to the 1M resistor. When the circuit fires, and BC547 turns on, the current through the LED is delivered by the capacitor, not the 9 V supply. The problem with trying to add an RC delay is that an RC circuit will gobble up energy (due to I[sup]2[/sup]R dissipation in the resistor), and there’s not much energy to spare in this part of the circuit - it’s all coming from the capacitor when the circuit fires, and you’re going to want to steer as much of this energy to the LED as possible. This is why RC delays only work when driven by a source with low impedance.
The left-hand circuit looks O.K., but the right-hand circuit should be redesigned if you want to incorporate a delay. The rectifier portion can stay, but instead of using transistors I would be inclined to use op-amps and/or comparators. These devices have low impedance outputs, and RC delays can be thrown on their outputs.
Even better, I would be very tempted to forgo the op-amp/comparator route and use two or three one-shots. In fact, a couple years ago I designed a circuit for triggering a high-speed camera when a light-beam was broken. I used some programmable one-shots and a few logic gates, and it worked great.
aerodave, I’d suggest a compromise; pick up one of the kits QED linked to and learn to code it yourself. As mentioned, you’d end up with a system that could easily be modified or repurposed, plus you’d have great fun doing so. You’ll be able to get plenty of help here and elsewhere on the 'net.
On a related note I’m just starting to play with my favourite bit of Christmas loot - an Xilinx Spartan 3A FPGA developer board As a starter for 10 I’m building a circuit that will allow me to reverse engineer the truth tables and logic functions for PAL/GAL(small programmable logic chips).
I repair old arcade games and whilst MAME have done a superb job in archiving ROMs (the program code and sound/gfx data), they are missing most of the PAL/GAL data. If these chips fail then you’re stuck with a dead board.
My intention is to build a system which clocks through all combination of inputs whilst monitoring the outputs. When complete it will reduce the IO states to a series of logic equations (for most PAL/GAL - I’ve not yet figured how to resolve PLDs which implement state machines). I can then store this data in a universal JED format which can be bundled with the game ROMs. This can be used in the future to reprogram new PAL/GAL devices, allowing dead boards to be brought back to life!
As a starter for 10 I’m building a circuit that will allow me to reverse engineer the truth tables and logic functions for