Slashdot Mirror
DIY Sound-Activated High-Speed Photography
eldavojohn writes "Have you ever wanted to catch the perfect photo with your SLR camera but couldn't time the shot just right? Photography enthusiast Matt Richardson brings us an instructional video over at Make Magazine that shows how to use some very basic breadboarding and an Arduino Nano to do some high-speed flash photography that is timed by sound instead of your finger hitting the button on the camera. He pops a balloon and smashes a wine glass to show some results. His code is available on Github, and you can find more of this sly hardware hacker on his YouTube channel."
Seriously, why do we need an Arduino to make a sound trigger?
Oh yes, all the cool kids use arduinos now, so if you make an electronics project without one, no one is going to read your article.
Seeing how light travels much faster than sound, my initial reaction is that this is a terrible idea.
One of our competitors trademarked the term "hypothesis". From now on, we will call them "boneheaded ideas".
Why is that worthy of a Slashdot submi... OMG ARDUINO, WOOT!
A large portion of the canon cameras support CHDK (canon hack devkit). One of the nice features included in this is motion detection. This requires no external components to get it working, as it uses the CCD for triggering. There are numerous scripts that use CHDK to implement delays, time laps photo sequences or other funky actions. Most of the supported cameras are fast enough to capture lightning.
!
Back then, a similar project used op amps to trigger a flash unit. It was an article in one of the electronics mags I saw back in the late 1970's, titled "Build the Thunderbolt". (I Googled it, but came up empty.)
You adjusted the timing of when the flash was triggered, by moving the microphone closer or farther from the sound source. You could also have added a 555 timer, if you needed a longer delay than was feasible with a longer distance.
It reminds me of a discussion at the Electronics firm I am consulting for. They needed to add a 1/2 second delay to the startup of a device in a new product. I suggested they add a 555 timer circuit. They looked at me like I had two heads. Their solution was to throw a microcontroller into the product. Come on, guys! It can be done with a 555, a cap, and two resistors. It's crash-proof, too. Whatever happened to K.I.S.S?
Willie...
DIY Sound-Activated High-Speed Pornography, but got utterly disappointed by the fact he used Arduino :(
This is a cool proof of concept and neat little hack. But sound really isn't the best way to do high speed photography.
Look at the balloon in the video, you see the end of the action. I think ideally we want to see the moment the pin pricks it. Same with the glass, we want to see the hammer smashing it, not the moment after when the pieces are falling.
The CHDK has been around for awhile and can produce flash sync at least up to 1/60,000th of second on some pretty cheap cameras.
Simpler, if you don't like hacking your firmware, with high end flash heads you can simply use high speed sync mode and shoot around 8 fps at up 1/8000th of a second. Experimenting with wide apertures and higher (but still perfectly good)ISOs you can do pretty amazing things. Especially with the latest off-the-shelf radio strobe controls. You might have to work a little to catch the absolute perfect moment, but that's usually not hard with the right setup. TFA is a neat hack, but probably not the best nor the most direct route to great high speed photography.
Operator, give me the number for 911!
For a long time, the famous Edgerton photos were a staple of physics textbooks. E.g., you could see the (huge) deformation of a tennis ball being hit by a racket. But the Edgerton images are all copyrighted, and it would be really helpful to have CC-BY-SA-compatible photos that could be used instead in places like Wikipedia. I'm the author of some copylefted physics textbooks, and I really haven't been able to find much that's useful. There's this category on wikimedia commons, but there's currently not much in it that's useful educationally. IMO there are a couple of things that would be useful in physics education: (1) an image like the tennis racket, showing how an object's center of mass accelerates even while it's in contact with another object; (2) an image like the bullet going through the apple, which I believe shows that the speed of sound in the apple is less than the speed of the bullet.
Find free books.
The guy should smash his own head. That would be cool. Should make a great meme too!
In 1973 I had a tobacco tin, a bent piece of foil and a strobe light, airgun and camera. shutter open, fire airgun, shockwave completes circuit, strobe backlights pellet. trial and error found the right spot. The next one I did was a flicker photometer, but it was cool being allowed to bring an airgun into school. Jerry
They were talking about *adding* a micro to a mostly analog product. A 555 with the C and two R's would actually be cheaper. It was not meant to add features or function, *only* a .5 sec delay.
Willie...
I'm surprised the piezo had enough output, without amplification, to trigger the chip.
Willie...
You mean like this?
And then you could have an arduino with one of the analog output pins changing the R value in the 555 delay loop? Right?
Dude, that would be awesome!
Of course, it all depends, and I've not seen your design, so it could be that a 555 is simpler. In other cases, a microcontroller may be simpler. What's cheaper also depends on the circumstances. Price at digi-key for a NE555 from TI is 29 cents, while their cheapest microcontroller (PIC10F200) is only 34 cents (both @ 100 pieces). The microcontroller also comes in SOT23-6, so it takes up less board space, not even including the C and Rs. And maybe the microcontroller, in the same circuit, can also replace some other functions as well.
Multiply it by thousands. Those chips all have to be flashed and tested. That's time & money. With the R and C, the 555 just works, out-of-box. Crash-proof, foolproof.
Willie...
Reiterating - this is a mostly analog product. It does not already have or need an MCU.
They were thinking of throwing one in, *ONLY* to add a .5 sec power-on delay. There are other chips dedicated to this same function, but the 555 is universal. (Or, at least, it USED to be.) 3 parts; the IC, a cap and a resistor. No need to flash and test. Piece of cake for most assembly houses.
The MCU would require at least a crystal or some other clock. There is also the potential for some RFI being generated, or the MCU locking-up due to RF exposure. (This product is meant to be used in an RF-rich environment.)
Willie...
How is making a noise faster than clicking a button would be?
Multiply it by thousands and those capacitors and resistors begin to add up, multiply it by millions and the printed circuit real estate plus solder needed for the passive components start having a price impact. Flashing and testing a PIC takes less time than inserting, clipping leads, and soldering a couple of components.
Anyhow, when was the last time you needed just a half-second delay and nothing else from a circuit? With a PIC, the connections are the limit. From six pins in an 8-pin DIP to 40+ or more from the bigger packages, you cover a vast range of applications.
The only thing I really don't like about current PICs is the programming. I dream of an age, hopefully not too far away, when there will exist 8 pin PICs that cost less than a dollar running Linux and programmed in Python.
That's not reiterating, as that wasn't in your original message. It's adding additional details after the fact. Details that are utterly irrelevant to the design principles I enumerated in my message. This, along with the balance of your reply, shows you can't tell the difference between design principles and blind dogma
I would expect an intelligent consultant to know whether or not it CURRENTLY is universal - that it used to be is utterly irrelevant. Again, you advocate blind dogma over knowledge.
I'm sure glad I no longer hire consultants - I encountered too many just like you.
The units being made are all surface-mount. No insertion or drilling or clipping. There are already dozens of parts. The "real estate" is a non-issue, as there is PLENTY of room on the board. This device was already developed, and testing was being done on a few production prototypes, when the need for a brief delay was noticed.
In this application, there is absolutely no question that the 555 would be the MUCH better choice.
Willie...
It *is* a reiteration, because in my original post, I said, "Their solution was to throw a microcontroller into the product."
This implies the following:
1) This product does NOT already have one. .5 sec delay.
2) It was being inserted for no other purpose than a
I think you missed that.
It is pure assumption on your part, to say that I am spouting some kind of "dogma". What "dogma", pray-tell? That using a common, inexpensive, very reliable, simple device is better than a highly sophisticated (comparatively speaking) microcontroller?
The 555 is a mature technology, with very well-known characteristics. It is also very stable. This product is intended to be used in an RF-rich environment. The MCU generates, and could be susceptible to, RF energy. This could be a problem.
There also could be occasional noise on the power supply feeding the product. This would only cause the 555 to recycle, thus reinitializing the product, but an MCU could lock up. If MCU were to lock up, that would force the user to power-cycle it. The 555 wins for these reasons. Nothing "dogmatic" about that. It's common sense.
You've formed a very negative opinion of me, my experience, and my integrity, based upon very little information. I hope this additional information has clarified why I said what I said. In all honesty, I didn't think it would have been necessary to go into such detail. Maybe next time, I should just post their schematic... no, I think not.
Willie...
The place I worked for was very much analog. My former boss worked for CBS Labs, and was absolutely amazing. He could design almost anything using op amps! I assisted with the design of a number of nice little products.
That company I worked for was bought by another that specialized exclusively in CPU-based devices. Gobbling-up a product line that is 99% analog has proven to be quite the paradigm shift for them, so I have been consulting and providing tech support to their staff for a while, now. I like them, they are good people, and are good at what they do. I am learning from them even as they are learning from me. :)
Sure, you CAN use a CPU, but why, when a basic analog device will do the same job? CPU's can crash. Code can have bugs that crop up. CPU's tend to generate radio frequency noise, and can be susceptible to nearby radio transmitters. (This happens to be an important factor with many of this company's analog products.)
Barring physical component failure, an R/C time constant just works. Every time. Use analog for what it is best at. Why not use CPU's for what they are best at: Computation, decision-making, and automation? Marry the two when needed.
We had a product that used a PIC and analog together very nicely. An elegant little device. It was designed before my time there, but I helped to redesign and improve the analog circuitry quite significantly. That redesign has already proven itself in a drastic reduction in the number of returned products. Can't argue with success! :)
Willie...
Not to disagree, but most PICs have internal oscillators these days.
Tiller's Rule: Never use a word in written form that you've only heard and never read. You will end up looking foolish.
That's fine... but it is still a source of RF, and could still be susceptible to the effects of external RF.
Willie...