[Swlug] Further interesting calculations

[Rhys Sage]

The concept is to make a microflash. Those are designed after a trigger signal to wait a user-specified period then to give a brief burst of light and to do that only once before reset.

The aim with the flash is to take pictures similar to those in my book: "High Speed Photography - it's not Rocket Science". Those were taken with a Canon 540EX flash that I sold about a decade ago. High speed photography is a case of opening the camera shutter and locking it open then using the flash to do the exposure. The photos are taken in absolute darkness bar the flash. The shorter the flash, the faster the action that can be frozen. A Canon 540EX has an absolute minimum exposure time of 1/38,000th of a second. The very fastest Xenon flash has a minimum exposure time of 1/100,000th of a second. Xenon cannot go any faster because although it is inert, it has an afterglow that takes a while to fade.

An LED flash can give exposures down to the millionths of a second as there's zero afterglow. With my 540EX, the best photo I could take of a .177 air rifle pellet travelling at 1200fps was quite blurry. The delay timer I have worked well and got the exposure time spot on. I want to build an all-in one unit.

I have an LED array I built 15 years ago with 227 white LEDs on it. I'm trying different resistors to get optimum brightness since the online resistor calculators are wrong. I should have put resistors when I built it but they're all in parallel so now I need super-duper 15W resistors. I have to order those.

The variation in the delays on the Arduino and Pi are a little worrisome but I should not be gauging equipment using its own timers so I have ordered a timer kit that I'll have to solder together. Looking at the pictures, that shouldn't take more than an hour to solder and assemble. 

So - the timing events needed.... From audio trigger to flash. This can be variable and could be anything from 1/10th of a second down to 1/50,000th. Nothing particularly fine needed there. I can always move the audio sensor back a bit to give myself more time delay while remembering the speed of electricity is 270,000km/s. The flash duration needs to be as small as possible variable up to 1/30,000th of a second and I'm hoping to go orders of magnitude faster than 1/100,000th.

For my own use, the fastest thing I might be photographing would be 9mm NATO which travels at up to 1,400 fps unless I can borrow something that'll shoot .220 swift at 4,665fps.

For the flash duration, reading Edgerton's book (and a few others) it seems that if I power (in their case an air-gap flash) the flash from a capacitor discharge then electrolytic capacitors are too slow and the type needs to be air-gap or mylar. I don't want to use an air-gap flash even though I have some of the kit to build one (don't ask) as they're a bit spicy in use, carrying around 50KV. The other downsides to air gap are copious quantities of ozone, intense UV light and radio interference. 

Sorry if I sound a bit boring here. I wrote a couple of books about it. Both could do with a massive re-write as they were written during one of the worst, most traumatic periods of my lfie.

So, having blustered profusely and bored everybody to death, I'll answer Colin's question...
I really need both bits of timing - they delay and the exposure to be predictable.

Rhys Sage