Highly accurate competition timing

[Raccoon]

As some people on this board can attest, I am already developing a 99.9% accurate timing mechanism for competitions that will be completely objective and highly accurate. More details as soon as I have the first prototypes out the door.

I'm certainly not one of those who can attest, and I STARTED this thread! What gives?!

(so out of the loop)

[BazookaMedic]

Why not rig the shackle to a switch mechanism so that when it opens it stops the time on that contestant. Its simple, easy to rig, and effective. The only human erroris in starting. It would just be like a light switch flip on flip off.

[cryptocat]

I built a little AVR-based timer for just this purpose over the last couple of weekends - time is displayed to 5 decimal places (10usec). I haven't calibrated or temperature-compensated my crystal, since I assumed that for the duration of a single picking run crystal error wouldn't matter that much.

Trigger methods I looked at:
1) Conductive: touching the lock with the wrench completes a circuit. This means that competitors must use the supplied, wired wrench.

2) Capacitive: touching the lock with the wrench, a tool or your body (or even getting near enough) can trigger a sensor. Not as simple as other trigger methods, may not be as accurate.
3) Optical: my favorite. Illuminate a couple of photodetectors - maybe use a green LED. Putting a wrench or a tool in the lock will cause a detectable drop in light levels. Simple, cheap, scalable to all kinds of mechanisms.
4) Electrical: some locks may come wired to detect interaction... this may be useful. Not many locks are wired for this though.

Stopping the timer can be done with a 3-position rotary switch on the tailpiece. "run" is position 2, rotate the plug to position 1 or 3 to stop. This is reasonably resistant to electrical bouncing - breaking the connection at #2 isn't sufficient to stop the clock, you actually need to connect to #1 or #3. An optical trigger can be used here too - a little bow-tie, propeller or flag can interrupt a beam of light, same as the start.

[cryptocat]

I forgot to mention: with optical triggering, I specifically chose to use a visible source so everyone knows when the trigger beam is live, and when it's been tripped. It's not quite as obvious if you use an IR beam.

I guess you could use a a conductive foil seal to trigger - breaking it with your tools starts the timer. Then you need to worry about consumables and carefully resetting it.

[adrenalynn]

Actually, the capacitive should be exceptionally accurate. When a conductive pick enters the lock, the R/C change will be huuuge, a little less so with rubber insulated pick handles (since the body is a large capacitor and you're removing it from the equation)

A sliding window should be pretty indicative. Whip this up an Arduino in ten minutes... If you're interested, I'll do so tonight and post a graph of results (using Processing)

[jgencinc]

What about using 2 ground straps, like you use to work on computers. one attached to hand, one attached to lock. When your hand touches the lock, resistance is lower and it starts the clock, when you drop the lock time ends.

[cryptocat]

Part of my approach was to minimize the amount of junk in the way. I'm sure someone would complain if I made them use a wrench with a wire on it, or made them wear a static strap. I wanted to be able to time anyone who shows up with a reasonable set of tools... sledgehammers and thermic lances are not reasonable for a picking contest.

[adrenalynn]

I'm wondering, for most locks, about just a simple electromechanical or similar. Some plug has to rotate at some point, or some hasp has to open or... Regardless of when the contestant says "Open!" - it aint _really_ open until it's... open, ya know?

So why not electrical contacts that the act of rotation or hasp springs bring into contact with one another? "Go" hit the button... tick tick [click] timer stops

[tballard]

Okay, I'll take a swing at this. Drop the idea of catching the exact moment the pick enters (or touchs) the lock, and instead include the time it takes to enter. Fancy sensors and optics are fun, but a good solution should be constructible by the average high school student with access to only say.... the crappy selection of parts at Radio Shack (ewww...) and a rudimentary "electricity is magic which runs around in wires" level of knowledge.

All which is needed is three microswitches and timing circuit. Two of the switches are configured so they have to be held down by the weight of the competitor's hands, and are placed far enough away from the lock that you can't hold the switches down and put a pick in the lock at the same time. When either (or both) of the switches are triggered by the competitor's hands (and thus tools) leaving the "at rest" position, the timer starts. When the lock opens, it triggers the third switch, stopping the timer.

With this scheme competitors can switch hands, tools, etc without needing an adjustment to the mechanism. The mechanism stays out the way of the lock, and the start-triggers don't have to exist in a matched pair with the lock. (Imagine a whole series of locks, each with only an "open" trigger and a connection plug. To be timed on a lock, plug the starting switch/timer assembly in, and go....)