[islandlabs] keycam microcontroller
toolfox at gmail.com
Fri Aug 13 20:02:05 EDT 2010
On Friday, August 13, 2010 05:47:36 pm Burns, William wrote:
> At/after the last party/meeting, we got a microcontroller working.
> At the moment it's programmed to detect the power light on a key-cam,
> turn it on (if it's off) and press the "shutter" button every 3 seconds.
That's cool! How does it detect the light? Optically? Sense wires across the
> It's an 8-pin DIP chip, so the chip (at least) is the same size as the
> 555 chip on the intervalometer that Joe built and already interfaced to
> a keycam.
> I've got a pair of 8-pin sockets, and some perf-board from Radio-Shack.
> I'd like to move the microcontroller from it's breadboard to the
> perf-board, and interface it to a key-cam.
> Are you available to meet?
No. Maybe if I had known about this sooner....
> If not, could you post info to the list about how the intervalometer
> interfaces to the keycam?
Sure! Here's what I learned about the keycam shutter button:
One side is system ground; the other goes to the on-board controller. Short
the two and the camera triggers.
A direct short to ground is not necessary. Using the 2N2907 PNP transistor in
my circuit triggers the camera without problems, even though the voltage drop
through the transistor means the signal line is at about 0.6 to 0.8 volts.
That level is within the controller's logic zero threshold.
Another observation: the signal line reads 3.3 to 3.7 volts (or whatever the
voltage level is). I didn't take a qualitative reading, only noted that there
was a positive voltage.
Maybe it was something like 2.9 volts. Whatever.
Most importantly, this voltage is only present WHEN THE CAMERA IS TURNED ON
AND READY! This might be an alternative way to determine camera state.
The flight camera currently has two wires -- green and white -- coming out to
an RCA socket; switch signal to the center conductor, switch ground to the
Can I ask something on a more administrative level? If you've got a new
circuit partially designed as of Wednesday night and needed my help, why
couldn't you fire off a quick email then (or the next morning)? You have my
direct e-mail address. I would have heard about it early enough to be of
Forgive me for being blunt, but it seems to me that: 1) a final build hasn't
been made, 2) the new device hasn't been fully integrated to the flight camera
yet, and 3) the software would need debugging and testing for proper detection
of all possible failure modes plus the ability to put the camera back into the
And you want to launch this in 36 hours?
My "inner engineer" is screaming, "No, no, a thousand times no."
What you've got is good, there's just not enough time to do it for this
launch. I recommend that if we want to launch this weekend, the capsule has
to be "locked-down" so that flight projections can be made. We can't forecast
accurately if the payload configuration keeps wiggling around. Besides, it
makes your tight-ass "Flight Dynamics Officer" not fit to be around people. If
not for me, do it for Amy...she has to put up with me. :)
Since I'm not available to "lock things down," I can drop off the canister,
cameras, battery holder, and parachute tomorrow morning for someone else to do
the final prep. Where? Bill's? Lori's? This will have to be in the
morning...the earlier, the better.
More information about the List