Photographing a launch with an event trigger.

Well I conducted some cheap and dirty experimentation with a Canon Digital rebel using a remote (wired) trigger.

Many cameras use the same connector for the wired shutter release, namely the 3/32" "mono" phono plug.

Completing the circuit releases the shutter, but there was a lot more delay than I had expected. In fact, by dropping an object, having it activate the trigger and then measuring how much it had fallen I found that the camera takes a picture about 75ms after the trigger is activated.

I set up an Estes Gnome rocket with an 1/2A3-8t engine (I think) and rigged the electrical contacts so that when the igniter fell, the bare wire would contact a piece of aluminum foil that was already in touch with the other contact.

Much to my surprise the rig worked, and the rocket used the 75ms delay to climb about 5 inches up the rod.

Time permiting, I want to make the system more robust by having a non-contact trigger (either optical or sound activated) and by having an additional, user-adjustable delay so that I can "catch" the rocket at about 6 to 15 feet of altitude. Also, I need more cable than the mere 5 feet I was working with here.

Nice job and experiment.

A good cheap and dirty optical sensor can be made. I have the plans around here somewhaere for a PlywoodCam..
Ah found the link again:
https://mypage.direct.ca/m/meisinge/index.html

Nice work!!!! Keep us posted on future experiments.

Interesting. I had another idea today and that was to wrap a coil around one of the conductors of the launch cable and connect that coil to a control box. When the ignitors break, the current through the launch cable should fall quickly. This could be detected and used to start the sequencing.

How about this - use the launch signal sent to the igniter to trigger the camera. Use a relay to control the closing of a circuit and that is connected to the camera (that way you aren't putting voltage to the camera, just open/closing a circuit). When you press launch button you can have the camera in multiple shot mode - and just hold it down through the launch to get pictures.

Edward

Originally posted by edwardw
How about this - use the launch signal sent to the igniter to trigger the camera. Use a relay to control the closing of a circuit and that is connected to the camera (that way you aren't putting voltage to the camera, just open/closing a circuit). When you press launch button you can have the camera in multiple shot mode - and just hold it down through the launch to get pictures.

Edward

Click to expand...

I'm guessing that the amount of delay between current start and actual liftoff of the rocket is too variable, so that I'll get results similar to hand-pressing the camera trigger.

I don't know how variable the ignitor-circuit-break to rocket-liftoff delay is, but I would assume it's very low.

Originally posted by FL_Flyer
Interesting. I had another idea today and that was to wrap a coil around one of the conductors of the launch cable and connect that coil to a control box. When the ignitors break, the current through the launch cable should fall quickly. This could be detected and used to start the sequencing.

Click to expand...

Close, you're on the right track. You would actually wrap ONE side of the current-carrying wire a few turns around an iron ring (or maybe even a bar), and also wrap around the iron a whole lot of turns of thinner wire. You would be creating a step-up transformer which will generate a low-current but high(er)-voltage spike in the secondary winding when the current in the primary is interrupted. That should be easily detectable by an appropriate circuit, such as triggering the gate of an FET like an IRF510 (available from Radio Shack).

You might even be able to use a small 120V-to-12V transformer, running the igniter current through the 12V secondary and picking the voltage spike off the primary.

But personally, whenever I get around to trying to set up a camera trigger, I'm going to go with the interrupted-light-beam concept, probably using a laser pointer as the source. The bottom line is not when the igniter burns, but when the rocket has moved.

Originally posted by TWRackers
But personally, whenever I get around to trying to set up a camera trigger, I'm going to go with the interrupted-light-beam concept, probably using a laser pointer as the source. The bottom line is not when the igniter burns, but when the rocket has moved.

Click to expand...

Good idea. Also, by adjusting the height of the detector, you can get better control of how far up the rod the rocket appears in the photo.

Or if you want a really simple, low-cost way of triggering, use a switch like this from Radio Shack



Mount the switch horizontally so that the body or a fin rests on the tab, or vertically on the rod so that the roller rests against the body tube. Either way, when the rocket moves, it releases the switch and triggers the camera.

Originally posted by BobCox
Good idea. Also, by adjusting the height of the detector, you can get better control of how far up the rod the rocket appears in the photo.

Or if you want a really simple, low-cost way of triggering, use a switch like this from Radio Shack

Mount the switch horizonatally so that the body or a fin rests on the tab, or vertically on the rod so that the roller rests against the body tube. Either way, when the rocket moves, it releases the switch and triggers the camera.

Click to expand...

And hopefully the switch doesn't melt down into a quivering puddle of molten metal and plastic from the exhaust plume .

Sound would be the most promising method, but would require working up (or modifying) a suitable sound-trigger circuit and adjusting it so it would not false-trigger due to other sounds. Of course, the adjustment method might not have to be electronic but physical by putting some padding over the microphone so only a really loud close sound would trigger it.

Years ago, I was going to try to work up an R&D project for detecting proper ignition of a cluster. For a possible shuttle model with three engines that all had to ignite at the right time, or it would crash. The model would be held to the pad until (or unless) all three engines ignited within a certain time window.

What I was going to do was to use switches attached to paddles under each engine nozzle. So when an engine ignited, the exhaust plume would press the paddle down. I think that would be a promising method to use for a camera trigger.

Unfortunately, the project fell thru. I had mentioned the project to John Pursley, and he mentioned an intriguing sensor method. He said that the plasma of the engine exhaust firing down inside of a hollow coil of wire could create a voltage or current that could be used as an ignition sensor. It was a lot more than I wanted to get into at the time. I mention it here just in case because it might be the ultimate engine ignition sensor.

Other sensors that had been considered were infra-red detectors, inside of metal tubes under the engines so outside light would not come in. I was concerned that a misfire might shoot out enough flame to be falsely assumed to be engine ignition. Thermistors were also considered. A bare thermistor would have probably worked, but would have been one-shots since they would have been burned up. And adding any heat protection to the thermistors to make them reusable would have delayed the reaction time unacceptably. Thermocouples were no good since they would be too slow to respond. Sound was also considered, but the sound of the first engine to ignite would have made it very difficult to not false-trigger the other sound sensors for the other engines.

Come to think of it, years ago Vince Huegele of HARA experimented with a self-triggered launch camera like this. I think he used sound as the sensing method.

- George Gassaway

Very interesting, George.

I'm going to take my time with this project. My first order of business is getting an oscilloscope. Something I should have done right out of college. Only in that way can I "easily" get a sense of the timing of all the events taking place; both on the rocket and within my electronics.

A false trigger wouldn't be such a bad thing for this project unless is occurs just before ( in milliseconds) before the real trigger because my camera has a >300ms delay between pics.

At first I was thinking of using a potentiometer to be able to adjust the delay (to account for rocket+motor combinations of different acceleration rates). Now I'm thinking of using multiple beam interrupts. Either 2 to get an average speed, or 3 to get an average acceleration. The delay can be calculated by the electronics in sub-millisecond time. A potentiometer could then be used only for fine tuning or perhaps a manual override.

I think the trickiest part for me would be to get a sense of what time periods are involved here, specifically how much does it take for the rocket to go from ...

ignition start to 1' up the rod.
1' up the rod to end of rod.
end of rod to 10' or 15' of altitude.

Even with infinitely fast electronics, the camera still has a delay of 75ms.

I used rocksim on a "bad case scenario" like an Estes Mosquito, and the accel was in the order of 5000 ft/s/s. Is that accurate? 150g + !!! My camera's 75ms reaction time would make this rocket too quick to capture automatically.

I like the little black switch method. It would be extremely reliable for detecting first motion, hard to trigger false starts, and beautifully simple. All you would need is to rest the rocket on a cheap, easily replacable actuator arm to keep the actual switch out from under the rocket.

Originally posted by FL_Flyer
I'm going to take my time with this project. My first order of business is getting an oscilloscope. Something I should have done right out of college. Only in that way can I "easily" get a sense of the timing of all the events taking place; both on the rocket and within my electronics.

Click to expand...

Unless you've got a LOT more cash lying about than I do, you might consider a USB oscilloscope from Parallax Inc (www.parallax.com). I have one, it's a nice little unit if you have a computer (preferably a laptop) with a USB 1.1 port free.

Disclaimer: I have no affiliation with Parallax Inc. other than they have too much of my money, and I have too much of their stuff. Someday I WILL fly an electronics payload running on a BASIC Stamp(tm).

Originally posted by FL_Flyer
I used rocksim on a "bad case scenario" like an Estes Mosquito, and the accel was in the order of 5000 ft/s/s. Is that accurate? 150g + !!! My camera's 75ms reaction time would make this rocket too quick to capture automatically.

Click to expand...

150 g's for a Mosquito is very believable. The proverbial "bat out of H-E-double-hockey-sticks". My Birdie is draggier than a Mosquito, but I have yet to see the thing actually leave the pad; I usually don't see it again until it bounces off the turf.

Incidentally, I guess I'm a "FL flyer" too, just from the other end of the state from you .

Originally posted by TWRackers
Unless you've got a LOT more cash lying about than I do, you might consider a USB oscilloscope from Parallax Inc (www.parallax.com). I have one, it's a nice little unit if you have a computer (preferably a laptop) with a USB 1.1 port free.

Disclaimer: I have no affiliation with Parallax Inc. other than they have too much of my money, and I have too much of their stuff. Someday I WILL fly an electronics payload running on a BASIC Stamp(tm).

Click to expand...

Thanks TWRackers. I looked at the specs for the USB o-scope, but I think it could "only" manage 1000 samples/sec. (I may be quite wrong on that). I think I'll need something that goes into the MHz ranges, just to leave plenty of margin. Also the traditional o-scopes usually have an adjustable DC power supply which I'll also need. Maybe I can pick up a used one somewhere, otherwise I've seen new ones online for about $350.