This simple circuit detects small changes in the earth's magnetic field to determine when a rocket has tilted to a horizontal attitude, presumably at apogee. It then fires a power transistor which sends current throught a pyrotechnic recovery ejection device. It is small, lightweight and inexpensive.
Building The Flux ComparatorYou can wire a power switch in series with the battery (or else hook up the battery clip every time you want to power up). Your switched battery leads go into the Bat terminals, with the positive side toward the Pyro terminals. Red is positive, black is negative. The Cont terminal block can go to an LED mounted on the airframe so that it can be seen when the rocket is assembled, or you can solder an LED onto the PCB in the D1 position.
The KMZ51 magnetic sensor is sensitive to ferrous metal objects, so you need to avoid steel mounting screws, threaded rod, U-bolt, eyebolts, launch rods, etc. in near proximity to the sensor. I haven't done any rigorous testing, but large steel objects within about four inches seem to cause problems. The stainless rods I've tested don't interfere with the sensor. Smaller steel parts (e.g. launch rods) need to be about an inch away. Of course, you can easily test it yourself. Being in the presence of large ferrous objects can throw off the sensor, so you'll occasionally need to press the reset button when the UP end is up.
And finally (with all due respect) mount the UP end up.
One of my favorite power sources is a 145mAh lithium-polymer battery. It which can deliver a lot of current without dropping the voltage very much. A 7.4V pack weighs only 10 grams, but requires a special charger. I also like the new Lithium-Iron-Phosphate (Li-Fe-PO4) 9V rechargeable cells. A 9V alkaline battery should work fine. Test your battery with your chosen pyro. Christmas light mini-bulbs pyros (around 4 ohms) can be powered fine by a 9V alkaline battery, or even (barely) an A23 (a.k.a. MN21) 12V alkaline security battery. Really, any battery with a voltage from 6V to 15V should work, as long as it can fire your pyro charges.
The version without a G-switch is easy to test: you just turn it over. (When you first build it, you need to hold it vertical and press the Reset button.) The pyro will trigger. You can put a light bulb or an LED and series resistor into the Pyro terminals and watch when it lights up.
But how do you test the G switch? Here's one idea. Attach the unit to a bicycle wheel, with UP pointing inward, between spokes near the rim. If the spokes are stainless steel, the flux sensor won't be fouled up, but steel spokes will need to be clear of the sensor by an inch. When the wheel is spinning, centripetal acceleration is velocity squared divided by radius. You need to measure the distance from the G-switch's location to the axle. One G is 32.1 ft/sec/sec, or 9.8 m/s/s. If you trust your G-switch's trigger threshhold, you can do an easier test: hook up everything and turn it on, hold it all in your hand with the UP end pointing toward you, arm extended, then swing it around like you're winding up for an underhand softball pitch. It's pretty easy to generate two G's this way. Once armed, you can test-tilt the device and check that the pyro test light glows.
Care must be taken when using steel launch rods. Aluminum rails pose no problem, unless steel rail connector hardware is nearby. If you use this unit in small-diameter rockets which use launch rods, you might want to consider bringing your own stainless-steel rod. I have heard of magnetic sensors getting fouled up by steel launch rods which have somehow gotten magnetized.
If you have the version without a G-switch, wait until it's vertical to power it up. It wouldn't hurt to press the reset button, if you can get to it. (Generally, this is not needed unless the sensor has been near a large ferrous object.)
Standby current drain isn't high, but as with all electronics, you don't want to leave them turned on for hours awaiting launch.
Last updated: Sep 23, 2007