Magnetic Apogee
Detection Sensor
This schematic is a modification of the
original
designs published by Robert Galejs in
Sport Rocketry magazine
Sport Rocketry, Sep/Oct 1999, pp 6-9
Sport Rocketry, Mar/Apr 2000, pg 18
Sport Rocketry can be obtained by joining
the
National
Association of Rocketry
His article can be read at
Essence's
Model Rocketry Reviews
Click HERE
for a direct link to that article
Honeywell
Magnetic Sensors
The HMC1001 can be obtained from
Newark
Electronics
Search for part #91F4713
After reading Robert's article,
I immediately wanted to build this project so I ordered an HMC 1001 sensor
and started breadboarding the circuit. As it turns out, there were
a couple errors in the schematic that appeared in Sport Rocketry.
I took Robert's design apart and did a bit of redesigning to make it work.
There has since been a correction published, but I had already come up
with a much simpler circuit that works perfectly. Instead of trying
to amplify then compare the outputs, I simply used a single op-amp as a
comparator right off the magnetic sensor. Like the original, this
drives a Field Effect Transisor to fire a flashbulb (or e-match) circuit.
I also found that the voltage
regulator really isn't neccessary. The components are rated for up
to 12 volts, so a 9 volt battery could be used to power everything without
having the 5 volt power supply from the regulator. Be cautious as
the HMC 1001 can be damaged if it sees more than 12 volts. According
to Honeywell, the circuit performs better with a 5 volt supply, so I have
left the regulator in this schematic.
Another option on the power
supply is to use the 544 battery. This is a small 6 volt battery
found in the photo section. The battery is a little smaller than
an "N" cell and measures about 1/2" diameter and 1" long. They are
also available in Lithium. I built a circuit without a voltage regulator
using a single Energizer 544 battery. It was able to fire the smaller
flashbulbs taken from flash cubes very well, but didn't have the power
to handle the larger AG-1 flashbulbs. Of course it was easy to simply
replace that battery with a standard 9 volt and it will fire the AG-1 bulbs
just fine. I haven't done any testing with electric matches or any
other type of ignitor.
For the serious weight conscious
project and to guarantee that you have the power to fire your ejection
charges, an idea I came up with here is to use two 544s in series.
Two of these little batteries end to end are just about a perfect match
to a single AA battery. The diameter is slightly less, but guess
what? They fit perfectly inside a piece of 13mm body tube, which
just happens to have the same outside diameter as an AA battery.
So cut a 1 7/8" long piece of 13mm body tube, slide in two 544 batteries,
and put it in a standard AA battery holder. You now have a 12 volt
power supply that weighs about 1 ounce compared to over 1 1/2 ounces just
for a single 9 volt battery. Not to mention the reduced size.
Pretty sweet, eh? If you use this power supply, you have to use the
voltage regulator circuit or you risk damage to the HMC chip.
Another noteworthy item
that came up during bench testing was that when you first power up or even
disconnect the circuit, there is a very brief surge of voltage that will
go through the flashbulb. I was using an LED on the bench during
my testing in place of the flashbulb, and it flashed every time you turn
the power on or off. This is apparently a leak through while the
capacitor is charging, or may be a quirk in either the op amp or the FET.
Either way, it could cause the flashbulb to fire, so I added a shunt to
the flashbulb (which you ought to do anyway, no matter what circuit you
are using). This could either be a piece of wire that you cut just
before leaving the rocket on the pad, or a jack with a shorted "remove
before flight" plug. I set mine up with a jack and plug setup.
Just make sure the plug is in place before connecting the flashbulb and
before turning on or off the power. Although I put a power switch
on mine (SPST slide switch at the positive battery terminal), you could
just install or remove the battery to power it up if the battery will be
readily accessable.
The push button switch in
this circuit is the set/reset circuit for the magnetic sensor. If
it is exposed to a strong magnetic field, the orientation of the sensor
can be changed. The HMC circuit has a built in coil to apply a small
correcting magnetic field. Before flight, you simply tap this switch
to discharge the small capacitor through that coil and make sure the sensor
is oriented properly. The sensor as wired in this circuit has to
be mounted so that the beveled top edge is pointing toward the nose cone
and the pins are pointing toward the motor. Do not put any ferris
metals or magnetic sources near this circuit! It would be best to
use plastic, brass, or stainless fasteners in the area where this circuit
is mounted. You should also locate the battery pack as far away as
possible. A stainless steel launch rod would be a good idea, too.
Final note here is on calibration.
I settled on a 200K resistance for my calibration. That value may
be different depending on where you are on the Earth and also due to differences
in individual HMC circuits, op amps, or power supplies. Make sure
to run a bench test with the components you are going to use in the final
cicuit to come up with your calibration value. It would also be possible
to tune it "loosely" and have a trim pot to finalize the settings in the
field, especially if you travel around to different launch areas.
You'll want a variable resistor of at least 100K to make this worth doing.
Here are some of the values I found with the positive angles measured up
from the horizontal and negative angles measured down from horizontal followed
by the compass heading:
Infinite resistance triggered
at 60°N and 25°S
10 M ohms 55°N and 20°S
1 M ohm 50°N and 5°S
470 K ohm 45°N and 0°S
220 K ohm 20°N and -20°S
200 K ohm 10°N and -30°S
147 K ohm 5°N and -45°S
100 K ohm -5°N and -60°S
<80 K ohms resulted in
no activation
The unit triggers very close
to horizontal in the East or West directions. For a field adjustable
calibration, I would probably go with a 147K resistor in series with a
100K trim pot. Also remember that these values were measured at N60.5°
W151° and your location will be different.
OK, here's the schematic.
It's hand drawn, but readable. I'm working on a neater computer drawing
of this circuit as well as a whole new page that has more information on
the theory of operation and lots of pictures of the parts I built and the
completed circuit. I've come up with several variations on this circuit
that I'll discuss and post schematics. I'll also put up some pictures
of the circuit board etchings I came up with. In the mean time, this
should give you something to work with.
Written by Scott Aleckson
Last updated on 6-14-00
Contact: aleckson@alaska.net
Aleckson's
Place Homepage