Remote controlled servo deployment
Introduction
We have all experienced the frustration and agony when one of our rockets ,that took
several hours to build, lawndarted and broke up. Not to mention the risk of hitting
a human target with a falling rocket.
After the successful design of the high-voltage pin puller, I wanted to have remote control
instead of a timer to deploy the parachute.
There are several different methods to detect rocket apogee, but none of them can beat the human
eye/brain combination. Even with a timer on board, you are always anticipating the moment at
which the parachute should deploy theoretically. With the remote control at your fingertips,
you are the only one to decide when and if the parachute should deploy.
Ideally, in a very large open field, you can wait until after apogee to deploy the parachute.
Depending on wind conditions, you can decide yourself at what height the parachute has to open.
This way, you are able to determine for yourself how many calories you want to burn chasing and
retrieving the drifting rocket. Even when there is only one single tree in sight,we all know by
experience that trees are rocket magnets and that your rocket will end up dangling in that
tree top. When you see the rocket heading towards the tree upon descent, you may decide to NOT
deploy the parachute by forcing away your finger tip from the remote control button.
The disadvantage of this remote control project is the cost: about 20 Euros for the servo and
another 20 Euros for the remote control electronics.
The easiest way to implement remote control is to use a commercially available transmitter /
receiver combination. This may be the standard remote control from a plane or car if you are
a RC enthousiast already. However, when using the standard hobby RC equipment, keep in mind
that the range between your transmitter on the ground and the rocket may be well over 100 meters.
One big advantage is that you should normally have a direct view path to the receiver in the
rocket. Unless you are launching somewhere in the middle of the city center, which is not the
best place.
When you are not in the RC hobby yet (as in my case) and want to build a remote control deployment for your
rocket anyway, this could be the solution for you. All you need are basic electronic skills
like soldering and maybe measuring a voltage. The electronics are kept as simple as possible
with no exotic components involved. If you are still interested, get yourself a bottle of your
favourite regional beer and settle down.
For the remote control transmitter / receiver pair, I chose a 433MHz solution from a company
here in Belgium called Velleman. Click on the image to get a link to some more info on the
receiver and transmitter.
The dimensions of the transmitter and receiver are very small.
With direct sight conditions as we typically have with launching water rockets, range should be
more than sufficient ( 200 meters is enough for me ).
Their original target application is remote control for garage door openers, etc. The transmitter
transmits a digital code at low baud rate. At the receiver output, a microcontroller decodes
this digital signal and when the code is right, the gate / lock is opened.
Have no fear. In my deployment design, I am not using a microcontroller, just ordinary
off-the-shelf components.
Remote control transmitter
Transmitter schematic diagram.
One of my primary requirements was that the transmitter should be small enough to fit into
your pocket. After all: you only need a single button = eject! Another requirement: simplicity !
The transmitter is powered by a small 12V battery.
The heart of the circuit is a good old CMOS quad Schmitt-trigger NAND gate: the 4093.
The NAND gates U1..U4 are the 4 gates inside this single chip. The chip is powered directly
from the battery voltage, no extra 5V or 8V stabilizer. This allows us to use the battery until it really
drops below acceptable range. Also, with 12V the transmitter range is bigger than with 5V.
The disadvantage of this simple power supply is that the pulse width and frequency generated
by the 4093 is supply voltage dependent. In practice this is not so critical.
U1 forms a basic oscillator with a frequency of about 60Hz. The square wave at the output is
differentiated by 2 x 100nF + the 2 x 100K potmeters. The positive edges of the oscillator
result in a negative puls at the output of U2 and U3 with an adjustable pulse width. The
pulse width should be adjusted between 1 ms and 2 ms. In practice: 1 pulse width is the
position of the servo horn in its standby state, the other pulse width is the position of the
servo when activated. With the dual-pole pushbutton it is possible to select between one pulse
width or the other going to inverter U4. U4 simply reverses the polarity of the pulses and
at the output we have positive pulses : this is the way they should go to the servo.
So far, this simple circuit may be used as a simple standalone servo tester. The pulses
coming out of U4 pin 10 may be applied directly to the servo control input. However, this is
only valid if the circuit would be powered at 5V. Do not apply 12V power supply or pulses
to your precious servo motor directly! The servo control pulses are applied at the data input
of the 433MHz transmitter together with a filtered power supply.
If you have some spare contacts on the pushbutton, a nice feature is a beeper that
is activated when you push THE pushbutton. The beep will be the evidence to yourself and
(especially) to a critical audience that you did not forget to push the button while your
beautiful rocket is heading face down for your lawn (aka lawn darting).
The first picture shows the transmitter housing with extendable rod antenna (62 cm = full wavelength
at 433 MHz) , on/off switch and 2 potmeters
for the pulse width settings: one setting is the normal position (servo does not pull the deployment trigger)
and the other pulse widht is the deployment position (servo pulls on the deployment trigger).
The second picture shows the inside: pushbutton, beeper, 12V battery, 2 potmeters, CMOS chip and transmitter module.
Remote control receiver
Receiver schematic diagram.
The receiver electronics are extremely simple. The power supply for the 433Mhz receiver and the
servo motor is split up by using 2 separate 5V stabilizers. Do not try to use just a single
stabilizer. Servo motors genereate spikes and other noise on their power supply lines. In order
to make sure that the receiver is not influenced by power supply pollution, use a separate
5V stabilizer. The battery I used is a normal 9V battery block. This is quite heavy, but
lasts for a very long time. Note that a rechargeable 4.8V RC battery pack may be used here
instead of the 9V battery and the stabilizers. However, the 433MHz receiver is quite picky:
the data sheet requirements for the power supply are 4.9V...5.1V. The 9V battery is quite
heavy. The very light 12V battery that is used in the transmitter is useless here because
of the high internal impedance of the 12V battery that limits its output current too much
for servo operation.
Mechanical setup
The remote control electronics is one thing. Translating electronic signals into mechanical
motion is another. That's where the digital servo motor comes into the picture.
Digital servo motors as used in RC planes / boats / cars / robots.. are relatively complex
devices internally but they are easy to control. Basically, all you have to apply to a servo
motor to make it work is the power supply = 5 Volt and ground and a control signal that determines
the angle of rotation. This control signal is a pulse width modulated signal with a frequency
of about 50Hz. Pulse width is somewhere between 1ms and 2ms. Anything in between will position
the servo motor shaft differently. If you are eager to learn more about servo operation, check
out e.g. this link:
Servo basics
Very important: there are different sizes and shapes of servos. The ones that are used in cars,
boats,.. are usually quite large and heavy but cheap. There are also very light, but more
expensive servos that are suitable for airborne applications, like the one in the picture:
a Futaba S3107:
I started off with a large and cheap servo, like the ones that are used in model cars.
Click on the picture to see the resulting setup:
Another very useful feature of servo motors is that they try to keep the shaft in the same
position even when an external mechanical force is applied to the shaft or servo horn.
In other words: as long as there is no mechanical force trying to move the servo horn,
power supply current is very low but as soon as an external force is applied, current increases
to keep the motor shaft in the same position. In my design, the servo horn is used to pull
a trigger only. In rest, current consumption is very low. It's only during the trigger pulling
action that the servo experiences some external force. This is only for a very short time
when the remote control button is pushed and the servo moves to its activated position.
The actual deployment mechanics are based on older designs published on the net called
VDTT (Vertical Deployment by Tomy Timer). The mechanical timer is replaced here by the servo
motor. One of my personal requirements was that it had to be very easy to prepare the rocket
again for a next launch. All I have to do is push down the plunger of the vertical ejection
mechanism until it hooks into the trigger mechanism. One important issue here is to keep
all forces perpendicular with respect to each other to make sure that they do not "see"
and influence each other. I'm far from a mechanical genius, so probably somebody will come
up with a better mechanism.
This is the same setup with the smaller Futaba servo mounted in the hole for the big one. Notice the difference...
The next two pictures give a good impression of the completed deployment unit. The wooden plunger has a diameter of
about 8 cm. The inner diameter of the bottles I use is about 10 cm. Plunger throw is about 5 cm.
All the electronics and a rechargeable battery are mounted on the back side of the deployment unit.
Note that there is a two way on/off switch: in the OFF position,
the battery + pole is connected to a 3 mm jack at the side of the unit that is used to recharge the 9 V battery.
The servo horn is connected to the trigger with a small length of nylon fishing line.
This way it is possible to position the horn in the normal rest position so that it does not pull the string and there
is still some slack. This ensures that current comnsumption is minimal. Only during the actual pulling action will
the servo consume some more current. In rest, there is no mechanical force on the servo horn.
The green wire is the antenna. It will be stuck to the outside of the deployment housing with duct tape.
Length of the antenna = 15.5 cm or 1/4 wavelength at 433 MHz.
Another alternative and smaller in size is
the helical antenna.
At long last, 6 months after finishing the prototype, we finally had the chance and the time to launch it.
IT WAS A BIG SUCCESS!!!
It deployed 7 times out of 7 launches without any problem. It was a very good feeling to watch the rocket launch, wait 3 to 4
seconds and then, at apogee, press the button and see the chute being ejected instantaneously!
Click on the thumbnails to see the bigger pictures.
Maiden launches!!!
Home
