MODEL RADIO CONTROL ELECTRONICS 
Norcim page 4 covers some R/C related circuits and
stuff together with some non-related stuff plus some history of radio control
beginnings and aerial kite photography……..We hope you enjoy………The Norcim
Team
ECONOMY ELECTRIC FLIGHT REVISITED more info on electric flight on a
shoestring.
A SIMPLE SERVO TESTER CIRCUIT an economy circuit to suit all
known servos.
FAST CHARGING OF NICADS AND NH BATTERIES circuit thoughts…a possible method.
A TRANSMITTER OUTPUT TESTER CIRCUIT a field strength circuit that works
at 10 metres.
FURTHER THOUGHTS ON FAST CHARGING output circuit thoughts of above
circuit.
A POWER RELAY SWITCH FOR R/C RECEIVERS switch up to 15 amps with this one.
LOST MODEL ALARM CIRCUIT on board piezo sounder circuit for
model aircraft.
SOME ANTIQUES ROADSHOW STUFF nostalgic beginnings of
proportional radio control.
AUTOMATIC CHARGER FOR GLOW-BATTERIES. Makes them last longer.
AERIAL KITE PHOTOGRAPHY superb active panoramic views from a
digital kite camera!
CARAVAN SECURITY LIGHT? What’s this got to do with model
control electronics???
FORWARD TO NORCIM 5 commercial receiver testing by
Radio Guru Dave McQue UK.RCC
FORWARD TO NORCIM 6 historical info on PPM radio
control inc original Mathers/Spreng system
ECONOMY ELECTRIC FLIGHT was
outlined in Radio1 and Radio2 pages.
Well
the picture shows one of the test models, a 
‘Graupner
Biene’, using the ‘Sun’ motor, on a typical climb to thermal height, and it
will do this twice on seven GPâ AA
nickel hydride cells from CPC electronics.com at just £1.00
per cell. With the motor costing just £2.30 from JPR electronics,
this shows just how inexpensive good electric flight can be. The AA cells give
around ten minutes of full power in the air using a Perkins Distribution
(most model shops) 8x5.5 electric flight prop. Increased climb rate can be had,
by using a Graupner 9x5 (most model shops) electric flight prop, but this
reduces ‘full power’ in the air to around 8 minutes.
The
following circuit shows a real simple speed controller used to power the Sun
motor. Many of these controllers have now been used with test models using Micron
receivers and all have 
performed
well. The few components of the controller assists home construction and can
produce a controller/motor component
package for less than ten pounds! The controller is based around a CD4001N quad
2-input NOR gate IC which can cost as little as 30p from your electronics
store! The real expensive bit is the MOSFET output transistor which could lay
you back around £2.50. The single MOSFET drives the SUN motor (including
standard 380 and 540 motors) using seven cells (8.4volts) with no problem. You
can even cut the metal tag off the transistor to make the controller more
compact, without overheating of the MOSFET.
The
input circuitry produces a saw-tooth waveform with each servo pulse. This can
be adjusted via trimmer R5 to trigger IC1A gate to produce a 50/50 mark space
output thus driving the motor at half speed. Changes in the pulse pulse width
cause the saw-tooth waveform level to change giving variable speed from fully
ON through variable to fully OFF condition. The circuit without the optional 5v
regulator is intended for use with a separate receiver battery. If the LM2940
is fitted then the receiver, servos and controller are supplied by the flight
battery.
The
simple circuit is more suited to receivers with full rail voltage output pulses
(usually Cmos decoder chips) and has been tested using Micron, Fleet and RCM+E receivers.
Using the three remaning gates of the
CD4001BCN in parallel gives adiquate drive for up to five
MOSFETs in parallel for ‘hot’ motors. (at the relatively low switching speeds!)
D2 and C5 provide good suppression of
motor interference for the Sun motor, which has no internal brush suppression
capacitors. For ‘Hot’ genuine electric flight motors just follow the
suppression instructions supplied with the motor.
IT’S SERVO TESTER TIME!
The following circuit offers manual
control of almost any make of R/C servo, using just a receiver battery (4.8 nicad) for power. The components should
cost you no more than a £1.50 or just over 2 Euros! and it should work with all
makes of servo. The circuit is a text-book ‘two transistor multivibrator’
providing an 18 to 20ms cycle time which suits R/C servos. The circuit is
however extremely lob-sided! In that one side produces around 17ms delay before
triggering the other side, which only stays on for 1.5ms! It’s this side that
the servo output is taken from. The 4k7 pot with knob, produces a pulse
variation from 1ms to 2ms at the output which can be used to test most known
servos. C3 provides sufficient damping of servo noise to let the circuit work correctly.
It’s worth mentioning that the hfe of the transistor used must be high, better
than 250 @ 1ma. So if substituting, then pick a high gain device.
THOUGHTS ARE NOW WITH FAST CHARGING
NICAD AND NICKEL HYDRIDE BATTERIES
There are several methods of fast
charging model car and ‘electric flight‘ batteries but the most popular system,
is called the ‘Voltage Peak Detect Method’. When NC or NH batteries are
charged, their voltage slowly rises to a peak (when fully charged) but this is
quickly followed by a small but significant dip in the battery voltage. ‘Peak
detect’ fast chargers, rely on their ability to detect this small voltage drop
when the nicad is full and terminate the charge before overcharge damage is
done. A simple ‘peak detect’ method for six or seven cell sub-C batteries is
shown. At switch-on, (connection to the 12v supply), the forward volts of D1
holds the inputs of the OP AMP apart, preventing any switching of the output.
Initially, the voltage across D1 (which is a Schottky RF diode) can be as high
as 0.3v but this quickly reduces within minutes to less than 0.05 volts because
of the extremely low current flow through the diode. C1 then, follows the
increasing nicad voltage (which can be as close as 0.02 volts when near to full
charge). At full charge, the nicad voltage begins to fall…..D1 soon becomes
reversed, (also IC1A inputs!) causing IC1A output to ‘flip’. This terminates
the charge current. The result is a fast, half to one hour, automatic cut-off
charge! There is no need to discharge the nicad first. Full charge is detected
even with a partly charged nicad. It is important that C1 is one of the
recent Aluminium Electrolytic low leakage type capacitors. (Rubicon TWL series
47UF16V-16TWL47MO811 Farnell order code 499-067). It’s worth mentioning that
the 1N6263 has a lower reverse leakage current than general purpose schottky
diodes. This helps at the peak detect point.
NEXT THOUGHTS ARE WITH CHECKING
This next circuit lends itself
not only for home checking but also club and quick model shop checks. The
circuit checks for correct power output of any 35 or 40 MHz radio
control transmitter is shown. These things are called ‘field strength meters’
and are a standard piece of electronic equipment in the service workshop to
check the output power of R/C transmitters. ‘Field Strength Meters’ (as they
are called) are usually based around a reasonable size sensitive 50uA moving
coil panel meter. These are now 
listed (Farnell) between £20 and £30
each (before circuitry!). This circuit is based around the National Semiconductor
LM661CN Cmos quad op-amp IC. The circuitry components should cost no more than
£4.00! and it has greater sensitivity than the standard meter type. Transmitter
output strength is shown by four Superbright red light emitting diodes. A
correctly functioning R/C transmitter, will illuminate three to four LEDs at a
distance of 10 metres away. Adjusting the length of the short telescopic aerial
will allow all LEDs to operate at a shorter distance for indoor checking. With
occasional use, a four AA alkaline battery lasts over a year (even occasionally
leaving the thing switched on)
The OA47 diode seems to work best but
more difficult to get. L1 needs to be initially adjusted to illuminate the
maximum number of LEDs at a range of 10 metres or so. Once set that’s it. The
Toko coil used is no longer manufactured but many are still in the pipeline and
there are alternatives. Remember, if you set L1 using a 35MHz Tx then the unit
will only check other 35MHz transmitters. If 40MHz Txs are to be checked, set
L1 using a 40MHz Tx. L1/C1 form a tuned circuit at 35MHz. A 35MHz Tx will
excite this coil and cause a resonance of L1. D1 detects this and a little
current flows at 35 million times a second! into C2. This increases the voltage
across C2 (slightly) in proportion to the power of the transmitter signal. The
LMC660CN is a Cmos op-amp and has little effect on the input circuit. The
op-amps are arranged as voltage comparators using the potential divider R1-R5.
The resistor values are selected to give a 3dB step between op-amps flipping
on. (each one showing twice the transmitter power output) So with a weak
signal, IC1D output will illuminate LED4. As the received signal gets stronger,
the remaining LEDs will illuminate in turn, until all four are illuminated.
FURTHER THOUGHTS NOW RE: THE FAST NICAD
CHARGER ABOVE
Fig 1 shows a possible output stage for charging
a 7v2. nicad. Note that this is not a ‘constant current’ circuit but should
work OK. Note that the 1R resistor will initially let through a charge current
of around 3 Amps which will slowly fall to 1 Amp or less at peak detection. The
MOSFET in this version would not need to be fastened to a heat sink as it is in
a fully ‘on’ condition when charging and will need to dissipate little heat.
The 1R or 2R resistor however will get very hot and would benefit from a heat
sink. The 1R resistor value should be used for sub-C nicads. The 2R2 resistor
is a better value with less current flow for 7v2 AA nicads. Fig 2 shows another
possible output circuit in which the charge current through the MOSFET (Q2) can
be set by adjustment of the 22K trimpot. A charge rate of 2 Amps is suggested
for 7v2 nicad batteries. Some care will be required with the setting of the 22K
pot and best to start with the wiper at the negative end of the pot. The charge rate should be reasonably constant
with this circuit, owing to the bias characteristics of MOSFETs. Note that this
circuit uses the MOSFET (Q2) to restrict the charge current flow and therefore
Q2 will get hot! AND will need to be clamped to a metal heat sink. IT MUST BE
POINTED OUT THAT ALTHOUGH THE ‘PEAK VOLTAGE DETECTION CIRCUIT’ HAS BEEN WELL
TRIED USED TO DATE, THE TWO ABOVE CIRCUITS FIG1 & FIG2 ARE ‘POSSIBLE
OUTPUT’ CIRCUITS AND ARE INCLUDED FOR EXPERIMENT.
A POWER SWITCH FOR R/C RECEIVERS
Fig 3 shows a simple circuit that can be
plugged into one of the receiver servo outputs to switch on and off an external
load of up to 15 Amps. The circuit is shown driving an electric flight motor
but for boat people the relay could switch several torch bulbs, or a high power
halogen lamp, or winch, or sound circuit. The dotted line shown from 
the relay back contact to ground
provides dynamic braking of the motor when using folding props with electric
flight. The input components form a text-book ‘diode pump’ circuit. The
trim-pot is set so that thin pulses from the transmitter pump up C1 voltage to just
get the Logic Level MOSFET, conducting. The current flow through the relay
however, is very small and well below that necessary to close the relay
contacts. But throwing the stick at the transmitter to get fat pulses,
increases the voltage across C1, allowing much more current flow through the
MOSFET and the relay closes. Both 5 volt and 6 volt relays have been used in this
circuit with windings of 55R to 90R. Notice that only the Rx pulse input and
the negative input is required from the receiver (The red positive input is
simply left unconnected). The current drain of the circuit on the receiver
battery is almost undetectable. At higher voltages of the external circuit, the
MOSFET (Q1) tends to limit the current flow through the relay to acceptable
levels. The MOSFET, RFD4N06L (or similar) is a TO 251 (I-Pak) device and may
get warm with a 12 volt external supply. This is normal. Several relays are
suitable…finder type SPCO 16A Farnell code 431-308…or smaller 12Amp
version finder type Low profile 21 series Farnell code 321-0224.
EVER LOST A MODEL IN A CORN FIELD???
Well this next 555 circuit will detect its position!
PIEZO SOUNDER DEVICES ARE SMALL AND
WEIGH JUST 5gms! These small units can be heard up to 100 metres away in quiet
countryside conditions, particularly if they are switched on and off! The
following circuit uses a 30p 555 timer IC, to drive one of these Piezo
sounders. The 555 is configured as a 50/50 astable pulse generator which sounds
the piezo thing around once a second. Transistor TR1 keeps shorting C2 with each
servo pulse, so that the 555 IC can’t get going. The piezo sounder remains
quiet during normal flight conditions. Should the model decide to land in a
corn or rapeseed field, then switching off the Tx allows TR1 to go open circuit
and C2 to begin its charge/discharge cycle enabling the 555 to switch the piezo
sounder on and off at around once per second. With careful listening, the model
can be located! Suitable piezo sounders are Farnell order code 927-181 or
927-119. The receiver input stage should suit most makes of receiver including
those with low voltage servo pulse output.
AND WHERE IT
ALL BEGAN!
Did you know that one of the first offerings
from the ‘Micron people’ was a pulse proportional system? Based around a ‘Mighty Midget’ electric
motor? Full kits were available from around 1962! The system used a
single electric motor to pulse both rudder and elevator 
control surfaces. Surprising
proportional flight could be achieved with this system, which was termed
‘Galloping Ghost’ (because of the noise of the flapping control surfaces!). One
of the flight demonstrations of the system, involved several passes through
soccer goalposts (at a medium throttle setting, as there was no throttle
control), which was quite an achievement alongside the standard ‘reed’
equipment of the day.
The receiver (shown left) used a
super-regen front end (only allowing one model to fly at a time!) was designed
by early R/C pioneer, Doug Bolton. Other R/C equipment of the day used
self-centre toggle switches on the front of the transmitter for full throw
of the control surfaces Joystick units had not been invented! So
joysticks for the ‘Micron people’ transmitter kits had to be individually made
from brass stock with piano wire wound springs! Two external pots provided the
in-flight trims. Early transmitter circuitry used a variable rate saw-tooth
generator (unijunction) followed by a varying level Schmitt trigger to generate
the mark/space and pulse/rate, pulses for the ‘mighty midget’ actuator. Later
circuitry used a curious diode gated two-transistor astable-multivivrator,
which gave a logarithmic effect to the rate channel 
(elevator). This improved the elevator
effect to perform outside loops, when using the ‘mighty midget’ actuator.
Unfortunately this effect could not be used with the later more popular
American Rand actuator.
Rapidly advancing multi-channel analogue
and digital technology saw the demise of this kind of simple pulse proportional
control but there is little doubt that these early systems began the trend into
true proportional control of model aircraft.
The photos show one of the early Micron
People ‘Galloping Ghost’ proportional systems, which is now forty years old!
When last tested with replaced nicad batteries…….It still worked!….and without
any doubt, it would still fly a model through those soccer goalposts!
AN AUTOMATIC CHARGER FOR 2 VOLT
GLOW BATTERIES
This simple circuit charges
‘part-charged’ two-volt glow batteries in an hour or two, with auto slow-down to
trickle charge. (The charge rate gradually reduces toward the end of charge, to
minimise battery ‘gassing’). The power input is taken from the 12-volt field
starter battery (or car battery). Both
input and output are polarity reverse 
‘safe’ and the green LED will only
illuminate with correct input and output connections. The Car Bulb LMP1
indicates charge, taking place and its illumination intensity shows the level
of charge. Bright is fast charge, Dull is slow charge and No illumination is
charge complete with ‘Trickle charge’. The glow battery can be left in the
trickle charge condition indefinitely (for a convenient time to unplug!) as the
trickle charge eventually drops to less than 25 milliamps. The LED US1 is used
as a voltage source for the base of T1, so care must be taken with substitutes.
The one used in the prototype was a ‘Kingbright’ superbright 5mm green! LED,
which gives a foreword voltage (Vf) of 2.3 volts (in this circuit). D1 adds
another 0.75 volts to the base of T1. The voltage across US1 can easily be
measured after connecting the input leads of the circuit. (red LEDs give a
lower Vf so don’t use). All of the other components are non-critical. NOTE….if
the glow battery is severely discharged, (to be avoided for long battery life!)
it may take a minute or two for the green LED to illuminate. The ‘Hawker Cyclonâ’ glow-battery quotes 2000 charge cycles
and 10 years life with sensible charging!
NOW… HOW DO THEY DO THAT!
Take a look at http://www.kapshop.com/full/
. It shows some of the latest digital technology taken from the hobby of
‘aerial kite photography!’ Wow, these effects are just like the best computer
game! You control the on board kite camera to look wherever you want around a
complete 360 degree panoramic view, I guess, (pi D squared and all that!) over
hundreds of square miles! You can telephoto in to near objects of interest too!
Real weird, but quite wonderful stuff, not to be missed! (takes a little time to download so be
patient, then use the controls at the page bottom)
BILL BOWDEN’S HOBBY CIRCUITS
Although not directly connected to ‘radio
control electronics’ Bill’s hobby circuits cover a wide variety of general
electronic circuits with masses of information including component spec
sheets and component suppliers. Many links to other sites are there too. Well
worth the visit. www.ourworld.compuserve.com/homepages/Bill_Bowden
CARAVAN SECURITY LIGHT?
OK! Not directly associated with model
electronics but as many modellers have a caravan, then this little circuit could
be of interest. The prototype was assembled on Veroboard and inserted in place
of the bulb in the awning light. It gives a bright looking light that can be
left switched ON for up to six months! Using just the internal caravan
battery. It gives the impression that the caravan is occupied which adds to
security. The prototype used 5mm Ulta-bright yellow LEDs from Rapid Electronics
at 15p each. They drop 1.85 volts each at 15 milliamps. The resistor is a 0.25
Watt type. There are now Ulta-bright white and blue LEDs which are being used
for front lights of bicycles! Providing the volt drop is similar then these
could be used or a resistor value change to compensate. Another idea was to
place the device in a small plastic box with wire connector, to plug into a 12
volt interior socket in the caravan. The unit could then be placed near the
front blinds or curtains to suggest there was an interior light on. Wherever
you park your caravan, this device will help with unwanted callers! For
Months!!
Please note more to come