As a small project, I decided to build my own light controller to run from the third channel on my Flysky reciever. Obviously there are a ton of off the shelf options but I figured, What’s the fun in that?
This whole project would/will be easier if it was PIC micro based. It would remove the need for much of what I have going on here and would be way more flexible. I may eventually do a pic version. I figured I would try an analog version first since I have a source of free parts. I stole most of these from scrap PC boards but they can be easily sourced from Digikey or Mouser.
Right now I can only confirm this will work with a Flysky receiver as I have no way to test others but, if my reading is correct then most receivers output pulses at a frequency of around 50Hz. The pulse width is the only true standard. Given the simulations I have ran, this should operate without any modification between 45 – 65Hz. Wider with some small component changes.
Here is the signal from the receiver when the Transmitter Switch is set “low” The output is a 1ms pulse at a 55hz rate. Pressing the 3rd channel button changes the width to 2ms. For those unfamiliar to how a servo interprets these signals, if you were to pull the trigger full reverse, this would equal 1ms pulse width. Full forward would be a 2ms width (unless of course you an own ESC that operates with the channel reversed). Center equals a 1.5mS pulse.
Turning the output pulses from something resembling a digital control to a switched analog output is was quite simple. I am using some parts I was able to pull from some used PCB assemblies I was ready to toss from work. Thus free. The schematic below represents how I designed and tested the circuit before actual prototyping. Using freely available Microsoft Windows software from Linear Technology, I simulated the design which allowed for me to tweak the component values before I breadboarded the design. For anyone into electronics or a hobbyist, you can grab a copy from the Linear Tech website,
If anyone is interested in simulating this with LTspice, I have a downloadable copy of my schematic. It’s slightly different from the schematic picture below because I had to simulate the output pulses from the receiver.
The idea is to turn this “Digitalish” pulse train coming out of the Flysky receiver, shown in green……into an “analog” voltage that could then be use as a switch. In my case for LED RC lights .
Shown to the right, the first two pulses are 1mS and yield approximately a 2volt peak. As the pulse train widens from 1 to 2mS, effectively switching from full reverse to full throttle or the action of the third channel switch in the Flysky GT-3B, the peak voltage of the saw tooth shown in red further increases. The change in amplitude is because the pulse width has doubled and allows more time to charge C1. The purpose of C2 is to filter out the ripple caused by the charge, discharge of C1, turning it into more of a recognizable DC analog voltage. The only problem with this is the the maximum voltage we can achieve is about 0.3volts full reverse and about 0.6volts full throttle.
The only IC used in the circuit at U1, is an On-Semi 33072 Duel Operational Amplifier. Because 0.3v and 0.6v is inadequate to use for comparison, the voltage is amplified through 1/2 of this duel OP-AMP. The input in green is fed into pin 3, The OP-AMPS Non-Inverting input. The resistor network at R5 and R6 set the amplifier up for a gain of 6. The blue line represents the output of the amplifier after further being filtered by C3 and a better signal to feed into the next stage, the OP-AMP comparator.
The final diagram is showing three signals. Blue represents the output of the amplifier after being filtered and fed into the comparator’s non-inverting input at pin 5. The green signal is the output of the voltage divider, R8 and R9, both 1Kohm. This signal is 1/2 of V+ and fed into the inverting input of the comparator to set a threshold value. As the input to the comparator at pin 5 passes through the green threshold value, the output of the comparator turns on. this is the red signal. And conversely as the blue analog voltage drops below the green voltage, it turns off. This “On/Off” signal is then fed into the gate of a FET that is able to handle high current and is the actual switch. the string of LEDs are placed in parallel with the FET. Each string has it’s own current limiting resistor sized to drive each string to roughly 20mA. The total power dissapated by this entire circuit, including LEDs is only 250mA. Most of which is the LEDs them selves.
The breadboard is a bit messy since I was a bit excited to test this. This is the first attempt at testing the circuit. Because of the simulation software, the first test was successful and other than changing to different capacitors. all values remained the same. The Flysky receiver and a 6v hump pack to drive the circuit was used in the first test.
After successfully bread-boarding the initial circuit and obtaining some smaller sized caps, the next step was to construct the circuit using perf-board. This proto ended being the circuit I ultimately installed into my Axial EXO. Before doing so I did one final test with the LEDs and resistors I was going to use to wire the RC.