Coaxial Motor Control With Gyro on a Helicopter

But since the gyro module outputs a PWM signal and the main throttle control outputs PWM signals, I can't mix them without decoding and converting to analog at least one of the two signals.

Can you break this down a little more? It is rather difficult to follow the description without a little bit more information. Perhaps a circuit diagram? Lastly, I take it that the gyro module has dual outputs for left, right, and level control?

A simple analog circuit, no. This should be trivial in the programming of either the chip that produces the PWM signal on the copter or handheld controller.

Thanks for getting back on this.

I have found that all coaxial hellis have proprietary radio/controllers. Why? Because the left right controls from the transmitter have to be mixed with the level/throttle in order to turn.

I don't have a schematic because the chip on board is rubbed off on all of the receivers.

I am trying to build (re-build) using a 9x transmitter and receiver and off the shelf RC motor controllers and gyros. The problem arises in the mixing of the left-right and throttle signals. Typically, in a regular heli, the throttle and left right (yaw) controls are separate. The gyro gets it's signals for control from the receiver, and then passes that on to the tail rotor swash-plate servo to alter the pitch of the tail rotor blades and the throttle signal goes directly from the receiver to the main rotor motor controller. The controller stick is in the neutral position between left and right rotation.

On a coaxial model, there is a need for on-board mixing and control. The gyro in the stock receiver is mixed with the throttle signal and a left command will lower the duty of the counter-clockwise rotor and increase the duty of the clockwise rotor causing the difference in torque to rotate the heli to the left. The gyro then locks the heli in that position when the command is back to neutral. In this respect, both gyros perform the same function, that is to control torque to maintain a heading unless given a turn command.

I want to find an off the shelf RC solution to controlling a coaxial motor system, but I can't find one that works on-board the heli.

The motor controllers in both systems are MOSFETs that have their gates driven by the PWM signal and their Power directly from the battery. (The receiver gets it's power from the connection to the main throttle control cable via the motor controller, but can be powered independently as in a glider).

Nobody I know of is currently making an off the shelf gyro/main motor controller for standard RC radios. I can do mods on the existing radio but I can't get the existing radio/gyro to work with an off the shelf RC controller and receiver.

Standard digital radio control systems for RC model systems use potentiometers on the stick connected to a monostable to create pulses of variable duration (pulse width modulation). The monostables in the transmitter are triggered by a counter which counts through the number of control channels the system has, and the output of all the controls are fed into a multiplexer which creates a series pulses representing the control signals for each channel in sequence, and a short reset pulse on completion of the sequence. This pulse width modulated stream is sent to the transmitter, and on receiving it on the receiver, it is sent to a counter which is reset by the short pulse, and a demultiplexer which sends the pulse appropriate to each channel to the appropriate servo. The servo has a monostable connected to a potentiometer triggered by the the start of the pulse from the reciever, and its output is gated with the pulse from the receiver so that the difference in the pulse duration dumps a charge into a circuit which drives the servo motor (the direction depending on which pulse stops first).

The problem with this setup for combining or subtracting the values corresponding to the pulse width is that the pulses in any two channels will occur at different times, so the only easy way to do it on the receiver side is by using analogue circuitry to store charge - I am not sure whether there is any standard product to do this. The easiest way to do this if you need to do a custom solution is therefore on the transmitter side by using one channel to control each of the the two coaxial contra-rotating rotor motors on the reciever side, and mechanically connecting the two control sticks on the transmitter (eg. by rotating the 2 axis control stick 45 degrees so that moving it up and down will collectively increase/decrease motor speed to make it go up/down, and moving it left or right will differentially change motor speed to effect a turn. Alternatively it may be possible to electrically wire the two potentiometers to achieve a similar effect. This assumes the gyro output is intended for a coaxial helicopter with two lift motor controls rather than a lift motor output and a tail rotor output.

I can't say I fully understand the question but from from what I think I understand I don't know of any simple analog circuit. I think a small microcontroller like a pic chip would be able to handle this however.

They run on 5 volts which you should have available and many have internal oscillators so you don't need any external components. The circuit would be a single chip something in the 8 to 16 pin range that would take in the pwm signals do the required math and output pwm signals.

Unless you have some background in microcontroller the complex part would be programming it. The chips in that range are often under $1 and chip programmers can be found for about $10.

If you don't mind a slightly bigger and complex circuit something like an arduino micro for about $20 I am sure could handle this and you wouldn't need a programmer as it is built in. You would still of course need to program it with the corresponding math for the task you want.

Arduino. That seems to be where I'm going to have to go.I have a couple of UNOs and I have lots of experience programming, I was hoping I could avoid doing it with a custom setup.

I might be able to do this with two gyros, one for each motor. But if I need to use an Arduino, then so be it. It will add some weight but not much, and it will be a little more complicated, but not much.

I wish you joy of programming an Arduino to do that job, but before you start, have a look at what is available already. Double Horse do a 2.4GHz receiver here

http://www.aliexpress.com/item/Double-Horse-Shuangma-spare-parts-DH-9115-22-2-4G-receiver-receiving-board-PCB/576032930.html

which might mate up to your existing gyro. I haven't investigated their transmitters for range, but binding another brand should not be too difficult.

That could be it. But I don't know if it is compatible with standard 2.4Ghz radios like the 9x. It is nearly indentical the 27Mhz receivers I have now, minus the long antenna. The proprietary control chip is turned 90deg too, and the crystal is missing.

The problem with the DH-radio is compatibility and flight time. The MOSFETs and H-bridge transistors are not heat-sinked and if you fly more weight and for longer, they overheat. I have removed mine and replaced them with heatsinked versions that I bolt directly to the aluminium frame of the heli.

The arrangement of fixed-channel transmitter and receiver pairs with matched crystals, which are the technology of the 27, 35 and 40 Mhz RC bands, does not apply to the current 2.4GHz band transmitters and receivers, which use frequency hopping to avoid interference. That is why binding with a specific receiver is necessary. I think you will find a tiny crystal on the board, but I have not worked out how the multiple frequencies are generated.

Wait a minute. They use frequency hopping. That's getting into the realm of spread spectrum techniques the military uses.

Indeed. Perhaps the Far East RC manufacturers got there first. It makes little difference when you have the only model for miles around, but when a band of enthusiasts gather together to fly it's a different matter.

With regard to the use of the Arduino - a bit of an overkill, but it does offer some interesting possibilities.

If you are going to use the Arduino, then I would suggest to rig up a timer or use the Arduino itself to measure the duration of the control pulses and gyro pulses, perform arithmetic on the input, and have the Arduino or a counter generate an output pulse of the appropriate length for each lift motor. This will need to be a real time computation though, so it has to be interrupt driven.

An even neater solution would be if you could get hold of the pulse stream coming out of the receiver before it goes to the demultiplexer. You could use the Arduino to get all the command inputs from the transmitter - allowing you to do far more sophisticated computations with the complete inputs to the system.

I researched the Arduino idea and found that it is limited to using only the combined signal from the receiver which I would have to tap into by modding the receiver. Apparently there is only one PWM input on the Arduino, and so all the functionality of the receiver is lost and replaced by the Arduino and a lot of code to de-mix all of the control signals.

I don't want to go there because it is a whole lot of work. I'm still looking at a simple control of the negative leads back to the battery from the motors as a possible solution. I'm thinking of using a couple of PNP MOSFETS which are normally conducting and using the gyros control signal to throttle down the motors on the backside.

Here's my understanding of how your controller works. You have at least 2 potentiometers that each generate a voltage that goes to another circuit which controls a pulse width output for that channel. (Multiples channels, I believe, are usually time multiplexed ,i.e., interleaved on the output.)

You have to take the two input voltages and create two new output voltages to feed to the pulse width modulation circuit so that:

Output A = Input A + Input B

Output B = Input A - Input B

where Input A controls the collective (both rotors) and Input B controls the differential (turning).

You should be able to build a circuit using a dual op amp to add and subtract these signals. Here is a link:

http://www.fatih.edu.tr/~aliadam/EEE201A/EEE201opampsCircuit.pdf

Wow. That link Rixter is food for thought. I have to do some homework now and see what I can come up with.

Update:

I may have found a control unit! It is a stability control for Helis and it has several rotor configurations, including a new one for twin rotors on a V22 osprey or Avatar gunship. It was less than 20 bucks so I ordered it and I am awaiting it's arrival. I will post the results and even a wiring diagram for those who are interested.

Would this be feasible?

LOL! The truth is that I need something like that only electronic. The similarity is real. It's like having a motor turn each wheel on a rear-end differential in opposite directions and then turning the normally stationary pilot shaft a bit in order to vary the realative speed of the wheels.

That is what I need the gyro PWM signal to do to the main rotor motors.

They used that arrangement at a disposable diaper plant. the whole drive line was split by the differential in order to shift the pad cutter for varying gaps between the pads. The phase shift was performed by a stepper motor linked to a processor using an optic sensor. This whole thing moved very fast.

I was on Youtube last night watching old Chevy and Military training films on planetary gears, hydraulic steering assist and other things and realized how closely some of the mechanics and hydraulics are to electronic circuits and controls. Lots to think about.