Proportional controller for microRC's (major mod)

[b007a]

Here is a project for someone out there with more time than myself ... build a controller that will give the microRC's proportional control. Here's my idea ... (I just give these things away ... I'm not sure if anyone would pay for them though):

1) remove the switches from the Tx PCB for left and right,

2) using two 555 timers build a PWM (pulse width modulated) waveform circuit for each,

3) here's the part I'm not sure of ... the steering on other larger RC's. If it is a pot, use the right resistance to change the pulse width ... more deflection in the directions of the turn will increase the pulse width. The frequency of the square wave will stay constant, but the amount of time the pulse stays high will increase with more steering deflection.

4) the output of the 555 (use the open collector transistor provided to gound (previous) switch input to the TX2 IC.

This scheme is base on the fourier analysis of a square wave. There is a DC component, a fundamental freq componenet, then all the add multiple harmonics of the fundamental. For those un-familar with Fourier Analysis: Fourier (some old dude a long time ago) said that any waveform can be broken down into individual frequencies of differing amplitude and phase. When they are all summed up the result is the original waveform. The DC component is proportional to the duty cycle of the square wave. So, if a PWM signal is low-pass filtered before the fundamental freq, then the result will be a DC value that is proportional to the duty cycle of the PWM square wave ... kind of a poor man's D-to-A converter. The corner frequency of the filter must be sufficiently low compared to the fundamental or the DC value will have a ripple that has the frequency of the fundamental. In our case, the Tx and Rx pair should act as the low-pass filter ... the rub is how to control the PWM signal. I was going to by a el-cheapo RC with a gun type controller, gut it and see how the steering wheel works. If the steering is jerky then the fundamental frequency must be increased. Normally I would control the low-pass filter to reduce the ripple below annoyance, however, we cannot control it in this case (it's should be related to the rate that data is sent to the car).

As for the frequency ... looks as though (re: datasheet for Rx2/Tx2 control IC pair) the repeat rate is approx 14 Hz (based on needing 4 W2's, and 64 W1's). So, make the period 100 times less or 0.14Hz. Why 100 times ... think logrhytmically ... 100 times would give a ripple that is 40dB below the DC value. Some trial and error would be benefical here, it is unknown at this point the order of the filtering action of the system.

For now I am only worried about the electrical side of this project. The packaging will come later ... this may be another challenage, because the capacitors will be huge (or a lot of little ones in parallel.)

If no-one bites ... I will design the PWM's if someone else will build the circuit, but you must build two (one for me) as my design time is not cheap.

Regards.

[hogjowlz]

there is a motor speed controller at allelectronics for 3 bucks.

my friend has a pwm box that we were able to test it with and it worked. i think a pwm would be better for throttle than steering tho.

[b007a]

I think the place for PWM would probably be both throttle and steering. The need would be depending on your setup.

I really did not state my needs very clearly ... just blurted out the design requirements.

I think that the point is mute, now. I found a car and controller can't remember the company that makes it though ... it was at Toys r Us. The car is smaller than the microSizers, so the parts are probably not changeable. Something else cool about the car ... it's controlled via infra-red. But the Tx/Rx IC pair are probably the same as the ones for the for the microSizers. So the controller could be modified to tx RF.

The project still may be fun.

regards.

[slugbugg]

I'm wondering if you have looked through the bit science forum.There are people working on a way to pulse the signal from the controller to simulate proportional control.

It sounds like you could put something together fairly easy.
It is a good idea but rumor has it that radio Shack is releasing a proportional control zipzap(ZIPZAP se - I'll believe when I see it).

I myself had thought about pulsing the signals almost a year ago but I really dont know enough about electronics to mess with it myself at this time, I have other area's of these little cars that I want to work with right now.