Decoding PWM

[Joosh]

Hi

Looking for some help from anyone that has experience with this, any useful info or even a ready made alternative. I am in the process of building a robot and require a higher than normal current DC motor driver. As these are quite expensive to buy I want to have a go at building one.

Basically I have a spektrum receiver that runs on 5-6V and outputs a PWM at 22ms, 1ms to 2ms, 1.5ms in 'neutral'.

What I would like to do is turn this in to a useful signal to drive h bridge transistors. Ideally by using cheap components (no arduino etc) and for it to be reliable.

I am wondering if there is a way to monitor the signal then output it in a PWM voltage that could operate PNP or NPN transistors depending on the input.

For example:
Receiver in 'forward': PWM 1.51-2ms ~3.5V
-> 0 to 100% duty PWM at 5v.

Receiver in neutral: PWM 1.5ms ~3.4V
-> 0 % duty.

Receiver in 'reverse': PWM 1-1.49ms ~3.3V
-> 0 to 100% duty PWM at -5v.

I am thinking some sort of timing op amp circuit?

Any thoughts from the more electronic savy or should I just start saving up my money for a proper driver? Lol.

Josh

 

[Bluejets]

Different signal.
Not what you think. RC signal is a span over 20ms.

 

[Alec_t]

If you are trying to control the speed of a powerful motor by PWM then your PWM frequency will need to be much greater than the 50Hz (20mS period) RC signal. Also, you will be increasing the duty cycle range from the present RC 7.5%-12.5% to a 0%-100% range. Those are not trivial tasks. Doing them without the use of a microcontroller (Arduino or other) would not be simple or cheap and would take a lot more than a "timing opamp circuit". How are your programming skills?

 

[BobK]

Alec,

He is not trying to control the speed, just the direction via a remote that is intended to drive a servo. Simply running forward if the pulse width is > 1.5 and backward if < 1.5 and stopped if nearly 1.5.

What you need is an integrator and a window comparator. On the falling edge of the signal the outputs of the comparator would be latched and would then control the motor direction until the next pulse is seen.

Bob

 

[Alec_t]

He is not trying to control the speed, just the direction

Maybe. So why the reference to 0-100% duty cycle PWM?

Receiver in 'forward': PWM 1.51-2ms ~3.5V
-> 0 to 100% duty PWM at 5v.

 

[BobK]

On rereading, I think you are right.

Much more complicated now. Only a microcontroller would make it easy.

Bob

 

[Joosh]

Hi

Thanks for your replies and input, appreciated. What I listed was best case. One speed may effect the handling too much for it to be used. Could possibly get away with two speed.

After having a look in to window comparators they seem feasible with a bit of logic, however the PWM frequency issue...

The motors are 24V, the common type you would see on wheelchairs. If you are sure the 22ms PWM is not enough to run the motors smoothly, then I will have to invest in a shiny new microcontroller and find some good tutorials!

Josh

 

[BobK]

Easier would be to use a remote that was not designed for something completely different.

Bob

 

[Alec_t]

22mS is not enough. The motor would buzz horribly at ~50Hz. From what I see online the typical PWM frequency for a motor of that sort is >1kHz.