8 replies to this topic

[Asymptot]

Hello,

I currently have this strange problem when I try to use the DFRobot MotorShield 1A (http://www.dfrobot.c...or_Shield_(L293)_(SKU:_DRI0001)) with my Netduino, still using the 4.1 firmware and framework :

When I use a dummy code like this :

while (true)
{

}

Both motors run (and if i don't plug any motor, the motor LEDs indicating motor activity are lit)
Nevertheless when i set the different PINs for more control, like this :

            PWM E1 = new PWM(Pins.GPIO_PIN_D5);
            OutputPort M1 = new OutputPort(Pins.GPIO_PIN_D4, true);

            PWM E2 = new PWM(Pins.GPIO_PIN_D6);
            OutputPort M2 = new OutputPort(Pins.GPIO_PIN_D7, true);
            
            E1.SetDutyCycle(50);
            E1.SetDutyCycle(50);

Then, both motors run during the boot/loading phase, and then stop.
It seems to me that for some reason, bothmotors run by default, and trying to set PWM on them turn them off...
I have no idea why this is happenning.

Does anyone has an idea of why this is happenning?

[Paul Newton]

Hi there! Most people find this! When the Netduino boots, all the IO lines are set as inputs with weak pullup resistors enabled. The internal resistors don't give enough current to light an LED, but they DO give enough to turn on logic gates like your motor driver. Its a very annoying problem, but there is method behind the apparent madness. Paul

[Asymptot]

Thank you, that explains why the motors run by default. I still don't understand why trying to set the PWMs completely turn them off... Any suggestion?

[Paul Newton]

It might be the period of the pwm signal. I am not sure what the default is. Try something less than 1000Hz and see what happens. Also, what happens with higher duty cycles? Paul

[Stefan]

What kind of power source do you use? Just so you know, I've written a library for using that shield, documentation @ http://netmftoolbox.... L298N and L293

[Asymptot]

Thank you for your answers! I use a 9V battery power source when i have the motors plugged, and onyl the USB one when I don't have the motors plugged, and just testing the shield looking what happens with the LEDs. I will test your library and also try different cycle values. Thank you for the suggestions :-)

[Chris Walker]

A few technical insights: The SAM7X MCU enables pullups on all pins by default during boot. Generally speaking, MCU pins are best left in pullup or pulldown state. This helps reduce power consumption among other things. The SecretLabs.NETMF.Hardware PWM class sets the PWM dutycycle to zero by default, as a safe initialization value. This will also turn off the pin pullups if they were activated, which could then cause the motors to stop running. The new Shield Base actually keeps the pins floating during boot, and also has control over power to the shield. The motor sceanrio was taken into account when designing the new Netduino Go and Shield Base. The design ensures a premium experience for new users and offers high design flexibility for one's projects. Chris

[Paul Newton]

I made some notes a while ago and created a Wiki page about using opto-isolators. On my robot buggy I am using two opto-isolators to prevent my motors running during reset.

Its fairly simple, cheap and works!

Regards - Paul

[Asymptot]

Thank you very much! I think i figured out my PWM problem. Any value of the duty in SetDutyCycle(duty) between 70 and 100 work well. 70 giving a very slow running motor. Under 70 it sometimes run, and sometimes doesn't. I plan of using it in a PID-corrected line following bot. If i keep using SetdutyCycle, it'll give me 30 distinct values, which is enough. Do you think it'd be worth it using SetPulse for more granularity? (then, i'll need to figure out what frequency to use first). Thank you again, Farouk