Raspberry Pi Pico Impressions

@whitneydesignlabs, you so right about how surprisingly small it is.  Very flexible form-factor as well, with the multiple connection options for the pins on around the perimeter.  You can surface mount the thing on another circuit board very easily without adding any height.  The thinness is striking.  The micro-usb and bootsel button are the only thing making this little board 3D, practically.  I've included a pic of the pico in my hand so you can get a sense for the scale.

Does the Pico run an RTOS of some kind? I may have missed it in all the material, but it doesn't seem to be mentioned.

Posted by: @larrytubbs

The micro-usb and bootsel button are the only thing making this little board 3D, practically.   

The other thing I noticed, and I think I am correct, is the USB connector appears to have metal tabs soldered through the board, increasing durability. I have not wrenched on it to find out, however... 🙂 One complaint I read about Arduino Pro Micros is that the USB connector is fragile, and has given people fits trying to re-solder it back on after accidentally ripping it off the PCB. On the left is the Pico, on the right is Pro Micro.

 Scott

Posted by: @yurkshirelad

Does the Pico run an RTOS of some kind? I may have missed it in all the material, but it doesn't seem to be mentioned.

I had to google up the acronym RTOS to learn what it was. My results seem to indicate the answer to your question is, yes. It seems pretty new, and scantily documented at the moment.

Github, Raspberry Pi Pico & FreeRTOS

The referring article

Scott

I can state unequivocally that the Pico can take a licking and keep on ticking. I was so excited to get mine that I made the classic noob mistake of soldering on the header pins out of alignment making it useless to use on a breadboard. So I had to de-solder all 40 pins, clean up the connections and re-solder new headers.  After all that mess, the Pico still works perfectly. Thanks to Paul at learn electronics YouTube for the push in trying out the SSD1306 OLED. I had a color OLED, but it works the same as discussed in the video.

Cheers!

--->Sean

I wish someone would bring out a version with pins soldered on - I don't have a soldering iron. 😐 

Posted by: @yurkshirelad

I wish someone would bring out a version with pins soldered on - I don't have a soldering iron. 😐 

Brand new from Pimoroni.

Cheers!

--->Sean

Posted by: @yurkshirelad

I wish someone would bring out a version with pins soldered on - I don't have a soldering iron. 😐 

I found that when I started getting back into MCUs I needed some tools and a soldering iron was key found a good one on Amazon for <$11. 

soldering iron

I picked up a pair of Picos on Saturday.  Impressive little MCUs!  I like the way you can re-assign the GPIO pins to perform different functions, eg PWM, I2C, SPI, SIO, UART, PIO.  So far I’ve used PWM to control the brightness of a couple of LEDs on a single slice.  I’m going to attempt to port some robotics sketches that I’ve written.  So I need DC motor control, LED control, analog sensors, piezo buzzer, IR emitter/receiver pairs. 

I was wondering if anyone has tried to max out the use of PWM slices in a single app?  With 8 slices I was thinking of trying: 1 slice for LEDs running at 1kHz, 1 slice for a dual DC motor controller running at 20kHz, 1 slice for a piezo buzzer running at 1kHz, and 2 slices for IR emitters running at 38kHz. Is this even possible without Bourbon?

Tom

@thrandell I have no experience with any of that, but IIRC re-assignable pins are also common to the arduino family so maybe also research there for similar issues and motivation. I haven't even unpacked some of my Picos yet, todo list too long!!!

Second Impression.

I got a chance to try out some more PWM settings on the Pico and I’m impressed by how much the SDK hides all the minutia involved in setting up frequency and duty-cycle.  I have to admit that I’ve been staring at PWM set ups on ATmega boards for some time now, but setting up PWM for the first time on the Pico was pretty painless.  I added an example to this post if anyone is interested.

One thing I noticed about the Pico documentation is that when it mentions “level” they are talking about duty-cycle and when they say “wrap” it’s the same as top which determines the frequency.  The RP2040 Datasheet has a nice short explanation of the PWM programming model.  So anyway, in this example I wanted to control a DRV8835 dual H-Bridge DC motor driver with two plastic gear motors attached.  The driver board uses two GPIO pins to determine the direction of the motors and two PWM attached GPIO pins to determine their speed.  I used the following formula to determine the wrap for a 20kHz phase correct carrier frequency.

wrap = f(clock)/2 * f(desired) - 1;  so 125,000,000/2*20,000 - 1 = 125,000,000/40,000 - 1 = 3,124

where f(clock) as it is used here is after applying the divider.  Clear as mud?

I did verify my wrap value using the formula in the RP2040 Datasheet and they matched. 🙂

Here is my set up code.

int main()
{
  gpio_init(12);                  // enable i/o 
  gpio_set_dir(12, GPIO_OUT);     // set gpio direction
  gpio_put(12, 0);                // drive gpio low
  gpio_init(13);                  // enable i/o 
  gpio_set_dir(13, GPIO_OUT);     // set gpio direction
  gpio_put(13, 0);                // drive gpio low
  gpio_set_function(14, GPIO_FUNC_PWM); // slice 7A
  gpio_set_function(15, GPIO_FUNC_PWM); // slice 7B
  uint slice_num = pwm_gpio_to_slice_num(14); 

  uint chanM1 = pwm_gpio_to_channel(14); // the PWM channel that
  uint chanM2 = pwm_gpio_to_channel(15); // controls the gpio

  uint32_t clock = 125000000;
  uint32_t f = 20000;
  uint32_t wrap = clock / (2 * f) - 1;
  pwm_set_clkdiv_int_frac(slice_num, 1,0);
  pwm_set_wrap(slice_num, wrap);
  pwm_set_phase_correct(slice_num, true);

  pwm_set_chan_level(slice_num, chanM1, 0); 
  pwm_set_chan_level(slice_num, chanM2, 0); 
  pwm_set_enabled(slice_num, true); 

The motors will respond to speed values in the -100 to 100 range.  Minus is reverse and the value represents the percentage of 100 that the level (duty cycle) should be set too.  Here is the incantation for that command.  I strip the negative sign off first and use it to determine the setting for the direction pin. "speed" as used here is an integer type variable.

 pwm_set_chan_level(slice_num, chanM1, pwm_hw->slice[slice_num].top * speed / 100);

This is all getting a bit dense, so I’ll stop my blabbing here.  

Tom