PICO lightsleep and pin hold

[davefes]

One benefit of lightsleep, on the ESP32, is that one can hold the pin state during lightsleep. This capability does not appear to be possible on the RP2040 devices. Have I missed something?

[GitHubsSilverBullet]

What makes you think it doesn't?

from machine import Pin, lightsleep

l = [x for x in range(50,200+1,10)]
idx = 0

pin = Pin(25,Pin.OUT)
while True:
    pin.toggle()
    lightsleep(l[idx])
    pin.toggle()
    lightsleep(l[idx])
    idx = (idx+1) % len(l)

[davefes]

I have since found a comment RP2040 dormant mode has all the RAM still powered, thanks Google.

And thank you for the code snippet.

[davefes]

I tried the code snippet and now I can't reclaim my PICO. Holding the boot selector while plugging it in says the .uf2 file is no longer there, but the program still runs and I can't stop it.

[robert-hh]

You're welcome. Once I have been in that jail too. The Pycom MicroPython port has the feature, that if a certain Pin is high at boot time, boot.py and main.py will not be executed. On some ports, you can implement a similar behavior by using the feature, that main,py will not be executed if boot.py fails. So if you check the level of a pin of your choice at the start of boot.py, and if it has a do not run condition, you can then raise an exception. Then, main.py will not run.
Unfortunately that is not implemented at the RP2 port. Can easily be added. I'll make a PR. Until then, you could for instance rename main.py.

[davefes]

I normally put a 2-5 second sleep at the beginning to give me a chance of breaking-in. I was too fast with the copy/paste :(

[peterhinch]

I do the same, but it doesn't make sense for micro-power applications that use deepsleep because the delay consumes power each time you wake. The solution I use is similar to that of @robert-hh but it doesn't require a PR:

from machine import Pin
from sys import exit
breakout = Pin(16, Pin.IN, Pin.PULL_UP)
if not breakout():  # Pull down for REPL
    exit()

[davefes]

Good point. I have read through @glenn20 mods for reducing current, but haven't had a challenging enough application ... yet.

Thanks for the code snippet.

[robert-hh]

Of course you can put that to the start of main.py. But since this mechanism of skipping main.py if boot.py fails is implemented on a few ports, I'll make a PR to extend it to all major ports.

[davefes]

@peterhinch, tried your suggestion as follows:

from time import sleep
from machine import Pin
from sys import exit


breakout = Pin(16, Pin.IN, Pin.PULL_UP)

if not breakout():  # Pull down for REPL
    exit()

led = Pin('LED', Pin.OUT)
led.off()

while True:
    led.off()
    sleep(0.5)
    led.on()
    sleep(0.5)

It works, but I have noticed an unexpected behaviour. I will start another thread.

[GitHubsSilverBullet]

I'd suggest a tiny delay between the breakout = Pin(…) and the if not breakout().
Depending on electrical charges an immediate check might not work as expected due to the high resistance pull-up (50kΩ…80kΩ)

PS. On reset the RP2040 has all inputs pulled down to ›low‹ and any existing capacitance is therefore discharged. Toggling the input to ›pull-up‹ will allow the capacitance to charge, but it might take a couple of microseconds to reach the Schmitt-Trigger threshold and thus (if read to fast) could still be read as ›low‹.

PPS. The inverse is true for capacitance between input and Vcc. They're fully charged and need to discharge, thus amplifying the "problem".

[davefes]

Sorry, I started another thread where I discovered that a short delay was required.