Your new topic does not fit any of the above??? Check first. Then post here. Thanks.

Moderator: igrr

User avatar
By chamathkv
#79176
schufti wrote:it is a design flaw. w/o connecting gpio16-rst the esp wakes but will not start user code.
with gpio16-rst it behaves like regular reset and starts up like on reset (not power on).
startup reason can be checked in code.


That should be how it works, but even though I connected GPIO16 to the RST pin, it still doesn't wake up from deep sleep. Or do anything after going to sleep for that matter. And as shown in my code, I am not using GPIO 16 for anything else either.
User avatar
By Pablo2048
#79177 It is not a design flaw IMHO - deepsleep timer is connected to GPIO16 so when you interconnect gpio16 with reset the CPU is "woken up". Many CPU has such kind of sleep state from which only way to wake up is reset...
User avatar
By Pablo2048
#79178
chamathkv wrote:That should be how it works, but even though I connected GPIO16 to the RST pin, it still doesn't wake up from deep sleep. Or do anything after going to sleep for that matter. And as shown in my code, I am not using GPIO 16 for anything else either.

Can you take a look into GPIO16 and reset pin with an oscilloscope?
User avatar
By schufti
#79181 it is definitely a design flaw as w/o this connection it does wake too, but doesn't start user code.
the connection to rst seems just to be a work around and a pretty q&d one as the 'direct connection' mentioned in the datasheet is the worst method thinkable.

what else would you call such behaviour?

some NodeMcu have too big capacity on rst, so the short low on gpio16 doesn't trigger reset ...
NodeMcu's are a collection of minor designflaws on their own ...