Hello guys

the M5Cardputer doesn't have that hold button until green LED lights up mechanism. To put M5Cardputer into download mode use the method described here.

If you want to enter download mode, press and hold the G0 button on StampS3 before turning it on, then release it after powering it on.

Thanks
Felix

Hello @jamstooks

@ajb2k3 is correct, the PbHub is for analog devices only and cannot be used for UART.

Unlike UART, I2C is a bus and can handle multiple units (as long as the units use different I2C addresses). So if you want to connect multiple I2C units to you M5Core2 you can use a Groove-T.
However as soon as some units you wish to connect share the same I2C address you need a different solution. You have two options:

• the units in question allow for the I2C address to be modified
• use a PaHub (like @ajb2k3 suggested)

As for connecting the UnitV2 to your M5Core2 via UART: I would suggest the Battery Bottom 2, which has been designed for M5Core2 and it gives you the UART port needed.

The COMMU module should work as well, but as it has been designed for the M5Stack (Basic, Gray, Go, Fire) there seems to be some incompatibility issues like while the COMMU module is stacked to the M5Core2 the M5Core2 cannot be flashed.
(Note: I don't have the COMMU module, so I cannot confirm that issue.)

Thanks
Felix

Hello @m5stack

thank you for letting us know about the latest UIFlow 1.7.3 firmware. I appreciate it.

It is however very unfortunate that the M5Core2 PaHub bug has not been addressed in this update. I'd appreciate if your engineers fix that bug in the next update.

Or, as stated already many times in many places, please open source the UIFlow code so we can help you fix issues like that.

Thank you for listening.

Felix

Hello @Surya

after long hours of searching and debugging I found the reason why I2C would fail. It starts to fail as soon as SSL is invoked. The SSL library uses an analog GPIO as input for its random seed. And in your code you used GPIO2 which is one of M5AtomS3 Groove GPIOs.

So to get rid of the I2C error try changing the GPIO used for random seed to GPIO10.

//SSLClient ethClientSSL(ethClient, TAs, (size_t)TAs_NUM, 2);
SSLClient ethClientSSL(ethClient, TAs, (size_t)TAs_NUM, GPIO_NUM_10);

Thanks
Felix

Hi guys

an M5CoreInk with ESP32 in light or deep sleep still consumes a couple of mA (which is to be expected given the hardware architecture). However M5CoreInk allows to shutdown ESP32 completely and later wake it up from the RTC timer or alarm.

For an M5CoreInk in shutdown mode I've measured a current (at the battery) of only about 2.5 uA which is pretty good. And yes, the ink display retains the last image.

Cheers
Felix

Hi guys

Here is an explanation as to why it may take some time to fix: M5CoreS3 uses the SPI bus in a special way to control the LCD: MISO (normally an input) is reconfigured to be an output and is used as DC line (instead of a dedicated GPIO for DC). Unfortunately to access the SD card (which shares the SPI bus with the LCD) MISO needs to be an input.

In case you might be interested. I found a way to do this switching back and forth between input and output for Arduino.

Thanks
Felix

Hello @Bipman

by default Port A is meant for I2C, but it can be used for I/O (Port B) or UART (Port C) as well.

Note: But please be aware, that it can be only one type of port at any given time. Also, since its default use is Port A (I2C) there are internal pull-up resistors (10 k) in place on both lines which might have a negative effect when used in a different mode.

If you need Port A and Port B at the same time there is the PbHUB unit which can be used to connect some of the Port B units.

If you need all three types of ports (A, B and C) together then I am afraid M5Cardputer is the wrong device to use as core. There are simply no more GPIOs available for a dedicated Port B and/or Port C. Please see schematic here.

Thanks
Felix

Hello guys

my M5Cardputer takes about 7 hours for a full charge from a completely discharged battery state.

Note: I soldered two LEDs to the charger IC CHRG and STDBY outputs and it switched from charging to standby after about 7 hours.

Thanks
Felix

Hi guys

even though I read the warning about not using certain types of chargers with M5PaperS3 I managed to do so anyways.

Suffice to say, M5Paper released some magic smoke and wouldn't start anymore.

I cracked it open from the back (as I think the eInk display is glued in) to get access to the board. (I do not recommend doing that as it is a destructive process.)

With the eInk display and touch disconnected I started to inspect the board.

I found the charger IC USB-C protection diode pack being dead from the look of it. I removed the charger IC (as it is not strictly needed to run) I replaced the USB-C protection diode pack and found the MosFet IC getting pretty hot. So I removed the MosFet as well and injected some voltage (4.2 V) after the MosFet which revealed that the 3.3 V regulator took some hit as well. I replaced the 3.3 V regulator with a similar one I took from an M5Atom.

With two ICs MosFet removed and the 3.3 V regulator replaced M5PaperS3 took a reasonable small amount of current. I also measured ok voltages for the eInk display so I dared to reattach the display and touch screen and yo and behold - M5PaperS3 is alive again.

Now I need to order some replacement ICs MosFet.

You can see the opened M5PaperS3 and the dead ICs here.

Thanks
Felix

Hi guys

I've started to code a simple touch keyboard. I am still experimenting but I thought I'll share what I have so far.

You can find more pictures here: https://www.gwendesign.com/kb/m5stack/m5core2/#simple-touch-keyboard

You can find the code here: https://github.com/felmue/M5Core2/blob/fm-touch-keyboard-1/examples/Basics/keyboard/keyboard.ino

Happy Stacking!
Felix

Hello @gordsh

it looks like this warning/error has been fixed a while ago - see here.

Thanks
Felix

Hello @achayne

I understand your frustration. However all IP addresses you provided are listed as spam and/or brute force attacks when checking on CleanTalk from which we have the information which IPs to block.

May I ask you to reach out to CleanTalk to have those IP addresses deleted from their database and then report back here so I can adjust our list accordingly?

Thanks
Felix

Hello @LordInchcape

not that I know of. That said, have you tried to see what Micropython code UIFlow2 produces?

Thanks
Felix

Hello @silentrain23

I think you probably just need to wait a bit longer for it to register.

From the AT Command document (link here) the 2 in the AT+CREG 0,2 response means: Not registered, but MT is currently searching a new operator to register to

Thanks
Felix

Hello @plackettsj1

the PPS 13.2 module is meant to be used with an ESP32/ESP32S3 based M5Core like M5Stack, M5Core2 or M5CoreS3 - see PPS documentation here.

Thanks
Felix

Hello @wsanders

you answered your question yourself - power off functionality is not implemented yet on M5PaperS3 and therefore RTC alarms do not work.

Some background (as I understand it). Some M5Stack devices - like M5Paper (non S3 version) - can not power off while being powered from USB so the current code in Power_Class does the following: it sets up both, RTC alarm and ESP sleep functions (e.g. deep sleep and light sleep). It then first attempts to power off - which works when powered from battery, so the code execution ends here. However when powered from USB, power off does nothing and the 'backup' ESP sleep functions take over.

So with M5PaperS3 and the current state of Power_Class when you call timerSleep(int seconds) it works because ESP sleep function is setup and can take over. However when you call the other timerSleep() functions nothing happens as the ESP sleep functions are not setup here and power off isn't implemented yet.

And when you call setAlarmIRQ(30) and powerOff() manually nothing happens either - same reason again - power off is not implemented yet.

BTW: you can still test your code though - during the 30 seconds you can double-click the side button to actually power off M5Paper and when the time is up it should power on again.

Thanks
Felix

Hi guys

Update: M5PaperS3 fully restored and as good as new. :-)

Thanks
Felix

Hello @nelll

have you tried to not use M5AtomS3 library and only use M5Unified library and replace all AtomS3. with M5.? Note: not tested myself - just an idea.

Thanks
Felix

Hello @nelll

it looks like the crash is happening when AtomS3.BtnA.isHolding() is called.

Now the strange part. When I modify below two lines in put_asleep() the crash goes away.

//    M5.Display.sleep();
//    M5.Display.waitDisplay();
    AtomS3.Display.sleep();
    AtomS3.Display.waitDisplay();

I can only guess that while put_asleep() isn't called anywhere it still confuses somehow the namespaces - but that is really just a wild guess.

Thanks
Felix

Hello @marnix201

thank you for reporting back. I am glad you figured it out.

If the broken PPS modules no longer show up in an I2C scan (which I guess also means they can no longer be updated via M5CoreS3) you could try to copy the STM32 firmware from a working PPS module to the broken ones.

You'll need STM32CubeProgrammer SW and STLink HW for that though. And make sure to not overwrite the last sector as it holds the calibration data. See here.

Thanks
Felix