Hello @stefan-h

I've tried the similar code as above with an M5CoreS3 and ENV III unit as well and it worked fine for me.

I suggest you try the latest available UIFlow2 firmware version which is UIFlow2.0.0. alpah-28 at the time of this post.

Thanks
Felix

@ajb2k3 thanks so much for that insight. Do you have an implementation of HV write mode for the Paper display in a different platform (UIflow etc)? Don't worry if not, because it seems that existing shutdown() is able to retain images according to @felmue's example.

Sorry, that was a mistake I made on earlier available information. and seamed to work with my fat fingers. I have rechecked, corrected and updated the document and its in the process of being re uploaded.

Hello,
I Got the GFX font header file (by converting the Roboto.ttf file). Loaded the font file using canvas.setFreeFont(&Roboto_Medium50pt7b) and then display the text using canvas.printf(textString.c_str());.
It is working fine and display the text properly but when I am giving some special characters äöü then they are not appearing on M5stack paper.
Can someone please guide me what I am doing wrong? Am I missing something?

Thanks

The link to my question can be found here

@ivka on what?

@ajb2k3

Thanks for your advice.

What I want to do is to power the camera with the M5Capsule alone using it's internal battery. It won't last long but it woiuld be easier to make simple demos with all in a nice little package. It's actually the main reason why I bought my M5Capsule.

I will consider adding an additional power supply.

The E-Ink Paper Display (EPD) on the M5Paper consumes power even when the screen is idle due to its underlying technology. E-Ink displays are different from traditional LCD or OLED screens in that they only consume power when the content on the screen is changed. Once an image or text is displayed on an E-Ink screen, it doesn't require any power to maintain that state. However, changing the content on the screen, such as refreshing to display a new image or text, does consume power.

In your case, when the EPD power is disabled, the display is essentially turned off, and it consumes very little to no power because it's not actively refreshing the content. When you enable the EPD power and the display is idle, it's likely in a state where it's ready to receive new content and is therefore consuming some power in anticipation of a refresh.

The power consumption you're observing is not necessarily a sign of a defective unit but rather a characteristic of E-Ink displays. If you want to optimize power consumption further for your low-power application, you might consider:

Reducing Refresh Rate: Minimize the frequency of screen refreshes. E-Ink displays are typically used in applications where the content doesn't change frequently. You can adjust the refresh rate to be as infrequent as possible to save power.

Full Screen Updates: Instead of partial screen updates, use full screen updates when you need to change the content. Partial updates can be more power-intensive.

Deep Sleep: Continue using the deep sleep mode of the ESP32 when the screen is idle. This will significantly reduce overall power consumption.

Optimize Code: Make sure your code is efficiently managing the display and other peripherals. Ensure that you are properly putting the ESP32 into deep sleep mode when it's not actively needed.

Hardware Modifications: Depending on your specific use case, you might consider hardware modifications to further reduce power consumption. This could involve custom power management circuitry or using different power supplies.

Hi @arzaman, Sorry to answer with a question, but I just want to check I've understood.

You want to accommodate the 5V supply into this sensor and associated 5V? logic output from it down to the 3V3 input of StickC, Core2, etc?

A couple of thoughts:
#1 StickC has a 3V pin out on the end opposite Groove port, so I guess you could go from the head end of the StickC to sensor (reduce all to 3V3).

#2 Copy the voltage divider circuit a lot of the M5Stack sensors use (they are quite cheap, so you could even use one as a donor and solder your sensor on to it).
https://docs.m5stack.com/en/unit/dual_button

#3 I've a feeling Seeed, DFRobot, Waveshare do a mini board to do such a thing, but probably equivalent to donor M5Stack part.

If you used the M5Stack Mini Dual Button Unit as a donor/connection board. I suppose you could have two inputs, pin 32 and pin 33, hide the pull up resistors, etc. Or the Proto units, perhaps an easier solder at home proposition.

Q) If it "can be powered from data line. Power supply range is 3.0V to 5.5V", would 5V power issue go away anyway?
https://shop.pimoroni.com/products/temperature-probe-ds18b20-arduino-compatible-adapter?variant=39365415829587
(NB Alt schemtic for circuit)

@teastain
Hi Terry, et al

thx 1M for your effort helping me!

I've just found the work around:
As per "M5Stack Library won't install on Arduino IDE", CitrusPeel identified the issue:

The package's name "MODULE_GRBL13.2" in library.properties should be "MODULE_GRBL_13.2" (underline instead of space before 13.2).

So

download M5Stack unzip it edit file "library.properties", line to "depends=M5Family,M5Module-4Relay,MODULE_GRBL_13.2,M5_BMM150" zip it and install zipped library restart IDE update now performed flawless

@m5stack: awaiting M5Stack (0.4.6) to be released soon, as not all users having the knowledge on how to apply this "patch".

Cheers, Marcus

@davincino Thanks for posting your findings.

This topic is old, but I will reply if it is found through search functions:

I found a solution, what can be applied with small external hardware modifications. It is possible to take some measurements on HAT interface's pin "5VIN".

Measured voltages:

If powered by USB:
~4.7V

Powered by battery and switched off:
~0.7V

Powered by battery and switched on:
~2.4V

Using voltage divider with two same resistors, we get usable logic values for HIGH and LOW ( ~2.3V and ~1.2V). The measured values are not inside garanteed logic range but as a workaround: it works well. Maybe it helps someone else...

You would be better off directly contacting M5Stack about this but as the tough is a core2 and core2 is still being added to UIFlow, I wouldn't think it would be EOL that soon.

To make this detection really reliable I had to change the RTC register check like this:

Wire.begin(21, 22); uint8_t reason = M5.RTC.readReg(0x01); // now it's safe M5.begin(); // check reboot reason flag: TIE (timer int enable) && TF (timer flag active) if ((reason & 0b0000101) == 0b0000101) { restartByRTC = true; Serial.println("Reboot by RTC"); } else { restartByRTC = false; Serial.println("Reboot by power button / USB"); }

Did you click on the "Erase" button in M5Burner before burning the correct M5Paper firmware?

@cognitive5525

Thank you for your insight and investigation.
I was facing same situation after my m5stack core2 lost its battery at all. Then I almost gave up development with m5stack since I'm not familiar to this ecosystem but it started restart again after your i2c cord had been tried.
Thank you so much again!!

@zalaking The schematic shows the following I2S pins on the CoreS3:

GPIO00 I2S_LRCK, (Mic MCLK)
GPIO13 I2S_DOUT/DAT0
GPIO14 I2S_DIN/DAT1
GPIO33 I2S_WCK (Mic)
GPIO34 I2S_BCK (Mic)