I accidentally flashed a program that configured the SWD pin as a GPIO so now I'm not able to flash my board. All my pins on my board are broken out so I can access them all. How do I reset my pin?
just keep reset low when debugger connects to the micro. Manually or set the option "connect under reset"
Related
The dev boards for the ESP32 family of MCU use CP210x (or similar) "FTDI" chips to communicate with the MCU when flashing. CP210x presents a COM port to the host computer which runs esptool.py, a script which implements the Esressif communication protocol. Rather then use a CP210x, I would like to utilize a SAMD21 in its place.
I have managed to receive logs from my ESP32, to a serial monitor on my computer, via the SAMD21, over USB. The SAMD21 connects to the ESP32 via UART, with the standard ESP configuration of 115200 8N1. I can manually enter bootloader mode by holding the correct pins low at boot, and I get the log back confirming the correct bootloader mode.
When I run esptool.py, however, the connection fails, and I get a timeout. Likewise the esptool "monitor" fails to acknowledge the same logs which my terminal emulator easily detects.
What is the communication protocol between the host computer running esptool.py, the CP210x FTDI chip, and the ESP32, and how can I emulate the CP210x with a SAMD21? The definition of "FTDI", "TTL" and "RS232" are all a bit fuzzy, as far as I can determine with research online, so if anyone has experience in this arena, I would be very curious to hear your advice.
The code I have on my SAMD21 is just:
#include <Arduino.h>
void setup()
{
Serial.begin(115200);
Serial1.begin(115200);
}
void loop()
{
if(Serial.available())
{
Serial1.write(Serial.read());
}
if(Serial1.available())
{
Serial.write(Serial1.read());
}
}
Where Serial1 is the UART and Serial is the USB Serial connection.
esptool.py expects a serial port (known as COM port on Windows) to communicate with the ESP32. It doesn't care if the computer has an old-style serial port, is using a USB-to-serial bridge with a proprietary protocol (FTDI, CP210x and the like) or is using the standardized UBS protocol for serial communication (USB CDC ACM). This is left to the operating system and the installed drivers.
However, esptool.py modifies the baud rate and uses the RTS and DTR signals to reset the ESP32. It is also rather susceptible to timing issues with regards to the reset. If you set the ESP32 in boot mode manually, you should be able to get away without these.
However, the most likely cause is that the Arduino CDC implementation does not implement flow control. If esptool.py sends more data than fit into the internal buffer, it is likely discarded, instead of sending a NAK back so the host computer can retry later.
I don't fully understand the SAMD21 Arduino core to verify it. If so, I don't see how you can make it work with an Arduino program. You would need to resort to some other framework for programming the SAMD21.
Update
After more tests, it turns out the USB CDC implementation of the SAMD21 Arduino core correctly implements flow control. So no data will be lost.
In fact, I was able to successfully upload code to an ESP32 module. The setup was:
Arduino code built with PlatformIO. Instead of Serial, I've used SerialUSB as I'm unsure how to control the project settings available in the regular Arduino IDE.
For the ESP32, I've used a ESP32-WROOM-32 module on a minimal board (reset and boot button, 2 pull-up resistors).
I've connected the board via GND, 3.3V, TX, RX directly to the SAMD21 dev board.
I've verified that I can see the ESP32 log output in normal run mode and the "waiting for download" prompt in bootloader mode (after pressing BOOT and RESET).
Arduino has multiple boards where the esp32 is on-board as WiFi adapter. To flash the esp32, there is a tool sketch called SerialNINAPassthrough in examples of the WiFi library, which should be uploaded into the main MCU of the board (SAMD21 on two of the official boards).
The SerialNINAPassthrough sketch handles the DTR and RTS signals sent by the esptool to reset the board into the flashing mode.
I have a device connected to a USB and it periodically sends data to my PC. During the time it is not sending data, the USB cable needs to be removed, else, the device won't work (a fail-safe feature, if the USB cable is connected and attached to pc, the device is in communication mode, if the USB cable is disconnected from pc, it is in stand-alone operation mode).
I was thinking to use the Devcon so I font has to use another hardware, I'll just disable the USB port during an operation mode, then enable it during communication mode. However, the periodic transfer of data can be set to every 5,10,15,30 minutes depending on the settings.
My question is, will it be okay to disable/enable the USB port periodically? Say every 5 minutes? Won't it cause any problem in the long run? Or would it be more efficient for me to use external hardware, a switch to connect/disconnect the USB? Appreciate any advice or thoughts.
If devcon works for you, I don't see why it would cause any problems in the long run. There is no inherent reason why the hardware should get damaged when you run some commands in your software to disable a USB port.
Using external hardware to accomplish the same thing would not be more efficient because you'd have to pay for the hardware and maintain it.
I have an nrf52 board I was able to connect to board via USB. I downloaded a hardware profile to check the functionality and the device acted like a BLE and I was able to communicate with it. But someone said to implement UART on the device now i am clueless about what this is?
there is this nordic toolbox which has UART example but my board won't connect to this why is that?
I wish this NRFtoolbox UART connects to my board how is that achievable?
it will not connect using standard configuration because in this mode it uses pin 5/6 for tx/rx as i remember, and these pins are connected to internal modem, you can not exchange data on these pins. I would suggest to use pins 2/3 for tx/rx.
I'm working with a commercial device which uses a STM32F103RBT6 microcontroller. I need to replace the firmware to customize some functionalities, but before I would like to dump the current one to be able to restore it if it's needed.
I connected properly the board to my ST-Link v2 programmer, the device is powered by the programmer and in fact when I plug it on the USB the device starts correctly. The problem is that I don't manage to connect to the microcontroller using the ST-Link utility, I always get a "can't connect" error (error message), trying with different settings.
I need to reset the chip before? How?
Thanks!!
Please check you have connected the JTAG/SWD signals from ST-Link debugger to correct pins on the target.
Also ensure the board is power.
Then check settings in ST-Link utility, check JTAG/SWD connection, speed and connect under reset which assert the NRST pin on the target in order to allow connection on JTAG/SWD pins.
If this doesn't work then it can be also the case that the MCU is protected against reading by setting an option byte (look for RDP2 in reference manual). If this bit is set then you won't be able to connect.
I cannot boot on P2041RDB-PB. All ethernet ports (5 ports) are blinking rapidly after I press poweron button.
I am trying to check on the serial using minicom, it doesn't have any output.
I am trying to boot from SD Card, it doesn't output anything at minicom, and I still got all ethernet ports blinking rapidly.
How should I resolve this problem?
What the kind of bootloader did you used? please make it specific your question. Firstly, you can try to boot your P2041RDB PB using U-BOOT it's free don't forget to compile u-boot sourcecode for P2041 processor. You can use JTag Debugger to deploy uboot binary to your p2041rdb. You can read the help section in uboot homepage on how to make uboot call Linux OS or other Realtime OS like u/coss-ii.
Permios Amigossss
My P1021 behaves the same, if I try to boot from MMC/SPI wihtout placing valid bootloader code. It is also not possible to connect via JTAG debugger because the processer is kept in reset.
I configure to boot from (NAND/NOR), then connect with JTAG debugger to flash uboot into SPI and then reconfigure the board to boot from SPI again.
This works for me.