We bought a RAK3172 and connected it to our circuit to test it.
It is being used in P2P mode and 30 modules are being tested.
If a problem occurs in the RS422 line, wireless communication is performed using the RAK3172.
RAK3172 is not used because RS422 line is usually used.
It was confirmed that the RAK3172 was operating by disconnecting the RS422 line from sometimes.
I tested it for about a month and there was no problem, but suddenly it was discovered that the wireless was not working.
After searching for the cause, the RAK3172 did not work.
The RAK3172 did not respond even when AT commands were sent.
Even if it is separated from our circuit and connected to USB TO SERIAL for testing Similarly, RAK372 did not respond.
RAK3172 module firmware version
RUI_351
RAK3172 module hardware status
3.3V power supply
RST: 10k and 100nf connection, MCU connection
BOOT: Connect with MCU
UART2: Connect with MCU
UART1: Connected to MCU IO, but not used and set to LOW.
RAK3172 The remaining PINs are not connected to anything.
21 out of 30 modules do not work
For 14 out of 21 badmodules, the RST pin went low approximately every 3.56 ms.
This symptom is that when the firmware is downloaded to the RUI_342 version, the
phenomenon of the RST pin falling low disappeared and normal operation resumed.
After that, I am using it after changing the firmware to RUI_351.
However, it does not work if you download only RUI_351 from the beginning without
downloading the firmware to RUI_342.
You must download RUI_342 for normal operation again.
7 out of 21 bad modules are OK with no RST pin going low
But UART1 is set to Low.
The one with this symptom does not work even if the firmware is changed.
How can I prevent the symptoms that do not work as above during use?
If it stops during use, it cannot be used.
And if UART1 is low, please tell me how to make it work again.
Hello, The topic is old but I have the same problem. RST pin pulls to ground at different times. It doesn’t have a specific pattern, sometimes it works for days, but then it goes to the ground and the device turns off and on. As you suggested there is only resistor and capacitor on the RST pin.
The latest firmware is installed but the same problem persists. The same problem persists in custom firmware.
I’ve purchased a lot of RAK3172 and never had such a problem with previous modules. But there are such problems with the modules I bought in January 2023.
I uploaded the schematic and layout. Technically, I should only see VCC voltage on the RST pin. But sometimes it is 0.7V, sometimes 1.2V. For this reason, the module behaves unstable and resets or does not start at all. When it is 0.7V, when I give external VCC from outside, the module starts to work.
Thanks for the (partly) schematic and the PCB layout.
There is no need to pull BOOT0 to GND. That is already done inside the module.
Where else is your RESET_RAK signal going? Your schematic shows only the pullup and capacitor. But on the layout the signal is obviously going somewhere else as well.
Did you check the solder points on all components of the RESET_RAK signal? Your fluctuating voltage on the reset pin can be caused by anything that is connected to it. Inside the module is nothing connected to it and the STM32 MCU cannot pull it down by itself.
Did you check the solder points of all components on the RAK_RESET signal.
As I said, the module itself cannot pull it’s RST down by itself.
What firmware are you using? Are you using our RUI3 based AT command firmware, a custom firmware written with RUI3 or are you using a firmware based on STM32CubeIDE?
I am using software based on MBED OS. As I said, we are actually not newbies on the RAK3172. We’ve been using it since it first came out. So far, we have delivered a total of 5000 devices. However, we are experiencing similar problems on 1000 modules we bought in January.
Hello, @beegee
Was there any resolution to this topic after handling it internally?
My company recently experienced a very similar problem with RAK3172 modules where some modules show a repeating reset that can be observed on the RST line.
The cyan signal in both images is RST, yellow is VCC.
The module seems to work correctly when forced into boot mode via the BOOT0 pin and can be programmed, but start resetting whenever we flash a RUI3-based firmware (even .hex files downloaded from the RAK website). As stated before, this only concerns some of the boards we produced.
The RST is only connected to a 10k pull-up and a 100nF capacitor, nowhere else, so the activity clearly comes from within the module.
Resolution: In my case a full flash erase and reprogramming (both with the STM32CubeProgrammer via UART) has resolved the issue for the affected modules.
Our hypothesis is that a power supply issue caused flash corruption and that in turn lead to the reset cycle.