Proper way to do UART re-init in the code?

[divergentti]

I made simple asynchronous driver for a Neo6M GPS module -> https://github.com/divergentti/Micropython-for-ESP32/blob/main/esp32-gps/GPS_AS.py

The code seems to perform quite ok, but every now and then communication is silent or the module keeps responding "GPTXT,01,01,02,HW UBX-G60xx 00040007 FF7FFFFFp" or GPRMC messages.

I presume this problem is related to UART somehow and I implemented self.moduleUart.init() at the end of the async loop, but that does not fix the problem. Is there other ways to reset the UART?

HW is ESP32 with esp32-ota-20230426-v1.20.0.bin

[2dof]

form what You said, is seems that Neo6M GPS module is resenting. What Vcc did to connected to the module (3.3 or 5 V)?

[divergentti]

Using 5V VCC seems to work ok.

I added RTC update from satellite (via GPRMC NMEA).

[2dof]

My suggestions:

• def checksum(nmeaword) supposed to return checkum status ( True/False) only, there is no sense to preprocessing
  data frame: cut-off '$' (start_code) return and then add again in reader() return. Then if checkum ok, perform decoding from whole frame.

[divergentti]

Thank you for the suggestion. I reworked and tested the code. Now simplier, less variables and simplified reader.

Do you have idea how to calculate yeardate properly? With modulo 7 I should be able to calculate weekday for RTC time setting.

OLD:
"I had to add a lot of checks to understand what is going on. I started making asynchronous code (without CRC checks etc) by following "NEO-6M GPS Module with ESP32 using MicroPython" inststructions at https://microcontrollerslab.com/neo-6m-gps-module-esp32-micropython/"

[2dof]

If You mean current date field from $GPRMC

then:
( “Doomsday Algorithm.”)

def weekday(y, m, d):          
        
	t = [0, 3, 2, 5, 0, 3, 5, 1, 4, 6, 2, 4]
	y -= m < 3
	return (y + int(y/4) - int(y/100) + int(y/400) + t[m-1] + d) % 7

will return day no: 0 (Sunday) ,1 (Monday) ....
, for example :

wekday(23, 5, 30)   
    wekday(2023, 5, 30)  

will return value 2 (i.e Tuesday)
note that y = 23 (2023) , whole year y=2023 will also work

[divergentti]

Thank you for your help. The code seems to perform just fine. No crashes, no module inits. I set latest reply as answer to original quesion.

73's

1 ---------MCU------------- 1
Memory free: 69872, allocated: 41296
Heap info [(240, 0, 0, 0)], hall sensor 65, raw-temp 51.7C
2 ---------SENSOR----------- 2
Temp: 24.6C, Rh: 32.8
Gas: 74961.0
Pressure: 1009.0
Waiting GPS fix...

--- RMC debug ---
GPSTime:
System time set: (2023, 5, 31, 8, 7, 57, 2, 151)
Data valid True:
Speed over ground: 0.573
Course over ground:
--- End of RMC ---

1 ---------MCU------------- 1
Memory free: 73072, allocated: 38096
Heap info [(240, 0, 0, 0)], hall sensor 66, raw-temp 51.7C
2 ---------SENSOR----------- 2
Temp: 24.8C, Rh: 32.5
Gas: 79148.1
Pressure: 1009.0
Waiting GPS fix...

1 ---------MCU------------- 1
Memory free: 66192, allocated: 44976
Heap info [(240, 0, 0, 0)], hall sensor 65, raw-temp 51.7C
2 ---------SENSOR----------- 2
Temp: 24.8C, Rh: 32.3
Gas: 82186.5
Pressure: 1009.0
3 ---------GPS----------- 3
Latitude: 60.358562
Longitude: 23.100435
Satellites: 08
GPSTime: 08:07:57
SystemTime: 11:08:03 and weekday: Ke

--- RMC debug ---
GPSTime: 08:07:57
System time set: (2023, 5, 31, 8, 8, 1, 2, 151)
Data valid True:
Speed over ground: 0.144
Course over ground:
--- End of RMC ---

1 ---------MCU------------- 1
Memory free: 69696, allocated: 41472
Heap info [(240, 0, 0, 0)], hall sensor 61, raw-temp 51.7C
2 ---------SENSOR----------- 2
Temp: 24.8C, Rh: 32.0
Gas: 86687.4
Pressure: 1009.0
3 ---------GPS----------- 3
Latitude: 60.358570
Longitude: 23.100435
Satellites: 08
GPSTime: 08:07:57
SystemTime: 11:08:06 and weekday: Ke

--- RMC debug ---
GPSTime: 08:07:57
System time set: (2023, 5, 31, 8, 8, 5, 2, 151)
Data valid True:
Speed over ground: 1.375
Course over ground:
--- End of RMC ---