Feather RP2040 RFM69 Packet Radio version 35 not supported

[michaeldye]

A recently-purchased feather has an rfm69 device that reports version 0x23 which fails a version check in the RFM69 constructor:

...
  File "/lib/adafruit_rfm69.py", line 316, in __init__
RuntimeError: Invalid RFM69 version, check wiring!

Version number extraction from REPL:

import time
import board
import digitalio
import adafruit_bus_device.spi_device as spidev
from micropython import const

reset = digitalio.DigitalInOut(board.RFM_RST)
reset.switch_to_output(value=False)
reset.value = True
time.sleep(0.0001)
reset.value = False
time.sleep(0.005)

buff = bytearray(4)
dev = spidev.SPIDevice(board.SPI(), digitalio.DigitalInOut(board.RFM_CS), baudrate=2000000, polarity=0, phase=0)
with dev as d:
  buff[0] = const(0x10) & 0x7f
  d.write(buff, end=1)
  d.readinto(buff, end=4)


print(f"{buff[0]:#0x}")

[jerryneedell]

Wow! Is this an Adafruit rfm69 board? Can you provide a link to it and to the datasheet for it?
@ladyada Are you aware of any change in the device ID?

The data sheet clearly indicates that it should be 0x24 --- was it ever 0x23?

[jerryneedell]

Apparently the SX1231 chip does indeed have a Revision ID of 0x23
https://cdn-shop.adafruit.com/product-files/3076/sx1231.pdf
while the RFM69HCW data sheet says it it 0x24 ????
https://cdn-shop.adafruit.com/product-files/3076/RFM69HCW-V1.1.pdf

Both data sheets are referenced in the guide for the RP2040 RFM69 and the Feather M0 RFM69

What is correct?

[michaeldye]

Yeah, it is one of two PID: 5712 boards I bought from adafruit.com on 4/16/24. The other of the two boards reports 0x24. I made the change in PR #53 to get them to both work with the lib.

[jerryneedell]

Yeah, it is one of two PID: 5712 boards I bought from adafruit.com on 4/16/24. The other of the two boards reports 0x24. I made the change in PR #53 to get them to both work with the lib.

Yes, thanks for the PR... It would be good to get some confirmation of a HW change.
Can you post pictures of the 2 boards? Are the RF modules different?
This will set back debugging discipline ;-) Failure of the ID check has always been attributable to a wiring issue ...

[michaeldye]

Left - board whose rfm69 reports version 0x23
Right - board whose rfm69 reports version 0x24

[michaeldye]

Apparently the SX1231 chip does indeed have a Revision ID of 0x23
https://cdn-shop.adafruit.com/product-files/3076/sx1231.pdf

p. 84 of that doc says there are limited supply chips with values 0x21 and 0x22 for that register as well.

[ladyada]

those are almost the same datecode, but the 0x24 is a few weeks later...so maybe they did revise something. we do test each board by send/recv data so it should work fine, we just need to tweak the value we check against in this library @jerryneedell could you do a PR?