Notes on preparing the JeeLink hardware and usage of this Integration

Feedback

A link to the instructions on how to prepare the JeeLink module would probably be useful. Since it's in the German Wiki and Home Assistant docs are English, here's a Google Translate:

https://wiki-fhem-de.translate.goog/wiki/JeeLink?_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=nui,sc

Most important section is:

------------ snip ------------

In order to be able to use the JeeLink with FHEM, a specific "firmware" (a sketch) must be installed on the JeeLink (with the Arduino software / development environment (IDE)). The general preparation for this action is independent of the required sketch and consists of the following steps:

Install the appropriate FTDI driver for Windows or Mac OS X , under Linux this is usually already available
Installation of the Arduino software for the platform used (Windows, Mac OS X and Linux are available)
Depending on the sketch, integrate the Jeelabs library into the Arduino IDE
Download the required sketch (from FHEM / contrib )
Connect the JeeLink to a USB port on the computer with the Arduino IDE
Start the Arduino software and flash JeeLink

Flash JeeLink from Arduino JeeLink flashing

Connect the JeeLink to a USB port on the computer with the Arduino IDE
Start the Arduino software
Select the serial port of the JeeLink
Settings: select "Arduino Uno" as the board in the tools
Select sketch with "Open file"
Click upload

------------ snip ------------

For clarity, note that this code should not be confused with the RFM12demo sketch that can be found preloaded on some JeeLink devices.

The default Arduino sketch configures the JeeLink for 866 MHz operations. Current La Crosse Technology sensors in North America use the 915 MHz band (eg. my CA80550 - https://www.lacrossetechnology.com/products/ca80550 ) and older gear used 433 MHz, so it seems firmware modifications are necessary in some cases. The JeeLink board itself is provided with a physical radio block specific to these frequency bands as well, so care must be taken when initially purchasing the JeeLink.

To test the JeeLink board you can use terminal software such as Screen. Configure the program to talk at 57600 bps.

For my Mac, this was how the command looked screen /dev/cu.usbserial-AL006S9K 57600

The device should respond with something like: [LaCrosseITPlusReader.10.1s (RFM69CW f:xxxxxx r:17241)] ... where xxxxxx is the frequency the board is to be operated at.

Using the commands found in the comments of LaCrosseITPlusReader10, Screen will show the output from the JeeLink in a readable format.

The JeeLink.net forums are down, but can be viewed at archive.org (https://web.archive.org/web/20170718153206/http://jeelabs.net/projects/cafe/boards/7). Links on the RFM12 board page aere currently broken - https://jeelabs.org/202x/rfm/ - and no direct link to the Wiki is available.

URL

https://www.home-assistant.io/integrations/lacrosse/

Version

2021.10.6

Additional information

As an aside, this might not be the best integration for people to consider when starting a new HomeAssistant implementation talking to LaCrosse sensors. I tried to get this setup to work at 915 MHz with the CA80550 sensors and failed to read them, but I'm certainly not an expert in these devices.

There are various projects that integrate Home Assistant with the rtl_433 library (https://github.com/merbanan/rtl_433), which can use RTL2832U-based devices in a software-defined radio setup. RTL2832U devices can be tuned to a much wider range of frequencies at the cost of requiring more computing power. (eg https://www.nooelec.com/store/sdr/sdr-receivers/nesdr-smartee.html )

This library was able to decode the CA80550 sensors on first attempt. It also seems to have a wider range of Lacrosse decoders available along with more sensor types in general. The rtl_433 community seems to be reasonably active (https://groups.google.com/g/rtl_433).

A possible rtl_433 -> Home Assistant integration to consider might be https://github.com/pbkhrv/rtl_433-hass-addons

home-assistant / home-assistant.io