WiFi Modules for BeagleBone Black rev. C

Hi,

I’d like to ask about the current examples of WiFi modules for BeagleBone Black rev. C – on the latest Debian version ( AM3358 Debian 10.3 2020-04-06 4GB SD IoT ). I’ve found some of the examples on the elinux.org but I’m not sure if they are up to date.

My initial success included doing a google search on various wireless driver names listed in the kernel, including realtek (etc). I scored an initial dongle that had an RTL8187 class chipset from “Plugable” model USB-WIFINT.

It didn’t have the ability to go into “access point” mode, which is something I really want. (think ability to connect and surf to its own web server without external influence)

Right now I’m trying to get one of Microchips ATWILC3000 class of modules going.

If you’re interested, I can say that I’ve had some success getting SPI1 to work, though it destabilized kernels 4.19 and 5.10 in the process. The latest drivers from Microchip did build with kernel version 5.15.49-bone-rt-r25, but right now I’m running into issues where it’s not properly initializing the SPI channel. Other modules that are available without having to solder down a module are Microchip’s WILC3000 shield and MikroBus’s WINC3400 module.

Just curious if you got 'access point" mode to work. I’ve been using the Black Wireless in development of a new piece of ag equipment. Since Wireless is not available, I’m reverting to the basic Black and trying to get it to bring up a wifi network of its own. Our equipment will be in the field and won’t have a wifi network around so we need the equipment to have its own. The WIreless was ideal for this but I digress. I have a USB dongle wifi working but it only connects to our office wifi. It doesn’t bring up its own. I’m on Linux 4.19 right now. Any ideas?

It got “back-burnered” while I got to wrestle with a number of hardware issues not related to it. I’m also doing a bit of shouting about the wifi action considering some of our target market segments, and some rightfully cagey IT security teams…but I digress.

I did manage to get the latest 4.19 directed drivers to communicate with it long enough to get the MAC address out of it. The driver suffered from being hardwired somewhere to communicate through the SDIO interface…the interface I’m using as a development filesystem. It crashed.

In getting that far, I did find that the device tree file am335x-bone-common-univ.dtsi needed a tweak.
The definition involving spi1 had spi-cpha set (around line 1385) which mal-configured it for the ATWILC3000.

I’m due to get my head re-wrapped around it in the immediate future, so there may be some updates soon.

1 Like