ADB Busboy: Adapter for USB peripherals over ADB

[Scott Squires]

Thanks for the comments everyone. I'm still working on the bootloader which has had a lot of road blocks and detours. It's important to get the bootloader ready because it is the way that you'll be able to update the software when I develop new features. The original one I intended to use turned out to be incompatible. I tried to get a commercial one to work, but it lacked support for USB hubs. So now I'm adding USB host support to the manufacturer's bootloader.

 

[defor]

Definitely watching this closely!

 

[Scott Squires]

So, a few things have happened since my last update. The good news is that my bootloader is basically working. There are a couple little tweaks left to implement. But it does its job of updating the board firmware.

I was being clumsy with my bench power supply and accidentally released the magic smoke. So I had to assemble a new board. During testing the new board, I found that the integrated usb hub works, but if I connect another downstream hub, it fails. At this point I don't remember if I ever tested that on my original assembled board. I did test it on the development kit board, and it worked. So I don't know if it's a problem in my board design or a defect on this board. The pcb has a slight pressure mark on one of the usb traces. Otherwise, I can't find any defect. If that were the problem, I don't think the onboard hub controller would work either.

The one empty pcb I have left has a pressure mark in the same spot. So to make eliminate that possibility I would have to order more pcbs. And they would probably have the same issue. It could be the hub controller I'm using. To test that idea I would have to redesign the board for a different controller. Ugh.

 

[CelGen]

So what is your price going to be now that BMOW is selling a competing (and overly ugly) product at $50 USD?

 

[Scott Squires]

A related topic came up in IRC today so I figured I should really post an update.
 
I solved the unreliable behavior with daisy-chained hubs. Some USB devices draw a lot more inrush current than others. Adding a capacitor to each port solves this problem.
 
I redesigned the PCB to be about 40% smaller, switched to an RGB status LED, upgraded the ESD protection diodes, and probably a couple other things I have forgotten about. I haven't ordered this new board yet. I want to review the layout again first. When I will get around to doing that, I don't know.
 

 

[68krazy]

Did someone say R G B ?!



Sounds like this is going to be a very refined product when it's done.  Looking good so far!

 

[Trash80toHP_Mini]

I've had KVM switches and conversions on the brain for a bit now. Wondered if there might be a better spot than the DIN port to hook something like this into a TTL switchbox setup?

 

[bbe]

A related topic came up in IRC today so I figured I should really post an update.
 
I solved the unreliable behavior with daisy-chained hubs. Some USB devices draw a lot more inrush current than others. Adding a capacitor to each port solves this problem.
 
I redesigned the PCB to be about 40% smaller, switched to an RGB status LED, upgraded the ESD protection diodes, and probably a couple other things I have forgotten about. I haven't ordered this new board yet. I want to review the layout again first. When I will get around to doing that, I don't know.

Please, let us know the salestart....

 

[timmerk]

Will this be basically the same thing as the ADB-USB Wombat (https://www.bigmessowires.com/usb-wombat/) ?

 

[Scott Squires]

It's almost the same as USB Wombat. The primary differences are :

* ADB Busboy has 2 USB ports

* ADB Busboy only allows you use USB peripherals on ADB computers, not the opposite.

* ADB Busboy has better power protection between the ADB host (computer) and USB bus.

 

[mietek]

What happened to ADB Busboy? Will you release the sources?

 

[Scott Squires]

I have pivoted the project in a different direction. Keeping quiet about it until it's ready, because knowing me, it could be a while.

 

[mePy2]

Updates? ...

 

[daanvdl]

My understanding is that the Kensington mouse had both USB and ADB electronics inside, and switched between them depending on what it was connected to. The intelligence was in the mouse, not the adapter.

ADB Busboy will work without drivers, and will work on all system software that uses ADB, all the way back to the Apple IIgs.

After buying a Kensington 64210 and the lack of an adapter cable I can indeed confirm in having two microcontrollers in the mouse. The breakout cable is just passive.
For completeness and future reference, I reverse engineered the pinout back to the two microcontrollers.
The Kensington 64210 MiniDin8 breakout cable pinout is:

Pin 1	ADB Data
Pin 2	ADB VCC (5V)
Pin 3	USB VCC (5V)
Pin 4	USB DATA-
Pin 5	USB DATA+
Pin 6	GND
Pin 7	NC
Pin 8	NC