I'm getting an S1 in 2 days as a new toy to play with after the gridseed and zeus.
Considering what is the control board, I believe something can be done, but will almost for sure need custom firmware, and a few components.
I'll elaborate when I'll have the control board in hands, but it's not like splitting the data sent to one board to feed 2 or more boards.
I won't have the software skills to do it, but I'll probably have the hardware ones, if it can be done.
I've got my ear to the ground on that ma' friend!
I am convinced that those ribbon cables are simply rx/tx lines from a USB chip on the controller board. In my scheme of things, I was actually looking at running the boards without the bitmain controller board and simply powering them up and connecting to my own laptop's USB via a custom USB chip (I am guessing there are going to be at least 4 tx/rx lines per board) that can handle 4 lines then looking at kano's driver (I already have a miner) to run on a windows PC.
But hey, mine is probably more suited to me and will be in the works for a while now, nevertheless I await your results in a couple of days when you get your S1 (I personally want to use the boards from an S1 that I will be upgrading with the S3 kit).
You are right on the usb thing.
There is only 1 USB line from the atheros chip, and it's routed to some kind of USB hub on the board.
Then, each header is used for temp sensing, fan driving, and 1 usb line.
But the same atheros chip is used in the TPLINK 3020. I guess now, you understand what I'm looking for?
Hacking a TPlink firmware with the cgminer that supports the bitmain chips/boards is the easiest and most cost effective way to do it.
Then, the usb interface can be easily made with a simple and cheap PCB or even a usb cable to whatever pin header they are using.
Controling the temperatures and the fans would be lost if you use a simple usb cable.
I hope my explainations are clear enough so someone on the software side can implement them.
I started to look at this a few hours ago with nothing else than the schematics and a few pics of the board, so I don't have all the details yet, and I need the boards to work.
^sounds like a good way to go, the tplink would be a lot cleaner of a build than an RPi and even cheaper.
One easy way to do this may be to compare with the U1/U2 devices, which hosted a singular chip and acted as a UART-USB bridge. My understanding is that the chips communicate data in a singular chain, or at least as 4 chains per PCB.
undervolted boards could likely run with airflow only and no heatsink, making a restored S1 system very cheap to build as an open-air if many boards vcan control together