If I had a c6100 to play with I'd have it done and do a write upNice find Chuckleb - USB traces on the c6100 motherboard! The solder point encircled in white will be ground. The others, in order moving away from ground, are signal positive, signal negative, and +5V. Who is brave enough to give it a try?
It's a multilayer PCB so you may not see the two data tracks, it probably goes straight to the chipset or hub IC, Is there any way to view that part of the board when its in the chassis? 5 minutes with a multi meter and an oscilloscope will tell you everything, even a multi meter will crudely give you an idea if the data line is present if very careful .I think the first generation of the C6100 was OEM'd to Supermicro, so this definitely could be based off that design. I have an AMD cluster at work that is the twin-twin setup, I could pull one of those boards and see if it has a usb plug inside (never bothered to look).
I think a lot of us would love to have you figure this out I have a spare board so would be glad to give it a try but I think there are a few pieces that I are missing that I don't know other than the USB plug.
I assume that the USB data passes back over the FP connector bundle above it? I didn't see any other connections between main board and the expander.
Cicrled is +5V most certainly as the as the usb connectors shield is grounded. But it's possible that there's no actual connection between interposer and motherboard.Nice find Chuckleb - USB traces on the c6100 motherboard! If the pins are in order then the solder point encircled in white will be ground. The others, in order moving away from ground, are signal positive, signal negative, and +5V. Who is brave enough to give it a try?
Update: Or of course it could be the reverse of the above - TheBay below says the circled is +5V and he seems to know what he's talking about. Easy enough to test.
If anyone has a multi meter its easy to find out, or I can do the same if I have a board to pay with. I could fit the components and send it back to someone to test. J13 is more than likely a header not an input as it too requires power via the capacitors, the trace goes elsewhere. Not between J13 and USB. The USB signal will be via the PCB gold fingers or "FB Con"
Yeah that sounds like an idea, initially all I would need is the interposer, would be great if someone in the UK/EU could send one, I'll probe it out and see where the data lines go. Then they can test it (won't cause any damage) I can just put any USB socket on it for testing, the proper ones will need to be ordered in bulk or will cost stupid prices for a single item + shipping from my supplier.I have a spare complete motherboard and can donate and ship the interposer for an experiment. It sounds like we just need to figure out the capacitor design first?
What about the "usb key" box on the interposer board? There's also an 8-pin connector hidden under the pci-e riser labeled USB.I like the SATADOM idea, so I looked for a USB port or header on the motherboard - there isn't one.
What about the "usb key" box on the interposer board? There's also an 8-pin connector hidden under the pci-e riser labeled USB.
And does anyone know what the white 4-pin IPMB connector between CPU0 and memory DIMMs is for? Could power possibly be drawn from there?
I've been trying to follow along, so sorry if this was already discussed and I missed it.