OpenCS

OpenCS will be an open-hardware (and open-source) console server project.

Status

• 2013-05: Too many other projects on the go, changing to dormant for now until I get some things done.
• 2013-02-03: Base for the transmit side of firmware is written but not yet tested. Receive side not written, and changing parameters not yet supported.
• 2013-02-02: Started working on firmware.
• 2013-01-30: Built first IO232 board, power up + programs OK, need to write firmware now.
• 2013-01-28: Ordered parts for the first IO232 board.
• 2012-12-31: IO232 board is here! Need to order parts and write a bunch of Verilog code to get this module working.
• 2012-12-25: Still waiting for board to arrive...
• 2012-12-05: IO232 board shipped.
• 2012-11: Submitted IO232 board.
• 2012-10-25: Finished routing BACK4VS Rev001, 4-slot vertical single-sided backplane. Now just waiting to submit all 3 to OSH Park when I can fund it.
• 2012-10-14: Finished routing CTLR Rev001, added mounting holes to be able to vertically stack IO232 and tagged Rev002. Backplane board still needs to be done.
• 2012-10-09: Schematic for CTLR done, started routing board.
• 2012-10-08: Finished schematic/routing/CAM for IO232. Working on CTLR now.
• 2012-10: Started putting design together

Specs

• 1U rack-mount chassis with 8 slots (4 front, 4 back) connected by an internal backplane
  • Slots will ideally all be equivalent so users can put power/ports where they want them, can be filled with:
    • a PSU (up to two per system)
    • a controller (up to two per system)
    • an expansion module (lets you chain more chassis, up to two per system)
    • an I/O module (gives you serial ports)
• SPI between modules with module select pins shifted by one by the backplane for each slot
• Fully-expanded base chassis would be two PSUs, two controllers, and 4x 8-port RS232 modules.
  • 2 controllers + 4 I/O modules + 2 PSUs in the base
  • 4 I/O modules in each expansion unit, up to 3 expansion units
  • Each slot on the base gets 4 module select lines, which it can use or give to another chassis
  • Total of 4*8 = 32 select lines needed
• Controllers will have an Ethernet port and two serial ports on them (console/aux)
  • Probably PIC32-based makes the most sense
• I/O modules will have up to 8 serial ports on them
  • Probably CPLD-based makes the most sense

Revision 1 Issues

IO-232

• 1uF caps not on BOM
• Footprints for cap arrays have way too little exposed copper
• Status LED is backwards

Commands

• 0x01 - MODULE IDENTIFY
  • No parameters
  • Response Byte 1 - Module Model ID
  • Response Byte 2 - Protocol version
  • Response Byte 3 - Firmware version high byte
  • Response Byte 4 - Firmware version low byte
• 0x02 - PORT SEND DATA
  • Parameter Byte 1 - Bits 0-2 - Channel to address
  • Parameter Byte 1 - Bits 3-7 - How many bytes to send (max 32)
  • Parameter Bytes 2-X - Data to send
• 0x03 - PORT MODE
  • Parameter Byte 1 - New port mode
    • 0x00 - Do not change
    • 0x01 - RS232
  • Response Byte 1 - Updated port mode
• 0x04 - PORT SPEED
  • Parameter Byte 1 - Baud rate
    • 0x00 - Do not change
    • 0x01 - 300
    • 0x02 - 1200
    • 0x03 - 2400
    • 0x04 - 4800
    • 0x05 - 9600
    • 0x06 - 19200
    • 0x07 - 38400
    • 0x08 - 57600
    • 0x09 - 115200
    • 0x0A - 230400
    • 0x0B - 460800
    • 0x0C - 921600
  • Response Byte 1 - Updated port speed
• 0x06 - PORT DATABITS
  • Parameter Byte 1 - New number of data bits
    • 0x00 - Do not change
    • 0x01 - 5
    • 0x02 - 6
    • 0x03 - 7
    • 0x04 - 8
  • Response Byte 1 - Updated number of data bits
• 0x07 - PORT PARITY
  • Parameter Byte 1 - New parity setting
    • 0x00 - Do not change
    • 0x01 - Even Parity
    • 0x02 - Odd Parity
    • 0x03 - Mark Parity
    • 0x04 - Space Parity
  • Response Byte 1 - Updated parity setting
• 0x08 - PORT STOPBITS
  • Parameter Byte 1 - New number of stop bits
    • 0x00 - Do not change
    • 0x01 - 1
    • 0x02 - 1.5
    • 0x03 - 2
  • Response Byte 1 - Updated number of stop bits