Milkymist One RC3 Known Issues

From Qi-Hardware
Revision as of 13:13, 20 January 2012 by Adamwang (Talk | contribs)
Jump to: navigation, search

Here is the known issues found during debugging and testing RC3

Contents

from contributors

  • From Joerg & Werner [4], for Adam's review
1. Short L3 to fix audio problem
2. R11 should be 0R otherwise we loose stereo separation for the CD input on the expansion header
3. joerg would put ~50 pF in parallel to R23 and R24, to filter HF and to absorb ESD (there's no ESD protection on line in)

Major errors

the intermittent loss of video input

  • the intermittent loss of video input, replace L19 with short connection by soldering on run3 boards. In RC4, will have no this part[5] with routing in wire to connect two planes under video decoder, see page 95, figure 41.

darn of never booted up

  • darn of never booted up about solutions from reset IC and diode:

  • All RC3 boards have been finished/verified with hardware patches that Sebastien Bourdeauducq and Werner Almesberger got this done: 1. R30 and R157 were changed to 10K ohm. 2. R60 removed. 3. C238 added 220pF. Note: the added diode in red color have been removed now. Thus is called fix2b circuit instead. See previous rework steps of h/w patches here during producing rc3 boards.

Proposed pre-rc4 reset schematic

  • Proposed pre-rc4 schematic for design verification on RESET to solve nor-corruption.
  • L1's footprint needs to be smaller and same as 0402.

Minor errors and improvements

  • Downsize the pin number of J3 to two 1x3pins 2.54mm pitch header footprints of Jx and Jy and DNP in rc4 and close to audio codec. See J3 redesign for rc4 on list.
  • to remove the need for the metal sheet, Joerg suggested using 5mm wide endless aluminium sticky tape and attach two stripes across the diagonals of the bottom PCB side. He would connect them to the spacer only on one side, in the middle crossing they would ideally be isolated by the glue. [6]
  • We are thinking about replacing the VGA connector with a DVI-I connector (single-link), for some notes see mail and irc log.
  • Werner propoased that adding another same pins of J21 to make a "U"-shaped and need to fully specify the expansion system and decided to use like female receptacle socket, so that DIY can build small circuit to plug on it.

switch to ADV7181C 

1. Remove L14 ~ L16, C202 ~ C207 entirely
2. Short L14 ~ L16
3. replace U21 with an ADV7181CBSTZ

add USB Power Switch

  • usb power switch

  • Draft USB power switch schematic for pre-rc4 design verification: see [2], [3], an agreement before rc4 for design verification: 1. Add an active low usb switch AP2142ASG-13 2. connect FLG1 and FLG2 on FPGA side for indicator feedback and connect each for a pull-up 10K ohm to 3V3. 3. pre-placing an existing C178 220uF still there then later determine if needs it after verification. Connects pin2 IN to 5V net not 3V3. 4. Connect existed expansion header J21 reserved pins to EN1 and EN2 to let fpga can be able to control usb power. 5. Connect another J21 two pins to two LEDs being as individual indicator.

  • Werner proposed adding LEDs to the ports to indicate [9]
demand = patch wants to use feature X but nothing seems to be connected
activity = an aperiodic flash every now and then
problems = maybe periodically blink

R4 draft feedback

1. delete FLASH_RESET_N on U22B.P22 and move SW_RESET_N up from U22B.P21 to U22B.P22


Links

Milkymist One Schematic Change History

Milkymist One RC2 Known Issues

LVDS, CML, signal conditioning/equalization, jitter/crosstalk, and high-speed transmission line topics, see here.

DVI V1.0

http://www.datapro.net/techinfo/dvi_info.html

Retrieved from "http://en.qi-hardware.com/wiki/Milkymist_One_RC3_Known_Issues"
Personal tools
Namespaces
Variants
Actions
Navigation
interactive
Toolbox
Print/export