Commit 7afec3b4 authored by tomaszwlostowski's avatar tomaszwlostowski

--no commit message

--no commit message
parent fac2c64a
MCH_MAIN:
MCH_MAIN review minutes
(add fixes done after sch review....)
Tomasz Wlostowski, 2009/03/30
----------------------------------
- reordered CPU LEDs and UTCA leds signals
- eth connector ground changed to ETH100_AGND
Review at CERN:
- pullups for JTAG lines
- changed mezzanine connector layout (2x70-pin kx14 -> 3x50-pin kx14 - easier to obtain and more flexible placement on PCB)
- DM9161: pulldowns added on RX_DV, RX_ER, DISMDIX.
- changed value of R18, R19 in FPGA MSEL pins (100k were too big)
- add current monitors for PSU
----------------------------------
My own observations and Greg's comments:
CPU:
- fixed VDDBU power supply (should be 1.2V)
- reordered CPU LEDs and UTCA leds signals for easier routing to CPU
- eth connector ground changed to separate ground plane
- removed plane_connects, 0R resistors instead (100m ethernet)
- changed UART routing to front panel (now UART0)
- CPU reset thru diode (D22) to avoid ICE resetting whole thing
- MAIN_RESET going thru hole fixed
- reversed uTCA status LEDs
FPGA:
- added internal netlabels
- removed DCLK connection to GPIO pins (not necessary - we are using SSC)
- pulldowns on FPGA DIN and DCLK lines to reduce ringing/overshoot (very long traces)
- reordered SW_JTAG and FPGA config lines
- changed UART routing to front panel (now UART0)
- CPU reset thru diode (D22) to make JTAG working :)
- additional pinswap information configured for clocks
- testpoints at spare pins (FREE_2V5, FREE_3V3) on bottom layer (uncoated vias)
- disabled partswap for resistor ladders due to partswap bug in Altium Designer
Clocking:
- added length matching for REFCLK and RBCLK (~500 mil tolerance -> ~0.2 ns propagation time)
- additional decoupling caps/ferrite beads for fanout chip
PHYs:
- changed downlink PHY->fabric A line mapping for easier differential pair routing from edge connector to PHY chips
- PHYs are now powered from separate plane with additional HF noise decoupling (inductor + bank of caps
WD:
- pinswap configured for WD GPIO pins
- changed Atmega footprint to QFN for space reasons
- Additional 0R resistors on SPI bus added, allowing for disconnecting main CPU
- pullup for PEN line added
- added analog MUX (4051) for supply voltage/current monitor
Power supply:
- additional filtering for current monitors (RC)
- changed downlink PHY->fabric A line mapping
- changed atmega footprint to MLF64 for space reasons...
- pinswap in atmega WD chip
- changed 28c16 eeprom to ds18b20 unique id chip
- routed atmega uart to programming connector for debugging purposes
TODO:
General PCB:
- fixed thermal pad layouts in PSU/PHYs
- fixed edge connector paste mask issue
- bigger holes for MCH stacking
- lots of footprint updates....
- drawn a nice logo and warning sign :)
- added ESD strip resistors
- MAIN_RESET going thru hole (fix)
- check the footprint for fuse (2920L25DR)
- fix thermal pad layout for TPS40055 and LM20123
- wider pads (outside direction) for 32768 Hz xtal
\ No newline at end of file
This source diff could not be displayed because it is too large. You can view the blob instead.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment