46 lines
1.5 KiB
Markdown
46 lines
1.5 KiB
Markdown
v2
|
|
==
|
|
|
|
this has major fixes for version 1:
|
|
|
|
- missing crystal resistor
|
|
- 3.3V dip using larger capacitor
|
|
- 6-40V feedback
|
|
- DFP over-current
|
|
|
|
v1
|
|
==
|
|
|
|
this is a complete re- board layout.
|
|
it does not fit an off-the selves enclosure anymore.
|
|
all SMD parts are on one side to allow easier SMT assembly.
|
|
|
|
fixes:
|
|
|
|
- 6-40V power input activated when UFP is connected
|
|
- use current limit on DFP (prevents USB to reset when DFP is shorted)
|
|
- 6-40V power input capacitor now lays down
|
|
- remove copper plane under crystal to prevent impedance error
|
|
|
|
errors:
|
|
|
|
- crystal still does not start
|
|
- none of the two possible OCS inputs (Dx04 XOR Dx05) work
|
|
- power dip causes reset when enabling all 7 DFP power outputs at once
|
|
|
|
v0
|
|
==
|
|
|
|
this version has been designed to fit in a 100x60x25mm enclosure.
|
|
this makes it very compact, and not all surface mount parts fit on one side.
|
|
this is not too much of an issue since it's a hand soldered prototype.
|
|
|
|
the prototype mostly works, with a couple of small errors:
|
|
|
|
- 6-40V power input is not activated unless the hub is already powered (through USB or 5V input)
|
|
- crystal oscillator doesn't always start. I think the underlying copper plane skewed the impedance/capacitance too much
|
|
- USB-A footprint has inverted pinout: just solder it on the other side and don't swap the data pins in the EEPROM configuration
|
|
- fuse on downstream port is not fast enough to prevent shorts and causes the hub to reset
|
|
- 6-40V power input capacitor is too tall and does not fit in the case
|
|
- one of the LEDS on port 3 is misplaced
|