...
- Remove R55, R56 - the PowerPath has them pulled high internally
- Fix silkscreen? Taiping Li
- QFN fets are good, they can probably handle up to 8A continuous without heating up too much
- The dual p-fets had an abnormally high rds(on) - paralleling two helped, but still unsure why it happened: got very hot
- Look into replacing them with flat-pack for better thermals (Taiping Li: Please link)
- Automotive fuses have some resistance across them, but it's reasonable for the currents we're expecting.
- Might want to run power as 18 or 16 awg to minimize losses since we're potentially looking at around 2m cables with multiple sets of connectors
- Paralleled n-fets seemed to work - needed to redo one
- We can actually apply the solder paste by hand
- Might want to add some thermistors on the board near fets
- Debug LEDs are connected to output fets - pins should be switched
- Connect AUX_IN port to net
Revision 2.1
Retrospective
- It doesn't seem necessary to have the ability to monitor current in both directions for the aux battery, as charging will occur prior to the power distribution board
- Remove R20, and use a 0 Ohm for R22 to set the reference voltage to 0V. This will allow the current sensor to output a voltage from 0...3V, and use the same scaling factor as the DCDC current measurement. The only difference is that it will be double the value as the shunt resistor is also double the value.
- The flatpack dual p-fets seem to run at 7-8A without heating up, which is a significant improvement over the previous revision