Yep.Peter wrote:Problem one: Only a fraction of the I/O pins usually has that feauture.
I have reels and reels of 4x and 8x SMD resistor packages. If I were going into that level of production, I'd also just get a pick and place machine so they're tidy and consistent, densely packed, and not taking my time.Peter wrote:Problem two: A mass-grave of series resistors. In the end, level shifters are probably easier to solder.
This too. People like you and Nasta know what to do in these situations. I just take the advice/schematics I'm given. Chances are, the Aurora II complete board replacement could be implemented in the FPGA too.Peter wrote:Problem three: The required resistor values tend to be too large to provide sufficient drive current for all the pins inside the QL. This is because the clamp diodes are usually "weak", and the resistance must be large enough for the overshoot coming from the QL.
I'm just not going to baulk at a $40 component when a $10 component can do 1/10th as much