Improvements for calibrating new Capes

[orgua]

Problems

• the capes get currently calibrated with an expensive keithley SMU
• the device has to warm up for > 30 min to be stable
• the operation itself takes about 10 min

Observations

• we derive gain and offset for each target port and the gain is mostly identical
• the zero-crossing (offset) varies for each target despite having a VLN (very low noise) 10 mV reference

Proposed Changes

• current cape design should not need an SMU for calibration (only for profiling)
• gain and offset could also be measured with a precision resistor on a special target-pcb for calibration (0.01% resistor + 10 uF)
• we can go even further and add an analog switch to include a cal-path on the cape or even parallel to each port (the traces to the port matter & are not quiet symmetrical)
  • this would allow periodic self-calibration to compensate for temperature changes during the year or even the day

References

• visualizations of cals (the folder contains more)
• current cal-tool