Gizmo Delay Check

Tried different delay settings and used 'scope to measure how much delay the gizmo introduced.

Coarse Setting

Expect step-size of 17.5 ps. Measure a stepsize of 17.86 +/- 0.01 ps so all seems OK:

Fine Setting

Strong non-linear behaviour seen. Need at least a third-order polynomial to fit it:

Should be able to use coarse settings to get within 17.8 ps of target time, then can use fine setting with approximate step size of 11.1 fs.

History: Delays Before Fix

In original code, all signals from parallel port were opposite of what they should've been (i.e. positive going should have been negative going).
We think this is due to the non-documented use of inverters.
This configuration showed the strange behaviour in the plots below, which was fixed when the signals were inverted.
Red and green lines in the coarse plot were taken at different times.