= Field(
default=1.0,
ge=0,
le=1,
description=(
"Multiplier applied to the injector Cd each time the spray constraints are "
"violated. 1.0 = OFF, which is the default and the physical answer. "
"WAS 0.95, and that is not a discharge coefficient -- a drilled orifice does "
"not flow less because its spray is long. Measured on the 8 kN ethalox point: "
"the loop shrank Cd from 0.6017 to 0.4658 (0.6055 * 0.95^5, exactly), a 23% "
"error that propagates straight into orifice sizing, and at a realistic Cd the "
"same hardware then makes +11.9% thrust with dP/Pc falling to 0.164, under the "
"0.20 chug floor. Worse, the feedback has the WRONG SIGN: lower Cd means lower "
"jet velocity, larger SMD and a LONGER evaporation length, so x* went from "
"0.1256 m to 0.1338 m while the limit it was chasing is 0.05 m. It never "
"converged -- it just exhausted max_iterations. An x* violation is an "
"infeasibility to report (it is in diagnostics as x_star), not something to "
"fudge the discharge coefficient for. Set below 1.0 only to reproduce a legacy run."
),
)
1.15.0