Exact numerical inverse of sgToPlato (the ASBC cubic).
platoToSg in units/gravity is a different published approximation, so
the two aren't exact inverses and a round trip through them loses ~0.05 °P.
Attenuation is defined on extract (°P), and predicting FG then reading the
attenuation back off it has to round-trip exactly. So this is a
Newton–Raphson inversion of the same cubic. It converges to ~1e-12 SG in
three iterations.
Exact numerical inverse of sgToPlato (the ASBC cubic).
platoToSginunits/gravityis a different published approximation, so the two aren't exact inverses and a round trip through them loses ~0.05 °P. Attenuation is defined on extract (°P), and predicting FG then reading the attenuation back off it has to round-trip exactly. So this is a Newton–Raphson inversion of the same cubic. It converges to ~1e-12 SG in three iterations.