pub const EXHAUST_VELOCITY_RANGE_M_S: RangeInclusive<f64>;Expand description
The effective exhaust velocity c = I/m_p a solid motor’s curve and propellant mass have to
imply, m/s.
Measured over the 1,708 ThrustCurve.org simulator files with both a parsed impulse and a
propellant mass in their motor’s ThrustCurve.org record, c runs 236 to 3,031 m/s, with a
median of 1,866 and 90% of them between 928 and 2,210 (the full table is in
docs/research/exhaust-velocity-guard.md). The record’s mass is not the one the bundled
catalog flies: since issue #295 each motor’s index masses are its default file’s header, and
the record’s is kept only for comparison (crate::CatalogMotor::stated). The bulk is APCP;
most of the tail below about 900 m/s is black powder (64 of its 83 files).
The bound rejects none of those 1,708. It is not a filter on propellant: it is there to
catch a mass-unit slip, grams given as kilograms, which moves c by a factor of 1,000: the
worked example in SolidMotor::from_envelope’s test lands at 1.8 m/s. c doesn’t depend on
the motor’s size, so sizes given in millimeters pass it. The headroom is real but not great at
either end. Over the record-mass survey both ends are data errors: the lowest, 1.18x above
the floor, is a Cesaroni E75 record giving 104 g of propellant for a 52 g motor, and the
highest, 1.65x below the ceiling, a misfiled curve. The lowest entry NAR’s list backs, the
Quest B4, is 1.85x above the floor. On the header masses the catalog flies, the highest is the
Ellis J228 at 4,128.18 m/s, 1.21x below the ceiling. A small motor whose recorded propellant
mass is too large for its impulse could fall through the floor; issue #11 records the
fallback, which is to apply the bound only above a couple of grams.