|
lob 0.13.0
Exterior ballistics library — API + technical reference
|
|
lob 0.13.0
Exterior ballistics library — API + technical reference
|
Wind and the Earth's rotation are the two corrections lob applies that act on the projectile's velocity rather than on the geometry of its path. Neither one changes the drag curve or the atmosphere. Both are resolved once when the solver context is built and then applied at every integration step, which is why they are cheap and why they are grouped together here.
Wind enters the equations in only two places, and understanding both of them explains most of what wind does to a solution.
Drag acts on airspeed, not groundspeed. The trajectory computes its drag from the projectile's velocity relative to the moving air. That single fact accounts for the entire effect of a headwind or tailwind: a headwind increases the relative speed the bullet sees and therefore the drag it suffers, and a tailwind decreases both. Neither is a force pushed on the bullet; both fall out of computing drag against the air rather than the ground.
Crosswind affects drag through relative airspeed, and spin adds jump. Lateral wind affects the trajectory through relative-air drag even without spin corrections (drag acts on velocity relative to the moving air). Spin effects add aerodynamic jump pitched at the muzzle, which is covered in Spin Drift and Aerodynamic Jump. That correction reads only the lateral component of the wind at the muzzle, and deliberately so: the lateral direction is perpendicular to the plane of fire, so tilting the shot for an inclined range does not disturb it.
The heading convention is the one shooters already use on a clock face. Twelve o'clock is a tailwind, six o'clock a headwind, three o'clock a pure crosswind from the right.
A single wind vector is the common case and is the default, but wind does not hold still over a long trajectory, and lob can represent that.
A profile is a list of wind points along the trajectory, each with a magnitude, a heading, and the height at which it was measured. A weather reading taken at ground level does not describe the air a long-range shot crosses at altitude, so each point's height matters.
Heights are normalized once, when the context is built, against a fixed reference height using the standard power-law relation for wind speed versus height over open terrain. Building the profile up front means the per-step query stays a simple interpolation rather than a computation.
At solve time the profile does two things. It interpolates between the points by down-range position, so a wind that changes along the shot changes where the bullet is. And it scales with altitude according to a shear exponent, which models wind increasing with height. The exponent defaults to no shear, so a caller who supplies a profile and nothing else gets a constant wind that varies by range only; shear is something you opt into.
Inclined fire has one subtlety. Height is measured against the plane containing the trajectory, not against flat ground, so a downhill shot is not quietly reading a different part of the wind profile than an uphill one.
A profile with a single point is exactly equivalent to supplying a uniform wind, down to the last bit, so the simple case is a genuine special case of the general one and not a separate code path with its own rounding.
The Earth's rotation deflects a moving body. Over the length of a trajectory the deflection is tiny, but it is systematic: a long shot aimed north drifts consistently in one direction, and lob accounts for it.
The correction is standard, derived from working in a rotating reference frame. What lob does with it is unremarkable and deliberately so. The shooter's latitude and azimuth determine three scalars that couple the three velocity components to one another. Those scalars are computed once when the context is built, and each step adds the corresponding acceleration to the velocity, after drag and before gravity.
The correction requires both a latitude and an azimuth. Supplying only one is not an error, it simply produces no correction, because the effect is a function of the pair. A shooter in Wisconsin who gives no azimuth gets no Coriolis, which is the correct answer to the question they did not ask.
lob does not model the related effect of a plumb line tilting away from true vertical with altitude. Where that would matter, its effect is folded into the reported deflection and elevation rather than being a separate term in the trajectory.
1.13.2 