
ID	0364
Class	KNOWN-LIMITATION
Severity	1
Status	ready

Models: NetStructure, Sphere, Disk

Found: 2026-10-12, feature wake-docs (plans/wake-docs-cleanup.md in fhsim_workspace), owner ruling of 2026-10-11

Evidence

doc/adr/0006-lumped-external-wake-per-object.md:61-62 (accepted text): "A NetStructure's own attached spheres and disks can sample the private field." Amended 2026-10-12: they cannot.
fhsim_environment include/fhsim_environment/BodyWake.h:176: BodyWake::Query() returns the object's own query only; a Sphere or Disk queries with its own BodyWake, which leaves out its own lumped field and composes no other object's private field.
fhsim_environment include/fhsim_environment/WakeCaster.h:271: the private field is reachable only through detail::WakeCaster::PrivateField(), which no SimObject other than the net uses.

Effect

A sphere or disk attached to a net (sinkers, floats inside or on a cage) samples the net's lumped field like any other object. The lumped wake starts at the net's rear face (ADR 0006, QF8), so a body inside the net volume sees no wake of the net's front panels.

Possible fix

Let a group of SimObjects share one wake: the net's attached bodies would then sample the group's private field. Wake groups (plans/wake-groups-simobject.md in fhsim_workspace) are the possible route. No implementation now (owner ruling 2026-10-11).

Test that would prove it

The limitation itself is pinned by tests/NetWake_Test.cpp NetWake.BodyInsideTheNetSeesNoWakeOfItsFrontPanels_MARE0364: a receiving BodyWake between two screens of a net sees the ambient current exactly where the net's private field has a deficit above 0.02, and behind the rear screen exactly the lumped field. A fix would change that test. A sphere at the centre of a cage, current through the cage: its sampled flow is below the ambient current by the private-field deficit of the front panels.

Risk

Changes results for any model with bodies inside a casting net.