Calibration Uncertainty in Siren Cosmology: The Floor That Event Count Cannot Lower

This paper asks a single question and computes its answer. Detector amplitude calibration error passes directly into luminosity distance; with a stated calibration uncertainty, what floor does it set on the Hubble constant, and how many events are needed before that floor dominates? Working in a flat LambdaCDM background and carrying the bounce-frame density ceiling explicitly — the Aldred Limit, rho_A = rho_Pl (m/m_Pl)^2/(8 pi C), with the neutron branch at 9.72e+54 g per cubic centimetre — the calculation gives the headline values N2 = 36, N1 = 144, N05 = 576, rhoA = 9.72e+54. At two per cent calibration the floor dominates after only a few dozen events; at half a per cent calibration several hundred events are needed before the floor is reached. Current catalogues are already close to the regime where calibration, not statistics, is the binding constraint on combined siren cosmology. Every number reported here comes from the deposited computation script, which states its assumptions in full. This study is simulated and computed from stated assumptions and published catalogue values; it is not a measurement.