2024ApJ...966..155S

The evolution of cataclysmic variables (CVs) is driven by period changes ($\dot{P}$), for which the long-venerable consensus is the magnetic braking model (MBM). The MBM has its only distinctive assumption being a power-law “recipe” describing the angular momentum loss (AML) in the binary, producing a single unique evolutionary track with $\dot{P}$ as a function of the orbital period. This required prediction can be used to test the most-fundamental assumption of MBM, but it has never been tested previously. In this paper, I collect $\dot{P}$ measures for 52 CVs of all types. First, 44% of the CVs have positive $\dot{P}$, with such being impossible in MBM. Second, even among the CVs with negative $\dot{P}$, their $\dot{P}$ measures are always more negative than required by MBM, with an average deviation of 110×. Third, three CVs have large chaotic variations in $\dot{P}$ that are impossible for MBM, proving that some unknown mechanism exists and is operating that dominates for these systems. Fourth, the MBM does not account for the long-term effects on evolution arising from the large sudden period decreases seen across many nova events, with this unaccounted effect dominating for the majority of nova systems and changing the sign of the overall evolutionary $\dot{P}$. Fifth, three recurrent novae are observed to suddenly change $\dot{P}$ by an order of magnitude across a nova event, with this being impossible in the MBM. In all, the required MBM $\dot{P}$ predictions all fail for my 52 CVs, usually by orders of magnitude, so the MBM AML-recipe is wrong by orders of magnitude.