There is another alog from Louis coming soon, but we have been looking at the PUM crossover and see that it isn't well described by the pyDARM model.  Since the crossover measurement has low coherence and can cause locklosses, today we did some investigation using the DARM OLG measurement.  The attached plot shows a DARM OLG measured with our normal A2L decoupling (P2L = 4, Y2L = 4.4) and with half those values (P2L = 2, Y2L = 2.2), you can see that the DARM OLG is different between 20-5 Hz.  

• P2L = 4 Y2L = 4.4: ref 0-2, 11-13, 3-5 (difference sweep amplitudes and frequencies) ref 17, 18, 19
• P2L = 2, Y2L = 2.2, ref 6, 7, 8, 14, 15, 16, and