I was also surprised by the irregularity of the secondary O2's - I need to replace them since they are original, but I think they are working properly. Since I never took a baseline measurement of my X5 / didn't really know what working O2 signals look like, I also grabbed the same data (using the same tool - TorquePro) from my wife's CX-9. The CX-9 run is much longer - so the cat's likely got much warmer. I was expecting a smooth line, what I got instead was something that looked a lot like my X5 – but it seems to then smooth out a bit – still with some good spikes.