As promised, last week Terry got out and did some test riding to see what the BMSK doees when you disconnect an O2 sensor.
Because he rides an R1200GSA which also uses a BMSK, as does the F800, but with two O2 sensors, he was able to directly compare the fueling delivered from side by side cylinders operating under the same conditions. And because he has dual LC-1s he was able to log detailed AFR information in realtime.
In my Wideband O2 thread in the Boxer forum I'm going to post a lot of additional detail. But here is a chart that shows the AFR differences between his left and right cylinders, one running Closed Loop (purple) with an O2 sensor connected, the other running Open Loop (black).
At the beginning of the chart on the left side the motorcycle is idling. Looking at the Black Line #2 with the disconnected O2 sensor, it appears that it is running steady (and after a BMSK reset) but a lot richer. However, as he drives off you can see that the black line fluctuates about 12-15% while that #1 cylinder with its O2 connected holds tightly to the target AFR and varies +/-3% (6% total).
(You can see the black and purple cylinders start to converge as the BMSK with twin O2 sensors pulls off a clever feat: it uses data from the connected cylinder to bring the disconnected cylinder into a partial alignment. When Terry and I first saw this, it was a big, WOW!)
The large fluctuation of the black line is consistent with my measurements on the R1150 in Open Loop and should be representative of the fueling data being sent to the Power Commander connected to the BMSK of the F800. I can't think of a reason why it would be different.
The thing I wondered since there appears to be multiple AFRs for any given RPM on the disconnected cylinder (the fluctuation being seen) how does the PC V figure out which AFR it puts into the Trim table? It seems likely that this varies depending on the exact moment the data is taken.