Well based on direct prior experience, it is safe to say almost no one outside of OEM firmware programmers knows WTF is going on inside that ECU. Perhaps that even includes Dynojet engineers, I don't know. I was very surprised to find all the additional things going on that relied on an O2 sensor inside the Bosch Motronic for the model year 2002. These are the slow(er) global adaptations to fuel type for example, that use O2 sensor input even in the non-closed loop sections of the map. On the BMW at least you can reset these ECU adaptations, it really makes a difference. Perhaps on our bikes you can disconnect battery to reset ECU and start fresh each time. The adaptations can mask a lot of what is going on.
As far as the O2 sensor, yes it produces a voltage, that is how they work (via chemical reaction). But my understanding is that basic voltage signal needs some signal conditioning and voltage offset so perhaps the proper resistor sends a conditioned constant voltage signal back to the ECU, whereas an open circuit sends something else. Again no one knows but I am beginning to really want to find out.
My subjective impression was that plugging in the simulator changed things. I plan on repeating that test in reverse later this week. I also tried the M419-002 map (DJ 2 brothers O2 disconnected) after swapping 2 Bros cans in. Hated it. Terrible throttle response, jerky, stumbled badly, idled poorly. Top end was good at say above 60% throttle. Then I tried Wally's smoothness/Holeshot map. That was awesome, very smooth, monster power mid-high. I suspect it was running too rich as under load (going up steep hills) at mid-to-WOT it felt like it bogged a tiny bit. Never felt that kind of power out of the old FJR. You could almost see the gas gauge go down however. I also do not have holeshots (still have the cat) and if you do a map compare it really is very rich up mid-high throttle/RPM's where you'd expect a lot more air. Anyway the low-mid range improvement was outstanding, just super smooth. I will try uselss pickles smoothness and basic Wally's smoothness (no holeshot), settle on one map with the 2 Bros pipes, and then try removing the simulator plug to see what that does. Then probably shut it down for winter.
Situation here is somewhat difficult, high altitude (my house is at 8k ft), lots of elevation changes, and getting cold (high 40's low 50's mid day, tens low 20's at night)). Ordinarily this time of year we'd be snowed under, but that is on hold for at least a week or two more. So I can do more science.
There is an option here that I am considering, which is the Innovative Motorsports LC1 wideband but as a data acq tool to look into some of these "WTF is going on issues" on the FJR. This is a wideband sensor that has some electroncis you can get with it and you can use it to record many outputs as well as just AFR. SInce the old sensor is just sitting there doing nothing might as well take it out & get some good data. There is a guy over on the Advrider forum (Roger04rt) who hacked one of those into his Bosch system on a r1150rt, sort of a homemade wideband, and it spoofs the ECU with the wideband output into running whatever AFR target closed loop he wants. As I said earlier the bosch runs mostly closed loop on that engine (no MAF or AFM). I have an older PCIII wideband on my r1150gs that does the same thing (targets AFR but also adjust fuel trim tables RPM vs TPS), the improvement is massive. However the LC1 is mainly an excellent DA tool and maybe someone needs to do that on the FJR to thin out the speculation a little. We'll see if Santa maybe comes through on this. I had never heard of this hardware til I started reading roger's stuff over on Advrider, out of curiosity trying to understand how the PCIII + wideband on my GS was working. He is a pretty smart guy and has really spent huge time on this issue with documentation.
https://www.innovatemotorsports.com/
edit: BTW I also thought the PCIII + wideband must do its closed loop on the PCIII, send a constant signal to the ECU for its O2 input, and run the OEM ECU open loop. Well, it definitely does NOT. It lays its fuel tables open loop on top of the ECU output, and the ECU runs closed loop just as before but off the PCIII's wideband signal, which has a programmable AFR target. This was confirmed by DJ.
