It turned out I didn't have mine as sorted as I thought in post #4. I still had noise issues that I struggled to get sorted. The eventual answer was to solder a resistor into the tach line and connect to the negative side of the coil. The resistor was a 680K Ohm unit, but I don't think that was critical. It just cut the noise spikes and let the Vapor read the signal properly. When I put the resistor in line I was unable to get a signal at all from the positive side of the coil and I had pretty much given up on making it work. My supplier had mentioned that the negative side of the coil is noisier than the positive side, so I thought I would try it as a last shot and then give up if that didn't do it. Worked like a charm, with a solid, steady signal. It has been working perfectly for close to 2000km since. At this point I will say I have it truly sorted and will likely ditch my OEM instrument cluster over winter and use the Vapor exclusively from here on.
If I ever use a Vapor on another bike I will head straight to the resistor lead and the negative side of the coil instead of wasting a bunch of time with half measures.
Mark
The input impedance of the Vapor-trail is probably pretty high as it is an input. Say 10-100K ohms. So when you put a resistor in series with that it has the effect of scaling the coil voltage by what is know as a voltage divider formed by the input impedance and the series resistor. If the input impedance is 100K ohms, then the voltages will be scaled by 100/(100+680)=100/780 or 0.12 . That is close to a reduction of 10:1 in the input voltage spikes.
I don't know why this is not included in to teh Vaportrail itself, unless they are trying to use a single input as either a high tension coil wire input of a primary wire input. The circuit I provided will limit the current and voltage to the range of -0.5<V<5.0 and should make the Vapor-trail happy as well.
Just a guess, since the vapor-trail has a sensitivity that requires this input conditioning, I would have a concern about damaging it with inductive spikes from the ignition. The circuit I provided would prevent overloading of the input signal which I suspect is the issue in the first place.