The "Condensors Question" was finally settled in 2009, after we all did lots of testing. The box will run either way, but we found that since so many of the spark advancers are worn on their inside diameters of the points cam, and/or the springs are/were weak from age, that [re]connecting the condensors stopped an intermittent 'weak spark' situation between 1500-2000 RPM on some bikes. The worst ones were the CB350/360 and the CB500/550, which have weak springs to begin with, and others with hi-mileage springs acted similarly.
What is happening is this: the weak springs or loose points cam "jitter" a little bit when the cam begins the advance off idle. At this stage, the maximum looseness is available to let it wiggle a little, and it physically bounces back a tiny bit from the spring tension of the points being "hit", until the weights pull it to full advance and mechanically lock the cam still again. This bounce is normally displayed in the mechanical points as arcing, so the coils don't discharge all the way (on points alone), making an acceptable spark. But, the MUCH faster transistors DO make the coils fire all the way, but then they jitter closes the points again too soon, snuffing out the spark, then they jitter open again, and another, weaker spark also occurs. This phenomenon is sometimes known as "spark jitter" or "timing jitter".
(Sorry, but that's the only way I know to describe it...).
At any rate, the result is: on these bikes, there was a perceptible stumble just off idle without the Box connected, and it got worse when the Box was added. Connecting the condensors along with the Box solved it at first, but I wasn't happy with just masking the problem. A more complete correction is: strengthen the springs a bit (like, cut off 1/2 turn and bend the ends of the coil out to install them again) because they have become too weak from years of annealing heat cycles. Or, get a less-worn spark advancer, and keep it greased next time? Many of them were not cared for quite this well...
The way to test yours, if you suspect this, is: using a good timing light and a warmed up engine, shine the timing light on the advance marks and rev the engine up until it goes to full advance. Then slowly back it down until it begins to lose that full advance, and note that RPM. Then slowly increase the speed again and see if it reaches full advance at the same speed. If it advances slower than it retards, that's good: it should also reach full advance NO SOONER than 2500 RPM. If yours hits full advance before about 2250 RPM, the springs are just too soft. This causes the intake valve(s) to spit back into the carb throats, which disturbs the flow and causes all sorts of symptoms, ranging from fouled plugs to poor low-speed power.
In general, most of the springs on bikes over 15k miles are too soft now. Once annealed, they have settled into their "new mode" and will stay there, but need to be tightened up a little. If you slow down this advance curve this way, you'll likely find the bike to have more torque when pulling away from a stop, as the slower curve also matches the ethanol burn rate we see in our fuels today. It also won't hurt if you have no-ethanol gas, because modern fuels burn slower than our old ones did.
