If you are serious about improving engine performance, you need a way to accurately evaluate the modifications you make. Typically, a dynamometer is used to measure rear wheel horsepower. Another way is to take your hotrod to the dragstrip. Neither option is perfect, but they are both better than trusting the seat of your pants. I prefer the dyno because there are less variables to manage. No launch, no reaction time, no tire spinning, no shifts, no wind, etc.
Another option is timed runs. If you measure the time it takes to accelerate over a defined set of parameters, you can use the data to evaluate whether your latest modification works as expected.
I use a camera to record my tachometer readings. Then, I can review the recordings and use a stopwatch to measure the time it takes to accelerate through fixed rpm points. I accelerate in second gear, start timing at 4000 rpm, and stop timing at 7000 rpm. That allows me to accomplish the test run without raising the ire of the local constable. There are a few locations close to my house that are perfect for collecting data under consistent and repeatable conditions.
I can go back to the videos months later to do comparisons of different modifications and/or adjustments. The video recording also captures my Air/Fuel ratio and oil pressure. It works well. I think it’s a great tuning tool, and more convenient than scheduling dyno time (cheaper too).
The camera doesn’t have to record a tach, you could record the speedometer. For instance, on a stock LS650 with a stock rear tire and gearing, 2nd gear at 4000 rpm is about 35 mph, and 2nd gear at 7000 rpm is about 60 mph. So, you measure the time it takes to accelerate from 35 mph to 60 mph. If you always compare the timed runs over those data points, taken under similar conditions, you should get a decent read on how successful your modifications are.
When you do a timed run, you start accelerating well below your first data point (i.e 3000 rpm), and continue to accelerate well past your second data point (i.e. 7500 rpm). That eliminates human error like reaction time, trying to start from a dead stop, wheelies, tire spin, balance, shutting down early, etc.. Since the run is confined to one gear (2nd), there’s no shift involved.
I have found the data I collect using this method to be invaluable. I initially thought that I would only be able to evaluate large changes. I thought that something like a jet change would not be significant enough to detect an improvement. As it turns out, this system will definitively show a change in performance (good or bad) resulting from a Main Jet two-sizes larger or smaller. That’s a pretty good tool. It will definitely help you get it dialed in.
You must overcome the natural tendency to “will” your modification to work. You want it to work so you may find yourself getting an itchy trigger finger. I can always tell when I’m starting to fall into that trap. I take a break for a few minutes and start over.