Testing
In some steps of tuning the result can be seen immediately on an instrument, or felt. But some tuning can only be verified by getting in the car, dragrace it and note the result. And that has to be done consistently. You can’t do it over 2 days if the weather changes, hot or cold air or even worse rain on the track produce different results. You have to try to make the surroundings consistent. And you have to make sure your driving is consistent. Consistent launches and changing gear at the same rpm every time is critical. What you’re looking for is not an absolute number, but it’s whether the last change you made resulted in a car that car was faster or slower.
So in theory, it’s about doing one change (only one!) and then testing. Never change more than one thing at a time, because you won’t know which change resulted in what. If it’s faster, try more of what you just did, and test again, if it’s slower go the other way.
Example: Setting the timing.
Go to the track, test the car, write down the result. Now the car is warm. Test again, write down the result. This is your base line. Then increase timing to 3 degrees earlier. Test again, write down the result. If it’s faster, do additional 3 degrees earlier. If slower, do 3 degrees later timing than baseline. Test again, write down the result. Continue until you get slower both earlier and later timing. Then set back the timing to the original setting and test, write down the result.
Now you’ll have a set of data points, plug them into a table in the sequence you did them. The reason to do this is that the car might have changed performance simply from the repeated use and heat build up, we want to eliminate that from the results.
Results
We see that the 0 result has improved by 1.5 MPH, that has to be eliminated because it must be attributable to other factors than the one we’re trying to test for. So we extrapolate the 1.5 over time and get to this result:
Results adjusted for outside factors
Then we sort the data by the degrees:
Then we plot it into a graph. If you’re into Excel you can get the program to calculate a polynomial trend line, as seen in graph below.
Based on this, we would set the ignition at 7 degrees early, that’s where the curve would be the highest if we had tested all settings.
This was an example of testing. It takes time, but doesn’t cost too much. And it produces a lot better results than listening to some guy who “knows” that an engine like that need at least -10 degrees to perform. You can set it to -10 degrees, and improve performance, and you’ll think he was right, but you’re losing out on some power. It could be the power that makes you cross the finish line a wheel length in front of him.
Note that we’ve used MPH, not seconds. When you have a track to test on, you would normally go there with a stopwatch. But trap speed is more accurate to get you to best performance. Why? Because getting consistent E.T. is much harder than getting consistent trap speed. If you hesitate at the start, or if you do more or less wheel spin it has a huge effect on your E.T., but when hesitating or when spinning the wheels, you don’t move that much, so you still have the same distance in which to accelerate even though your start may have been bad. Besides, fumbling with a stopwatch while driving can be dangerous, a brief look at the speedometer is quicker.
If this process seems long and tedious, there are a few tricks on how it can be shortened, see section Tuning an ignition, part 1