Over the past couple of years, and increasingly, and super intensively over the last few months, I have been learning to understand what makes our cars tick on a theoretical level, and have been building up an understanding of the link between theory and practice. My ultimate goal is to get away from the industry standard trial and error system of setting up a car. Going to the track, trying a number of different combinations of things, and picking the one that works the best. I believe, that in order to be able to develop and set up a car to be the best it can be, it has to be done in theory, and on the computer. Someone has to know the theory behind the setup, and why every measurement and every component is the way it is. Only then can the reasons for changes in handling be determined, and when on track performance improves, it can then be optimised on the laptop in theory. Then back to the track to see if it improves in the real world too.
Someone has to know the theory behind the setup, and why every measurement and every component is the way it is. Only then can the reasons for changes in handling be determined.
When working purely from a trial and error standpoint, it is actually impossible to optimise the setup. The reason is that the basics of vehicle dynamics, roll centre locations and movement, and camber change amounts in roll and dive are not obvious from just looking at a car. With trial and error, it would be impossible to figure out, that in order to move the setup further in the same direction, which during track tests improve the car, one would need to move front links 1mm this way, rear links 1.5mm that way, adjust camber 0.5 degree this way, 0.75 degrees that way, but on a theoretical level this is possible. Why most (any?) other manufacturers aren’t doing this beats me.
Why most (any?) other manufacturers aren’t doing this beats me.
To prove to myself that I actually have learned something, and am not completely insane, I performed some experiments back home. Because I have been working on 1:8th scales for so long, I have all the measurements in my head, and I can’t look at anything without already having some sort of preconseived ideas, for what works. So I decided to experiment with 1:10th scale. It had been a while since I worked on my 10th scale, and I did not remember 10th scale geometry, and was able to set the car up puerly from a theoretical standpoint.
I got the Schumacher 4wd as it is extremely adjustable, and analysed the geometry on the computer. I started out with the stock setup and made changes I had come up with in theory. I ended up with a much different setup to stock, and it worked. I always believed that vehicle dynamics are vehicle dynamics, and what works in 1:8th scale works in 1:10th scale too. So far it looked like I was right.
Testing some crazy stuff because my computer told me to do so.
However, I wasn’t satisfied with this, because I was driving the car, and as such I could not help but influence my thinking based on what I felt when driving, due to all the years of trial and error setup testing. So I decided to go next level. I would work with a local racer, a youngster who can drive but doesn’t understand setup, and someone with a car I wasn’t familiar with. So I started working with my Junior Team driver Juho Rajaniemi, and his Durango 4wd. I again analysed the geometry, and came up with a setup. This proved to be rather exciting, as it was almost completely different to what all the Durango drivers were running. I couldn’t wait for him to try it.
if the laptimes vary, it’s because of driver error, not because the car did something erratic.
1st test wasn’t good. He didn’t like it. Watching the car it looked better, but something was wrong. I asked to see the car, I checked it over and re-set camber, front toe and rideheight which were a bit off. Juho tried it again and loved it. Camber, and the last few small toe out, and rideheight adjustments made a big difference. The importance of the small things on a good setup is something I need to write a blog about one day. I have been wondering about that for years, and I think I finally figured it out.
Anyway, I was happy, because I took a car I had not driven, created a setup for it in theory, it ended up being very different to the setups drivers were running, and the driver of the car loved it. Now for the ultimate test. Juho’s first race with his new setup.
I took a car I had not driven, created a setup for it in theory, it ended up being very different to the setups drivers were running, and the driver of the car loved it.
As the icing on the cake, against all expectations, he TQed and won his first ever winter national, against the best drivers in Finland. He had shown speed before, but nothing like this, so it definitely looked like I was on the right track.
What does this super long intro have to do with a new WE setup? Well, I came up with the setup completely on the computer, in snowy and cold Finland, and I had to wait a couple of months before I got to try it! Now what I would like all of you to do, is to try this setup, as accurately as you can, and provide me with feedback. What is good, and what is bad, the more precise the feedback is, the more it will help. Then I will go back into “science mode” and figure out what to do next. One day I will post version 2.0 and we try again! This is only the first step. It’s not perfect, but it does work very well indeed!
I will post pictures of the car with the setup here later. Promise!