Improving grip through data analysis

Hello KP members, how are you?

A year and a half ago I started analyzing GPS_LatAcc values to see whether adjustments being made to the kart´s setup were resulting in increased or decreased lateral g-forces. This helped me to confirm the perception of an increase or decrease in grip, mainly in the rear end.
But I’ve always noticed that the GPS_LatAcc values in the right turns on a clockwise circuit are always greater than the GPS_LatAcc values in the left turns and that always intrigued me.
As the track where I run is sometimes open for counterclockwise practices, I decided to compare the values of GPS_LatAcc and I noticed that even in the counterclockwise direction, the values of GPS_LatAcc are greater in turns to the right than in turns to the left. However, in the counterclockwise direction, the difference between the values for the right and for the left is smaller.
For example, in the clockwise, the average of GPS_LatAcc in the turns to the right is 2.1g while in the turns to the left the average is 1.8g. In the other hand, in the counterclockwise direction, the average of the values of GPS_LatAcc to the right drops to 2.0g and the average in the curves to the left rises to 1.9g.

I thought that the difference between the average of turns to the right and to the left was due to the temperature of the tires, which are higher in the outer tires, but this was not confirmed in the counterclockwise direction.
Can anyone tell me why this happens? Any relation to the weight distribution as the engine I use is really heavy (30kg)?
Additionally, I would like to ask if you guys use any other data to analyze grip.

Important information:
I run in the senior category with 4 stroke engines and 183kg (kart + driver).

I have a low confidence level in the GPS derived forces given the position accuracy is around two feet.

Open to correction, not I’m not sure how much utility there is in that data. I bought a race box mini which has a gyro in it, I think that would be a better way to go for good force based data.


I agree with James, we are talking small margins on honestly not super accurate GPS. How does the apex speed compare when the track is run in the other direction?

There can be a lot of static in the lateral G readings unless you’re on a very smooth track and you’re a very smooth driver.


In the description for the Mtchron5, Aim says:

Different Inertial Platform, at 9 axes. This will help with the development of many useful math channels.

So, they have a three axis accelerometer, the axis gyro and three axis magnetometer.

I don’t know if and how they fuse these units with the GPS readings…

1 Like

21 posts were merged into an existing topic: Mychron Gyro, Force sensors Discussion

Split the topic up since we didn’t really answer Ademir’s question, but that separate discussion was an interesting one…

1 Like

Could it be a difference in the track surface, bumps, camber\banking? It can be the nature of the turns too, you won’t necessarily see the same peak G on all turns

1 Like

I think this has more to do with it than anything. Unless the tack is symmetrical your are going to have different approach speeds and cornering speeds depending on multiple factors like straights before the corners, multi-corner complexes and corner types. At NTK lap times ;change dramatically between directions. I would assume that the lateral G’s would be different as well.

1 Like