why do karts pull so many lateral g's?

[Tim.Wright]

The main reason why karts have such good times around tracks and such a high grip is very simple and nobody has mentioned it: karts have a very low CG. This means that the transfer of weight is very low compared with a car. A kart runs at centimeters from the ground.

I wouldn't agree with that at all. The reason a low CG gives better grip is that is reduces load transfer and load transfer reduces the available grip by a few percent due to tyre load sensitivity. Now consider the fact that karts have a geometrically imposed, 100% load transfer on the rear axle and you will see that the all the advantages of a low CG are effectively thrown away by having to lift the inside rear wheel.

I was working on a tyre spreadsheet which I'm using for my own project yeterday. I have a table of vehicle mass and tyre information which I'm using to get a feel for how much tyre (width specifically) is used for a given car mass by all the large manufacturers. I decided to add a Kart and an FSAE entry in there, the results are pretty conclusive:



This is why karts are able to corner so hard. There is simply so much more contact patch area per kg of vehicle mass compared to any road going vehicle.

To be fair, I have only used tyre tread width as a measure of contact patch area so this is nothing more than an indication but a pretty strong one at that.

[Ciro Pabón]

Well, that probably is a very good reason. On the other hand, the true contact patch depends on the normal force more than on the width. As for the few percent, I completely disagree: you can notice when you drive that the rear kart wheel on turning(and both rear wheels on braking, as I said) give you a lot of grip. That's not true with a racing car. If you pull 2 g and assume you are having a .8 m CG height in a car with, let's say, a 2.5 m wheelbase, you get:

Delta W= 2*.8/2.5 = 65%

You're left with a meager 17% of the weight of the car on the rear wheels. Now, you lose almost 30% of the grip you "transfer".

On a kart you have a C.G. of perhaps .2 m and a wheelbase of 1 meter, with:

Delta W=2*.2/1 = 40%

You're left with 30% of the weight of the kart on the rear wheels. This means you have around 30% more grip. That's not a few percent.

[Tim.Wright]

There is something wrong in your calcs Ciro...

Load transfer doesn't change the total amount of load (i.e. Fz_left + Fz_right) on each axle. It only moves it from the inside to the outside wheel. So if you have a car/kart with a 50% mass dist, and excluding aero force, you will always have 50% of the mass on the front axle and 50% on the rear.

[Jersey Tom]

Indeed, sustained 2g's is rare..but possible (on very soft compounds) and seen. Actual transient sometimes goes up to 2,5G's and i'm speaking of well calibrated telemetry.

It can be well-calibrated and still fool whoever is looking at it. Roll and/or yaw can make for pretty good blips in a lateral acceleration trace.

Where else would a 25% boost in cornering capability be coming from.

[tuj]

Thanks all so far, really appreciate all the insights on carts.

Can anyone tell me how they developed that cart with downforce with no wings? Did they just get the bottom sealed with a good fence/perimeter and add a small diffuser?

I have a table of vehicle mass and tyre information which I'm using to get a feel for how much tyre (width specifically) is used for a given car mass by all the large manufacturers.

Thanks Tim, this seems to confirm what I suspected, the large contact patch relative to the vehicle size and weight makes them go so fast.

There is nothing more humbling than setting a really good autox time in a sports car and then watching a 12yo SMASH your time in a 50cc kart.

[Tim.Wright]

Can anyone tell me how they developed that cart with downforce with no wings? Did they just get the bottom sealed with a good fence/perimeter and add a small diffuser?

I don't know for sure, but some categories do run wings. Here I'd believe that 2G could be possible.


On a normal kart, I simply don't believe its possible. Any data acquisitions which don't translate the lateral acceleration reading to the CG are going to have peaks induced by the yaw rate.

Show me a genuine, sustained 2G at the CG of a non aero kart... then we have something to discuss.

[Ogami musashi]

Can anyone tell me how they developed that cart with downforce with no wings? Did they just get the bottom sealed with a good fence/perimeter and add a small diffuser?

I don't know for sure, but some categories do run wings. Here I'd believe that 2G could be possible.
http://upload.wikimedia.org/wikipedia/c ... ilveen.jpg

On a normal kart, I simply don't believe its possible. Any data acquisitions which don't translate the lateral acceleration reading to the CG are going to have peaks induced by the yaw rate.

Show me a genuine, sustained 2G at the CG of a non aero kart... then we have something to discuss.

What you ask here is impossible...even for cars. Telemetry in karts in located on the front floor pan most of the time that is LOWER that the CG...because in a kart the CG is actually influenced a lot by the driver and the total CG of the system is located around driver waist (of course it depends on the size of the driver).

And actually if you look at telemetry from superkarts they rarely reach 2G neither. Actually sprint karts( non aero karts) are as fast as superkarts in most corners, the aero in superkarts is mainly there to compensate for the greater weight.

You are free not to believe this is possible to sustain 2G's in a kart and actually you'll indeed have trouble finding a video of a real sustained cornering in kart for the reason i mentionned earlier (i.e: tyres load in karts is not steady at all) but almost all well placed telemetry onboard videos show average G values between 1,4-2g's (not talking about spikes).

And the tyres engineers (both from michelin and pirelli) i talked to seemed to say it is fairly possible at least in F1.
Considering the very low per unit area load on a kart, i don't see it too far stretched.



But coming back to my original point, i think some of you are misleaded. I didn't say a CAR (so with suspension, stiffer chassis, radial tires etc..) would be faster than a kart around kart track, i said a KART with suspension,stiffer chassis, radial tires etc... would be. That implies similar weight for the kart!

Of course a 1500kg car is slower around an autocross track that a 180 kg kart with about twice the yaw rate, better cornering and favorable power/weight ratio.
Now put a 180kg kart with aero, suspension, radial tires etc..i bet it will be even faster.

[Tim.Wright]

Can anyone tell me how they developed that cart with downforce with no wings? Did they just get the bottom sealed with a good fence/perimeter and add a small diffuser?

I don't know for sure, but some categories do run wings. Here I'd believe that 2G could be possible.
http://upload.wikimedia.org/wikipedia/c ... ilveen.jpg

On a normal kart, I simply don't believe its possible. Any data acquisitions which don't translate the lateral acceleration reading to the CG are going to have peaks induced by the yaw rate.

Show me a genuine, sustained 2G at the CG of a non aero kart... then we have something to discuss.

What you ask here is impossible...even for cars. Telemetry in karts in located on the front floor pan most of the time that is LOWER that the CG...because in a kart the CG is actually influenced a lot by the driver and the total CG of the system is located around driver waist (of course it depends on the size of the driver).

Its not impossible, you use the yawrate and velocity measurements to correct the lateral acceleration so that its back near where the CG is. Given the style of driving I have seen in karts (very agressive turn-in) I'm not surprised that you see massive numbers for latacc when the sensor is located pretty much at the front axle. Its massively contaminated by the yawrate (or more correctly yaw acceleration).

[Tommy Cookers]

Moto GP machines have a lot of tyre contact area relative to their weight
we are told they lean to eg 62 or 63 deg, backcalculated from accelerometer data presumably

some leaning moment is used working against wheel gyroscopic reaction even at the apex
sometimes there is clearly track camber, but not always
some lean is traded against the rider when at or after the apex he moves from his hanging inboard position
but this still looks to the human eye rather like 2g cornering ?
as often it appears to that eye that amount of lean still occurs at the apex, when acceleration in yaw has ceased
and before the rider moves

[Tim.Wright]

I don't know about bikes, but I wouldn't think they would be cornering at 2G. Typically corners are where they lose out to cars. They weigh about the same as a FSAE car but have a much smaller contact patch because they only have 2 wheels.

[tuj]

I may have been wrong about the 2G lateral for karts, that was told to me. Online I'm seeing regular lateral g of 1.5 however.

[Ogami musashi]

Its not impossible, you use the yawrate and velocity measurements to correct the lateral acceleration so that its back near where the CG is. Given the style of driving I have seen in karts (very agressive turn-in) I'm not surprised that you see massive numbers for latacc when the sensor is located pretty much at the front axle. Its massively contaminated by the yawrate (or more correctly yaw acceleration).

The same systems can be used for cars and karts (I.E smarty cam). Appart from the fact kart have more erratic cornering performance there's no indication the same system would be non reliable on kart and reliable on cars because it is the same software and hardware.

Here is one video from super KF. It was back in 2009 the top non gearbox category in kart and was run as a world championship. The track is Lonato and you have hairpins, you can see the kart are around 2g's mid corner:

http://www.youtube.com/watch?v=ai4x1Y7zYtM

http://www.youtube.com/watch?v=BQrnxqEhFwA

And you have to consider this is only the two first laps in the first video and the first lap on the second and in karting tyre warming if of course not allowed.

This is run on super soft tyres.

[Tim.Wright]

Its not impossible, you use the yawrate and velocity measurements to correct the lateral acceleration so that its back near where the CG is. Given the style of driving I have seen in karts (very agressive turn-in) I'm not surprised that you see massive numbers for latacc when the sensor is located pretty much at the front axle. Its massively contaminated by the yawrate (or more correctly yaw acceleration).

The same systems can be used for cars and karts (I.E smarty cam). Appart from the fact kart have more erratic cornering performance there's no indication the same system would be non reliable on kart and reliable on cars because it is the same software and hardware.

Here is one video from super KF. It was back in 2009 the top non gearbox category in kart and was run as a world championship. The track is Lonato and you have hairpins, you can see the kart are around 2g's mid corner:

http://www.youtube.com/watch?v=ai4x1Y7zYtM

http://www.youtube.com/watch?v=BQrnxqEhFwA

And you have to consider this is only the two first laps in the first video and the first lap on the second and in karting tyre warming if of course not allowed.

This is run on super soft tyres.

In both videos, there is an offset of 0.5g to the left when the car is driving in a straight line.

This is the only reason you are seeing sustained 2g cornering inthe left hand corners. To the right, there are only momentary peaks over 1.5g, but its really impossible to tell anything useful from these graphics. Don't you have some traces over time or distance to look at?

It means the accelerometer isn't mounted level. I hope this wasn't the well calibrated telemetry you were speaking of...

[Blanchimont]

Moto GP bikes are close to the 2G during cornering. With the 64 deg angle of the COG of the bike and the rider together from the video below, you can calculate a lateral acceleration of

tan 64° = 2,05 (G).

So the technology to reach friction numbers of 2 is available and surely can be used in karting, too.

[youtube]http://www.youtube.com/watch?v=J73XRDGPcpE[/youtube]

[Ogami musashi]

Its not impossible, you use the yawrate and velocity measurements to correct the lateral acceleration so that its back near where the CG is. Given the style of driving I have seen in karts (very agressive turn-in) I'm not surprised that you see massive numbers for latacc when the sensor is located pretty much at the front axle. Its massively contaminated by the yawrate (or more correctly yaw acceleration).

The same systems can be used for cars and karts (I.E smarty cam). Appart from the fact kart have more erratic cornering performance there's no indication the same system would be non reliable on kart and reliable on cars because it is the same software and hardware.

Here is one video from super KF. It was back in 2009 the top non gearbox category in kart and was run as a world championship. The track is Lonato and you have hairpins, you can see the kart are around 2g's mid corner:

http://www.youtube.com/watch?v=ai4x1Y7zYtM

http://www.youtube.com/watch?v=BQrnxqEhFwA

And you have to consider this is only the two first laps in the first video and the first lap on the second and in karting tyre warming if of course not allowed.

This is run on super soft tyres.

In both videos, there is an offset of 0.5g to the left when the car is driving in a straight line.

This is the only reason you are seeing sustained 2g cornering inthe left hand corners. To the right, there are only momentary peaks over 1.5g, but its really impossible to tell anything useful from these graphics. Don't you have some traces over time or distance to look at?

It means the accelerometer isn't mounted level. I hope this wasn't the well calibrated telemetry you were speaking of...

There's no offest. The straigthline at lonato is actually a very short one till the start finish line, then it angles right before a kink. The karts are actually turning.