4 ms·

If it was below the ground, the car would roll in the other direction.

meheleventyone 5y ago
Gravity acts downwards so having a low center of gravity resists rolling forces. This works in reality as well with vehicles with a high center of gravity being at increased risk of rolling over those with a low center. Games often take this a step further as physics engines let you manually specify where the center of gravity is so you can make it artificially low compared with the actual mass distribution.
- dTal 5y ago
  You haven't addressed the point. Rolling moment is proportional to the distance of the center of gravity from the ground, in either direction. If the center of gravity is exactly at ground level, the car cannot be rolled. Moving it lower, into the ground, increases the rolling tendency again - in the "wrong" direction.
  - meheleventyone 5y ago
    This is not correct because gravity only acts downwards. This means a lower center of gravity will always be more stabilizing for a vehicle the right way up. As the car rolls the center of gravity is pushed outward. When it’s too high it quickly passes over the pivot point of the wheels and the car will roll as gravity is applying additional moment about the wheel. When it’s lower than the pivot point which is not usually physically possible it’s always resisting the roll until the vehicle is basically nearly upside down. Draw it on a piece of paper if you need to convince yourself. This is an incredibly common trick to stabilise rigidbody physics simulated vehicles in games. Edit: I took some time to setup OBS on my non-work machine and make a little video explaining what's going on (apologies for kids in the background and that for some reason my voice only comes out of the left channel): https://youtu.be/qvMbdzbaAMQ https://youtu.be/qvMbdzbaAMQ
    - snovv_crash 5y ago
      In your video you show the force in red being applied somewhere up top. What is this force? What applies it? Your force sideways is actually from the ground, on the tyres. This acts on the CoG (or better to call it Center of Mass), so if this CoM is above the tyres you get a torque, because the mass doesn't instantly accelerate, but resists the force. This torque then needs to be countered by having more force on the tires on one side than on the other, which when you take into account things like suspension, causes body roll, or in the extreme case the car lifting wheels on one side completely off the ground. If you put the center of mass below the car then think what happens not during turning, but during braking. The center of mass will swing forward underneath like a pendulum (think of what happens to something hanging from the rearview mirror during braking) and if you took it really far down you could even get the car to do a wheelie while braking, which is of course ridiculous. If you want the car to be completely flat during turns, acceleration and braking then you need to put the center of mass at the ground level. This way when you turn and the ground puts a horizontal force on the car at the contact points with the road (ie. the tyres), it goes through the center of mass and there is no torque generated. No torque = no body roll.
      - meheleventyone 5y ago
        The torque to roll the car in my example is provided by an external force, the dark brown/red arrow being applied to the top right of the 'car'. This is why the car would still roll even if you had the CoM at the ground level. I'm simplifying the true situation (suspension, roll due to the forces involved in turning and so on) to isolate just the impact of CoM position on body roll. So you need a force to induce roll otherwise the car would just sit still. If you don't believe me you can download Unity, UE, Godot or the rigidbody simulator of your choice and just setup the experiment. Of the engines mentioned UE is perhaps the easiest to set this up with a straightforward car example.
        snovv_crash 5y ago
        Well sure, if an external object hits your car above the CoM then it will roll. I suspect we're talking past each other, I was referring to the handling characteristics, not the response to an impact.

Gravity acts downwards so having a low center of gravity resists rolling forces. This works in reality as well with vehicles with a high center of gravity being at increased risk of rolling over those with a low center. Games often take this a step further as physics engines let you manually specify where the center of gravity is so you can make it artificially low compared with the actual mass distribution.

dTal 5y ago
You haven't addressed the point. Rolling moment is proportional to the distance of the center of gravity from the ground, in either direction. If the center of gravity is exactly at ground level, the car cannot be rolled. Moving it lower, into the ground, increases the rolling tendency again - in the "wrong" direction.
- meheleventyone 5y ago
  This is not correct because gravity only acts downwards. This means a lower center of gravity will always be more stabilizing for a vehicle the right way up. As the car rolls the center of gravity is pushed outward. When it’s too high it quickly passes over the pivot point of the wheels and the car will roll as gravity is applying additional moment about the wheel. When it’s lower than the pivot point which is not usually physically possible it’s always resisting the roll until the vehicle is basically nearly upside down. Draw it on a piece of paper if you need to convince yourself. This is an incredibly common trick to stabilise rigidbody physics simulated vehicles in games. Edit: I took some time to setup OBS on my non-work machine and make a little video explaining what's going on (apologies for kids in the background and that for some reason my voice only comes out of the left channel): https://youtu.be/qvMbdzbaAMQ https://youtu.be/qvMbdzbaAMQ
  - snovv_crash 5y ago
    In your video you show the force in red being applied somewhere up top. What is this force? What applies it? Your force sideways is actually from the ground, on the tyres. This acts on the CoG (or better to call it Center of Mass), so if this CoM is above the tyres you get a torque, because the mass doesn't instantly accelerate, but resists the force. This torque then needs to be countered by having more force on the tires on one side than on the other, which when you take into account things like suspension, causes body roll, or in the extreme case the car lifting wheels on one side completely off the ground. If you put the center of mass below the car then think what happens not during turning, but during braking. The center of mass will swing forward underneath like a pendulum (think of what happens to something hanging from the rearview mirror during braking) and if you took it really far down you could even get the car to do a wheelie while braking, which is of course ridiculous. If you want the car to be completely flat during turns, acceleration and braking then you need to put the center of mass at the ground level. This way when you turn and the ground puts a horizontal force on the car at the contact points with the road (ie. the tyres), it goes through the center of mass and there is no torque generated. No torque = no body roll.
    - meheleventyone 5y ago
      The torque to roll the car in my example is provided by an external force, the dark brown/red arrow being applied to the top right of the 'car'. This is why the car would still roll even if you had the CoM at the ground level. I'm simplifying the true situation (suspension, roll due to the forces involved in turning and so on) to isolate just the impact of CoM position on body roll. So you need a force to induce roll otherwise the car would just sit still. If you don't believe me you can download Unity, UE, Godot or the rigidbody simulator of your choice and just setup the experiment. Of the engines mentioned UE is perhaps the easiest to set this up with a straightforward car example.
      - snovv_crash 5y ago
        Well sure, if an external object hits your car above the CoM then it will roll. I suspect we're talking past each other, I was referring to the handling characteristics, not the response to an impact.