0-60 of ct dual fs vs model x dual

[scottf200]

For true torque vectoring you need steering angle as an input - for example if you have the squircle cut to the left, the right/outside motor should be turning slightly faster than the left/inside motor.

So as you are implementing your torque vectoring, you might be able to do it right at the motor controllers, but you still need to know what the commanded steering wheel angle is. You can almost certainly poll the current steering angle off the ethernet ring at a much lower frequency than what you need from the the VSS' (Vehicle Speed Sensors) since you won't be turning the steering wheel that quickly, but you definitely need the steering angle as an input to the calculation of what speed to command at each wheel.

Interesting and good points. That is great input for torque vectoring and "cornering", however, I think we were mainly talking about traction control in terms of a straight line (ala drag race track or off-road taking a "line").

Sponsored

 

[JBee]

Because there are independent motors in the rear on the AWD Tri-beast did/can Tesla use 'regen braking' instead of the physical brakes for the logical "traction control"?

I also didn't know if the type of motor played a factor in any of this?
[oddly in contrast to the Tri-beast, the S/X plaid use permanent on the two rears per the manual]

PM and induction motors will have different torque curves, max rpm, and efficiency curves. So each of those will have optimised gearing. The PM is normally optimised for range at cruise speeds.

In fast acceleration the levels of traction change with velocity, where on start you have more traction in the rear at lower speed and this then moves forward the faster you go. So to get a good launch you might want to mix the gearing up accordingly, but in the CT a high top speed is not necessary, so in general the gearing ratio would be higher overall, giving more torque at low speeds. In most high performance versions, Teslas are traction limited, not power limited, so keeping the gearing ratio high means fast 0-60 and good towing and off road torque. You xan see how the CB runs out of speed in the Carwow Lambo test.

So because of this higher ratio gearing the motors will have quite a bit of control authority over torque, also because they are doing a 15 or so turns before a wheel does one. That is also good for regen as the motor will have more braking power at low wheel rpm.

So technically yes, regen could be used for traction control, but it is different between the AWD and the CB, so we have to differentiate the two here.

In the AWD, with an open diff on launch, the motor torque will be modulated according to the ABS wheel rpm sensors, if they register slip, it will decrease motor output until the next control loop cycle, and if it is slip on one side, the traction control will modulate in some brake force on the wheel that is too fast. There would be an ongoing control loop that balances motor torque, individual wheel slip, and brake traction control which as traction increases with speed, less brake TC is needed and more torque and power can be applied.

But in the CB (or any other DM rear) only the front axle does this, in the rear the two motors simply regulate torque in the motor controller directly, for each wheel individually. No brake TC or regen is required, the motor just needs to know if it's slipping in comparison to the other wheels, and modulates torque accordingly, which is way faster, and most importantly the wheel that has traction does not need to reduce acceleration at all, unlike the open diff AWD using TC. At low speeds in a straight line, a locked rear diff on the AWD will help get it off the line better as well, but you would want to turn it off straight after launch to maintain steering.

Both of these obviously require steering angle input for directional control, but this is where things change at speed. If you hold a steering angle, say on a drift pad, the AWD will struggle to maintain even torque on each wheel with different wheel rpms. There's two reasons for this, 1) is that the hydraulic brakes themselves can't respond fast enough to compensate for slip in real time and have a latency that is slower than the changes in traction 2) any brake event also changes the torque on the other wheel through the mechanics of the differential. Which in return results in the motor cutting torque as it only has one wheel of traction. In which time the slipping wheel gets traction again, the brakes releases, the torque increases and so on. Each time this happens with TC in alternating traction situations the latency builds up, meaning that full available torque cannot be applied over the shortest amount of time, so acceleration is reduced.

This is also the same in reverse, in that ABS brakes nearly always increase braking distance on a dry high traction road in comparison to a locked wheel. But ABS let's you steer, and is sometimes better in the wet or low and changing traction, where a driver is physically incapable of modulating the brakes fast enough. Further ESP/VSC uses the ABS and steering angle to do a "brake" version of torque vectoring for directional control, in that it simply brakes to the side it needs to to get the vehicle under control. Some VSC, even on ICE also mix in throttle, so the outside wheels can be accelerating whilst the inside decelerates.

Note however, in neither situation does regen come into play for acceleration in either the AWD or CB. In the CB rear during a drift, you could technically regen one side to reduce rpm and increase the other side, but that would mean that you had a wildly spinning wheel, that was not spinning at the same average as the other three.

You can of course also have a limited tank turn with the rear motors spinning in opposite directions. But this is hardly useful in driving, except maybe to manage drifting.

Thinking about the CB rear two motor setup some more, in low speed ops and low rpm, there's actually no reason why it couldn't have a Electro-mechanical clutch that locks the two motors together like a diff lock. That would get rid of the TV latency and be the same as a rear locker, but without an actual rear differential. Seems like a simple low cost solution.

 

[REM]

Lol why? Cause it looks cool?

I've had the x for 20k MI and still seems fast enough for me. I had a 2017 model s plaid as a loaner for a bit and sure it was the fastest car I've ever driven, but peeling out from a stop light like a bat out of hell got old after a while. Don't think i need to pay an extra $20k for that party trick.

4680s have much better thermal mass properties.

Sponsored