Can somebody please explain to me how Tesla AWD works? Thank you.

[JCtx]

Hey gang, tried to find specific technical info on this, and zero luck. I'd like to know how my car works. Since the rear motor is a permanent magnet one (a better motor), and the front an induction one, I assume the car is RWD unless it needs front traction, right??? Or does it send 'x' percent of traction to the front axle all the time (even if not needed), like FWD-based Audis (and other vehicles) do?

And finally, what's the deal with traction on each axle? Do the electric motors act like an open differential (traction on just one wheel), or like a limited-slip differential (traction in both wheels)? And do both front and rear axles behave the same way (both LSD, or both open), or not?

That's what I want to know gang. Or allow me to ask the same question in a more practical way, in case somebody already tried that : If you floor the throttle on a wet or sandy street, would the car leave 2 tire marks (1 front/1 rear), 3 tire marks (2 front/1 rear, or 1 front/2 rear), or 4 tire marks (all 4 tires have traction)? That'd also tell me what I need to know. Thank you.

 

[Sophias_dad]

Both diffs are open, but you can get LSD's for them:

Limited Slip Differential (LSD) for Tesla Model 3 and Tesla Model Y - Unplugged Performance

Its not the electric motor acting like an open differential, its the electric motor driving the differential. IIRC its around a 9:1 reduction between the motor and drive axles.

I can't answer the front to rear question absolutely, but I'm sure at high throttle inputs it won't wait for the rears to start to slip before the fronts get power. This is also adjustable via software, and I suspect Tesla has done some fine tuning over the years. There are some snow-traction videos around from the early days that clearly show the rears spinning, then the fronts spinning, then the rears, and so on.

I'd wager that you'll see four VERY small tire marks, but only because the traction control is so good. I've personally tried an experiment(not intended to be the one you asked for), where the fronts are on good pavement and the rears are on gravelly sand, stomped on the gas and the rears might have left 3" of spin marks. The computer can dial back the power every 54th of a revolution(I think, since its 9:1 and a 6 (effective) pole motor, which on an 82.7" circumference AWD tire is 1.53". There's really no telling how much slip occurred in those rear wheels, but there wasn't a giant pile of sand behind them.

 

[JCtx]

Thanks a lot for the explanation, but can I assume the car works in RWD only while cruising and such? And I figured the car had no LSDs, but it's good to get confirmation of that. The good news is the car apparently has some form or 'torque vectoring', by your comments, where the brakes are applied individually to power the other wheel on the axle if needed. So that's good enough for most people, myself included. No need to make the car even heavier with LSDs. It'd be great to have Torsens, but they're the heaviest LSDs, although they never wear out. Thanks again.

 

[MasterC17]

You are correct in that while cruising power is put down through the more efficient rear motor. So on the highway at 70mph the front motor won't be pulling any power.

The Model 3 has a mechanically open differential, but it can be manipulated with software (torque vectoring, think "e-diff"). The car can also induce oversteer by sending more power to the rear motor, or if it rotates too much send power back to the front motor.

The system in practice works really well depending on what you are trying to accomplish.

 

[GWord]

You can infer a fair amount about their theory of operation from this video test:

 

[DwightKSchrute]

I think this video explains fairly well how Tesla awd works.

 

[The805]

Hi Folks:

I saw this thread, and hopefully one of you who are much more familiar with AWD control systems than me can provide some insight. I have a M3P stealth. I have a fairly steep driveway (~10-11% grade) that has a 70-80 degree turn on the uphill. On occasion, my car momentarily loses traction (I hear the wheel momentarily spin, but not sure which one) as I complete the turn. I never had this occur with any of my previous ICE AWD cars and don't think the Model S that I previously owned did this either. I was also able to repeat the behavior consistently on a friend's uphill driveway by backing up in reverse. However, in that case his driveway was graded, not paved. Is this expected behavior with the AWD system?

 

[MasterC17]

Hi Folks:

I saw this thread, and hopefully one of you who are much more familiar with AWD control systems than me can provide some insight. I have a M3P stealth. I have a fairly steep driveway (~10-11% grade) that has a 70-80 degree turn on the uphill. On occasion, my car momentarily loses traction (I hear the wheel momentarily spin, but not sure which one) as I complete the turn. I never had this occur with any of my previous ICE AWD cars and don't think the Model S that I previously owned did this either. I was also able to repeat the behavior consistently on a friend's uphill driveway by backing up in reverse. However, in that case his driveway was graded, not paved. Is this expected behavior with the AWD system?

There are so many factors there is no way we can know exactly what is happening without being present with the car. My guess, however, is that instant EV torque in combination with the tires on the car is responsible for the traction loss. Regardless, if there were a problem the car would have already told you.

 

[JCtx]

Thank you very much for all the input folks; greatly appreciated. I guess it's better to have open differentials, as not to saddle the car with even more weight, when it's not really needed, as evidenced by the videos above. Thanks again.

 

[pt19713]

Thanks a lot for the explanation, but can I assume the car works in RWD only while cruising and such?

This is from my Model Y but I'd imagine the 3 would be very similar, if not the same.
I had a discussion with another Y owner who told me I was wrong and that the Y was pretty much AWD all the time. I said it wasn't, and that it was primarily RWD unless there's a loss of traction or spirited acceleration.

This is a datalog of a 33 minute drive. Speeds were kept under 65 mph, mostly around 55 mph.
The teal is the torque from the rear motor, the pink is the torque from the front motor. The blue is accelerator percentage. As you can see, the front motors barely kicked in. They activated 15 times, of which 12 were barely a blip. The three times were they provided a small amount of torque was when the accelerator pedal was at around 40-50%. Under most efficient driving conditions, the front motor won't be active unless the throttle is around 20-30% minimum.

 

[designrs]

What about font regen braking percentage with AWD? Is it significantly greater regen? Better tire wear due to regen distribution front / rear?

 

[pt19713]

What about font regen braking percentage with AWD? Is it significantly greater regen? Better tire wear due to regen distribution front / rear?

I'll have to do a specific datalog for this. This drive I posted had brake regen set to low due to the poor road conditions at the start of drive, so it might not be a good indication for what you're looking for.

 

[pt19713]

Here's the graph anyway, with the brake regen to low. I guess it'll be good to compare against standard brake regen.

Similar to the acceleration, the front motors were barely active in regards to regen purposes. Note, there's an oddity at around the 11 minute mark, which is showing an outlier for the front motor regen.

Also interesting to note, the brake regen when set to standard is 78 kWh max when the battery is fully warmed up. It appears max brake regen in low is half that. I had a few dots on the power bar, but I think it's safe to assume max brake regen would've been around 39 kWh but was only able to achieve 36 kWh.

 

[JCtx]

The

This is a datalog of a 33 minute drive.

Thank you. That pretty much confirms our cars work on RWD most of the time, at least when driving normally. It'd be interesting to see the graph with full regen (which is what 2021s do) braking. It might only use the rear motor as well.

 

[pt19713]

The
Thank you. That pretty much confirms our cars work on RWD most of the time, at least when driving normally. It'd be interesting to see the graph with full regen (which is what 2021s do) braking. It might only use the rear motor as well.

I hope to have this graph later tonight. I'm going to test drive a 2021 M3P this afternoon so I'll drive the Y there and datalog it.

 

[MagnusMako]

Another oldie but a goldie shows the AWD system of the 3 compared to that of Audi's. The level of slip control is exceptional and I'm sure has been fine tuned even further in software updates. The Model Y has an "off-road assist" mode that adds even more granularity in torque adjustment for getting out of rugged terrains.

 

[pt19713]

Thank you. That pretty much confirms our cars work on RWD most of the time, at least when driving normally. It'd be interesting to see the graph with full regen (which is what 2021s do) braking. It might only use the rear motor as well.

Here's the graph comparing front and rear motor brake regen in Standard mode. Essentially no real difference under normal driving conditions versus Low brake regen. The only time I could get the front motors to generate any regen was being aggressive with the accelerator and letting off completely. As long as I feathered it, even aggressive, all of the brake regen was from the rear.