Tesla Model 3 Dual Motor AWD - 4x4 test on rollers

[Snow Drift]

Its not actually that impressive and I fully expect it to be easily capable of those challenges.


Any Haldex systesm Gen 3 and up (VW etc), Audi Quattro, Subaru symmetrical and several others will be capable of completing all of those tasks. I am guessing that BMW's X drive and MB's 4Matic should also complete the challenge, though maybe with more fanfare than the Haldex, and the two symmetric systems. Not to mention any real 4X4 system, especially with any differential locks.

Haldex is garbage. They are FWD biased with unnecessary turns the system due to a transverse engine layout. They would not do as well in this test since barely any torque goes to the rear. MB uses three open differentials and cannot lock (fail), and BMW X Drive uses a chain to turn a forward shaft (fail). The only systems worth buying are Subaru manual transmissions (Viscous-Coupling Continuous and STI DCCD), and Audi longitudinal (TorSen center...not TT or A3, which is Haldex). Subaru Active (CVT/AT) system is the best clutch-based center system, because it always sends power back 60/40 and has a longitudinal engine layout (Symmetrical). A 4x4 would need to slow down, lock the center differential, and then disengage above 30 mph. The benefit of Subaru DCCD is the ability to have 3 mechanical differentials, including the driver controlled center, which can fluctuate the amount of lock as 59:41 to 50:50, to allow enough slip for high speeds. I'd rather that than needing to lose traction with a RWD based 4x4 until I slowed down to a crawl.

Yeah, It's certainly not impressive at all. It performs worse than is technically possible by a considerable margin. It performs about as well as a high end modern AWD system that's well tuned, but they have a huge advantage with two motors on separate axles.


I assume they don't think it matters all that much, and its biased conservatively to minimize brake wear and maximize efficiency. Reasonable design goals, but i'd imagine they can do a fair bit better. At least they're likely to give everyone an update if they do improve it with time.

The test is very impressive. Outside of Subaru and some Audi, almost every other "AWD" system would fail as most don't transfer torque properly. The vast majority of the offerings are two-wheel drive with a tiny amount of torque going to the secondary axle.

A car with a symmetrical AWD system would just spin all the wheels.

If you are referencing Subaru, that's not what Symmetrical AWD means. Subaru "Symmetrical" means the engine is longitudinal, and due to being a Boxer engine (H - Horizontal) the weight is evenly distributed on both front wheels. Longitudinal engine sends power straight to the center differential/transmission without losing as much energy in 90 deg bends like a transverse engine with a transfer case. The evenly distributed weight over the front wheels aides in traction.

 

[EVDRVN]

I would like to see a snow button that optimizes for those conditions without the rear end pitching like it has in snow tests.

 

[Snow Drift]

I would like to see a snow button that optimizes for those conditions without the rear end pitching like it has in snow tests.

Slip Start allows you to temporarily disengage the throttle reduction and wheel braking to allow wheel slip to break the car free in deep snow or mud. A Snow button like in an ICE actually only reduces throttle input and is the opposite of what you need in this test. A VDC off / Traction control off or Subaru X-Mode, is similar to Slip Start and is helpful in this test...but Tesla does not allow full traction control off.

 

[SammichLover]

I would like to see a snow button that optimizes for those conditions without the rear end pitching like it has in snow tests.

??? Ug, that sounds like it'd be horrible to drive.

Caveat. I've not driven in snow yet, unfortunately. I have hill climbed, at something of an angle to the natural grade, over mud and grass though. I found the sliding around and the rear occasionally moving sideways pretty natural feeling, for what it's worth.

 

[Daniel in SD]

You're asking for 4x4, not AWD.

Because they control the bias via different motors, there isn't that differential in the middle that a single engine vehicle has that creates dynamic evening out so they do it programmatically by not bothering with the front motor at times (for efficiency reasons) until there is a need.

If there is equal traction on all the wheels and you send equal torque then all the wheels will slip at the same time. An open differential splits torque equally.
4x4's usually have no center differential (though my Lexus has a lockable Torsen diff). That gives you equal RPM to the front and the rear. Traditional AWD systems with an open differential give equal torque to front and rear. Tesla's system sends all the power to the rear motor until it senses slip. I understand why Tesla does it the way they do but I think they should add a snow mode.

 

[EVDRVN]

??? Ug, that sounds like it'd be horrible to drive.

Caveat. I've not driven in snow yet, unfortunately. I have hill climbed, at something of an angle to the natural grade, over mud and grass though. I found the sliding around and the rear occasionally moving sideways pretty natural feeling, for what it's worth.

People have said it kicks in and the rear goes sideways for a bit then it straightens and that could be scary on a tight road.

 

[Daniel in SD]

If you are referencing Subaru, that's not what Symmetrical AWD means. Subaru "Symmetrical" means the engine is longitudinal, and due to being a Boxer engine (H - Horizontal) the weight is evenly distributed on both front wheels. Longitudinal engine sends power straight to the center differential/transmission without losing as much energy in 90 deg bends like a transverse engine with a transfer case. The evenly distributed weight over the front wheels aides in traction.

I stand corrected. What I meant was an AWD system with a center differential.

 

[ahfunaki]

Wow, that is awesome. Thank you for the video!

 

[SammichLover]

People have said it kicks in and the rear goes sideways for a bit then it straightens and that could be scary on a tight road.

If your nerves aren't up for a bit of a wiggle now and then you probably should have stayed home. Having no tolerance at all for slip angle is nanny overkill.

 

[SammichLover]

If there is equal traction on all the wheels and you send equal torque then all the wheels will slip at the same time. An open differential splits torque equally.
4x4's usually have no center differential (though my Lexus has a lockable Torsen diff). That gives you equal RPM to the front and the rear. Traditional AWD systems with an open differential give equal torque to front and rear. Tesla's system sends all the power to the rear motor until it senses slip. I understand why Tesla does it the way they do but I think they should add a snow mode.

Another 10% efficiency penalty in a situation that is likely the most challenging efficiency scenario the vehicle will face seems like a poor idea. Or at the very least a niche idea, which I guess would just put it down the software programming priority list. Maybe we'll see it some day.

 

[Daniel in SD]

If your nerves aren't up for a bit of a wiggle now and then you probably should have stayed home. Having no tolerance at all for slip angle is nanny overkill.

Disagree, my Subaru had zero nannies and handled snow better and I could easily get it sideways if I wanted.

 

[Daniel in SD]

Another 10% efficiency penalty in a situation that is likely the most challenging efficiency scenario the vehicle will face seems like a poor idea. Or at the very least a niche idea, which I guess would just put it down the software programming priority list. Maybe we'll see it some day.

It would probably be way less than that since very little efficiency is lost to the motors in winter conditions. Also inefficiency equals heat which is used to heat the battery in winter conditions.

 

[SammichLover]

It would probably be way less than that since very little efficiency is lost to the motors in winter conditions. Also inefficiency equals heat which is used to heat the battery in winter conditions.

The induction motor is still going to being costing that 10% extra electricity cost, winter summer or fall. True it might not be as bad if they are able to recover significant portions of that though, but I don't know how good the waste heat recovering is on the front motor? If it is mostly into to the other portions of the motor that'll be tougher to capture with the coolant loop (as well not nearly as important to capture, since it'll be better passive cooling serviced in other conditions). Being in front of the pack it could still wash the battery pack with a little heat but I won't count on that for much.

 

[SammichLover]

Disagree, my Subaru had zero nannies and handled snow better and I could easily get it sideways if I wanted.

Did you even read his post? He's talking about that not being a thing. I suspect the reports are also based around that thinking, too. Fear of any lateral movement at all.

 

[ZOMGVTEK]

The test is very impressive. Outside of Subaru and some Audi, almost every other "AWD" system would fail as most don't transfer torque properly. The vast majority of the offerings are two-wheel drive with a tiny amount of torque going to the secondary axle.

Yes, for a AWD vehicle, it performs exceptionally compared to its competition. However, they have such a huge benefit by having two independent drive units. For the hardware they have on board, they are not pushing the limits of whats possible by any stretch. It looks like they didn't put significant effort into it since they don't feel its necessary. Which I would agree with, but it would be cool to see it dominate since its technically possible.

Maybe they should put a bit more effort into the auto wipers first though...

 

[Daniel in SD]

Did you even read his post? He's talking about that not being a thing. I suspect the reports are also based around that thinking, too. Fear of any lateral movement at all.

It's not the lateral movement that's the problem, it's the rotation. The car handles like a RWD car in the snow until the nannies kick in and it's not smooth at all. I would like it to handle like an AWD car in the snow.

 

[Daniel in SD]

The induction motor is still going to being costing that 10% extra electricity cost, winter summer or fall. True it might not be as bad if they are able to recover significant portions of that though, but I don't know how good the waste heat recovering is on the front motor? If it is mostly into to the other portions of the motor that'll be tougher to capture with the coolant loop (as well not nearly as important to capture, since it'll be better passive cooling serviced in other conditions). Being in front of the pack it could still wash the battery pack with a little heat but I won't count on that for much.

There's only one coolant loop so unless the front motor is air cooled it should be able to recover the heat. In winter how much of the energy is going to the wheels? I always thought significant portion was going to heating the battery and cabin. That's why short trips have such poor efficiency in the winter, it has nothing to do with the motors.

 

[SammichLover]

It's not the lateral movement that's the problem, it's the rotation.

That's what I meant, the lateral movement of the rear "kicking out". And it looks like he wants it to never rear its scary head.

Ug.

The car handles like a RWD car in the snow until the nannies kick in and it's not smooth at all. I would like it to handle like an AWD car in the snow.

Already addressed. You probably also want it to get to the intended destination. 10% matters, and that's what running that front drive all the time will cost.

 

[SammichLover]

There's only one coolant loop so unless the front motor is air cooled it should be able to recover the heat.

Not if the coolant isn't running though the drive unit where the efficiency losses, the heat, is centered. More distance between them means a lot of the heat will radiate out instead.

The coolant loop will do a great job on the power electronics but if the extra heat loses are out in the rotors not so much. I'm not sure where that'll be on that motor.

Plus once you've gotten the system warmed up you probably aren't going to need unnatural waste, anyway.

 

[ZOMGVTEK]

I didn't take the 3 through a winter yet. I had a Leaf, and I generally would average around 2 mi/kwh over winter. Some trips would be as high as 2.4 if the conditions were mild, bad weather and deep snow was 0.8-1.0, which is 1000wh/mi...

That's with slow speeds in whiteout conditions, heat cranked, defrosters, A/C and all the stuff blasting. And of course the pedal was to the floor any chance, as you do. But the major issue was the resistance heat, low speeds, very high rolling resistance, and lots of sitting at lights with huge idle losses. So in those conditions, I don't see marginal losses from driving the second motor to be a significant factor. And my commute is 20 miles anyways, so I don't care about a kWh here and there.