Disappointing traction control - is it me?

[mswlogo]

How I think of it is this (and it may be entirely incorrect):

The synchronous motor here is likely very "coggy" to my understanding and wants to not spin if no power is applied. In order to make it spin, you have to be driving it at a very precise speed (synchronous). You can tell the difference in speed between the two powered wheels and see if one is spinning more than it should (implying slipping because of the open differential) but otherwise it's hard to tell if it's slipping or not. It's not just applying torque, it's basically selecting an exact speed (which I imagine is selected in the context of accelerator input and current speed).




Your not-a-cure-all statement is not incorrect, but:

1. If I have AWD, please let me use both the front and back axle so I, the driver, can utilise all available traction to minimise slippage.
2. EVs have instant torque but extremely gentle throttle application (which would be fine in any other vehicle, including older Teslas) shouldn't result in high torque that causes the wheels to slip. The driver isn't asking for high torque with their foot.

Fine and reserved throttle control results in 0% power being applied to the front motor unless the rear is slipping, and that rear slippage could have been prevented if the front was already engaged.

To be fair, you've also explained throttle and traction control in ways that doesn't exist in reality which is why people aren't listening. There is no traction control for regen, but you claim there is. You claim backing off the throttle applies front motor power, but it does the opposite. You claim driving at 60mph (and accelerating there as fast as possible) is a sensible way to drive in winter because it gives you more traction. None of these are true, and why you're seeing resistance to your statements.

I have no doubt that the way you are driving is "fun", but I'm looking for safety and confidence instead.

Correction again.

The rear wheels do not need to slip get power to the front wheels. But you might, unavoidably/unintentionally, encourage them to slip in order to get power to the front wheels by unnecessary acceleration. That can very easily happen on slippery roads. If what you said was true you'd have to burn rubber to get the 0-60 AWD numbers cars get from both motors. And you know that's not the case.

This is partly @StealthP3D argument. That he feels you can instantly engage front wheels (as needed) with the prefect feathering of the throttle enough to get power to the front wheels without the rear wheels slipping. That simply is not true all the time, and not easy to do, nor should you even have to think about it. It's a no brainer to get front wheels going on dry. It's even a no brainer on wet. But it's not so easy on snow/ice and behavior does not work.

We want preemptive power going to all wheels, not wait for acceleration of any kind.

 

[Saghost]

How I think of it is this (and it may be entirely incorrect):

The synchronous motor here is likely very "coggy" to my understanding and wants to not spin if no power is applied. In order to make it spin, you have to be driving it at a very precise speed (synchronous). You can tell the difference in speed between the two powered wheels and see if one is spinning more than it should (implying slipping because of the open differential) but otherwise it's hard to tell if it's slipping or not. It's not just applying torque, it's basically selecting an exact speed (which I imagine is selected in the context of accelerator input and current speed).

There's an encoder on the front of the motor package that gives you the rotor position precisely in real time - necessary so you can set the drive inverter field timing, because you not only have to match the speed of the rotor, you have to be in the correct phase with it to get the power delivery or recovery you want.

But the torque is tied directly to power and current flow through the motor too, so at any given moment the car knows: The speed of all four wheels, the speed and current position of the rotor, and the torque being delivered/absorbed by the motor. Obviously one wheel spinning faster is a dead give away to spinning, but seeing the torque drop should tell it, too. And if it isn't powering the front wheels forward, seeing the rear spinning faster than the front would be an easy sign, too.

 

[StealthP3D]

But again, my goal is not to get around "as fast as possible" in Winter! I want to arrive as confidently and safely as possible. I want no slippage, and if I can't get that, I want minimised slippage.

I'm telling you how to achieve that but you are apparently not listening! I only brought up "maximum speed" driving in order to help you understand how the throttle mapping is tied to the traction control programming. Obviously, I'm not saying you should drive at maximum speed!

Hmm. Your experience is significantly different from most others in this thread. I think most others have had regular AWD instead of Performance though. I wonder if it actually is different somehow?

I really doubt the AWD and P Models are mapped with different strategies for slippery conditions. The power difference between the two wouldn't even come into play on snow and ice. I would assume only in Track Mode would there be a significance difference. Anything else would either be a mistake on Tesla's part or it would require a lot of additional testing and validating which I'm pretty sure they didn't do.

I also have fairly new (this season) Winter tires. They've also been driven on fairly gently because I hate to burn through expensive soft rubber, so I don't think my problem is this "sawtooth" thing you mentioned (I also doubt it would be extremely noticeable to the point of necessitating this thread, but it's the first time I've heard of this phenomenon).

Obviously, I have no idea how hard or soft you drive. I will say this: I think the easily available and prodigious silent torque of the AWD Model 3's hides how much power peeople are applying through their tires. I know this because P3D owners report they quickly become accustomed to all that power and, after a couple of months it doesn't feel that fast anymore. Everything is relative. Remember, the P3D puts Supercar like torque through the tires and the AWD is not very far behind! It's so silent and easy to use one tends to become accustomed to it and think of it as normal.

But I must point out what you are noticing for "very fast-acting traction control" is no more than "very fast updating stats on regen power". Regen does not have traction control. The reduction in regen power is only because less torque is being transmitted to the drivetrain because your tires are sliding. You won't fully lose steering ability thanks to the induction motor up front, which won't fully lock up with regen alone. This is probably the reality behind your statement regarding directional control, and nothing related to intentional backoff of regen strength.

No offense as this is a friendly and sharing conversation, but it's clear you have no idea how regen actually works. It certainly does dial regen back to let the tires roll easier when the tires are sliding excessively. There is no way the car would stay on the road doing what I've done with full regen and no cold battery. I can feel it ease off regen to regain traction. I'm basing this off a lot of "testing" I've done entering icy downhill hairpins with full regen. When it starts sliding it backs off regen to regain traction.

First, letting off the accelerator transfers no power to the front. That's entirely false, verified by various people with the Scan My Tesla app. If anything, backing off actually applies regen in the front while providing forward torque in the back still, which seems odd IMO. The car is also simply not designed to accelerate when you lift off the throttle, because absolutely no car would ever be designed to do the opposite of what the driver requested with their foot.

I think you misunderstood my earlier comment. I will use more words to describe exactly what I'm talking about:

The car is accelerating up a sheet of ice and the driver is requesting far more power with the throttle than the tires could hope to handle while still maintaining traction. In this situation the back end is very loose, the back tires are spinning faster than the vehicle speed (not by a huge amount but enough to make the back end loose). This is the condition I describe the traction nanny as interpreting as "oh, so you want to play, do you?".

As the driver reduces throttle (but doesn't eliminate all throttle input) the back tires hook back up and more torque is applied to the front wheels. This causes the car to accelerate faster than in the earlier "tail loose" condition. More torque is applied to the front wheels as now the traction control is in the "zone" that the tires are micro-managed against slip. This is where you want to be for the most control and the most acceleration. And if you are slipping excessively it's a natural driver reaction to reduce the throttle. Pushing the pedal further makes the car accelerate slower as it transfers more power to the rear wheels and reduces power to the front to maintain directional control in the "tail-happy" condition. I have experienced this literally hundreds of times and it is completely different (better) from any AWD I've ever driven in the slippery stuff.

Second, the graph actually has all the relevance to this discussion. As the OP, I can absolutely make that claim as well because I know what I was bringing up.

No, the graph has the relevant information for the conditions in which the graph was recorded. The rear tires were not traction limited so you will not see what I just described above. I have a lot of experience driving this car and many others in these conditions and I know what my tires are doing.

My whole point is that the Model 3 relies on the rear losing traction before the front engages (or that's how it seemed, and indeed is now confirmed). I don't want to lose traction before I get more, I want to have all the traction available to control the car's vector at all times confidently and safely in a sure-footed manner.

Again, you are not listening to what I'm saying. What you describe is when you apply FAR too much throttle for conditions. The nannies are interpreting it as you wanting to play. If you don't want to play, simply drive with less throttle input, the TC will maintain perfect traction AND it will apply more even torque front to back (because it will detect micro-slippage in the back). This strategy maintains directional control at all times, even if you step far too hard on the accelerator (but, I admit, you do need to have a reflexive counter-steering input if you are going to be asking for that much throttle on very slippery surfaces). You can solve your problem by simply being more aware of low traction surfaces and not requesting grossly excessive amounts of power.

I don't want to get to 60mph as fast as possible on ice on public roads. Heck, I don't ever want to reach 60mph on ice because that's just too fast to be able to control the vehicle if something unexpected comes up. Even if I had studded tires I wouldn't imagine driving that fast in terrible conditions.

I'm not recommending that you do drive 60 mph on icy surfaces. Although there are places that I regularly do this with good sightlines and plenty of room to deal with the unexpected so I know how the car behaves. I was simply describing the cars behaviors under a wider range of driving scenarios than you experience. Studs are not a cure-all, they actually work worse on very hard, cold ice than a good studless winter radial.

Perhaps it was lost that this is a daily driver car for us. It needs to leave town via some curvy parkways and other roads, travel a highway with plenty of corners, turn off up a hill, navigate another section of curvy highways, then navigate a curvy road to get to work. Repeat in reverse to get home. Straight-line performance is not at all my priority. Even in the dry and grippy, I'm not rocketing to my destination as fast and aggressively as physically possible because even that isn't safe with all the dirt and crud on the roads.

It's not lost on me that this car is a daily driver for you and that you want to be as safe as possible. That's exactly why I recommend you focus your attention on your throttle control. Because the fact of the situation is that you are driving a very high-performance car and I don't think you are fully appreciating that. A small press of the throttle is still requesting substantial torque. You need to learn to modulate the initial part of the throttle. Fortunately, the Model 3 has much finer digital throttle control than that on my 2002 Ducati's with EFI. There are only a total of 128 different throttle positions recognized by the Electronic Control Unit. This bothers me to no end when I'm on ice because I can actually feel the throttle graduations. It's not quite fine enough. And, no, I don't purposefully ride it on ice but mountain roads can have patches of ice early in the morning and late in the evening, even in the summer but especially late in the riding season. The Model 3 throttle is very easy to modulate precisely but you might need to break bad habit developed on cars without so much power available. I know from driving on ice my entire life that the position of your right foot on the throttle is important for good throttle control. I suspect you are using a foot position that does not provide adequate throttle feel and repeatability. I have a certain relative that does this. He repeatedly "stabs" the throttle to maintain an (almost steady) speed and doesn't even notice he's doing it. It's a bad habit he developed decades ago on cars without instantaneous throttle response and much less power. Now that he drives a modern Volvo with great throttle response (for a gas car) he still does it. Thankfully, he stays at home when the roads are icy because he has terrible throttle control.

Surely forward motion can also be designed for safety in Winter conditions? Even if it's being a "Snow Mode" option like we've been talking about. Certainly beats everyone having to learn the very specific finesse required to drive a Model 3, which seems to also imply driving at reckless acceleration and speeds which aren't appropriate for icy public roads.

I've said this before on this forum repeatedly - I have no problem with people requesting different throttle mappings from Tesla if they think they could benefit from additional throttle mappings. That's fine. But just because it doesn't work exactly how you think you need it to work is no excuse for misrepresenting how it actually works. The front wheels do receive plenty of torque (contrary to multiple claims here by people who don't understand how it works at all). The graphs they hold up as "proof" that the front wheels are not getting any torque (or very limited torque) are not even applicable to ice driving (because they weren't recorded on slippery ice).

Exactly this. No reason to decide that people that drive in Winter conditions should be slipping 4/12 months of the year as an intentional design.

Sigh...Who said you should be slipping? I'm saying you shouldn't be slipping if you don't want to be slipping. Remember, throttle control. On the other hand, anyone accelerating as fast as possible on compact snow/ice is going to have some micro-slipping going on. And yes, I know you are not interested in accelerating as fast as possible, but that doesn't change the reality of how it works. It's true that a smooth hard tire that can't deform has the most traction with no slippage but we don't drive on tires like that in the real world.

 

[camalaio]

Correction again.

The rear wheels do not need to slip get power to the front wheels. But you might, unavoidably/unintentionally, encourage them to slip in order to get power to the front wheels by unnecessary acceleration. That can very easily happen on slippery roads. If what you said was true you'd have to burn rubber to get the 0-60 AWD numbers cars get from both motors. And you know that's not the case.

This is partly @StealthP3D argument. That he feels you can instantly engage front wheels (as needed) with the prefect feathering of the throttle enough to get power to the front wheels without the rear wheels slipping. That simply is not true all the time, and not easy to do, nor should you even have to think about it. It's a no brainer to get front wheels going on dry. It's even a no brainer on wet. But it's not so easy on snow/ice and behavior does not work.

We want preemptive power going to all wheels, not wait for acceleration of any kind.

Sorry, I feel we're devolving into semantics to the point where I can't communicate clearly anymore :S

I'm 100% agreeing with what you said, but it's tucked in specific words. "Fine and reserved throttle control results in 0% power". Reserved. As in if one is not applying throttle in such a way that the front motor would engage via "goosing it" (because to do so would intuitively result in a loss of traction, so why do this?). I know there's a point at which the front motor engages even if the rear has full traction. What we know (at least, I think we're in agreement on this?) is that this threshold is so high that in icy conditions, the rear end would slip before sufficient throttle is applied in order to engage the front motor. And of course at the time the rear slips, the front motor is now assisting in a mild amount. What we're both agreeing on is that "full-time" AWD is a much better solution than what we currently have.

Please correct me yet again if I am totally missing something.

There's an encoder on the front of the motor package that gives you the rotor position precisely in real time - necessary so you can set the drive inverter field timing, because you not only have to match the speed of the rotor, you have to be in the correct phase with it to get the power delivery or recovery you want.

But the torque is tied directly to power and current flow through the motor two, so at any given moment the car knows: The speed of all four wheel, the speed and current position of the rotor, and the torque being delivered/absorbed by the motor. Obviously one wheel spinning faster is a dead give away to spinning, but seeing the torque drop should tell it, too. And if it isn't powering the front wheels forward, seeing the rear spinning faster than the front would be an easy sign, too.

You're absolutely right of course. I oversimplified the phase/speed/power story.

And fair, seeing torque drop should be an indicator. Even if they don't have this directly, it seems possible they'd have a way to imply it (speed increase vs. accelerator input vs. car angle (for uphill/downhill)). Actually throws me back to the old smart cars... they measured torsion in a shaft in order to tell the amount of torque making it to the road. Kinda cool, but had issues.

 

[StealthP3D]

But the torque is tied directly to power and current flow through the motor too, so at any given moment the car knows: The speed of all four wheels, the speed and current position of the rotor, and the torque being delivered/absorbed by the motor. Obviously one wheel spinning faster is a dead give away to spinning, but seeing the torque drop should tell it, too. And if it isn't powering the front wheels forward, seeing the rear spinning faster than the front would be an easy sign, too.

I was under the impression that each motor drove two wheels (left and right) through an open differential. So how can "power and current flow to the motor" tell the car the speed of the left and right wheels independently? I can't see any way to do this without employing wheel speed sensors on each wheel (or at least one wheel so the motor speed combined with one wheel sensor can be used to derive the speed of the other wheel). I suspect, in order to know the speed of every wheel independently, they simply use a wheel speed sensor on each wheel, no?

 

[mswlogo]

OK, I found a work around for the Model 3 AWD. Unfortunely your efficiency might slip, but car car won't.

 

[mswlogo]

I was under the impression that each motor drove two wheels (left and right) through an open differential. So how can "power and current flow to the motor" tell the car the speed of the left and right wheels independently? I can't see any way to do this without employing wheel speed sensors on each wheel (or at least one wheel so the motor speed combined with one wheel sensor can be used to derive the speed of the other wheel). I suspect, in order to know the speed of every wheel independently, they simply use a wheel speed sensor on each wheel, no?

There are wheel speed sensors on each wheel. Any car with ABS has them.

They control left vs right by pulsing the ABS of the wheel that is spinning faster than the one on the opposite side.

 

[StealthP3D]

There are wheel speed sensors on each wheel. Any car with ABS has them.

I was responding directly to this comment by Saghost:

But the torque is tied directly to power and current flow through the motor too, so at any given moment the car knows: The speed of all four wheels, the speed and current position of the rotor, and the torque being delivered/absorbed by the motor.

It implies the speed of each wheel is known via power and current flow through the motor.

 

[camalaio]

No, the graph has the relevant information for the conditions in which the graph was recorded. The rear tires were not traction limited so you will not see what I just described above. I have a lot of experience driving this car and many others in these conditions and I know what my tires are doing.

Exactly. Under reasonable (not heavy) acceleration, the front motor is not being used and therefore the Model 3 is effectively acting as a RWD vehicle.

Come on man. I confirmed, as the one who started the thread, that the information was correct and relevant to the topic and what we're trying to discuss. You can't just say "No" like I'm not the authority on my own thought process and reasoning.

The nannies are interpreting it as you wanting to play.

I ignored this initially but sorry, this just isn't a thing for the AWD version, and to my understanding just not at all outside of Track Mode. It doesn't let you "play" or have fun. Again, perhaps the Performance is different (but we both doubt this). I can't imagine Tesla would intentionally program their mass-market vehicle to default to fun instead of slow lumbering safety, when most people don't know how to have fun in Winter conditions (because it's terrifying for the vast majority of people, not fun).

If you don't want to play, simply drive with less throttle input

This is going backwards a bit, but please re-read my original post. You will see that I pointed out that moving extraordinarily slowly results in the rear end slipping out. I'm perfectly capable of applying light throttle and so are most people on this forum -- the pedal input for the Model 3 is actually very conservative, predictable, and linear. Getting the power out of it actually requires fairly decent pedal pressure, then it's unleashed. Our Honda Fit is actually more touchy on the first 5-10% of throttle because of the low gearing and aggressive throttle curve (gotta make that econobox feel fast), yet I can drive that on pure ice with much less slip. I've driven a truck with a Hemi and that was significantly more twitchy to the point it was difficult to drive, but still manageable. I know how to moderate the very predictable and gentle throttle curve of the Model 3, especially as its one of the easiest vehicles I've ever driven to do so.

Blaming everyone for being too heavy on the throttle is unproductive and quite apparently not the issue. As owners of Model 3s we are well aware of the power under our right foot. Even friends and family that have been convinced to test drive it manage to do so very gently, because it is actually easy to drive this car gently. If it wasn't, we'd hear the reports constantly and Tesla would likely be forced to do something about it. You're right that all that's required to drive gently is to press gently, and that is what I am doing despite the strange behaviour of the car.

I've said this before on this forum repeatedly - I have no problem with people requesting different throttle mappings from Tesla if they think they could benefit from additional throttle mappings.

We're not asking for a different throttle mapping, we're asking for more (full-time) usage of the front motor. We have a different throttle mapping already - Chill Mode, which I am currently using. That doesn't change the fact the front motor is not being used full time.

The front wheels do receive plenty of torque (contrary to multiple claims here by people who don't understand how it works at all. The graphs they hold up as "proof" that the front wheels are not getting any torque (or very limited torque) are not even applicable to ice driving (because they weren't recorded on slippery ice).

They are entirely applicable until the rear loses traction and engages the front a small amount (which as we can see by Bjorn's video and earlier in this thread, it is indeed a small amount). Again, the point is that the system is waiting until it either receives a larger throttle input (unproductive and likely to cause loss of traction) or senses the rear slipping (which it detects poorly as in my original post).

The goal and what we're asking for, again, is full-time AWD to assist with the vehicle vector while minimising the possibility for slippage. Many like me bought the AWD assuming it would be a good AWD system for snow, not effectively RWD when driving cautiously.

OK, I found a work around for the Model 3 AWD. Unfortunely your efficiency might slip, but car car won't.

SOLD, where can I throw my money?! Unfortunately it doesn't snow enough here for that, just ice -.-

It's true that a smooth hard tire that can't deform has the most traction with no slippage but we don't drive on tires like that in the real world.

Ok. So this is the flex that drag cars (your point) take advantage of in order to get more launch traction:

https://static.carthrottle.com/workspace/uploads/posts/2015/04/zpujfny-554030b9bda29.jpg

Are you seriously suggesting that winter tires on our Model 3 are doing something equivalent when driving gently? I'm not saying we are driving around on hard metal drums, but it is absolutely not true for passenger vehicles in common winter conditions that slippage is beneficial. And again, if it were true, it would only be applicable to straight-line launch scenarios and not all-around grip for cornering etc.

 

[camalaio]

I was responding directly to this comment by Saghost:

It implies the speed of each wheel is known via power and current flow through the motor.

To clarify, they were saying that the motor power is known, and when combined with other info they also know the motor rotor positions (sensors) and the individual wheel speeds (their own sensors).

 

[StealthP3D]

Exactly. Under reasonable (not heavy) acceleration, the front motor is not being used and therefore the Model 3 is effectively acting as a RWD vehicle.

Come on man. I confirmed, as the one who started the thread, that the information was correct and relevant to the topic and what we're trying to discuss. You can't just say "No" like I'm not the authority on my own thought process and reasoning.



I ignored this initially but sorry, this just isn't a thing for the AWD version, and to my understanding just not at all outside of Track Mode. It doesn't let you "play" or have fun. Again, perhaps the Performance is different (but we both doubt this). I can't imagine Tesla would intentionally program their mass-market vehicle to default to fun instead of slow lumbering safety, when most people don't know how to have fun in Winter conditions (because it's terrifying for the vast majority of people, not fun).



This is going backwards a bit, but please re-read my original post. You will see that I pointed out that moving extraordinarily slowly results in the rear end slipping out. I'm perfectly capable of applying light throttle and so are most people on this forum -- the pedal input for the Model 3 is actually very conservative, predictable, and linear. Getting the power out of it actually requires fairly decent pedal pressure, then it's unleashed. Our Honda Fit is actually more touchy on the first 5-10% of throttle because of the low gearing and aggressive throttle curve (gotta make that econobox feel fast), yet I can drive that on pure ice with much less slip. I've driven a truck with a Hemi and that was significantly more twitchy to the point it was difficult to drive, but still manageable. I know how to moderate the very predictable and gentle throttle curve of the Model 3, especially as its one of the easiest vehicles I've ever driven to do so.

Blaming everyone for being too heavy on the throttle is unproductive and quite apparently not the issue. As owners of Model 3s we are well aware of the power under our right foot. Even friends and family that have been convinced to test drive it manage to do so very gently, because it is actually easy to drive this car gently. If it wasn't, we'd hear the reports constantly and Tesla would likely be forced to do something about it. You're right that all that's required to drive gently is to press gently, and that is what I am doing despite the strange behaviour of the car.



We're not asking for a different throttle mapping, we're asking for more (full-time) usage of the front motor. We have a different throttle mapping already - Chill Mode, which I am currently using. That doesn't change the fact the front motor is not being used full time.



They are entirely applicable until the rear loses traction and engages the front a small amount (which as we can see by Bjorn's video and earlier in this thread, it is indeed a small amount). Again, the point is that the system is waiting until it either receives a larger throttle input (unproductive and likely to cause loss of traction) or senses the rear slipping (which it detects poorly as in my original post).

The goal and what we're asking for, again, is full-time AWD to assist with the vehicle vector while minimising the possibility for slippage. Many like me bought the AWD assuming it would be a good AWD system for snow, not effectively RWD when driving cautiously.



SOLD, where can I throw my money?! Unfortunately it doesn't snow enough here for that, just ice -.-



Ok. So this is the flex that drag cars (your point) take advantage of in order to get more launch traction:

https://static.carthrottle.com/workspace/uploads/posts/2015/04/zpujfny-554030b9bda29.jpg

Are you seriously suggesting that winter tires on our Model 3 are doing something equivalent when driving gently? I'm not saying we are driving around on hard metal drums, but it is absolutely not true for passenger vehicles in common winter conditions that slippage is beneficial. And again, if it were true, it would only be applicable to straight-line launch scenarios and not all-around grip for cornering etc.

It makes me think your AWD might work differently from my P when you claim that the front wheels get no torque under "reasonable", not heavy, acceleration. Because I've been on a 6% grade stopped in heavy wet snow a bit deeper than the underfairing and gently pushed the accelerator (thinking "no way is this going to work, the snow is too deep and too saturated with water") and the car simply climbed straight up the hill without obvious slippage and zero drama. It just slowly tractored right up.

It's not like my winter tires are made for deep snow (they are Pirelli Sottozero II's). No way did it plow up that hill with rear wheels only!

To address your last point re: tire micro-slipping for maximum tractive power. I'm not talking about sidewall flex for maximum traction. It has to do with a slight scraping action of the tread on the driving surface. On polished ice or wet ice this doesn't work. But frozen ice is made of ice crystals and so is snow. When the tire micro slips it can pack the crystals together for more traction. If the tire slips too much it polishes the traction surface which is bad for traction and acceleration. But a little micro-slippage is good. The same thing applies to dry pavement because it is irregular. It takes energy to deform the tread surface around the irregularities. This manifests as a little more traction if it can be managed so as to not fully break loose.

 

[mswlogo]

It makes me think your AWD might work differently from my P when you claim that the front wheels get no torque under "reasonable", not heavy, acceleration. Because I've been on a 6% grade stopped in heavy wet snow a bit deeper than the underfairing and gently pushed the accelerator (thinking "no way is this going to work, the snow is too deep and too saturated with water") and the car simply climbed straight up the hill without obvious slippage and zero drama. It just slowly tractored right up.

It's not like my winter tires are made for deep snow (they are Pirelli Sottozero II's). No way did it plow up that hill with rear wheels only!

To address your last point re: tire micro-slipping for maximum tractive power. I'm not talking about sidewall flex for maximum traction. It has to do with a slight scraping action of the tread on the driving surface. On polished ice or wet ice this doesn't work. But frozen ice is made of ice crystals and so is snow. When the tire micro slips it can pack the crystals together for more traction. If the tire slips too much it polishes the traction surface which is bad for traction and acceleration. But a little micro-slippage is good. The same thing applies to dry pavement because it is irregular. It takes energy to deform the tread surface around the irregularities. This manifests as a little more traction if it can be managed so as to not fully break loose.

It's not your P. I had the same car you have. P behaves exactly the same as he described. I was also mostly in Chill mode.
But I should have experimented with Track Mode (just for the fun of it) in the snow.

And just because you CAN plow through snow with all wheels going (by punching it, and it will go reasonable straight) . Doesn't make it a great AWD system. Try going up a slippery hill when you're stuck behind traffic and can't accelerate enough to engage the front motor. You'll be a RWD car. And, no you don't want to wait until rear wheels slip to engage the front. You can't do a steady "crawl" up a slick hill with the Model 3.

Try a Model 3 AWD on a Snow/Ice test track. Then try a Model X. Both with similar tires. With the same driver. Not some pro.
I think you'd see a big difference. And the Model X could be better too. It's biased to use the front motor (which helps a lot) but it should have the same option to engage all motors all the time.

 

[tandemnw]

I entirely get we all took physics 101 and learned about static and kinetic coefficients of friction.

I’m sorry StealthP3D only has “life experience” evidence of why microslips enhance traction.

Instead of writing off that life experience based on the spherical cows of physics 101, let’s actually try to assume that StealthP3D might be on to something here and research a bit more.

In real life, traction on ice is not a 2D interaction. It’s a 3D interaction across the substrate, a thin water film above the substrate, and the tires ability to clear out that water to hit the substrate. Have you noticed all winter tires are aggressively siped? That slippage is using the siping to clear away the water film.

Try driving on racing slicks in the rain and you’ll get my drift. I’ll take the kinetic coefficient of rubber on road with water removed, over the static coefficient of rubber on water, any day.

In reality, this is just scratching the surface. Friction of tires on ice is an area of active research involving materials science and fluid dynamics. This is well beyond physics 101, we’re just starting to model and understand those life experiences observations in the field.

Fun reading for the bored.

https://aip.scitation.org/doi/10.1063/1.5041049

http://www.iaeng.org/publication/WCE2011/WCE2011_pp2381-2384.pdf

 

[Magnets!]

Not sure how to reconcile the many snow videos on Youtube and this post. Here is another.

 

[StealthP3D]

It's not your P. I had the same car you have. P behaves exactly the same as he described. I was also mostly in Chill mode.
But I should have experimented with Track Mode (just for the fun of it) in the snow.

And just because you CAN plow through snow with all wheels going (by punching it, and it will go reasonable straight) . Doesn't make it a great AWD system. Try going up a slippery hill when you're stuck behind traffic and can't accelerate enough to engage the front motor. You'll be a RWD car. And, no you don't want to wait until rear wheels slip to engage the front. You can't do a steady "crawl" up a slick hill with the Model 3.

Try a Model 3 AWD on a Snow/Ice test track. Then try a Model X. Both with similar tires. With the same driver. Not some pro.
I think you'd see a big difference. And the Model X could be better too. It's biased to use the front motor (which helps a lot) but it should have the same option to engage all motors all the time.

I have a pretty dim view of Chill Mode in the snow and ice the two times I tested it in winter conditions. I couldn't wait to get back to standard mode. It introduces a delay in the throttle which is exactly what a driver does not want when modulating the throttle.

There really is no way to reconcile these stories that the Model 3 does not apply torque to the front wheels until you punch it. That would make a very lame-asss winter driving experience but all the professional drivers that have tested it praise it. There really are only a handful of aliases here that continue to represent it as a sucky system. I have to chalk it up to either blatant FUD or lack of winter driving experience. Those are the only two things that could make any sense to me who has put my Model 3 through a full (and very snowy winter) in the North Cascades and who has seen all the praise heaped on the AWD/Performance Model 3 by professional drivers and testers who test cars for a living.

It's the same handful of people (I suspect one or two might have multiple aliases) who keep coming back with these bogus claims again and again. They use charts and graphs to try to act like they have proof but they have no answer when you point out that the charts and graphs have no applicability to snow and ice driving (because they weren't recorded in the snow and ice). They have no answer as to why so many professional car testers praise it. They have no answer as to why it works in deep snow climbing steep hills.

Meanwhile, the body of AWD and Performance Model 3 owners continue to praise the way the car handles slippery roads. I'm so glad I sold my AWD Mazda CX-5 to buy this!

 

[Saghost]

I was under the impression that each motor drove two wheels (left and right) through an open differential. So how can "power and current flow to the motor" tell the car the speed of the left and right wheels independently? I can't see any way to do this without employing wheel speed sensors on each wheel (or at least one wheel so the motor speed combined with one wheel sensor can be used to derive the speed of the other wheel). I suspect, in order to know the speed of every wheel independently, they simply use a wheel speed sensor on each wheel, no?

You are correct. The motor feeds aren’t related to knowing the wheel speeds.

All modern cars have a sensor measuring the speed of each wheel for the ABS system and that is the data I was expecting to integrate from the wheels.

I’m sorry if my glossing over that detail to keep the post focused created confusion.

 

[diamond.g]

The interesting thing about all of this is the utter lack of complaint about the TC System by our RWD brethren (sistren?).

 

[Daniel in SD]

The interesting thing about all of this is the utter lack of complaint about the TC System by our RWD brethren (sistren?).

I bet the RWD system feels much more refined. It doesn’t constantly switch between the front and rear motors. Everything is relative, most people are comparing it to other AWD vehicles. I bet the RWD is very good relative to other RWD cars.

 

[StealthP3D]

I bet the RWD system feels much more refined. It doesn’t constantly switch between the front and rear motors. Everything is relative, most people are comparing it to other AWD vehicles. I bet the RWD is very good relative to other RWD cars.

Daniel, please refrain from guessing when you have no idea what you're talking about or any experience with the topic at hand. I find it telling that one of the primary proponents of the wacky theory that the Model 3 AWD doesn't work well in the snow lives in San Diego, CA!

Having driven both the RWD and the AWD Performance on compacted snow/ice and on glaze ice, I can tell you that while the RWD is far superior to any other RWD car I've driven, it does not "feel more refined" than the AWD P version. The RWD's superior performance vs. other RWD cars is undoubtedly due to the combination of a very balanced weight distribution front to rear and it's superior traction control.

But both the AWD and RWD can claim these benefits. However, the RWD does not feel "more refined" than the AWD version, undoubtedly because it only has half of its wheels driven and thus is incapable of handing off the torque from the front to the back as required by localized conditions. The very capability that you theorize would be responsible for a lack of refinement is the very capability that makes it feel so refined (the ability to hand off torque front to rear as needed). You appear to erroneously believe that the handing off of torque from front to rear is a binary thing while in fact it is infinitely variable and can be proportioned as needed and seamlessly. For the best demonstration of this do not mash the accelerator to the floor but press it to the appropriate position for conditions, If you feel obvious slipping, reduce the throttle and find the sweet spot where the car feels "hooked up" this is where the fastest acceleration and the least tail-wagging will happen.

The fast-acting traction control on the AWD can accelerate more quickly and more smoothly precisely because it has more driven wheels when traction is spotty. The acceleration of the RWD under the same conditions is more variable and spotty (but still amazing for a RWD car). It's really hard to get into trouble with either of them compared to other cars I've driven but the AWD will outclimb the RWD on steep slippery hills without any spinning or detectable slippage when the surface is so slippery a light throttle is a necessity. The difference is night and day. This completely blows apart the theory that the front wheels are not driven until after major slippage is detected!

I'm not talking out of my butt here as we live on top of a steep hill (well over 6% grade) that was severely iced up for over two weeks last winter. I drove both cars on the same hill and there was NO comparison. I'm still impressed with how well the RWD did (for RWD) but, no, it's not "more refined", LOL! In corners it's the same story. Putting torque to all four wheels as needed is better than only having the rear wheels driven.

Please stop the whacky theories that have no basis in fact.

 

[StealthP3D]

The interesting thing about all of this is the utter lack of complaint about the TC System by our RWD brethren (sistren?).

Ummm, no.

If it's made by a company called "Tesla" you will find people complaining about it and saying it's no good!