HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
So the input is the flange?
Yes. If you had two motors I imagine they could both drive the same gear. But why even bother locking them together when they could be independent (without a differential)?
Sponsored

 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Threads
4
Messages
3,213
Reaction score
3,403
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
I hope it's obvious that I am being entirely tongue in cheek here. No one would think that there would be a differential with two independent motors. It would have no function and there would be no way to install it (other than as I have suggested with a second differential). But I was curious about the devices in the photos. A differential by its very nature has 3 input/outputs. The pictures show two shafts so the common mode must be the flange. I have trouble picturing this device in a vehicle.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
I hope it's obvious that I am being entirely tongue in cheek here. No one would think that there would be a differential with two independent motors. It would have no function and there would be no way to install it (other than as I have suggested with a second differential). But I was curious about the devices in the photos. A differential by its very nature has 3 input/outputs. The pictures show two shafts so the common mode must be the flange. I have trouble picturing this device in a vehicle.
It's inside the gearcase with a helical gear bolted to that flange.

My rhetorical question was more directed at Crissa's contention that the Plaid drivetrain might have two motors and a differential, all in the rear. It would be physically possible for two drive motors to both drive that helical gear on one differential but then the question arises, why not just have one motor that is longer, bigger and more powerful?
 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Threads
18
Messages
4,752
Reaction score
6,129
Location
Australia
Vehicles
Cybertruck
Occupation
. Professional Hobbyist
Country flag
All good points. Except I'm not sure how important the last one is to actual off-road performance. Especially in an EV that has a more balanced weight distribution that a front-engine vehicle.

With an EVs more balanced weight distribution, you are likely to be ascending a steep incline when you need maximum traction. Thus the rear wheels are doing ~80% of the work. This means a little light brake application (by computer) to the slipping front wheel is all that's required to make it 4x4 instead of 4x3. Done correctly, this is actually more effective than most LSD's and has important safety (on icy roads), efficiency (on tight, high traction trails) and tire wear advantages (in most situations where you might want to use a locking differential) over a front differential lock.
Ideally we'd have 2 independant motors up the front as well for off-road torque distribution! :-D

I can agree that in many situations traction control on the front will suffice on a TM. But on a dual motor you'd be back down to a 2x4 with two diffs. With a diff and having to brake costs a little bit of power (that the brake absorb until forces reach equilibrium) but more importantly causes lag because the brake is a reactive control method not a pro-active control method.

From what I can tell on Teslas from TFL testing etc, the traction control is a bit more responsive because they seem to have a faster sampling rate wheel rpm sensors and in particular better motor rpm control than an ICE with gearbox etc. But it still has noticable lag, meaning one wheel can and will provide forward force while another won't. That in turn means you won't have maximum forward force to climb and incline etc at low speeds. (But you might with momentum).

In fringe cases having offset forces can also be of benefit, for example when trying to dig yourself out of a sand hole where the sand has the right wet semi dense consistency. You quite literally just inch along like that until you get some form of floatation on virgin ground.

In my over 20 years of experience driving cars (Disco, Land Cruiser, Amorak) with traction control off-road it has been the case that brake modulation is combined with throttle control. This works if you have undulating roads where wheel travel limits grip but the surface has some grip and you don't need to carry momentum.

In sand, which is Australia's premier feature on the beach and many off-road tracks, its floatation that becomes a problem not traction. Traction in sand just digs you a nice hole so that your vehicle bottoms out. :)

There's two ways to create floatation:
1) have the right tyres that are not low profile, and can be depressurised to create a larger soft and plyable contact patch that conforms to the sand and;
2) enough momentum so that you mostly stay ontop of the sand instead of trying to drive through it.

This is where traction control provides little to no benefit on a open diff. In fact because of the throttle intervention of the traction control it is very difficult to maintain floatation through momentum. To which I typically either turn off traction control and lock the centre diff, (and ESP that also intervenes on the slightest cross track killing the throttle and momentum in the most untimely of ways) or select sand traction control mode that allows some wheel spin and more importantly gives more available throttle.

However, driving down the same stretch of beach with lockers on is even easier, and requires much less throttle in comparison (with the same car) because the forward force is consistent and un-retarded and most importantly with no lag.

In detail lag is bad because there is a response time for the traction control system to respond to wheel slip. This is made up of wheel rpm resolution, signal and cpu time (smaller fraction) and brake caliper response time and modulation, which in turn varies on manufacturer.

Imagine two opposite wheels, if the rotation is not synchronised (which I'm yet to see on a TCS) one wheel is rotating trying to push against another that is not rotating because it has grip. Typically its the wheels that are not turning that have the most grip. Or one rotates and slips even just quarter of a turn before the TCS stops it causing a ledge to form in front of the tyre which it then has to climb on it next turn. This is the downside of TCS in that throttle reduction combined with lagging brake interventions reduces forward force and therefore momentum.

Torque vectoring (like in a Evo) or locking diffs, or quad motor EV, through being able to control drive torque to each wheel makes all of this mute, because the wheels will all spin synchronised at the desired rpm and (except for the diff lockers) will even adjust each wheel according to steering input. So when cornering in sand it essentially forces the car to yaw by wheel rpm not steering angle. (Like a reverse ESP). This gets rid of nearly all the lag resulting in a smooth continuous force that can also create momentum, which in turn makes it easier to overcome track inconsistencies that would otherwise stop you with a TCS intervention.

So from my perspective a SM/DM needs at least one locker/lsd and ideally a TM would also have one in the front for those that take sand seriously. Don't get me started on what the difference is when towing an RV in sand, which is very common here, and that together with the very spread out supercharger availablility (1 in 2,500,000sqkm) a TM is the only option for rural WA.

As for weight distribution having a heavy ICE up front going up a hill is actually better because the front wheels get to do more work. Typically same size EVs will be heavier than a ICE, which doesn't help getting up a sandy track, but in the case of the CT I'm willing to make that sacrifice given I should be able to generate enough momentumwiththe TM.; -) Hopefully its only a 3-400kg more and the longer wheelbase and ground clearance will also help except for rampover angle. Will be fun to test.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
Ideally we'd have 2 independant motors up the front as well for off-road torque distribution! :-D

I can agree that in many situations traction control on the front will suffice on a TM. But on a dual motor you'd be back down to a 2x4 with two diffs. With a diff and having to brake costs a little bit of power (that the brake absorb until forces reach equilibrium) but more importantly causes lag because the brake is a reactive control method not a pro-active control method.

From what I can tell on Teslas from TFL testing etc, the traction control is a bit more responsive because they seem to have a faster sampling rate wheel rpm sensors and in particular better motor rpm control than an ICE with gearbox etc. But it still has noticable lag, meaning one wheel can and will provide forward force while another won't. That in turn means you won't have maximum forward force to climb and incline etc at low speeds. (But you might with momentum).

In fringe cases having offset forces can also be of benefit, for example when trying to dig yourself out of a sand hole where the sand has the right wet semi dense consistency. You quite literally just inch along like that until you get some form of floatation on virgin ground.

In my over 20 years of experience driving cars (Disco, Land Cruiser, Amorak) with traction control off-road it has been the case that brake modulation is combined with throttle control. This works if you have undulating roads where wheel travel limits grip but the surface has some grip and you don't need to carry momentum.

In sand, which is Australia's premier feature on the beach and many off-road tracks, its floatation that becomes a problem not traction. Traction in sand just digs you a nice hole so that your vehicle bottoms out. :)

There's two ways to create floatation:
1) have the right tyres that are not low profile, and can be depressurised to create a larger soft and plyable contact patch that conforms to the sand and;
2) enough momentum so that you mostly stay ontop of the sand instead of trying to drive through it.

This is where traction control provides little to no benefit on a open diff. In fact because of the throttle intervention of the traction control it is very difficult to maintain floatation through momentum. To which I typically either turn off traction control and lock the centre diff, (and ESP that also intervenes on the slightest cross track killing the throttle and momentum in the most untimely of ways) or select sand traction control mode that allows some wheel spin and more importantly gives more available throttle.

However, driving down the same stretch of beach with lockers on is even easier, and requires much less throttle in comparison (with the same car) because the forward force is consistent and un-retarded and most importantly with no lag.

In detail lag is bad because there is a response time for the traction control system to respond to wheel slip. This is made up of wheel rpm resolution, signal and cpu time (smaller fraction) and brake caliper response time and modulation, which in turn varies on manufacturer.

Imagine two opposite wheels, if the rotation is not synchronised (which I'm yet to see on a TCS) one wheel is rotating trying to push against another that is not rotating because it has grip. Typically its the wheels that are not turning that have the most grip. Or one rotates and slips even just quarter of a turn before the TCS stops it causing a ledge to form in front of the tyre which it then has to climb on it next turn. This is the downside of TCS in that throttle reduction combined with lagging brake interventions reduces forward force and therefore momentum.

Torque vectoring (like in a Evo) or locking diffs, or quad motor EV, through being able to control drive torque to each wheel makes all of this mute, because the wheels will all spin synchronised at the desired rpm and (except for the diff lockers) will even adjust each wheel according to steering input. So when cornering in sand it essentially forces the car to yaw by wheel rpm not steering angle. (Like a reverse ESP). This gets rid of nearly all the lag resulting in a smooth continuous force that can also create momentum, which in turn makes it easier to overcome track inconsistencies that would otherwise stop you with a TCS intervention.

So from my perspective a SM/DM needs at least one locker/lsd and ideally a TM would also have one in the front for those that take sand seriously. Don't get me started on what the difference is when towing an RV in sand, which is very common here, and that together with the very spread out supercharger availablility (1 in 2,500,000sqkm) a TM is the only option for rural WA.

As for weight distribution having a heavy ICE up front going up a hill is actually better because the front wheels get to do more work. Typically same size EVs will be heavier than a ICE, which doesn't help getting up a sandy track, but in the case of the CT I'm willing to make that sacrifice given I should be able to generate enough momentumwiththe TM.; -) Hopefully its only a 3-400kg more and the longer wheelbase and ground clearance will also help except for rampover angle. Will be fun to test.
I hear what you are saying about the disadvantages of using brake based traction control, particularly on ICE vehicles, but also on current EV implementations.

However, I would suggest that the primary problem is not in any delay that is inherent to the technology of braking the slipping wheel, but that current implementations are not tuned for maximum performance in an off-road environment for reasons of lack of sophisticated enough software to really take proper charge at the proper times (without becoming obnoxious or adding wear and tear and lack of efficiency at inappropriate times).

There is no significant additional delay due to the braking force being reactive, at least not one greater than the loss of perfect traction in every turn that every locking differential experiences by the nature of locking differentials. In deep sand you will experience this as a sinking of the vehicle when tight turns are attempted. Sure, it will probably keep going but it's not ideal torque delivery in turns. An electronic, brake-based "locking differential" can outperform a real one because computers can react essentially instantaneously. The problem with current implementations is a lack of aggression in terms of when and how quickly to apply the brake (s). They wait to make sure you are actually stuck before they redistribute torque. This is not necessary if there was a dedicated mode. With a mechanical locking differential it must be selectable too. It can't be on by default.

I'm hoping Tesla overcomes this with a selectable electronic "locking differential" mode that aggressively modulates the brakes to keep both tires applying constant torque. It's entirely possible with wheel speed sensors of sufficient resolution and supporting software. In fact, it's inevitable. The only question is whether Tesla will lead the charge.
 
Last edited:


JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Threads
18
Messages
4,752
Reaction score
6,129
Location
Australia
Vehicles
Cybertruck
Occupation
. Professional Hobbyist
Country flag
You might get the electronics and software to a insignificant lag time, but brake pressure modulation is a bit harder. Hopefully they can make it much better. Best part is no part, so I'm wondering at what point we might actually see the brakes being removed instead :)
The Plaid can already accelerate faster than it can decelerate, it's just a question of having quad motor for ABS/ESP and then trying to twist the regulators arm.

What I'd really like to see is them using SLAM computer vision for track and road placement to augment the wheel and accelerometer sensor inputs to for example detect yaw drift and ruts, wheel slippage, track drift, bumps etc. Plus real time suspension adjustments like leaning into corners, predictive bump detection, bounce mode etc like Merc has:



That with 20" travel would be epic offroad because then you could sync the suspension downforce with the traction control.
 

intimidator

Well-known member
First Name
TJ
Joined
Nov 16, 2020
Threads
11
Messages
675
Reaction score
919
Location
Alexandria, VA
Vehicles
2023 Ford Lariat Lightning
Country flag
We all have seen the patent UI mock-ups - 610 miles of range! Shocking, surprising? Even unbelievable? Just a random number in a mockup? How does the CT get there in a package that weighs the same (or less) than a Ford ICE F-150 and cost $70k (or less)?

Elon had stated in the reveal that they didn't "cheat" with the Cybertruck - they stayed within the dimensions and weight of the F-150 which weighs 5700 lbs max.
I think the 610 on the screen mock ups was a random #.

I will be delighted if they ever deliver me my 500 mile range CT. (I am not going to dream that it might really be 610 miles)

500 mile range for a large vehicle like the CT will be a great accomplishment in 2023.

Maybe in 2030 we can talk longer ranges....
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
You might get the electronics and software to a insignificant lag time, but brake pressure modulation is a bit harder. Hopefully they can make it much better. Best part is no part, so I'm wondering at what point we might actually see the brakes being removed instead :)
The Plaid can already accelerate faster than it can decelerate, it's just a question of having quad motor for ABS/ESP and then trying to twist the regulators arm.

What I'd really like to see is them using SLAM computer vision for track and road placement to augment the wheel and accelerometer sensor inputs to for example detect yaw drift and ruts, wheel slippage, track drift, bumps etc. Plus real time suspension adjustments like leaning into corners, predictive bump detection, bounce mode etc like Merc has:



That with 20" travel would be epic offroad because then you could sync the suspension downforce with the traction control.
I don't think we will ever see brakes removed from a road going vehicle - they are the most cost effective way to stop in a hurry. Using motors to stop would require almost as much electricity as it took to accelerate to the same speed.

What vehicle(s) have you driven with a front locking differential?

What did you use it for?
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
Fixed it for you.

-Crissa
That's fine for light or moderate braking but to stop quickly, near the limits of tire adhesion, requires stronger magnetic fields than a drive motor can produce organically. The coils would need to be energized and the power would necessarily come from the battery. The ability to brake at the limit of adhesion is required for safety in the event of the unexpected, a deer or elk jumping onto the highway, an errant human, etc, so brakes are not going away for this reason.
 


Ramojarocho

Member
First Name
Omar
Joined
Aug 3, 2020
Threads
2
Messages
22
Reaction score
27
Location
USA
Vehicles
Cybertruck 09 C300 4matic, 98 Chevy 1500
Occupation
Automotive Industry Muth Mirror Systems
Country flag
Yeah, I hadn't known that. Apparently they have a separate app to track energy usage.

I'm not sure how they manage to not include a driving history in the calculation; the only way to diagnose battery health is to gauge power used. Which would naturally fluctuate based upon the user as well. That's why my Zero's displayed range fluctuates wildly, because it doesn't know the temperature or anything, just miles, time, voltage, amperage.

But Tesla knows more about your accelerator pedal position than you do, so there's that, too.

-Crissa
just want to meet Crissa someday! we all can meet with our CTs and talk in person!
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Threads
4
Messages
3,213
Reaction score
3,403
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
When using regenerative braking the power returned to the battery is m*a*v. The battery can safely accept charge only up to a certain rate - about that of the numerical capacity of the battery. For example my X has a 100 kWh battery and regen is limited to 70 kW. Thus the limit on the deacceleration my car can provide is a = C/p where p = m*v, the momentum. If the car is going 60 mph the most regen braking I can get is 0.115g. Realizing that the car can pull 3 or more times that during acceleration it is apparent that more than 0.115 g will be needed from time to time. That's what the friction brakes are for. To get more regen braking we would have to supply additional loads (other than the battery) to aborb the energy generated ala a diesel electric locomotive. That energy would be wasted as heat and so BEV don't have resistive dump banks.
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Threads
126
Messages
16,211
Reaction score
27,070
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
That's fine for light or moderate braking but to stop quickly, near the limits of tire adhesion, requires stronger magnetic fields than...
No.

Regeneration only requires a load, and you can produce a load exceeding the tires' adhesion easily. And the limit on the battery is only part of it... Your load can be anything, really, even resistors.

-Crissa
 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Threads
18
Messages
4,752
Reaction score
6,129
Location
Australia
Vehicles
Cybertruck
Occupation
. Professional Hobbyist
Country flag
So with the Plaid they are already reaching the adhesion limits with acceleration and they are accelerating faster (avg 1.14g) with the motors than the friction brakes can deccelerate (1.11g) according to the Motortrend test:

https://www.motortrend.com/news/a-c...model-s-p100d-ludicrous-acceleration-run/amp/

That means the drivetrain itself can already handle the braking forces required, which to be honest is impressive. But they've been approaching that for a while with 60-0 braking times in the mid 2.5seconds for any decent car. So if you can accelerate in that time you can also brake in that time.

Like Adjelange pointed out the limiting factor is the recharge rate on the batteries. A resistor bank could easily accommodate this, theres about 0.3kwh of energy lost though from 60-0 unless you store that for use as interior heating (if required). Depending on the cooling rate of the controller/motors you could even dump some in there by varying the timing and use its thermal mass to absorb the heat (just a software change). This could also be a "backup" dump load should the battery fail. Worst case...

But we know that the newer batteries will have a much higher C rating for charging than previous models like MX or M3. With V3 supercharging at 250kW and a M3 battery with 75kWh thats a C rate of 3.3. If you would apply that 2170 cell rate to a TM CT with a 200kW battery you'd get around 660kW of available charge rate (@Adjelange whats the CT deceleration g's then?). Or in this case a safe rate to regen at which in turn would already be enough to cut about 75% out of the 0.3kWh heat generated.

However, with the tabless 4680 cells, because of their lower internal resistance you could easily double that input value, especially so for intermittent and very short periods for an emergency brake. (I only have about 4-5 emergency brake events a year, mostly kangas so cycle degradation should be minimal given the rarity) Thats 1300kW or so that the batteries could handle, which is in turn more than the Plaid train is officially rated for.

So overall with the ongoing trend to even more EV performance there will relatively soon be a case for deleting the brakes, or at least relegating them to a emergency backup system only. I expect the CT to have much more aggressive regen. Also helps to achieve range with a load.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Threads
9
Messages
4,481
Reaction score
9,453
Location
Washington State
Vehicles
2010 F-150, 2018 Model 3 Perform, FS Cybertruck
Country flag
So overall with the ongoing trend to even more EV performance there will relatively soon be a case for deleting the brakes, or at least relegating them to a emergency backup system only. I expect the CT to have much more aggressive regen. Also helps to achieve range with a load.
I already consider the friction brakes on my Model 3 mostly as emergency backup. And, yes, the Cybertruck Tri-motor should have stronger regen but don't expect that much from the single motor, it's likely it will have less effective regen than most current EV's due to it's higher weight and large wheels which will absorb some of the regen of it's singular motor only affecting the rear wheels.

Even if it were possible to regen to the limits of tire adhesion on dry pavement without energizing the coils (it depends upon what kind of motor is selected), common sense says there needs to be a backup friction brake in the event of electrical failure.

Don't expect friction brakes to go away anytime soon.
Sponsored

 
 




Top