ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,364
Reaction score
2,557
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Yes. It takes more than 385V to push current into a 385 V battery. If there is no DC stepup and assuming that motor voltage is proportional to dø/dt then the motors must be producing a bit more than 385V at 5 mph, 3850V at 50 mph and 6160 V at 80 mph. I am sure that's not the case.





Advertisement

 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Messages
55
Reaction score
53
Location
6000
Vehicles
Cybertruck
Country flag
@ Hauling
Current regen rate and profile doesn't reflect system capability, rather what is acceptable for the customer in terms of driveability. In that high declaration low rpm braking can cause uncomfortable cogging and an abrupt stop.


@Adjelange
Motor rpm is controlled by phase switching frequency not volts.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
@ Hauling
Current regen rate and profile doesn't reflect system capability, rather what is acceptable for the customer in terms of driveability. In that high declaration low rpm braking can cause uncomfortable cogging and an abrupt stop.
It already causes an uncomfortable stop for those who have trouble adapting to one-pedal driving, that's why Tesla offers a low regen mode.

I live in a coastal climate with moderate weather and my car is almost always regen limited at 75% state of charge unless I've driven it for an hour to warm up the battery. That tells me the regen limit is not due to drivability concerns but battery longevity.
 

larryboy31

Well-known member
First Name
larry
Joined
Apr 29, 2020
Messages
51
Reaction score
60
Location
springfield NE
Vehicles
2001 Dodge turbo diesel ext. cab,2016 Nissan versa, 2021 Subaru cross trek
Occupation
retired
Country flag
It already causes an uncomfortable stop for those who have trouble adapting to one-pedal driving, that's why Tesla offers a low regen mode.

I live in a coastal climate with moderate weather and my car is almost always regen limited at 75% state of charge unless I've driven it for an hour to warm up the battery. That tells me the regen limit is not due to drivability concerns but battery longevity.
When you take your foot all the way off the "gas" and slow down faster than an ICE car do vehicles behind you get any warning? Like brake lights or something.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
When you take your foot all the way off the "gas" and slow down faster than an ICE car do vehicles behind you get any warning? Like brake lights or something.
Yes, brake lights come on at a deceleration threshold set by regulations. So if you quickly remove your foot from the accelerator the brake lights are on before you hit the brake pedal.
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Messages
4,362
Reaction score
5,073
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
Crissa, please read what I wrote. I said the motor cannot break the tires loose at all speeds, only slow speeds. If used in a regenerative...
It already can. The amount of resistance to break the tires loose is about an abrupt change in wheel speed, which the motors are certainly capable of.

And modulating quickly is something it already does better than the friction brakes - that's what the motor controller is doing.

...motors must be producing a bit more than 385V at 5 mph, 3850V at 50 mph and 6160 V at 80 mph. I am sure that's not the case.
Your math is wrong by an order of magnitude again. But yes, the motor produces sufficient voltage, which then goes through the motor controller, which then rectifies it (much like it inverts it to make the wheels spin).

...at a deceleration threshold set by regulations. ...
No. There is no regulation on regenative or engine braking activating brake lights in the US.

So what happens when to start regen varies by vehicle brand.

-Crissa
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,364
Reaction score
2,557
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
@Adjelange
Motor rpm is controlled by phase switching frequency not volts.
I am not sure what you are referring to here. My recent comments have been regarding regenerative braking. If one closes all the switches in the bridge a PM motor acts as a generator as it is moving magnets past coils. This generates and EMF whose magnitude depends on the speed at which the magnet passes the coils. If the motor is turning slowly the EMF is small. if it is turning fast the EMF is large. The rpm is, at this point, determined by the speed of the vehicle.

In dynamic breaking you connect the coils to a load e.g. a resistor. This causes (AC) current to flow in the resistor and coils. That current creates a coil magnetic field opposing the permanent magnet field and torque is applied to the rotor in a direction such that the car slows. Speed depends on the amount of current and is controllable by changing the size of the resistor (and hence the current).

In regen it is a little different. Instead of a resistor the "dump" is a battery. To get the battery to take current the AC from the generator must be rectified and that means that the switches in the bridge must be opened at the right time such that the current is unidirectional and that means that the switches must cycles at synchronous speed. By "space vector control", that is, opening the switches in the proper phase relative to the rotor position you can put the stator magnetic field vector any where you want it relative to the rotor and by changing the duty cycle (PWM) you can control it's magnitude but at all times the phasing of the switches must be synchrononous. To increase the speed of the motor you put the stator magnetic field ahead of the rotor magnetic field. This puts a torque on the rotor in the direction of travel causing the motor to speed up. As it speeds up the controller has to follow the rotor and the switching frequency goes up. To slow down you phase the switches to put the stator field behind the rotor field. This causes torque in a direction that slows the vehicle. Again the controller must stay phase locked to the wheel.

What I don't understand is that in motor mode the battery voltage has to be bigger than the back EMF at whatever speed you want to operate so current will flow into the motor. For generator mode the machine EMF has to be higher than the battery voltage so current will flow into the battery.

Everything here is with respect to PM motors because the CT is, in all probability, going to use them exclusively. Control of induction motors is different. A PM motor must not slip. An IM motor can. Thus one can control an IM motor by simply changing the frequency of the drive. The more slip, the more torque and the motor speeds up or slows down depending on whether the drive is faster or slower than the rotor.
 
Last edited:

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
It already can. The amount of resistance to break the tires loose is about an abrupt change in wheel speed, which the motors are certainly capable of.
It already can what? Break the tires loose at any speed? No, it can't.

And modulating quickly is something it already does better than the friction brakes - that's what the motor controller is doing.
The motor controller does modulate regen based on tire slippage. However, it is not nearly as precise or quick as friction brake modulation. I know this because I have used one-foot driving with full regen to drive an icy, snowy, downhill highway with switchbacks while being in controlled slides most of the way by using the throttle pedal to modulate regen. Regen braking does not modulate braking between left and right wheels as friction braking does so it does not modulate better or more quickly. It's actually very slow in modulation and allows a lot more slippage than anti-lock brakes which modulate in quick pulses.


No. There is no regulation on regenative or engine braking activating brake lights in the US.

So what happens when to start regen varies by vehicle brand.

-Crissa
No, the illumination of brake lights and when they illuminate is standardized. You will find that all cars sold in the U.S. abide by UNECE harmonized regulations for vehicle standardization. A manufacturer that doesn't conform to these regulations is at increased legal risk in the event their car is rear-ended. So you will find all regenerative braking systems are tuned to fall within these regulations.

1623330084178.png


The entire document can be found here:
Microsoft Word - R13hr2e.doc (unece.org)
 
Last edited:

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
You don't slow down faster than an ICE car.
That's true, at least regen does not slow you down faster than an friction brakes are capable of when braking at the threshold of traction. But I think the question was referring to slowing down more quickly than the cars around you that are using normal braking to come to a stop (for example, when approaching a red light).

In practice, EV drivers modulate the accelerator to achieve the desired amount of regen braking to match the deceleration of traffic in front. It's like pushing on the brake to slow down except the throttle pedal is used in reverse. The further you let the throttle pedal release, the more regen braking is delivered.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,364
Reaction score
2,557
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
The motor controller does modulate regen based on tire slippage. However, it is not nearly as precise or quick as friction brake modulation.
This is just contrary to common sense but I do not have any knowledge of design detail or any real world data to back that assertion up. The reason it violates common sense is that the "inertIa" of the stator magnetic field, which is what determines the torque, can be moved to any angle in microseconds. Obviously this is much faster than any mechanical part of finite mass can be moved with an actuator capable of finite effort.


I know this because I have used one-foot driving with full regen to drive an icy, snowy, downhill highway with switchbacks while being in controlled slides most of the way by using the throttle pedal to modulate regen. Regen braking does not modulate braking between left and right wheels as friction braking does so it cannot modulate better. That is just false.
It does not modulate between left and right in a X or S because there is only one motor. All torque vectoring in these cars is done by differential braking. Common mode thrust or drag can, of course, be supplied by the motor(s) one in front and one at the rear. Even so I find, in the very few times I''ve driven in icy, slushy conditions, that the mechanical vectoring is quite good. There are, of course, limits.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
This is just contrary to common sense but I do not have any knowledge of design detail or any real world data to back that assertion up. The reason it violates common sense is that the "inertIa" of the stator magnetic field, which is what determines the torque, can be moved to any angle in microseconds. Obviously this is much faster than any mechanical part of finite mass can be moved with an actuator capable of finite effort.


It does not modulate between left and right in a X or S because there is only one motor. All torque vectoring in these cars is done by differential braking. Common mode thrust or drag can, of course, be supplied by the motor(s) one in front and one at the rear. Even so I find, in the very few times I''ve driven in icy, slushy conditions, that the mechanical vectoring is quite good. There are, of course, limits.
Yes, differential friction braking kicks in to modulate torque left/right at low speeds when having trouble making forward motion. This transfers more power to the wheel with traction. For whatever reason, differential friction braking never kicks in under regen braking.

I agree that from a technical perspective the regen could be modulated almost instantly. However, no Tesla delivered to date has more than one motor per axle. What that means is fast modulation of regen braking at the limits of traction is not possible because it cannot be expected that the right and left wheels will loose or gain traction at the same time because all regen braking has to be transferred through an open differential. This explains why regen braking modulation is MUCH slower and softer. To do otherwise requires a dedicated motor for each wheel.

It's still much more capable than any ICE car I've driven.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,364
Reaction score
2,557
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Yes, differential friction braking kicks in to modulate torque left/right at low speeds when having trouble making forward motion. This transfers more power to the wheel with traction. For whatever reason, differential friction braking never kicks in under regen braking.
I wonder what your basis for thinking that is. Certainly given the way Tesla's cars handle in icy conditions there is differential braking going on. If you say it never happens when regen is happening you are saying that regen is turned off when ever torque vectoring is needed. I just don't believe that.

I agree that from a technical perspective the regen could be modulated almost instantly. However, no Tesla delivered to date has more than one motor per axle. What that means is fast modulation of regen braking at the limits of traction is not possible because it cannot be expected that the right and left wheels will loose or gain traction at the same time because all regen braking has to be transferred through an open differential.
As soon as slip is detected in a wheel the brake on that side is applied slowing that wheel. At the same time the common mode torque is reduced so that the vehicle doesn't yaw due to differential thrust. Or something like that. I don't know the details of the design or algorithms.


This explains why regen braking modulation is MUCH slower and softer. To do otherwise requires a dedicated motor for each wheel.
You feather the skinny pedal, you feel the car slow from the regen (you get to know what that feels like) and you can feel the wheels slipping and recovering. The car goes strait even though the substrate is slippery. Thus it appears that differential braking and regen do occur simultaneously.
 

HaulingAss

Well-known member
First Name
Mike
Joined
Oct 3, 2020
Messages
397
Reaction score
565
Location
Washington State
Vehicles
2010 Ford F-150, 2018 Tesla Model 3 Performance
Country flag
I wonder what your basis for thinking that is. Certainly given the way Tesla's cars handle in icy conditions there is differential braking going on. If you say it never happens when regen is happening you are saying that regen is turned off when ever torque vectoring is needed. I just don't believe that.
I may have overstated the situation by using the word "never", it might be more accurate to say that in situations requiring differential braking the car reduces the strength of regen until it senses differential braking is not required. This is a temporary setting that will fade stronger or weaker based on conditions but it has a certain amount of persistence beyond a single loss of traction event. What happens in slippery situations is regen become more limited but it doesn't just suddenly go away, it fades in and out to not surprise the driver. This reduces or eliminates the need for differential braking until the brake pedal is used. And the differential braking that may be happening before regen is reduced is very subdued compared to when I turn the car on low-regen and descend using friction brakes. What I mean by that is under regen braking, while sliding down hills, entering hairpins, etc, the car will yaw left and right quite a bit, more than when pressing the brake pedal and feeling the anti-lock switching sides. The amount of regen fades such that it cannot slow as much as friction brakes alone. Friction brakes, especially when modulated by the anti-lock system, can slow the car more effectively because they are more aggressive at the limits of traction. This causes a little wall of snow to build up in front of each tire that regen braking alone does not do (because I have felt regen fade away as the computer determined it was too slippery for full regen).

In short, my primary basis for saying that differential braking is not nearly as effective under regen is the way the car reduces regen and brings friction brakes into the equation which then provide more effective yaw control (and also allows speed to be scrubbed off faster since the car has reduced the amount of regen available in general in response to the conditions). I fully admit this may just be a matter of how Tesla decided to tune the systems to work together and may not be a limitation of the technology.

As soon as slip is detected in a wheel the brake on that side is applied slowing that wheel. At the same time the common mode torque is reduced so that the vehicle doesn't yaw due to differential thrust. Or something like that. I don't know the details of the design or algorithms.
You're right, it does do that but it is not as responsive as friction braking alone (in terms of reducing yaw). I think this has to do with the amount of torque being applied through the differential by the regen braking messing up the sensitivity of the anti-lock system/stability control. Tesla even advises to turn regen to low in such situations.


You feather the skinny pedal, you feel the car slow from the regen (you get to know what that feels like) and you can feel the wheels slipping and recovering. The car goes strait even though the substrate is slippery. Thus it appears that differential braking and regen do occur simultaneously.
I have hours driving down steep grades in mountain storms so active the plows can't keep up. Sometimes the underbody of my Model 3 is dragging through the snow hard enough to feel the deceleration and loss of traction as the heavy snow packs under the car. It performs amazingly well compared to all my gasoline powered ski cars (Subaru's Volvo's, VW's) but differential regen braking is not amazing like it is with regen turned off. That said, it's still good enough that I would only turn regen to the low setting if it was glaze ice. I like the challenge of sliding down off the mountain with as much regen as the car will give me. In such conditions the AWD Model 3 puts ICE cars to shame. But if I were racing an ICE car off the mountain I would turn regen to low because it makes the stability and traction control systems less effective.
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Messages
4,362
Reaction score
5,073
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
It already can what? Break the tires loose at any speed? No, it can't.
That it has limits to avoid this is no evidence that it can't.

No, the illumination of brake lights and when they illuminate is standardized. You will find...
...That you don't know how to read that document.

First, that's the ECE regulations, not DOT. They do not apply to vehicles in the Americas, they're for the EU.

Second, most of it is saying that the brakes need to trigger the lights at a certain threshold. For instance, on a motorcycle, brake lights are often triggered by a pressure sensor, rather than an activator switch. That's what it's describing here.

Third, where it does describe regen, it describes a braking force much larger than say, Tesla uses normally, for the 'must'. And yet they do the 'shall' and they do it in the US as well. (Also, a force large enough to break the tires free.)

-Crissa
 

Advertisement





 


Advertisement
Top