additional regen braking from the brake pedal finally?

[BerTX]

I'm not sure imperceptible is the right word here. If you have a D, changing some of the load from friction braking to regen braking means you'd be moving it from front-of-car to front-and-back; for non-D, you'd be moving the braking work from front to back.

Good point. I guess it didn't sink in for me that I'm getting regen on my front wheels on my Insight, which would not be much different from applying the brakes, whereas the single motor Tesla is applying regen braking to the rears. I remember people pointing out issues with driving in icy conditions and strong regen, but since I don't drive in icy conditions, I skipped over it.

Seems like the dual motors could be tuned to overcome this difference in brake feel/action pretty easily.

 

[Danal]

Test for yourself while using cruise control or by viewing brake lights.

I have. I can pop an S out of cruise with no mechanical braking. A very light press on the brakes is all it takes. No braking engagement, and/or so little mechanical braking as to meet the pragmatic definition of "none", and cruise is long gone.

Everyone is entitled to their opinions, and I'd never try to convince you to "like" or "dislike" the pedal movements that I prefer in my other car. Nor am I stating that a Model S can (or even should) exactly emulate that behavior. All I'm saying is that, in that other car, I like the behavior. I may come to like the Model S way better over time. I do like the stronger deccel effect (feeling) in the Model S from full "lift the go pedal" regen.


At the same time as we have our various opinions, there is only one set of facts. And the statement that a Model S cannot sense the brake pedal movement prior to the engagement of mechanical braking is not in alignment with reality. I just drove the car a few seconds ago. Details:

1) Put the car in cruise at 30 MPH
2) Press the brake just enough to make the brake lights come on (visible at night on concrete behind the car)
3) Cruise instantly ends & car begins regen deceleration (because the gas pedal is up and cruise is off, not because the brake pedal is pressed per se)
4) Regen decel is medium strong at 30MPH, and central dial indicates 30kW or so of regen. Brake lights are still on.
5) As regen tapers to zero, so does deccel feeling. Brake lights are still on.

Car took few seconds to go from 30MPH to about 5 as the regen tapered away. The remaining 4 or 5 to 0 takes additional 17 to 22 seconds very level street. The last 1MPH to a full stop seems to take forever. Brake lights have been on since step (2) above. If I get on a slight down hill, the car actually rolls at 2 to 4 MPH as long as the hill lasts. Not stopping. Brake lights on, rolling along at 3MPH or so.

I can only state these facts for one 2015 VIN P85D on 6.1.blah.179 software. That car? It is dead easy to turn on the brake lights and knock the car out of cruise with no effective mechanical braking action.

 

[wk057]

I have. I can pop an S out of cruise with no mechanical braking. A very light press on the brakes is all it takes. No braking engagement, and/or so little mechanical braking as to meet the pragmatic definition of "none", and cruise is long gone.

Everyone is entitled to their opinions, and I'd never try to convince you to "like" or "dislike" the pedal movements that I prefer in my other car. Nor am I stating that a Model S can (or even should) exactly emulate that behavior. All I'm saying is that, in that other car, I like the behavior. I may come to like the Model S way better over time. I do like the stronger deccel effect (feeling) in the Model S from full "lift the go pedal" regen.


At the same time as we have our various opinions, there is only one set of facts. And the statement that a Model S cannot sense the brake pedal movement prior to the engagement of mechanical braking is not in alignment with reality. I just drove the car a few seconds ago. Details:

1) Put the car in cruise at 30 MPH
2) Press the brake just enough to make the brake lights come on (visible at night on concrete behind the car)
3) Cruise instantly ends & car begins regen deceleration (because the gas pedal is up and cruise is off, not because the brake pedal is pressed per se)
4) Regen decel is medium strong at 30MPH, and central dial indicates 30kW or so of regen. Brake lights are still on.
5) As regen tapers to zero, so does deccel feeling. Brake lights are still on.

Car took few seconds to go from 30MPH to about 5 as the regen tapered away. The remaining 4 or 5 to 0 takes additional 17 to 22 seconds very level street. The last 1MPH to a full stop seems to take forever. Brake lights have been on since step (2) above. If I get on a slight down hill, the car actually rolls at 2 to 4 MPH as long as the hill lasts. Not stopping. Brake lights on, rolling along at 3MPH or so.

I can only state these facts for one 2015 VIN P85D on 6.1.blah.179 software. That car? It is dead easy to turn on the brake lights and knock the car out of cruise with no effective mechanical braking action.

Approaching 40k miles of Model S driving, I can tell you that anywhere past the brake-lights-on portion does apply friction braking in all three Model S I've driven. Since the car doesn't have any digital knowledge until this point, any regen application tied to brakes would need to be *after* this point. This is what I was getting at earlier. Especially with the P85D. I can hold the P85D still on a slight incline with barely any force on the brakes past the sense point. This is actually super noticeable on my P85D right now since light application of the friction brakes at low speeds is causing one of my front brakes to squeal very annoyingly. (Tesla will be looking into it)

The point is, shifting regen to this spot would make no sense at all. Adding "new" (impossible/additional/etc) regen to here would also make no sense since friction+regen = loss of efficiency vs just regen. The nearly nonexistent travel between the time the brake pedal engages the brake lights and the time it can hold the car at a stop is so insignificant it would be pointless to have regen in this area.

Except in the case where the user setting is geared towards someone unfamiliar with EVs (like the Low regen setting) it makes no sense at all to have any regen ties to the brake pedal. As for such a thing being imperceptible, definitely possible with the D models. Equal regen force could be applied to each motor. Not as efficient, but doable. For the non-D it would be technically different, but I doubt many, if any, would notice.

 

[billarnett]

Wk047, can you explain why you believe that no more regen force is possible with the current hardware?

- - - Updated - - -

I did a little experiment this morning. I stopped on a slight down sloping road and put the car in R. Releasing the brakes let the car roll down hill slowly with a small amount of regen showing on the power meter. But the interesting thing is that when I pressed on the right pedal (usually called the accelerator) I got much more regen and the car slowed down correspondingly. Driving the motors in reverse did NOT consume any energy so long as the car was still moving forward.

What this tells me is that it should be possible for the software to bring the car to a complete stop without using the friction brakes and without using any energy. IMHO, this would be a significant improvement; one foot driving would become possible in all but emergency (or racing) situations. No doubt some would find it unfamiliar so making this behavior optional might be appropriate. Or it might be OK to just bundle it with CREEP=OFF.

 

[wk057]

Believe it or not, it takes power to create the magnetic field that demands regen/negative torque. There is a balance between that power cost and the regen possible at X RPMs with Y force. At low speeds this becomes diminishingly small. At higher speeds we'll easily hit a design limitation more quickly (around 60-70kW for non-D, 80-90kW for the P85D, not sure on the 85D (no data)). To capture power from regen there is a specific circuit in the inverter to do so. This circuit is going to be sized according to how it will be used, in this case about 70kW max, or about 1/5 to 1/3 of maximum output power (depending on trim level). It really is a matter of maximum current, which can vary a bit, but I won't get into too much craziness this morning.

There is also the charge taper for battery safety and longevity. This applies to regen also. If you ever noticed, the dashed line limiting regen matches supercharger taper almost exactly? Only below ~X% SoC (don't have the exact number in front of me, but I want to say mid 60% area) is full regen usable as it is now anyway, especially on the P85D. The dashed line disappears after ~35kW is usable, but the limit still exists above that until the battery is ready for more.

Can the motor be used to bring the car to a complete stop? Sure, technically. Would this be regen, per se? Not a chance. It would definitely take more energy to do this than the motor would get back.

As for your regen in reverse, while moving forward, after pressing the accelerator, I think you might actually be experiencing a bug there. If you saw regen on the meter above ~60kW I'd be even more intrigued, since it'd most definitely be a bug. I'll have to try it, but when you press the accelerator in this case the car will have to demand torque in the opposite direction with the use of energy. I'm not sure how fast you were coasting in this case, but, it is more likely power was being used to slow the car towards the reverse direction than any significant amount of regenerative braking occurring. I can probably try this out later today because that is interesting behavior, UI wise.

 

[billarnett]

Believe it or not, it takes power to create the magnetic field that demands regen/negative torque. There is a balance between that power cost and the regen possible at X RPMs with Y force. At low speeds this becomes diminishingly small. At higher speeds we'll easily hit a design limitation more quickly (around 60-70kW for non-D, 80-90kW for the P85D, not sure on the 85D (no data)).

I was clearly seeing more regen when pressing the right pedal in R than I was with neither pedal pressed. So either it's actually capable of much more regen at low speeds than we ordinarily see or there's a bug in the software.

To capture power from regen there is a specific circuit in the inverter to do so. This circuit is going to be sized according to how it will be used, in this case about 70kW max, or about 1/5 to 1/3 of maximum output power (depending on trim level). It really is a matter of maximum current, which can vary a bit, but I won't get into too much craziness this morning.

80 or 90kW would be better than the current 60 (though I sometimes see about 65).

There is also the charge taper for battery safety and longevity. This applies to regen also. If you ever noticed, the dashed line limiting regen matches supercharger taper almost exactly? Only below ~X% SoC (don't have the exact number in front of me, but I want to say mid 60% area) is full regen usable as it is now anyway, especially on the P85D. The dashed line disappears after ~35kW is usable, but the limit still exists above that until the battery is ready for more.

Are you sure about that? I'm pretty sure I see 60kW of regen at 90% SoC. (P85D)

Can the motor be used to bring the car to a complete stop? Sure, technically. Would this be regen, per se? Not a chance. It would definitely take more energy to do this than the motor would get back.

So all the kinetic energy is being converted to heat in the motor/inverter? Can you explain why some of it can't be recovered by regen? Sure, it takes a bit of energy to set up the appropriate fields in the motor but if the stator field is out of phase with the rotor field in the right direction energy should flow back in to the battery, no? Regardless of rpm.

As for your regen in reverse, while moving forward, after pressing the accelerator, I think you might actually be experiencing a bug there. If you saw regen on the meter above ~60kW I'd be even more intrigued, since it'd most definitely be a bug. I'll have to try it, but when you press the accelerator in this case the car will have to demand torque in the opposite direction with the use of energy. I'm not sure how fast you were coasting in this case, but, it is more likely power was being used to slow the car towards the reverse direction than any significant amount of regenerative braking occurring. I can probably try this out later today because that is interesting behavior, UI wise.

I was going very slowly, probably no more than 15 mph. I did not see the power meter go above about 40kW in the green.