Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Regenerative brake settings and lead feet

This site may earn commission on affiliate links.
Why is coasting the most efficient in general? If you coast to zero from 60 mph, all the energy has been lost to drag.
The primary task of a car trying to maintain velocity is countering drag. Therefore the kinetic energy stored in the car is 100% efficient in that task, it is not a "loss". The regen efficiency, whatever it is, is well under 100%.
 
The primary task of a car trying to maintain velocity is countering drag. Therefore the kinetic energy stored in the car is 100% efficient in that task, it is not a "loss". The regen efficiency, whatever it is, is well under 100%.
If you coast down to zero miles per hour, none of the original kinetic energy will be available to accelerate you back to speed. It will all be lost as heat to the atmosphere. If you regen to zero you'll have some of the original kinetic energy to reuse.
 
If you coast down to zero miles per hour, none of the original kinetic energy will be available to accelerate you back to speed. It will all be lost as heat to the atmosphere. If you regen to zero you'll have some of the original kinetic energy to reuse.

You can have some of that energy to reuse, you just won't have traveled as far as you would have if you coasted to a stop.
 
  • Like
Reactions: dgpcolorado
I haven't seen any numbers from tesla, but rimac says regen efficiency can be as high as 96%.
I don't believe that. There are too many places to lose energy on a roundtrip from motor/generator to the battery and back — that energy has to come back out of the battery — for me to believe a number like that. But it isn't my field of expertise, to put it mildly. There has been some discussion of Tesla regen efficiency here at TMC but I don't know if it has ever been "settled".
Why is coasting the most efficient in general? If you coast to zero from 60 mph, all the energy has been lost to drag. The power the car is losing to drag goes up as the speed cubed. If you coast at 1/2 the speed you are dissipating 1/8 the energy per unit time. There's probably some optimum combination of regenerating to a lower speed and then coasting. If the car was 100% efficient, you wouldn't want to coast at all. Any time spent above zero miles per hour would just be wasting energy to drag.
"There's probably some optimum combination of regenerating to a lower speed and then coasting. If the car was 100% efficient, you wouldn't want to coast at all."
Just so.

In the example of coasting gradually to zero, the regen scenario might include slowing with regen then using energy to maintain a speed to get all the way to the stop. Otherwise you would fall short in distance because the regen would reduce momentum to a number too low to overcome friction on the way to the stopping point. The roundtrip from regen to battery and back to the motor, to maintain speed, is lossy.

In simple terms, if coasting to a stop would need some additional braking then some regen is clearly beneficial. If coasting to a stop would involve stopping before the intended stop then some coasting is likely beneficial. In a sense, single pedal driving usually makes this moot because we feather the go pedal to add just enough regen to keep speed and momentum at the desired level. In effect, we are doing both at the same time and trying to draw a line between coasting and regen is splitting hairs. Drag at slow speeds, say 30 mph or less, is small so coasting can be efficient at such speeds. Coasting is less efficient at much higher speeds. I think that you are right that it would be more efficient to regen to a lower speed (reduced drag) and then coast from there, assuming that momentum is sufficient to get to the destination.

For my use case, I coast when I approach the bottom of my big hill and manage to coast the next mile, including over the next rise, before returning to regen to come to a stop at a stop sign. It is remarkable how well the Model S coasts in such a situation — low drag and high mass — and how poorly my neighbor's CUV coasts in identical conditions. I'm not trying to be maximally efficient since I also want to keep my speed up because I don't have the patience to drive that stretch at 15 mph or something like that, so it is a compromise. The no-coasting alternative is to use regen to slow and power to maintain a steady speed as needed to clear the next rise, which is what I do when there is a car in front of me and I can't let my speed and momentum build coming off the hill. [Or when I spot the rare Sheriff's Officer keeping tabs on speed: the stretch is posted at too low a speed limit, for complex reasons.]
 
Last edited:
There are too many places to lose energy on a roundtrip from motor/generator to the battery and back
That's not round trip. That's just converting the kinetic energy of the car back into electrical energy. I haven't seen any numbers from tesla going in the other direction either, but inverter/induction motor efficiency is around 91% for mid-range power. So the round trip would still be 87%.Throwing in some losses for the differential, it's still around 80% recovery.
 
Last edited:
I test drove (and ordered) a 90D and commented on how little the car would coast with my foot off the gas. The salesperson directed me to the regenerative brake settings and I changed it to low. She said with this setting, the brakes wear out much faster (believable) and total mileage is negatively affected (questionable for my driving style).

For people who are always pressing either the gas or brake, I can see keeping regenerative on high helping a lot, but for those of us who understand the nonlinear dynamics of traffic and the benefit of rolling up to traffic lights and leaving a gap in stop-and-go traffic, I can't imagine how this setting could help. Surely someone whose goal is to maximize the amount of time spent driving with neither the gas nor brake pressed would do better never expending the energy rather than inefficiently recapturing it.

Any theoretical insight or real world observations about range between the two settings?

To answer your question, a definition of coasting is necessary. In an ICE vehicle with an automatic transmission that applies no engine breaking, when the driver's foot is off the accelerator, no power is applied and the car will coast as far as initial velocity, road inclination, mechanical friction and air resistance allow. Taking one's foot off the accelerator to coast well before arriving at a stop sign saves brake wear and reduces energy use and brake wear compared to applying power until near to the stop and then applying brakes. (It is also likely to anger most following drivers.)

A Tesla with full regen can also be made to "coast", simply by reducing accelerator pedal pressure to the point where the energy consumption meter shows zero kW (after a while, you can feel that point without having to look). At that point, there is neither power being applied to the motor nor (slightly inefficient) regenerative braking occurring, and the car will coast as far as initial velocity, road inclination, mechanical friction and air resistance allow. The energy consumption of the Tesla would actually be slightly better than that of the ICE because a coasting ICE is still burning a little fuel to keep its idling engine running, while a "coasting" Tesla would consume no power at all.

Without regen, you lose the capability to control downhill speed without braking while capturing energy back to the battery, yet you gain nothing on the flat or uphill. As to mileage under your driving style not being negatively affected by low regen, that would only be true in a flat land where regen would not be used to control downhill speed.
 
If you coast down to zero miles per hour, none of the original kinetic energy will be available to accelerate you back to speed. It will all be lost as heat to the atmosphere. If you regen to zero you'll have some of the original kinetic energy to reuse.

What you are forgetting is the energy used up to the point of applying regen. Consider two identical cars traveling at 60 mph: at point A one removes power and coasts to a stop. The other car in its turn continues at 60 mph for some distance beyond point A and then applies regen in order to come to a stop at the same point where the coasting car had stopped. The continued energy consumption to remain at 60 mph longer and the inefficiency of energy returned to the battery by regenerative braking would result in more energy consumed by the second car.

Of course, the energy subsequently consumed performing a launch to escape a fusillade of bullets from the road rage-consumed drivers behind the coasting car would more than make up for the difference. :)
 
  • Like
Reactions: dgpcolorado
What you are forgetting is the energy used up to the point of applying regen. Consider two identical cars traveling at 60 mph: at point A one removes power and coasts to a stop. The other car in its turn continues at 60 mph for some distance beyond point A and then applies regen in order to come to a stop at the same point where the coasting car had stopped. The continued energy consumption to remain at 60 mph longer and the inefficiency of energy returned to the battery by regenerative braking would result in more energy consumed by the second car.

Of course, the energy subsequently consumed performing a launch to escape a fusillade of bullets from the road rage-consumed drivers behind the coasting car would more than make up for the difference. :)

If you're going to impose the completely unreasonable constraint on me that both cars come to rest at the same location, I would suggest regen to a much lower speed immediately and then continue to the destination instead of zooming up to the final destination.
 
Last edited:
I'm not understanding your point @TIppy, and I think my view is the same as @brucet999 etc., perhaps you can clarify?

Here's a hypothetical example:

I have 3000 feet to a junction, and I'm doing 40 MPH.

1) If I coast in Neutral I will slow down to, say, 10 MPH by the time I get to the junction, at which point I will brake.

2) Or I can maintain 40 MPH and at 300 feet, say, lift off and using full Regen get to, say, 8 MPH at the point at which i have to brake for the junction.

(I have no idea what the actual distances would be to coast-to-10MPH and Regen to 8MPH)

In (2) I will have used more energy to continue travelling at 40 MPH for (3000 - 300) - 2700 additional feet, and then I will regain (say 80%) of my energy using Regen. Actually might be less than 80% because I maintained a higher average speed over the 3000 feet, so more energy to push the air out of the way at the higher speed, but I did save some travel time (which might be important ...)

So to my way of thinking Coasting saves energy compared to Regen, but takes a bit longer.
 
I'm not understanding your point @TIppy, and I think my view is the same as @brucet999 etc., perhaps you can clarify?

Here's a hypothetical example:

I have 3000 feet to a junction, and I'm doing 40 MPH.

1) If I coast in Neutral I will slow down to, say, 10 MPH by the time I get to the junction, at which point I will brake.

2) Or I can maintain 40 MPH and at 300 feet, say, lift off and using full Regen get to, say, 8 MPH at the point at which i have to brake for the junction.

(I have no idea what the actual distances would be to coast-to-10MPH and Regen to 8MPH)

In (2) I will have used more energy to continue travelling at 40 MPH for (3000 - 300) - 2700 additional feet, and then I will regain (say 80%) of my energy using Regen. Actually might be less than 80% because I maintained a higher average speed over the 3000 feet, so more energy to push the air out of the way at the higher speed, but I did save some travel time (which might be important ...)

So to my way of thinking Coasting saves energy compared to Regen, but takes a bit longer.
Why wait to regen until later? Regen at 3000 feet and approach the junction at a slower speed.
My original question was sincere. I really haven't worked out what the best approach is. It just seemed to me that you're burning kinetic energy a lot faster at higher speeds, so if you regen down to the avg speed of the coast you would use less of the initial kinetic energy to get to the junction and get there at the same time. Regen to a slower speed would save even more energy.
 
Last edited:
If you start Regen 3000 feet out and reduce speed from 40 MPH down to, say, 25 MPH, and then at at hat point coast, you won't have enough residual speed to get to the junction - or am I missing something? Thus to my way of thinking any Regen will start later than any Coast [and use more energy], but of course you can maintain power until later, and then a short Regen followed by a Coast, or maintain power until the last moment and then full regen down to 8 MPH (or even maintain power longer still, and use both Regen and Brakes)

My coast-to-a-junction (or Bend) is never perfect, of course, so in practice even if I try to coast from " far out" I will wind up using some Regen to compensate for having started my coast too late (or some Energy if I started to early).

I also need to allow for any following vehicles - they won't be too thrilled if I coast from 3,000 feet out!

My Hypermiling started over a decade ago on the school run, a boring repetitive short-ish journey, where landmarks became familiar in terms of "With speed around XX MPH then Coast will be perfect for cornering speed for next bend".

Of course Regen is so much more Eco than "Accelerate then Brake" that I'm not really sure if any possible saving is actually worth having; for me its mostly a mental excise and a challenge to judge exactly when to start Regen.
 
  • Like
Reactions: dgpcolorado
If you start Regen 3000 feet out and reduce speed from 40 MPH down to, say, 25 MPH, and then at at hat point coast, you won't have enough residual speed to get to the junction - or am I missing something? Thus to my way of thinking any Regen will start later than any Coast [and use more energy], but of course you can maintain power until later, and then a short Regen followed by a Coast, or maintain power until the last moment and then full regen down to 8 MPH (or even maintain power longer still, and use both Regen and Brakes)

My coast-to-a-junction (or Bend) is never perfect, of course, so in practice even if I try to coast from " far out" I will wind up using some Regen to compensate for having started my coast too late (or some Energy if I started to early).

I also need to allow for any following vehicles - they won't be too thrilled if I coast from 3,000 feet out!

My Hypermiling started over a decade ago on the school run, a boring repetitive short-ish journey, where landmarks became familiar in terms of "With speed around XX MPH then Coast will be perfect for cornering speed for next bend".

Of course Regen is so much more Eco than "Accelerate then Brake" that I'm not really sure if any possible saving is actually worth having; for me its mostly a mental excise and a challenge to judge exactly when to start Regen.

You use some of the energy you captured with regen to keep the car at the avg speed. At 1/2 speed you're dissipating 1/8 the power, so even keeping the speed constant at 1/2 uses less energy over the same distance. Plus you still have some kinetic energy when you get there to regen into the battery.
 
Last edited:
If you're going to impose the completely unreasonable constraint on me that both cars come to rest at the same location, I would suggest regen to a much lower speed immediately and then continue to the destination instead of zooming up to the final destination.

"Unreasonable constraint" of coming to rest at the same location?! Removing variables is the basis of scientific comparison.

"Zooming up to final destination"? I postulated no acceleration, only the normal driving case of not slowing down as early as coasting would require, thus requiring some sort of braking - in this case regen.
 
"Unreasonable constraint" of coming to rest at the same location?! Removing variables is the basis of scientific comparison.

"Zooming up to final destination"? I postulated no acceleration, only the normal driving case of not slowing down as early as coasting would require, thus requiring some sort of braking - in this case regen.

I was joking about it being completely unreasonable. I guess I should start using imojis. But it is different than what I was talking about. You have introduced a new variable, the average speed while coming to a stop. So now we have to specifiy what would be apples to apples. If you want to lower the average speed and therefor energy, do the regen first and travel to the final location at a lower speed.
 
Last edited:
You use some of the energy you captured with regen to keep the car at the avg speed.

Thanks. I've got it now :)

1. Straight-line coast from XX MPH to, say, 10 MPH
2. Full regen to reduce speed by, say, half, then maintain that speed and then full regen to 10 MPH
3. Maintain XX MPH until the latest moment and then Full Regen to 10 MPH

All tests starting and finishing at the same point.

What is the difference in energy used? Answers by tomorrow's lesson please :D
 
Thanks. I've got it now :)

1. Straight-line coast from XX MPH to, say, 10 MPH
2. Full regen to reduce speed by, say, half, then maintain that speed and then full regen to 10 MPH
3. Maintain XX MPH until the latest moment and then Full Regen to 10 MPH

All tests starting and finishing at the same point.

What is the difference in energy used? Answers by tomorrow's lesson please :D

Not practical, but might help. Full regen to 2 mph, then 2 mph to final point. Although it would take longer to get to the final point, very little energy would be expended at 2mph. Since kinetic energy is proportional to speed squared, regen would essentially capture all the kinetic energy at 2 mph. Then if you did full regen to 1mph, it would take twice as long but would use 1/4 of the energy required at 2 mph. In the limit it would take you forever to get there, but you would capture all of the energy.
 
Last edited:
What is the difference in energy used? Answers by tomorrow's lesson please
Let's go down to zero speed. It only changes the limits of integration and makes the numbers easier to compare.
We're only interested in relative efficiency, so let's call the energy dissipated by the linear ramp as 1unit of energy.

Keeping the speed constant at 1/2 the peak speed of the linear ramp results in 1/2 as much energy being used over the same time/distance. We're using the average speed for the constant, so the time and distance will be the same as the linear ramp.

The amount of energy used over the same distance goes up as the speed squared, so the constant full speed case will use 4 times as much energy as the 1/2 speed case. That's 2 times as much as the linear ramp.

1/2 speed: 1/2 energy unit
linear ramp: 1 energy unit
full speed: 2 energy units
 
Last edited: