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

Regenerative braking

This site may earn commission on affiliate links.

TEG

Teslafanatic
Moderator
Aug 20, 2006
22,101
9,490
Many hybrids and BEVs have a feature called "Regenerative Braking". This was Recently discussed on the Tesla Motors Blog.

It appears that the Roadster may activate regen when you let up on the accelerator pedal, but may not add any extra regen based on brake pedal pressure.
(In other words as soon as you start to put pressure on the brake pedal you start to engage the friction brakes).

Apparently many other BEVs did tie the regen to the brake pedal, so it would be interesting to understand why Tesla chose not to do this.

Also, people have been asking if Tesla could add variable regen controls to their cars... Apparently the regen profile changes depending on 1st gear or 2nd gear selection, but otherwise is not intended to be driver adjustable.

Here are some examples of places that mention brake pedal tied regen on other BEVs:

In an account of driving an EV1, John wrote:
"..time to employ my usual high miles per charge techniques. I switched off the coast-down (we EVers call this off-throttle regen) features, so that I could let the EV1 glide along on slight down hill runs without the car trying to slow down putting a small bit of juice back into the batteries (I'd save this feature for later, as when exiting the freeway and for congested traffic stop and go driving)..."

If you go to the EV1 FAQ, and go down to the bottom there is a section called "What is regenerative braking". They mention " 'Regen' is the process of making the EV1's motor work like a generator when the brake pedal is pushed, or when the vehicle is coasting with the 'coast regen' on."

First Peek Inside the (AC Propulsion) eBox:
"When I applied the brake pedal, the regenerative braking was smooth. When I pushed the pedal further, the transition from regenerative to mechanical braking was seamless."

snl101.myers.richman.010704_small1.jpg

Discuss...
 
Tesla tidbits from the web:

http://mn.eaaev.org/resources/Tesla Presentation Notes.pdf
"• Regen is controled by removing your foot from the accelerator pedal.
• Regen has 2 levels: Max in low gear, and reduced in high gear (although, it is fully
proportional).
• Maximum limits on regen are similar to agressively downshifting an ICE sportscar.
• Note that regen is inverse of the drive wheels (i.e. By using the motor, and the car is
rear-wheel-drive). Therefore, because the rear wheels don't provide majority braking, a
maximum of about 80% of the regen opprotunity can be realized."

http://jalopnik.com/xml/comments/188590
"
The Tesla Roadster has regenerative braking but it works differently than the typical hybrid. Hybrids apply regen braking when the driver steps on the brake pedal. The Tesla Roadster applies variable regen braking based on accelerator position. As the driver eases off the accelerator the regen braking is applied. Regen braking is also a function of the gear selected (2 speed manual gearbox with electric shifing and no clutch). First gear has higher regen braking than second. In driving, it feels very much like a standard transmission gasoline powered car. You can actually come to a complete stop without touching the brake pedal. Also, for those who wonder, the brake lights are computer controlled and based on measured deceleration rather than the standard switch on the brake pedal."
 
Yeah...I wish there were better descriptions of this too. Maybe you just have to experience it.

Does the accelerator essentially function as the stop and go pedal in circumstances other than those requiring heavy braking? How sensitive is it to small variations in foot position?

If the stop light ahead is red, and I want to preserve momentum while the light goes to green, what behavior do I expect from the car as I ease off the accelerator? Does it begin to slow more rapidly the less pressure applied? If so, would this behavior be similar to the variable regen some people are hoping the car will implement?

And, of course, there's the much-discussed brake light question...
 
Brent said:
Yeah...I wish there were better descriptions of this too. Maybe you just have to experience it.

Does the accelerator essentially function as the stop and go pedal in circumstances other than those requiring heavy braking? How sensitive is it to small variations in foot position?

I haven't driven it yet, but. . .    I thought the Jalopnik description seemed pretty apt -- like driving a manual transmission car and using engine braking.  If the Tesla simulates that behavior, I'll be happy.

Although, at first I might miss the "burble" of the V8 when I back off the throttle. That sounds so cool.
 
Is it the case that the most efficient driving in stop-and-go traffic would use the least amount of regen?

I have the sense that any amount of electricity regen might make is not enough to offset the cost of acceleration. To be sure, if you have to stop, you might as well take advantage of it. But slowing well in advance of a red light in hopes it goes green might still provide the better usage than driving to it quickly and then braking.
 
Brent said:
Is it the case that the most efficient driving in stop-and-go traffic would use the least amount of regen?

I have the sense that any amount of electricity regen might make is not enough to offset the cost of acceleration. To be sure, if you have to stop, you might as well take advantage of it. But slowing well in advance of a red light in hopes it goes green might still provide the better usage than driving to it quickly and then braking.

I think you are probably correct. 

Since regen recovers only 80% of the energy of deceleration, not having to decelerate more than that owed to rolling resistance keeps more juice in the battery compared to yo-yo accel/decel.
 
Since regen recovers only 80% of the energy of deceleration

This is only half the story. When those 80% of recovered energy is used again it goes through another conversion with 80% efficiency giving you total 64% percent of original energy. More than a third of original energy is lost during deceleration/acceleration. It pays to plan ahead, regen is there for unplanned situations.
 
I went back and reread the blog entry on regen. . . It was well written, well explained.

Then I went to read the comments, and it seems like several people still didn't understand. There may be some who will never "get" this until they experience it.

I like the idea of having regen controlled solely by the throttle and mechanical brakes controlled solely by the brake pedal. That's simple, that's predictable. You always know exactly what you are telling the car to do. And again, it shouldn't be much of a leap for anybody who has experience driving stick.

Yet, there are always a few who insist on putting a computer between the driver and the car, and letting the computer decide what kind of braking to use. Tsk. . . That may be okay in a Prius, which is a pretty sedate people-mover, but for a sports car it's just not right.
 
WarpedOne said:
This is only half the story. When those 80% of recovered energy is used again it goes through another conversion with 80% efficiency giving you total 64% percent of original energy.

Nah, for purposes of comparison it's pretty much the whole story.

I know that the conversion process is only 80% efficient in either direction, and both directions equal .80 times .80 = .64

What I was saying is that putting only 80% of the recovered kinetic energy back into the battery means that only 80% of deceleration is on the same footing as the original battery charge from the AC mains.  Both the original charge and the recovered charge pass through the  same 80% loss to accelerate the vehicle whether from recovery or coal fired power plant.

OK?
 
When Tesla mentioned "because the rear wheels don't provide majority braking, a maximum of about 80% of the regen opportunity can be realized."

I don't think they were talking about 80% efficiency of the regen system, but 80% efficiency of using rear wheel regen only (since more braking force typically is needed on the front wheels when slowing down).

So we have 80% * 80% = 64% efficiency when turning rear wheel slow down regen into battery power, then 80% of that back to the wheels for acceleration, so I think we are down to about 51% efficiency now.

In practice when comparing Toyota vehicles which have Hybrid and non-Hybrid variants, we found something like 30% improvement in city mileage (when using a comparable ICE with and without Hybrid assist). That is in part because the driving does involve some use of friction brakes which don't recapture any of the energy.
 
Regen

When Tesla mentioned "because the rear wheels don't provide majority braking, a maximum of about 80% of the regen opportunity can be realized."
QUOTE]

I'm sure I'm also missing something in that, if regen would be more effective from the front wheels, then could we say that approximately 60% of regen would be possible to recover from front wheel braking and 40% from rear wheel braking? Wouldn't this represent the maximum regen opportunity? If so, how does 80% fit this?
 
So we have 80% * 80% = 64% efficiency when turning rear wheel slow down regen into battery power, then 80% of that back to the wheels for acceleration, so I think we are down to about 51% efficiency now.

I somewhat disagree. Your numbers are worst case.

The 80% factor when referring to the rear-wheels counts only during traction limited "opportunities" where the deceleration rate exceeds the rear wheels' ability to provide the decel forces all by themselves. If one doesn't need the front wheels to supply any retarding forces to slow the car in a low-g deceleration, all the energy removed by slowing down can try to recharge the battery (less the 20% anti-perpetual-motion tax).

Tesla's statement is that only 80% of all the available opportunities to recover momentum will be below the rear-wheel-only traction limit (same as driving situation limit). The remaining 20% will be lost through the necessity of using the friction brakes as well as regen to slow the car in more extreme maneuvers.

When Tesla says that "because the rear wheels don't provide majority braking, a maximum of about 80% of the regen opportunity can be realized," they are talking about the whole universe of deceleration opportunities, from all-out panic stops to mildly "breathing" the throttle to slightly slow the car.

Slowing the car via regenerative motor braking can be a 100% opportunity right up to the point where the traction/stability controls intercede to maintain traction at the rear tires. The exact point will vary with road surface conditions, cornering loads, etc., that influence the surplus traction available (not needed for other vehicle control purposes) at the drive wheels. If one drives the car by planning ahead to not need the friction brakes and only using throttle lift to slow down, all driving done in this manner will be 100% regeneration capable. If one habitually uses the friction brakes for a majority of deceleration, the regen opportunities drop precipitously. It all has to do with the driver interacting with traffic and roadway configurations. Drivers that ride the friction brakes all the way down hills can expect very poor regen recovery.

This is actually an argument for regen programming to be as forceful (extreme) as can be safely tolerated by a wide scope of drivers.

Your Mileage May Vary,

Mike - Cincinnati
 
... Drivers that ride the friction brakes all the way down hills can expect very poor regen recovery.
This is actually an argument for regen programming to be as forceful (extreme) as can be safely tolerated by a wide scope of drivers.

Yes, I think that is close to the heart of the matter.

As I said on the Tesla blogs, driver habits from auto-trans ICE cars are not as conducive to maximizing regen. If you expect your vehicle to coast "off throttle", and only ever use the brake pedal to slow down, then much regen will be missed.

If, on the other hand, you are used to a high revving manual trans car and take advantage of engine braking then you will make better use of the roadster's regen system.

I suspect a problem that Tesla is having are that some customers want auto-trans like driving with minimal regen to allow coasting. Since the brake pedal doesn't control the regen it cuts into range. I think they must be trying to engineer some sort of compromise with a bit more "off throttle" regen than the "coasters" request, but less than the "engine brakers" would want.

=================================================================

Regarding the rear regen only... Traction conditions can dictate how much rear bias you can give when slowing down. I wonder if the Sweden tests had them find that too much rear regen was unsafe on snow/ice? I wonder if the changes in regen programming had anything to do with the reported range reduction.

I really don't know if it is safe to invoke heavy regen on the rear wheels up to the point where some sort of traction control has to intervene. I suspect that Tesla are becoming experts in this area based on their ongoing testing.

I still think my intuition could be correct that front wheel regen would be required to be able to maximize all the regen opportunities.
 
I think someone once mentioned on the Tesla blogs that the ESS is capable of capturing all the regen energy that they can feed it. You don't really need ultracaps for that.

There seem to be plenty of technologies that can charge & discharge fast enough to provide plenty of acceleration and regen. I think the hope is that ultracaps will someday help in these other areas:

#1: Lower weight
#2: Lower cost
#3: More durable (can be charged/discharged more times)

Until they provide a significant advantage in some of those categories they will still not replace chemical batteries.
 
I'm not an engineer, so I won't get into the discussion about weight distribution, ultra-caps, and such. But I do drive 50 miles a day on rural and urban roads. Since Greg Solberg's post, I've spent much of that commute bemoaning the fact that my gas tank doesn't fill back up when I slow down. I've decided how I would like the braking profile to feel in the WhiteStar or BlueStar (whichever I end up with). This is not appropriate for the Roadster because the Roadster is designed for performance.

First, my rationale. I drive-by-cruise. I keep a thumb on the cruise control and use it on any stretch of road more than a few blocks long. I also accelerate and decelerate using the cruise control. But cruise control has a couple of huge limitations. First, it doesn't work at low speeds. And second, there is no way to adjust the rate of acceleration (deceleration). When I can't use the cruise for whatever reason, I find I am constantly tapping the accelerator or brake to keep the car coasting at 55 (ahem) or keep it from idling into the vehicle in front of me. So what I want (me alone, I'm not saying this is right for the rest of the world) is to eliminate the cruise control and replace the accelerator and brake with an accelerator and decelorator.

A nod to my brother the engineer who insists that deceleration is acceleration because acceleration is any change in velocity. Duly noted. But I've got to differentiate the pedals somehow. To his point - both pedals will behave in the same manner with the only difference being the direction of the flow of energy.

Here's how I see it.

The position of the pedal determines the rate of acceleration. Push the accelerator to the floor and hold it; the car will accelerate to it's top speed as quickly as (practically) possible without losing traction and spinning it's tires. Push the decelerator to the floor and hold it; the care will slow to a stop as quickly as (practically) possible without losing traction and going into a skid. Take my foot off either pedal, and the car remains at a constant speed - a.k.a. no change in velocity.

Neither pedal directly controls the speed of the vehicle nor the direction of the flow of energy. Apologies to tonybelding, but I do want a computer between the pedals and the motor - we've got that anyway with all modern ICEs. I want the pedals to inform the system of my intentions and the system will then respond by sending power to or from the motor/generator as necessary. If I have my foot off the pedals and am rolling downhill at 35 mph, the car should remain at 35 mph, pulling energy off the wheels and putting it back in the battery.

Caveats, of course. First, there have to be mechanical friction brakes at the bottom end of the decelerator in case of system failure. Besides, I don't think I'd trust regen in an emergency stopping situation when I can't remember where the hand brake is. And I realize that the driving experience will be different from a non-electric vehicle. But I don't think this type of behavior would be so remarkably different that a person with two cars would have any problem adjusting.

--
Mark Tomlinson
 
Does this imply that we want two cruise settings (rather than more pedals) which work as one.

Constant Power Cruise, which works for flat sections and going up inclines (obviously at a lower speed than on the flat) the advantage here is maximizing range by keeping power consumption constant. However, when the car goes downhill the system flips over to....

Constant Speed Cruise, which maintains whatever speed you were used to on the flat sections of your journey and regens any extra energy back into the battery.

Alternatively, you could choose to select the Constant Speed Cruise for all gradients if you prefer not to travel more slowly uphill.
 
Does this imply that we want two cruise settings (rather than more pedals) which work as one.

Constant Power Cruise, which works for flat sections and going up inclines (obviously at a lower speed than on the flat) the advantage here is maximizing range by keeping power consumption constant. However, when the car goes downhill the system flips over to....

Constant Speed Cruise, which maintains whatever speed you were used to on the flat sections of your journey and regens any extra energy back into the battery.

Alternatively, you could choose to select the Constant Speed Cruise for all gradients if you prefer not to travel more slowly uphill.

Keeping in mind, of course, that I'm interested in the WhiteStar where safety and economy are weighted over performance, I'm not sure why two cruise settings would be needed. I would envision the vehicle as always being in "Constant Speed Cruise". Let the car decide how much power is running to or from the wheels depending on conditions. Uphill at 35; power running to the wheels to keep it moving in the right direction without slowing. Downhill at 35; power running from the wheels to keep it from picking up speed. Flat, dry pavement at 35; small amount of power, as needed, running to the wheels to keep up speed.

Changes in speed are determined by the length of time and position the foot is kept on a pedal. Falling behind on the freeway? Tap on the accelerator a couple of times to adjust your speed upwards. The drive train firmware then decides how to get you there - you may be going downhill already so the firmware decides you need less regen. Need an emergency stop? Slam your foot on the decelerator while screaming at the vehicle in front of you and leaving fingerprints in the steering wheel. There's friction brakes at that end of the decelerator. But nonetheless, the firmware is sucking as much power as it can get off the wheels without sending you into a skid.