Regen losses mean it is best to coast rather than regen with the brakes.

[Recumpence]

Hello All,

 

I wanted to start this thread to give a very brief explanation of regen versus coasting.

 

First, it is extremely important to remember that a 100% brake score does not mean you recouped 100% of the kinetic energy and converted that to battery storage. What it means is 100% of all of our slowing was achieved through regen and 0% was from the friction brakes.

 

Second, when using regen, there are many losses convering the kinetic energy into electrical energy in the battery. There are losses through the motor/generator, losses through the inverter, losses through the transmission, tires, CV axles, wiring, and battery internal resistance. The motor/generator loss is in the range of 15% to 20% most likely. The inverter would be similar loss (maybe 10% to 15%). The transmission loss will be a minimum of 10% (probably much more). The IR of the battery and wiring adds more losses (though probably a lower % of the other losses).

 

Third, you have all those same losses when turning that electrical energy back into kinetic energy for accelleration when the light runs green again.

 

I have heard it said in electric car forums that only about 30% of your kinetic energy is converted to electrical energy. This seems deplorable. But consider, without regen, 100% of your energy is wasted as heat when you brake. So, a 30% return is huge compared to 0%.

 

The take-away is this;

 

It is generally better to coast longer and regen less if you are truly looking for the best FE possible. There are exceptions to that rule, but in general, coasting is far better than running farther up to a stop, then regenning more.

 

Matt

Edited July 6, 2013 by Recumpence

[viennacoup]

The battery display shows the battery charging during both coasting and regen.  The only way regen could have an advantage over coasting would be if it charged the battery at a much greater rate than mere coasting.  Personally, I have reached the conclusion that it is inefficient to give up any momentum using the brake pedal.  I am a coaster.

[plus 3 golfer]

Yes, I've said this before. Coast, coast, and coast more.  From a previous post of mine:

 

Several times I've recorded the SOC while achieving 100% brake score.  I thought I reported this before in posts but can't locate it.  IIRC, a rough calculation showed that the change in SOC was capturing only 50 - 60 % of the initial KE + PE at the start of the regen braking until the stop. There's drag + drivetrain losses + conversion losses + load on the HV battery that affects the estimate.  

I recorded speed, time, SOC (for numerous stops from about 60 mph to 35 mph) at a frequency of about 2/3 second intervals and used topo maps, gps to adjust for elevation change (even a small change in elevation has a good bit of potential energy to be captured).

 

Of course, we don't know the exact conversion efficiencies of every component but my rough estimate was that only about 50-60% of the total energy of the car was recovered and usable to propel the car when getting 100% brake score vs coasting. This was based on actual measured change of SOC when storing the energy through braking and then assuming that the losses associated with use of the stored energy would be about the same as when storing the energy.

 

Even if one used conservative estimates of the efficiency of 0.95 for the major components like motor / generator, the inverter and the drivetrain, one would get an overall effective efficiency of about 73% = (0.95)^6  Drop the efficiencies down to 92.5% and the overall efficiency drops to 61%. If each major component were 90% efficient, the overall efficiency would be 53%.

 

Here's one of the recordings I used and posted previously when testing grill covers. 

 

 

 

 

[MtnMarty]

Near my home there's a long downhill to a stop sign. I always use the regen brake lightly on that hill since I will have to come to a complete stop in any case. Even along this about 400 yard stretch to the stop sign I am able to charge the battery quite a bit. And since this hill is only around the corner from my home, I usually need to recharge the battery anyway due to EV+ the trip before. Otherwise, I agree completely...coast as much as possible.

[C-Maxgo]

Thank you for the very informative information Recumpence.   :worship: 

[Sparky]

It is generally better to coast longer and regen less if you are truly looking for the best FE possible. There are exceptions to that rule, but in general, coasting is far better than running farther up to a stop, then regenning more.

 

Matt

When you say "coasting" do you mean coasting with the transmission in neutral (N) or coasting in drive (D)?  Coasting in drive (without any brakes) is using regen and charging the battery.   This condition also causes the car to decelerate more than coasting in neutral. 

 

I live near a 3 mile section of freeway that is a slight downgrade.  In drive I must apply slight power to maintain the speed limit.  In neutral the car will coast at the speed limit for the entire 3 miles. 

[Recumpence]

Yes, then the same losses exsit when accellerating. So, roughly 30% of the energy from cruising is captured and returned back into kinetic energy through regen and accellerating again.

 

Matt

 

 

Yes, I've said this before. Coast, coast, and coast more.  From a previous post of mine:

 

 

I recorded speed, time, SOC (for numerous stops from about 60 mph to 35 mph) at a frequency of about 2/3 second intervals and used topo maps, gps to adjust for elevation change (even a small change in elevation has a good bit of potential energy to be captured).

 

Of course, we don't know the exact conversion efficiencies of every component but my rough estimate was that only about 50-60% of the total energy of the car was recovered and usable to propel the car when getting 100% brake score vs coasting. This was based on actual measured change of SOC when storing the energy through braking and then assuming that the losses associated with use of the stored energy would be about the same as when storing the energy.

 

Even if one used conservative estimates of the efficiency of 0.95 for the major components like motor / generator, the inverter and the drivetrain, one would get an overall effective efficiency of about 73% = (0.95)^6  Drop the efficiencies down to 92.5% and the overall efficiency drops to 61%. If each major component were 90% efficient, the overall efficiency would be 53%.

 

Here's one of the recordings I used and posted previously when testing grill covers. 

[KAL Cmax]

If you are in neutral you do not have any regen at all.

 

The reason we have to give it some gas downhill is to "bypass/disengage" the regen stage. The C-Max automatically regens when you let off the gas.

[MtnMarty]

Wow...I've learned a lot here and this is an example. Who knew you got regen while coasting....

[plus 3 golfer]

In neutral, the output shaft is disengaged from the wheels and no power flows through the transmission..The electric motor does not provide power to or hold the final drive and the final drive can spin freely.

 

Coasting to a stop in neutral or drive is better than braking to a stop. My 50-60% efficiency assumes coasting in neutral vs braking and allows drag to slow the car down. But coasting in neutral is against the law in many (probably most if not all) states as one loses an important element of control - throttle. The regen load when coasting in gear is suppose to simulate engine braking.

 

The issue I have with coasting in neutral in "normal" driving is that it is not practical as it takes way too long to slow down and pi$$es off the drivers behind and adds a significant amount of time to ones trip. Coasting in drive while not as efficient as neutral does add some time to the trip but usually doesn't pi$$ drivers off as frequently. But coasting in drive allows more of the drag load to slow down the car than by applying the brakes (where one introduces more drivetrain and conversion losses into the mix).

 

If I had the time and desire, I could easily run tests to record data and estimate the difference in energy of coasting in neutral vs drive. :) But since I don't coast in neutral (and am not about to start), I have no use for the results. ;)

 

One point I forgot is that when in neutral one has no regenerative braking. So, if you have to brake one will want to shift back to drive to capture some of the energy instead of using the friction brakes.

Edited July 6, 2013 by Plus 3 Golfer

[salsaguy]

there is also some loss due to heat from the battery. batteries work better when cooler not hot.

[fotomoto]

 

First, it is extremely important to remember that a 100% brake score does not mean you recouped 100% of the kinetic energy and converted that to battery storage. What it means is 100% of all of our slowing was achieved through regen and 0% was from the friction brakes.

 

 

I haven't seen the info on the CMax yet but all the other hybrids I've owned had a transition from electric regen to friction brakes around the <10mph zone.  Holding the car at a stop via the electric side of the hybrid equation would be incredibly inefficient.

[bigalpha]

Are you saying it's better to coast to slow down so you don't have to accelerate from a stop, even if you regen that energy from braking?

[Jus-A-CMax]

Coasting, whats that? :shift:

[plus 3 golfer]

Are you saying it's better to coast to slow down so you don't have to accelerate from a stop, even if you regen that energy from braking?

If I understand you question correctly, the answer is YES because conserving the kinetic energy in the car is better than regeneration. Coasting can be done "in gear" or in "neutral" (freewheeling).  It's always better to coast if one has to slow down than to apply the brakes to slow down.  This is because more of the stored energy in your car (kinetic energy) will be used to overcome the drag forces on the car than converted to electrical energy which involves a considerable efficiency loss during the conversion process.  If one has to come to a complete stop, one has to apply the brake (more regen braking and eventually friction braking). Otherwise, the car will continue to move forward.

 

In Ford's videos on hybrids, they say the car captures 90% of the available braking energy AND then sends it back to the battery to be stored.  So, 0.90 is likely the efficiency of the traction motor, the inverter is likely around 95% efficient (large PV solar inverters are around 96 / 97%).  Li-on battery efficiency is around 98%. I would assume the I²R losses in the cabling to be less than 1%. Since during regenerative braking the traction motor (usually referred to as MG2) is used and speed is generally slower than cruising, drivetrain losses vs freewheeling for the e-cvt transmission is likely very small so let's use 5% (this could be included in the 90% Ford states in the video).  So, a better guess at the overall regen braking efficiency vs coasting (freewheeling) might be between (0.9x0.95x0.95x0.99x0.98x0.99x0.95x0.95x0.9) = 57% and 63 % removing the drivetrain losses.  Again, this is very close to my estimate based from recording SOC and using speed to calculate KE while braking and getting 100% brake score.  The exact number is not material.  The point is that there are considerable losses in regenerative braking.  On battery temperature affecting these losses, the presumption is that the total conversion process (from charging to discharging) takes place over minutes and thus battery temperature remains very stable.

 

So, if one coasts in gear but does not apply the brakes, the regeneration efficiency would still be around 60% (maybe less if there are more drivetrain losses).  But, if one applies the brakes, one slows down quicker because a larger % of the kinetic energy (KE) is being converted to electrical energy over time.  In essence, the additional kinetic energy that was converted to electrical energy by applying the brakes could have been used for overcoming drag with "no regen efficiency" losses when continuing to slow down or accelerating again.  So, by starting ones coasting in gear earlier, one maximizes the KE and improves FE.  This is why it makes sense to try to time traffic lights by coasting and not have to stop or virtually coasting to a stop.  Yes, friction brakes will eventually grab below a certain speed which to me seems to be in the 5-7 mph range but there's not much KE left in the vehicle below 10 mph.  KE is a functional of the square of the speed (KE = !/2x(Mass)x(velocity)².  So at 20 mph, there's 4 times as much KE than at 10 mph.

 

Coast, coast and coast more for better FE (time and driver following permitting). :)

[bigalpha]

Holy crap, thanks for the dissertation!   I usually already coast to lights as much as possible for this very reason.  I've been doing it for years in my other cars too.  When coasting, is the car regen-ing electricity for the battery essentially the same way as the brakes do, just starting later in the chain of components?  If so, then we bypass some of the extra inefficiencies, right?

[fotomoto]

  When coasting, is the car regen-ing electricity for the battery essentially the same way as the brakes do, just starting later in the chain of components?  If so, then we bypass some of the extra inefficiencies, right?

 

You're confused by the misnomer term of "regenerative brakes".    Coasting/slowing down is accomplished by having the electric motor reverse and work against the drivetrain which causes two things: driveline drag and electrical charge (the electric motor is now a generator).  This all happens inside the transaxle not on the wheels or brake rotors, etc. the way traditional brakes function.   Adding light brake pedal simply makes the motor/generator work harder and thus deliver more regen charge back into the battery.  Add even more brake pedal pressure (hard braking) and the system will blend into traditional friction brakes.  

 

http://www.myfordmag.com/askford/regenerative-braking

Edited July 7, 2013 by fotomoto

[ScubaDadMiami]

First, it is extremely important to remember that a 100% brake score does not mean you recouped 100% of the kinetic energy and converted that to battery storage. What it means is 100% of all of our slowing was achieved through regen and 0% was from the friction brakes.

I thought that it just represented an efficiency percentage, starting at 100 percent of the maximum potential energy that could be captured, and then declining from there.  Friction brakes can also be applied, though gently, arriving at the gauge still saying 100 percent, because the light braking didn't result in a decline in the amount of charge that the system was capable, as constructed, of capturing.

 

Am I mistaken?

[ScubaDadMiami]

I haven't seen the info on the CMax yet but all the other hybrids I've owned had a transition from electric regen to friction brakes around the <10mph zone.  Holding the car at a stop via the electric side of the hybrid equation would be incredibly inefficient.

That seems to be what is going on, if the brake regen display shows what actually is going on.  Being a coaster, there are many times that I end out having to lightly apply the brakes as I drift up to a light or stop sign.  I notice that the regen circle disappears, and the charge arrow changes to point downward when the C-MAX slows to almost stopping.  It is somewhere around 10 mph, but I am not exactly sure of the speed.

 

I just wasn't sure if it is the way that the display shows, or if it reflects what is actually going on. 

[hybridbear]

First, it is extremely important to remember that a 100% brake score does not mean you recouped 100% of the kinetic energy and converted that to battery storage. What it means is 100% of all of our slowing was achieved through regen and 0% was from the friction brakes.

This is incorrect. The brake score measures the % of the potential energy that could be recovered that you actually recovered. The friction brakes engage at lower speeds. The electric motor has limits to its performance and it is not used to slow the car down at low speeds because the kinetic energy is likely too low to be able to recover much based on the charging losses discussed above.

 

I thought that it just represented an efficiency percentage, starting at 100 percent of the maximum potential energy that could be captured, and then declining from there.  Friction brakes can also be applied, though gently, arriving at the gauge still saying 100 percent, because the light braking didn't result in a decline in the amount of charge that the system was capable, as constructed, of capturing.

 

Am I mistaken?

You're exactly right

[plus 3 golfer]

Focusing on a precise definition of brake score is not the intent of the thread.  What matters are the losses associated with capturing and reusing the captured energy from regenerative braking vs coasting either in neutral or in drive.  If you have to use the brakes to slow down from moderate speeds, shoot for 100% brake score.  Otherwise, shoot for "no brake score" by coasting or 100% by coasting and only applying the brakes just prior to stopping. That's the message I think recumpence is trying to make.

 

Here's a few ways Ford has described regenerative braking and such seems to be consistent every time I find something on Ford and regenerative braking.

 

 

This remarkable technology helps you achieve maximum energy-efficiency. It captures more than 90 percent of the energy normally lost as heat during braking, and recycles it to recharge the battery.

....

Brake Coach

This feature, found in the instrument cluster screen, helps you optimize the use of the car's regenerative brakes. It displays a graphic that shows the amount of energy captured each time you stop, so you can make adjustments to help maximize your range.

 

So, I assume as one begins braking and as long as the friction brakes are not applied but for the last part of slowdown before completely stopping, one gets 100% brake score - that's easy.  But what algorithm does Ford use when the friction brakes are applied during the slow down process.  The applicable data could be captured continuously and thus a very reliable estimate of the amount of energy captured each time one applies the brakes could be computed once the car comes to a stop OR Ford could take short cuts and do a "ballpark" estimate. :)  So how close is a score of 75% to capturing 75% of the available energy that would be lost to friction braking but for regen braking.

 

I'm interested in knowing what the word "captured" means and where is the captured energy "measured".   IMO, to use the word "captured", the measurement has to be at the output of the traction motor. Otherwise, no energy is "captured".  If the field winding are open, no energy is captured.  This makes sense as Ford states they capture over 90% of the energy normally lost in heat and recycle it to the HV battery.  This would indicate a traction motor efficiency of about 90+%.

 

[plus 3 golfer]

After doing searches / reading papers on regen braking.  It appears that regen braking strategy may employ use of the front or rear friction brakes to accomplish a "safe and smooth braking feeling" comparable to cars without regen braking especially when no throttle is applied to simulate engine braking and also when the brake pedal is depressed as opposed to simply using the traction motor 100% of the time.   Also, I found this efficiency curve below for an electric traction motor.  It shows a max.efficiency of about 92%.

 

So again, do we really know what 100% brake score means. I would assume that since this is trying to provide a measure for the driver, that friction braking embedded in the control algorithm would be excluded and only friction braking initiated by the driver would be detrimental to the brake score.

 

Also, regen braking strategies were road tested in one paper and the results of two are shown below.  I believe the regenerated energy is the energy used to propel the car and thus the overall efficiency of the regen braking was about 52%.  Of course the C-Max could be different. :) 

 

 

Control strategy------------ Recoverable energy (kJ) -------Regenerated energy (kJ) -------  Regeneration efficiency (%)

 

Good-pedal-feel strategy ---------80.253-------------------------- 37.846---------------------------------- 47.16

Coordination strategy-------------- 84.015 -------------------------43.342----------------------------------- 52.32

 

[fotomoto]

Well one thing I do know is the ford engineers got the transition RIGHT.  Damn if I can tell when it happens; very smooth.   :thumbsup:  :salute:

[plus 3 golfer]

Well one thing I do know is the ford engineers got the transition RIGHT.  Damn if I can tell when it happens; very smooth.   :thumbsup:  :salute:

Your right, very smooth. I didn't give any thought to use of the friction brakes but it makes sense. Think about having the rear wheels cut out going around a corner on icy / wet roads if you would simply use the traction motor to brake the front wheels when one applied the brake pedal to max regen braking.

[viajero]

In neutral, the output shaft is disengaged from the wheels and no power flows through the transmission..The electric motor does not provide power to or hold the final drive and the final drive can spin freely.

 

I don't think this is strictly true.  Based on various documents on Prius forums, the motor/generators are still geared to the wheels and still spin when in what the shifter calls "neutral".  The motor/generators are electrically disconnected, so they offer no extra torque or drag.  I'm being kind of pedantic here as the end result is essentially the same.

 

Theoretically, you could achieve the same thing when in "drive" gear by feathering the gas pedal just right so that the battery was neither charging nor discharging.  In practice it would be hard to achieve that perfect balance, and I wouldn't want to pay that much attention, anyway.

 

The counterintuitive thing about a power split hybrid is that nothing mechanical ever happens when you change between N, D, L, and even R.  All the gears are still connected exactly the same in all four cases.  The gearshift just tells the computer how to feed electrical energy in or out of the two motor/generators.  Even when you select reverse, the motor is still geared to the wheels the same as in drive.  The computer just changes the phasing of the AC power and causes the big motor/generator to turn backwards.