1. Attachments are working again! Check out this thread for more details and to report any other bugs.

How much energy does regnerative breaking save?

Discussion in 'Gen 2 Prius Main Forum' started by tom1l21, May 16, 2009.

  1. tom1l21

    tom1l21 Member

    Joined:
    Mar 24, 2009
    268
    13
    0
    Location:
    East Longmeadow, MA
    Vehicle:
    2005 Prius
    I was wondering what the ratio of miles generated from a gallon of E10 gas as well as the energy recouped from regenerative sic[coasting] is? Anyone have any ideas? I threw around some calculations and came up with a rough estimate of 13%, but I think it is actually half of that b/c NiMH battery is about 50-60% efficient (not sure if this is true or already calculated for when the MFD gives you the energy recovered in kwh).
     
  2. john1701a

    john1701a Prius Guru

    Joined:
    Jan 6, 2004
    12,747
    5,243
    57
    Location:
    Minnesota
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Advanced
    You'll discover that you don't actually use the brakes as often.

    With a traditional vehicle, brakes are really the only way to slow down since the transmission is designed to reduce drag.

    Prius isn't designed like that. When you lift your foot off the accelerator-pedal, the smaller of the two electric motors engages and generates electricity. The resulting drag slows the car down, without pressing the brake. RE-generating doesn't occur until later when you do. That switches from the small motor to the big one.
    .
     
  3. tom1l21

    tom1l21 Member

    Joined:
    Mar 24, 2009
    268
    13
    0
    Location:
    East Longmeadow, MA
    Vehicle:
    2005 Prius

    I keep calling it regenerative braking, I actually meant regenerative coasting. Does that change anything?
     
  4. efusco

    efusco Moderator Emeritus
    Staff Member

    Joined:
    Nov 26, 2003
    19,891
    1,191
    9
    Location:
    Nixa, MO
    Vehicle:
    2004 Prius
    Model:
    N/A
    Technically, any regen ultimately costs you energy..but compared to a car with friction brakes it does recapture some. I've seen numbers like 2-4% tossed about, but I don't know that that means much. If you're on a highway trip and don't use the brakes except when you stop your trip you'll have recaptured but a very very tiny fraction of the energy used to get you to your destination.

    If you are on a city trip and accelerate then stop at every block you'll recapture a much higher percentage of energy.

    But it all goes back to driving the car as if you have no brakes at all, anticipate traffic slowing, stops, turns, etc. so you don't need to touch the brakes much if at all...that will save you more energy than any regen.
     
  5. tom1l21

    tom1l21 Member

    Joined:
    Mar 24, 2009
    268
    13
    0
    Location:
    East Longmeadow, MA
    Vehicle:
    2005 Prius
    Yeah, of course there are so many factors which will determine how much energy you recapture such as your driving style as well as the terrain. I modeled it after a rolling hill commute that I have which is 50mi round trip. Driving with hypermiling in mind (P+G, gentle stops to lights/stop signs, and coasting down hills, I calculated around 15 green 50Wh cars on the MFD over 50miles at an average of 50mpg. Since the numbers worked out pretty well with 1gallon of of gas for the round trip (50 miles @ 50mpg), I calculated the energy in 1 gallon of e10 and checked the efficiency of an Otto cycle and found it to be around 35% and did a rough estimate of 25% efficiency which includes the ICE efficiency minus Cd as well as Fr (rolling friction). Again, a very rough estimate. This number came out to about 31.75MJ per gallon which compared to 4.32MJ recovered from regenerative coasting+braking(15*50Wh). This ratio of 4.32/31.75 is about 13.6% but this is where I wasn't sure if you have to multiply 4.32MJ by an efficiency of NiMH which I found to be about 50-60%, therefore the overall ratio was 13.6%/2 or around 6-7%.

    Not sure if I am way off base, but seems that it could be pretty reasonable. So out of the 50mi during my commute which used exactly 1 gallon of E10, around 3 mi were provided soley on energy stored from the regenerative braking and coasting.
     
  6. JimboK

    JimboK One owner, low mileage

    Joined:
    May 1, 2006
    2,817
    187
    49
    Location:
    Chesterfield, VA
    Vehicle:
    2005 Prius
    It doesn't address your question (I'll just have to take your word on your calculations), but I can offer this to reinforce Evan's point that completely avoiding regeneration saves more energy than regeneration does. (Not that regen "saves" any anyway -- it just allows less to be wasted. "Recapture" might be a better term. But I digress.)

    These are from a recent morning commute:

    [​IMG]

    [​IMG]

    I drive to avoid regeneration as much as possible. Still, some regeneration is inevitable -- I have 13 lights and 3 stop signs in a 13 mile commute -- and an MFD completely free of regen icons is very unusual. Even here, there was some regen, apparently insufficient to register on the screen. But I choose a route to maximize pulse and glide opportunities (and minimize regen), I've learned the timing of the lights, and I leave home a few minutes earlier than necessary when traffic is lighter.

    Now, if you learn to drive like this, imagine all the time you'll save on the calculator by rendering all your math irrelevant. ;)
     
  7. Matt Herring

    Matt Herring New Member

    Joined:
    Apr 28, 2009
    421
    51
    6
    Location:
    North Andover, MA
    Vehicle:
    2010 Prius
    Model:
    II
    Jimbo,

    Nice results...I'm in the 61-62 mpg range (minus a 800 round trip highway commute that bore poor results for me...plenty of work to do)...you set the bar pretty high.
     
  8. tom1l21

    tom1l21 Member

    Joined:
    Mar 24, 2009
    268
    13
    0
    Location:
    East Longmeadow, MA
    Vehicle:
    2005 Prius
    Yeah I do realize that limiting regen coasting and breaking is the way to go to maximize mpg bc there are energy losses with conversions. I guess I was doing the calculation with the best case scenario where P+G is at its best. There are some very steep hills so there are portions where I have to brake to stay within the speed limit. Also, you can't roll up to a stop sign at 5mph when its a heavy commuter route :)
     
  9. donee

    donee New Member

    Joined:
    Aug 15, 2005
    2,956
    197
    0
    Location:
    Chicagoland
    Vehicle:
    2010 Prius
    Model:
    III
    Hi Tom...,

    In suburban (non-highway) driving, it depends on how hilly your terrain is. If its very hilly, it will save allot of energy, because downhill the battery soaks up a bunch of energy.

    The actual regeneration efficiency is 85% or so. But reusing the battery for accelleration is so poor that the overall reuse efficiency is only in the 40's % in that scenario. If you use the battery for cresting shallow hills, or cruising along sub 40 mph or moving the car around like in a parking lot, the efficiency of reuse will be much greater.

    You will not that is how the car is programmed too. Put any kinda foot into the accelerator pedal, and on pops the engine.
     
  10. SageBrush

    SageBrush Senior Member

    Joined:
    Jun 4, 2008
    11,627
    2,530
    8
    Location:
    Southwest Colorado
    Vehicle:
    2012 Prius v wagon
    Model:
    Two
    I don't anybody outside of Toyota (or maybe hobbit) has a clear idea how much energy is saved in regen braking vs friction braking. Just because wear at the brake is not occurring does not mean all (or even most) of that energy is going into the battery.

    If you are up for a bit of math, you can calculate as an integral the friction work the car is subjected to in slowing from say 40 to 0 on a level road with no arrows on the MFD. That result can be used to isolate the work the regen system pulls out of the kinetic energy the car disposed of when slowing from 40 to 0 with the foot off the brakes or go-pedal. Lastly, certain gauges read the battery SOC directly.
     
  11. EZW1

    EZW1 Active Member

    Joined:
    Aug 20, 2008
    722
    80
    7
    Location:
    Phoenix, AZ
    Vehicle:
    2016 Prius v wagon
    Model:
    Four Touring
    I know this much from experience: When I have to brake, I do but I gently press the pedal early to start the regen process. The regen process charges the battery much faster than coasting. Here in AZ we use the A/C almost all the time. The A/C draws the battery down quite fast which causes the ICE to run much more and the MPG to drop. Regen braking helps recharge the battery faster and reach the 2 red bars far less often than I would if I don't regen.
     
  12. KTPhil

    KTPhil Active Member

    Joined:
    Jan 14, 2005
    1,379
    20
    0
    I figure battery losses both charging and discharging only leave about 30% max energy capture. This was analyzed numerically a couple of years back on the forums and this was the consensus of the max ability. It is significant, and is partly the cause of the higher mpg in the city than on the highway.
     
  13. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    What is the basis for that 40%? Have you got example efficiency of the stages to illustrate? If the regen was 85% efficiency that would mean 47% efficiency for the rest to get 40% overall.

    When I've done electric acceleration from a stop and long EV glides on the flat (1 to 1.5 miles) the apparent efficiency has seemed higher than that. Granted that any regeneration/re-use path will result in efficiency below 1.0 (100%), so it is to be avoided all else being equal.

    Battery temperature during regen is also a consideration. A considerable factor in Winter is the loss going into heating the battery from what I can tell looking at Ken's graphs and explanation. Drives me nuts when the car is apparently eager to force charge the battery well into the green just to get the battery to temp.
     
  14. donee

    donee New Member

    Joined:
    Aug 15, 2005
    2,956
    197
    0
    Location:
    Chicagoland
    Vehicle:
    2010 Prius
    Model:
    III
    Hi Shawn,

    Attila has a website where he recorded braking and electric accelleration currents and voltages. I used those numbers and calculated the 1/2 MdeltaV^2 energies. Its on a Prius Chat post somewhere. I said in the "40's". That was I remembered off the top of my head. Which means anywhere from 40 to 49 % . I will look for it after work.
     
  15. grand total

    grand total Member

    Joined:
    Sep 9, 2007
    154
    14
    0
    Location:
    ON, Canada
    Vehicle:
    Other Hybrid
    Model:
    N/A
    Prius engine uses Atkinson cycle not Otto. Maybe need to modify your calculation.
     
  16. donee

    donee New Member

    Joined:
    Aug 15, 2005
    2,956
    197
    0
    Location:
    Chicagoland
    Vehicle:
    2010 Prius
    Model:
    III
    Hi Tom,

    Not sure where you are getting that 35 % for a Otto engine. A Turbo diesel is with the typical 20:1 compression and expansion ratio is on the order of 40 %. The Prius Atkinson engine with the 13:1 expansion ratio varies between 20 and 35 % depending on load. Argonne tested it at 25 % efficient at 12.5 hp which is outstanding for such a low-load. My understanding (may be wrong) is the best a non-turbo charged Otto engine does is like 25 % over a narrow load and RPM range, and at low load they are as bad as 10 %. Note that the Otto engine is worse than half as efficient as the Prius Atkinson/Miller engine at low load. If you operated the Otto engine at optimum load and RPM, your car would be doing 90 mph before long. And you would waste that efficiency gain in heating the air.

    Its this low-load poor efficiency of Otto engines that makes battery recovery of energy worth-while. This is the whole purpose of a Hybrid drivetrain. At least, that was what I was taught by my professor in 1979.
     
  17. donee

    donee New Member

    Joined:
    Aug 15, 2005
    2,956
    197
    0
    Location:
    Chicagoland
    Vehicle:
    2010 Prius
    Model:
    III
    Hi All,

    Here is a link to the thread about Prius regeneration from Attila's measurements. Its on Prius online:

    PriusOnline.com • View topic - Regeneration

    The numbers indicate 81.5 % energy recovery from optimum braking by the electric action of the hybrid system. And 66.6 % of the electricty out of the the battery makes it into car motion using electric accelleration. With battery storage of NiMH batteries typically at 85% effiiciency, the overall efficiency of reuse is about 46.2 %.
     
  18. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    donee,

    Thanks for that link and analysis.

    I'm not sure how the numbers would come out in all this, but noticed something odd in the acceleration data (graph): the amperage goes negative at between 42 and 43 km/hr. Yet the velocity continues to increase, so I'm wondering if everything between 42 and 50 km/hr in the acceleration phase should be discarded--did the ICE turn on? This would put the acceleration and deceleration quantities on different velocity basis (which can be accounted for), but would eliminate seeming incoherent data.

    How is the overhead of the electrical system of the car accounted for in your calcs? Is the baseline VA subtracted out? (Looks to be about 1.8 A @ 230 V.) This won't be large, but it should increase the overall efficiency by 3-4% if I'm eyeballing it correctly.

    Does the kinetic energy of the rolling wheel/tire figure in anywhere? (One needs an estimate of moment of inertia for this.) I'm not sure that it will change the values in each direction much at these speeds, but without it the higher speed kinetic energy of the car is slightly higher than shown. Without it the acceleration efficiency value will have a negative bias (too low) while the regeneration efficiency will have a positive bias (too high.) So this is will be most important to the relative values. Overall I wouldn't be surprised if it neatly offset the system operating power requirement.

    I'm not sure what to make of the SOC...how can that be converted into kwh so that it can be included, perhaps as a crude check on the battery efficiency assumption?

    It's too bad that a repeat wasn't done in the other direction. That would theoretically eliminate any minor elevation effect that might be throwing off one value or the other.

    Hmm...in thinking about this some more, the rolling resistance and aero losses should also be subtracted out. The losses are real, but have nothing to do with the efficiency of the energy recovery as they would be present even if it was ICE acceleration and friction brake deceleration.
     
  19. ken1784

    ken1784 SuperMID designer

    Joined:
    Dec 18, 2003
    2,940
    1,359
    67
    Location:
    Yokohama, JAPAN
    Vehicle:
    2010 Prius
    Model:
    N/A
  20. Celtic Blue

    Celtic Blue New Member

    Joined:
    Aug 4, 2008
    2,224
    139
    0
    Location:
    Midwest
    Vehicle:
    2008 Prius
    Ken,

    I see a problem/disconnect between the way the test was done, and what it would mean in real world braking. In the real world one seeks to maximize energy conserved/recovered vs. expended for a given distance, not speed or time. So rather than doing a test only of energy expended/conserved over differing time periods, one should be testing increasing braking force over a fixed distance. The quantity regenerated perhaps matters less than the total expended over the time.

    For example, if I need to stop 1/4 mile ahead I could glide until near the end and brake harder. Or I could start more gentle braking sooner. The distance covered would be the same. This would account for all of the other losses to identify the energy optimum.

    There are several losses that will skew the results of a test that does not control for distance. The ones that come to mind first are: 1. Aero (a loss than over a given distance increases linearly with speed) 2. Rolling resistance (function of time...so it will have more effect at lower speed.) 3. Electrical system overhead (function of elapsed time.)

    It occurs to me that for this sort of testing, one would really only need to determine the effective watt-hour change in the battery charge for EV only. However, I question how accurate of a measurement SOC might be. Afterall, it is trying to measure the product of an electrochemical state which will be dependent on localized temperatures, etc. I would expect it to have considerable lag compared to voltage or amperage which should be near instantaneous. So one might wait an additional 30 seconds or more stopped to determine a "final" SOC value--but this reference idle time should be fixed for all tests.

    And of course, differing starting speeds will likely give different optimums.