TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Toyota Prius braking problems

Discussion in 'Off Topic' started by stopcrazypp, Feb 4, 2010.

  1. doug

    doug Administrator / Head Moderator

    Joined:
    Nov 28, 2006
    Messages:
    15,914
    Location:
    Stanford, California
    True in that in involves how the software deals with the interaction of the regen and friction brakes with the ABS.

    I believe, however, that the Roadster is performing as designed. That doesn't mean it's necessarily the optimal solution, though.
     
  2. stopcrazypp

    stopcrazypp Well-Known Member

    Joined:
    Dec 8, 2007
    Messages:
    7,047
    Yes, it seems like the system is supposed to perform this way. The only complication that the Prius and the Tesla adds is regen braking, but it should still be managable.

    Also I found a similar complaint for the WRX: it was found not to be a design/safety fault by the NHTSA, and Subaru release a similar ECU recalibration to mask the sensation of braking loss.
    http://www.scoobymods.com/02_03_mt_wrx-t5701.html
     
  3. Tdave

    Tdave Member

    Joined:
    May 16, 2008
    Messages:
    721
    Location:
    Columbia, MD
    I'm really glad the Roadster associates the regen with the accelerator, not the brake. Why aren't these other manufacturers doing it that way? It's easier to design, and easier to drive.
     
  4. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    I don't understand what you're saying here...could you elaborate?

    Since this is a more advanced audience I thought I'd explain, a little, why this braking transition issue is more noticeable on the 2010.

    With the Prius 2G the regen maxed out at about -100 amps, that, apparently, has been boosted up to around -125 amps (maybe more?) in the Prius 3G (at least the ones sold before January). Thus, the stronger regen gave the benefit of more pronounced braking before the friction brakes kick in. That's great, except that when you hit a bump and the skid control ECU senses a wheel spin differential or possible loss of traction it transitions to friction...that transition takes about 1/2 second...since you were braking pretty hard with regen, then the regen suddenly stops before friction kicks in that decrease in deceleration rate becomes very noticeable...even some experienced 2G Prius drivers have described the sensation as startling or unsettling.

    I'm not convinced that this transition, unsettling though it might be, causes any significant increase in braking distance or that it's a safety hazard (and thus shouldn't be a "recall"), but I think it's unsettling enough and in light of the current accelerator recall hysteria that Toyota will be forced to have a recall to reduce maximum regen amperage and thereby reduce the transition impact.
     
  5. Tdave

    Tdave Member

    Joined:
    May 16, 2008
    Messages:
    721
    Location:
    Columbia, MD
    #6 Tdave, Feb 5, 2010
    Last edited: Feb 5, 2010
    You elaborated well on the Prius issue. The Ford Fusion has a similar issue with the regen/friction braking transition, according to reports I've read yesterday/today.

    My point was that those vehicles have an additional technical design hurdle to overcome because they tie both the regen and friction braking to the brake pedal, and need to control both with that one control pedal, including the transition between the two.

    The Roadster in contrast ties the regen to accelerator lift, and the brake pedal is solely used for friction brakes. This is a more elegant design, easier to engineer, and has operator advantages most of us here appreciate.
     
  6. TEG

    TEG TMC Moderator

    Joined:
    Aug 20, 2006
    Messages:
    17,252
    Location:
    Silicon Valley
    This situation was discussed at length in the past on this forum and the TM blogs. Rather than dredge up those old posts, here is a brief summary from my perspective:

    It boils down to expected accelerator pedal behavior.

    In a high end sportcar like the Roadster (e.g.: a Ferrari), many people are used to a high revving performance engine with a manual transmission where letting up on the go pedal produces strong engine braking and associated slow down. So the Roadster can get away with fairly strong regen just by lifting the "torque pedal" which is familiar behavior to manual transmission sportscar drivers.

    A car like the Prius, on the other hand, is trying to attract family car drivers who might otherwise be used to something like an automatic transmission Camry or Taurus. Drivers used to automatic transmissions don't expect there to be engine braking when they ease up on the accelerator pedal.

    I am sure Toyota could have engineered the Prius to do regen on accelerator pedal lift up, but too many customers would have found that unexpected/unusual/unfamilliair.
     
  7. Tdave

    Tdave Member

    Joined:
    May 16, 2008
    Messages:
    721
    Location:
    Columbia, MD
    I say tough. Get used to it working the right way, like the Roadster.
     
  8. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    Interesting...
    There is some mild regen with accelerator pedal lift in the Prius (around -15amps give or take based upon speed and battery SOC). But you can vary the amount of regen when braking up to a max of around -100 to -125 amps under ideal conditions, though I tend to stay in the -40 to -80 amp range for most of my common braking situations.

    So, if I'm understanding this right, with the Roadster maximum regen is "coasting" with the foot off of both brake and accelerator--correct? Any idea what kind of regen power that is? It would seem like it would be fairly limited compared to the Prius-style system. Does regen cut out as soon as the brake pedal is depressed?

    Finally, in the Prius 1G friction braking and regen were simultaneous...with mild braking there was very little friction braking, but it was there, with harder braking the friction and regen increased, but eventually the regen would stop and you'd be on friction only. Thus, with the 1G Prius there was no 'transition' like there is with the 2G and 3G. In an effort to increase the effectiveness of regen succeeding generations of the Prius increased regen, stopped friction until the max regen threshold was exceeded or for "emergency" conditions (which includes bumps, skids, uneven braking surfaces, uneven wheel spin rates, etc.)

    So, I don't really consider the Tesla system better...it may be smoother and avoid the transition, but it limits driver control of regen and probably reduces the overall effectiveness possible out of the regen system. It's obviously simpler to implement an smoother...maybe even "safer"...so I'm not bashing, just saying I wouldn't call it better.

    Indeed, if Toyota can find a way to smooth that transition form high regen to friction I'd have to proclaim that system better b/c of the ability to maximize regen and minimize friction brake use/wear.
     
  9. doug

    doug Administrator / Head Moderator

    Joined:
    Nov 28, 2006
    Messages:
    15,914
    Location:
    Stanford, California
    #10 doug, Feb 5, 2010
    Last edited: Feb 5, 2010
    I wouldn't necessarily call it more elegant, but it certainly is more simple and is my preference. Regardless, the issue at hand is similar: how traction control deals with regen.

    Perhaps it's even more dramatic in the Roadster where the regen is likely stronger and you're less likely to have your foot ready over the brake pedal. If TC engages there's a sudden loss of regen which makes the driver experience positive jerk (decrease in deceleration), which could be unsettling.

    Do you have these numbers in terms of power? With that and the mass of the car one can make an estimate of braking force and acceleration.
     
  10. ChargeIt!

    ChargeIt! Member

    Joined:
    Jul 24, 2008
    Messages:
    619
    Correct ... touch neither pedal (what you call coasting) ... and you get maximum regen. But I am a bit surprised at the high amp numbers quoted for the Prius max regen (100A, 125A). The Roadster tops out in the 80s and feels VERY strong. But then again, the amount of deceleration (force) is not determined by amps alone. You also need to know the voltage and vehicle mass, since decel is a reduction in total kinetic energy.

    No, regen continues (under normal conditions, i.e. there's no transition) when you apply the brake pedal. The caveat is, regen gets interrupted -- as explained in other posts -- if the Traction Control software finds something amiss, which could include slippage, or the ABS being activated during strong friction braking.
     
  11. donauker

    donauker Member

    Joined:
    Sep 5, 2006
    Messages:
    755
    Location:
    Pennsylvania
    After several years of driving my Lexus hybrid with the Toyota approach, when my Tesla finally arrived it only took minutes of driving to consider Tesla's approach far superior. I only drive the Lexus when I absolutely must and I now find their approach to regen very uneven and difficult to maximize efficiency.

    The regen on the Tesla is quite strong and it is very easy to learn one pedal control for most normal stop and go driving. With reasonable alertness the brake is really only needed for holding at stops, aggressive driving and of course dealing with idiot drivers.

    I have never experienced a loss of regen in the Tesla that would remotely resemble what happens in the Lexus. The Toyota approach is to totally turn off regen on traction loss and transition to friction brakes which takes some very noticeable amount of time. They do not restore regen until you release the brakes and reapply them. I have found the reapplication of regen with my Tesla to be so smooth and quick that it is hardly noticeable much like the limiting of torque on wet roads during acceleration. It is certainly much smoother and quicker than traditional ABS.

    The road at the entrance of my development has a cobblestone section with a rather abrupt edge right in the area that you are stopping. This will almost always give a brief feeling of brake failure in the Lexus and will only be noticed as a bump and a flicker of the TC light when only using regen on the Tesla.
     
  12. bolosky

    bolosky Member

    Joined:
    May 5, 2009
    Messages:
    422
    The maximum regen on the Roadster is about 40kW, which is roughly -100A with the Roadster's ~400V system. Taking your -125A number and Wikipedia's claim of 288V for the latest generation Prius, the Roadster's about 10% more than the Prius.

    However, the software doesn't always use maximum regen when you let up all the way on the go pedal, it depends on the speed. 40kW at low speed would be quite large deceleration, but it's much more gentle at highway speeds, so it reduces regen power as you slow down.

    Regen doesn't stop when you use the friction brakes, they both work in tandem.

    My experience with the Roadster regen was that by the third traffic light on my test drive, I could pretty much anticipate what it was going to do and use it properly. I imagine that if you don't expect it it might be pretty confusing, though.
     
  13. Tdave

    Tdave Member

    Joined:
    May 16, 2008
    Messages:
    721
    Location:
    Columbia, MD
    Regen on the Roadster is a fairly consistent -0.2 g of deceleration, which as has already been mentioned generates about 40kW of power.
     
  14. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    Sorry, I don't. And the Prius III numbers are estimates b/c noone's decoded the CAN data yet to determine the actual thresholds.
     
  15. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    #16 efusco, Feb 5, 2010
    Last edited: Feb 5, 2010
    Thanks guys, that clarifies things a lot.
    I think that, if I had my way, I'd still prefer a 'free-wheel' option when there was no pressure on any pedal with regen controlled by the brake pedal. But that may be because of the way I hypermile my Prius.

    During my test ride in the Tesla last month I specifically asked the driver about getting used to the regen on accelerator lift...he actually had to play with it during my ride to answer the question b/c he'd adapted to it so much and had just learned to feather the accelerator from throttle-on, to 'glide' to slowing w/regen.

    I can see how this would be desirable in that there is no transition at all b/w coming off throttle and moving to brake (if single foot driving).

    Thanks for the education and information.

    One last thing...someone mentioned being surprised at my -125 amp number...I can assure you this occurs, but I've never seen (in my Prius 2G) much more than -100 amps sustained for more than 10-15 seconds and braking b/w -20 to -80 amps is MUCH more common for any normal braking and any hard braking kicks out regen completely. One has to be going pretty fast and slowly but firmly apply brakes to see the -100 amp number.

    Just 'cus I can, here's a screenshot I just took of my Prius CAN-View data log. You'll notice the 2nd line has the max/min amps, and the 4th line the max/min kW. It's been ages (maybe a year or more) since I last did a reset, so this is historic data.
    4332469061_7dae617c64_b.jpg
     
  16. Tdave

    Tdave Member

    Joined:
    May 16, 2008
    Messages:
    721
    Location:
    Columbia, MD
    Why does hard braking cut out regen on the Prius? I see no reason to do that.

    On the Roadster regen is always there no matter how hard you're pushing on the brake pedal. The friction brakes are always just used to suppliment the regen force.
     
  17. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    I have wondered the same thing.
     
  18. doug

    doug Administrator / Head Moderator

    Joined:
    Nov 28, 2006
    Messages:
    15,914
    Location:
    Stanford, California
  19. efusco

    efusco Moderator - Model S & X forums

    Joined:
    Mar 29, 2009
    Messages:
    4,590
    Location:
    Nixa, Missouri, United States
    Interesting:
    from NHTSA.gov Don't know how to link to the exact report, but it's pretty easy to search for. Of note, this was just submitted today for a July 15th, 2009 incident.

     

Share This Page