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

2010 brake system

Discussion in 'Gen 3 Prius Technical Discussion' started by bwilson4web, Jan 30, 2010.

?

Details on braking anomoly

Poll closed Feb 13, 2010.
  1. Happens in the beginning of braking

    19.4%
  2. Happens after braking for a bit

    79.1%
  3. Had a load in back of Prius

    1.5%
  4. No significant load in Prius

    70.1%
Multiple votes are allowed.
  1. Airbalancer

    Airbalancer Active Member

    Joined:
    Aug 6, 2009
    892
    73
    0
    Location:
    Cobourg, On, Canada
    Vehicle:
    2010 Prius
    Model:
    N/A
    Can you mark you bottom scale of second, just to mark sure I am reading correctly
    is each line a second, or each darken line a second?
     
  2. Sagitar

    Sagitar Junior Member

    Joined:
    Jan 18, 2010
    59
    14
    0
    Location:
    Ampthill
    Vehicle:
    2010 Prius
    Model:
    V
    Where do I put "It's never happened to me"?
     
  3. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus

    The vertical dark lines are each second. The dashed lines are 0.100 seconds. BTW, I uploaded the spreadsheet so feel free to fix up the charts.

    My next steps are to:

    1. convert acceleration into velocity and distances
    2. retest using identical speed bumps on an adjacent lane (weekend traffic flow)
    Bob Wilson
     
  4. Airbalancer

    Airbalancer Active Member

    Joined:
    Aug 6, 2009
    892
    73
    0
    Location:
    Cobourg, On, Canada
    Vehicle:
    2010 Prius
    Model:
    N/A
  5. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    Well I was surprised after integrating the X-axis acceleration:
    [​IMG]
    Wow, I was not expecting to find evidence of a very small, acceleration during the very first part of the braking anomaly. It does seem to shadow the struts expanding but I have no idea how that might couple to the very slight acceleration.

    Bob Wilson
     
    4 people like this.
  6. tpfun

    tpfun New Member

    Joined:
    Oct 4, 2009
    791
    54
    1
    Location:
    Oh Never Mind,CA
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    If you best fit a smooth curve on the acceleration data during the speed bump (212 -212.6), is there a decrease in braking force during that time ?
     
  7. adrianblack

    adrianblack Member

    Joined:
    Sep 26, 2009
    447
    187
    0
    Location:
    Los Angeles
    Vehicle:
    2010 Prius
    Model:
    III
    I'm sure you had the zero point set on the accelerometer when putting it in the car and calibrating it, but if, for instance, the car was on a slight incline during the event, the spring rebound could be interpreted as slight forward acceleration. The integrated x-axis speed increase coincides perfectly with the rebound (and hitting there are speed "increases" as you go over the speedbumps too) that I think it must have been caused by the accelerometer being affected by the car being slightly off flat.

    Of course, not seeing the terrain where this happened, I could be so far off. :)

    Now, either way, you certainly had ZERO slowing down for that 800ms, and I could see how that would feel very scary!

    Once you do your speed calculation, I am going to be interested to find out exactly how far your traveled without slowing down.

    Something else, do you have an ELM327 deice where you can dump all CAN traffic? I would love to find the passive PID's for the car's accelerometers. Using your accelerometer and the dumped traffic, we should have a better lead finding the CAN data for the G-force sensors by correlating the data.

    The car's G-sensors most likely sample at 79fps (because this data is used by the skid ECU) so the resolution should be decent.
     
  8. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus

    This is how I had it mounted for the Friday data:
    [​IMG]
    I got the units on Thursday, the same day we learned a fix is coming. So my first efforts have been to check them out, understand their operation and possibly see if I can capture anything of use. The Friday event was quite a surprise and in fact, I would have blamed our poor quality, pavement.

    But I agree, it will take time to get the mounting location(s) aligned with the vehicle axis and center of gravity. Still, I'm racing the clock. So I drove around today with a second accelerometer in the rear, inches from the latch. I synced their times and I'm hoping with parallel recording I can use differential acceleration over a know X-axis distance to calculate pitch and yaw.
    The problem is the guards get a little 'anxious' when they see people with cameras taking photos of the gate area ... the armed guards.

    I hate to say this but it really felt like an ordinary, rain-slick, Huntsville street. We're way beyond being scared by now ... those folks moved out a long, long time ago.

    It is already there, a minimum of 7.6 m/sec. for ~800 ms. ... a distance of just about 6.08 m. of constant speed. If you pickup the spreadsheet, looking at the deceleration rate after the 800 ms., the difference in stopping distance could be calculated.

    No generic CAN monitors. I'm leaning more towards independent monitoring systems so it will work in any model of Prius, NHW11, NHW20 or ZVW30. But we can chat about that another day.

    Bob Wilson
     
  9. bighouse

    bighouse Active Member

    Joined:
    Sep 7, 2009
    822
    126
    1
    Location:
    Guerneville, CA
    Vehicle:
    2010 Prius
    Model:
    IV
    I'm very impressed. Good work Mr. Wilson.

    I've thought more about this issue from my perspective and after reading some of the NHTSA reports more carefully (which, btw, there are now over 400 on) and have come to my conclusion that it happens to me when three or four criteria are met:

    1. The exact right speed. It seems to happen somewhere around 9 to 12 mph.
    2. "Medium" brake pedal pressure is being applied.
    3. I am breaking while turning my steering wheel.Happened twice to me while turning into my parking space alongside my road during the left turn after a right turn- ogee type manuever.
    4. I'm doing all the above while driving atop an irregular surface. In this case it's a change from asphalt to compacted gravel.

    Fortunately for me it's a rarity. I've never been able to duplicate it willingly and it's only happened twice in hundreds of times of making the exact same manuever in the exact same location. So, I really think it has to happen when all the above factors are in "alignment".

    Perhaps you might try your experiment while in a small, slow turning operation while braking, it seems common in the NHTSA reports.

    Thanks again! I'm very intrigued by your results. I like reading graphs!
     
  10. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    13,527
    4,057
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2018 Tesla Model 3
    Model:
    N/A
    Nice work. I really appreciate you collecting data for this.

    I've only had it happen once in 2400 miles. I would love to see how those that can repeat this do after the software upgrade.

    Is there a way to use my iphone as the acclerometer?
     
    1 person likes this.
  11. fuzzy1

    fuzzy1 Senior Member

    Joined:
    Feb 26, 2009
    17,039
    10,013
    90
    Location:
    Western Washington
    Vehicle:
    Other Hybrid
    Model:
    N/A
    About what speed did you hit the speed bump? I'm trying to estimate 'extra braking distance' caused by this, and compare it to both braking distance at this level of decel, and to hard braking.

    My attempt to integrate the acceleration data to get a speed profile isn't scaled correctly. And of course there is no zero speed reference.

    I hope more of these events can be recorded, to put some bounds on decel losses and dropout times.
     
  12. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus


    Feel free to check my work:
    Near as I can tell:

    • ~8 m/s - initial contact with speed bump
    • ~7.6 m/s, ~655 ms duration - end speed bump event
    • 7.6 m/s, ~700 ms duration - no speed delay, ~5.32 m
    • resume deceleration
    [​IMG]

    I used the stop at the gate to find the integration constant for the zero point:
    [​IMG]

    I don't think the reduced braking from the speed bump can be counted against the reduced stopping distance. However, the 700-800 ms. of effectively no stopping force directly relates to extended stopping.

    I have to admire the irony of a "speed bump" leading to an extended stopping distance.

    Bob Wilson
     
    2 people like this.
  13. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    OWCH! I wished I'd thought of it! Good idea!

    I'll check the iTunes store to see what is available for my iPod Touch:

    • Accelerometer Data ($3) - just downloaded to begin testing, looks promising!
    Bob Wilson
     
  14. adrianblack

    adrianblack Member

    Joined:
    Sep 26, 2009
    447
    187
    0
    Location:
    Los Angeles
    Vehicle:
    2010 Prius
    Model:
    III
    OMG -- too simple! LOL

    Looks like the Pro version $8 might be the one to get ... 100hz sample rate, and it stores the data on the iPhone/iTouch as a CSV which you can grab via Wifi.
     
  15. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,068
    15,372
    0
    Location:
    Huntsville AL
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    I agree. The $3 version works but is limited in the number of samples and if you start another application, it looks like the data is lost. Still:

    Timestamp,AccelX,AccelY,AccelZ
    0,0.126785,-0.670151,-0.742599
    0.100172,0.144897,-0.633926,-0.742599
    0.200286,0.144897,-0.670151,-0.778824
    0.300247,0.126785,-0.670151,-0.742599
    0.400363,0.126785,-0.688263,-0.742599
    0.500524,0.144897,-0.652039,-0.724487
    0.600505,0.144897,-0.706375,-0.742599
    0.700671,0.126785,-0.670151,-0.688263
    0.800808,0.163010,-0.688263,-0.742599
    0.90085,0.163010,-0.688263,-0.706375
    1.00079,0.144897,-0.706375,-0.760712
    . . .

    So I've got some really nice, dedicated acclerometers with high resolution, 160-320 samples per second. But while one is out, I'll be able to use the iPod Touch as a second. With three accelerometers located at three locations, I should be able to generate all six metrics ... NEAT!

    I can use the less expensive accelerometer connected to axles and use the iPod for body motion. . . . hummm

    Bob Wilson
     
  16. hitechboy

    hitechboy New Member

    Joined:
    Sep 19, 2009
    128
    6
    4
    Location:
    Toronto, Canada
    Vehicle:
    2010 Prius
    Model:
    N/A
    Wow! Bob what a work! I am your fans from now on!:yo:
     
  17. adrianblack

    adrianblack Member

    Joined:
    Sep 26, 2009
    447
    187
    0
    Location:
    Los Angeles
    Vehicle:
    2010 Prius
    Model:
    III
    What's cool is if you buy the app and you have an iPod and a Touch, you can have it on both. It seems possible to buy 8gb 2nd gen iPod touches on Ebay for about $130-ish ...

    Oh, the app "Axeleron" seems to log to files as well, and it samples at up to 100hz and is free. (At the moment, at least.) It also logs your location and you can email the CSV to yourself.
     
  18. Rhino

    Rhino New Member

    Joined:
    Oct 5, 2009
    460
    41
    0
    Location:
    U.S.A.
    Vehicle:
    2010 Prius
    Model:
    III
    Purely technical question.

    Has anyone tried the new Prius with the new software vs. the old 2010 Prius? Is there a mileage different (i.e., mpg difference)? Or is the fix so minimal that mileage is unaffected.
     
  19. barbaram

    barbaram Active Member

    Joined:
    Jan 24, 2006
    911
    70
    9
    Location:
    Trenton, NJ
    Vehicle:
    2010 Prius
    Model:
    IV
    Having had a 2006 Prius, AND as a locomotive engineer who has to deal with slip/slide systems that do not always respond as expected, this problem has not caused me too much trouble. My training demands that I always be in control of my vehicle. I guess that has spilled over to my driving style. I've felt the brakes slip a bit under bumpy conditions. Common sense tells me to drive with more caution.
    Guess I'm not a typical driver.....:eek:
     
  20. adrianblack

    adrianblack Member

    Joined:
    Sep 26, 2009
    447
    187
    0
    Location:
    Los Angeles
    Vehicle:
    2010 Prius
    Model:
    III
    This app works, but the timestamp data has a resolution of 1 second -- and it seems to drop samples, at least at 100hz. Too bad, since it seems to work otherwise.