ICE load vs. IGN: pulse & glide to maximize MPG

Discussion in 'Gen 2 Prius Technical Discussion' started by 2009Prius, Jan 6, 2011.

  1. 2009Prius

    2009Prius A Wimpy DIYer

    Joined:
    Mar 25, 2009
    2,693
    461
    63
    Location:
    USA
    Vehicle:
    2009 Prius
    I have an impression that for best fuel economy one would pulse at IGN 14 where the ICE loading is supposed to be at its maximum, although I haven't seen real data plotted out, just claims from experienced hypermilers. I finally got the ICE load reading implemented in my PriiDash(TM) project and did a quick "splat plot" of the ICE load vs. IGN:
    [​IMG]
    The data are plotted with ~ 67% transparency to show the frequency of occurrence of the many overlapping data points. Even though the outer envelope of the data points shows a prominent peak of ICE loading around IGN 14 as mentioned above, the trend of the majority of the data points is much less certain - and this seems to agree with the observation during driving that the ICE load does not seem to depend very strongly on IGN, and it seems that IGN 12 and 13 are at least as efficient as 14 if not slightly more.

    Some more observations (that also agree with past experience and/or common sense): ICE loading is maximized when climbing a hill or accelerating from low speed, and it droops down when the car reaches some steady high speed - time to start gliding.

    Happy hypermiling and drive safely. :)
     
  2. 2009Prius

    2009Prius A Wimpy DIYer

    Joined:
    Mar 25, 2009
    2,693
    461
    63
    Location:
    USA
    Vehicle:
    2009 Prius
    More data from two other trips seem to support the observation above that ICE load does not depend strongly on IGN:

    [​IMG]

    [​IMG]
     
  3. Mrn

    Mrn New Member

    Joined:
    May 8, 2010
    13
    1
    0
    Location:
    Europe
    Vehicle:
    2010 Prius
    Model:
    Three
    Hi. I am interested in analyzing the "happy mode" of the Prius, discussed under point 7 here. In short, happy mode is when the "ICE can drive directly the wheels", without energy loss between MG1 and MG2. That of course happens iff there is close to zero energy flow between MG1 and MG2.

    To see, when that "happy mode" kicks in, and if it exists at all, it would be nice to have the current (A) values measured at MG1, MG2, and battery when driving at different speeds.

    Questions:

    1. Is there a way to measure the breaking power of MG1 at different speeds? (happy mode probably happens at zero breaking power).
    2. Is there a way to measure the current (A) to/from MG1, MG2 and the Hybrid Accu at different speeds?
    3. Does somebody happen to have measurement data at different speeds?
    Different speeds: let say 30, 45, 60, 75 mph.

    Thanks in advance!
     
  4. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    Hi Mrn. You can use Torque app on an Android device with the custom PIDs.

    The goal is to adjust the accelerator pedal so that MG1 and MG2 power is near zero.

    [​IMG]
     
  5. 2009Prius

    2009Prius A Wimpy DIYer

    Joined:
    Mar 25, 2009
    2,693
    461
    63
    Location:
    USA
    Vehicle:
    2009 Prius
    Does "MG1 and MG2 power is near zero" correspond to the superhighway mode (IGN 14)? I don't go on highway much at all these days and last time I looked it seemed impossible to get MG1 and MG2 power near zero, but I was pulsing and gliding, not using superhighway mode.

    My general feeling these days is that Prius is pretty well optimized so there is a wide range of accelerator pedal positions that are equally efficient.
     
  6. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    No, at highway speed (65 mph), for MG1 and MG2 power to be near zero, gas engine has to be around 2,800 rpm. From what I have seen, IGN 14 is possible around 1,280 rpm.
     
    1 person likes this.
  7. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,111
    997
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    That's interesting. If you put the ICE RPM at 2800 rpm and the speed at 65 MPG you get MG1 operating pretty much as a stalled motor.
    See Toyota Prius - Power Split Device

    The thing about "happy" mode is that it minimizes the circulating energy loss between MG1 and MG2 but it tends to correspond to a fixed "gear ratio" so it doesn't necessarily optimize the ICE efficiency. I've read that sometimes it's necessary to waste circulating power on MG1 and MG2 (eg so called "heretical mode" with MG1 motoring and MG2 generating) to optimize the overall efficiency including that of the ICE.
     
  8. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    Right, for constant speed. However, the "happy" mode is useful to accelerate at highway speed and the glide with an arrow from the battery (stealth).

    For constant speed, Prius will be in heretical mode almost all the time at highway speed. The exception is steep hill or acceleration.
     
    1 person likes this.
  9. jpadc

    jpadc Type before I think too often

    Joined:
    Sep 9, 2008
    343
    30
    0
    Location:
    Indianapolis and Northern Wisconsin
    Vehicle:
    2008 Prius
    Model:
    N/A
    You just have to show us that screen shot to make those of us with our little 4 inch phone screen jealous now don't you :rolleyes:
     
  10. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    I recorded this as a starter. I have a second video with IGN Timing Advance at higher speed. Will upload it tomorrow.

     
  11. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    Ok, here is the second part at higher speed. I meant to do high speed but the traffic prevented me, only reaching 65 mph momentarily.

    I think only a hardcore HSD would understand what the heck I am talking about. Yea, I do need a better setup because it is not safe to do this while driving.

    BTW, I forgot to add engine load. IGN Timing, MPH, MG1 and MG2 plus others.

     
  12. Mrn

    Mrn New Member

    Joined:
    May 8, 2010
    13
    1
    0
    Location:
    Europe
    Vehicle:
    2010 Prius
    Model:
    Three
    @usbseawolf, thanks a lot for the videos!

    Theoretically the following should hold:
    MG1 [HP] +MG2 [HP] - losses* = power to/from the ACCU [HP] + losses*.

    However the above equation does not seem to be completely valid:
    please see below extractions of time, MG1, MG2 and ACCU values..

    Analysis of the 2nd video: (values: MG1..MG2 ..BATT)

    0m49sec: +1.6, -3.9, 0.0 => sum of losses: 2.3HP
    1m07sec: -2.1, +0.6, 0.0 => sum of losses: 1.5HP
    3m45sec: 0.0, +10.6, 11.4 => sum of losses: 0.8HP
    5m42sec: +7.1, -9.1, 0.0 => sum of losses: 2.0HP
    8m20sec: 0.0, -10.7, -6.7 => => sum of losses: 4.0HP

    I see 3 possible reasons for the above:
    1) the values shown by Torque have some level drift,
    2) the values diplayed by Torque do not correspond to one single moment, ie. the MG1 value is not measured at the same instance as MG2, only displayed at the same time.
    3) there is a 1-2HP loss between MG1, MG2 and the Accumulator.

    * If it turns out that neither 1 nor 2 is the case, then it would be interesting to analyze where that 1-2 HP disappears.

    Questions:
    1) is there a way to check the validity of the values shown by Torque?
    2) @usbseawolf, could you please make some graphs of MG1, MG2 powers and that loss, as a function of speed, and as a function of the power from/to the battery?
    3) Is it possible that MG1 feeds MG2 through the battery, and therefore the loss is introduced by the 650V->200V and 200V->650V transformations?
    4) Is there a single pair of cable between the inverter and the battery, or two pairs?

    UPDATE:
    Analysis of the 1st video:
    battery charging mode:
    0-60sec, ICE 1300rpm, 1 to 2 HP loss between MG1 and BATT

    Car standing, ICE off:
    3m00s and 5m10s: BATT shows 0.6HP. Why?

    EV-mode:
    3m20sec - 3m:30: MG2 shows only 1HP when moving the car ?!?
    Is it possible?

    deceleration:
    5m:00s MG1: 0.0, MG2: -30.5HP, BATT: -26 HP => total loss of 4HP acc to Torque
     
  13. Mrn

    Mrn New Member

    Joined:
    May 8, 2010
    13
    1
    0
    Location:
    Europe
    Vehicle:
    2010 Prius
    Model:
    Three
    ...
    I have found this:
    "One issue with the MG torque readings is that the resolution is too coarse. So a small non-zero torque value would show up as zero. For example when I glide the torque readings almost always show zero even though there are some HV currents flowing."
    Source here.
     
  14. usbseawolf2000

    usbseawolf2000 HSD PhD

    Joined:
    Sep 22, 2004
    14,487
    2,931
    0
    Location:
    Fort Lee, NJ
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    MG1 and MG2 horsepowers are from the HV ECU. It is derived from the RPM and Torque they produce. It is not from the current x voltage it consumed. It could be consuming electricity and making maximum torque but at 0 RPM, the gauge will report 0 hp. Something to keep in mind. The electric motor/generator and inverter losses are also in there.

    I think MG1 power goes to the power control unit then to MG2. It won't go to the HV battery unless there is a surplus or the battery needs to be charged.

    I don't have the entire HV battery pack voltage, just the current. So I added up the voltage from all 14 battery blocks. Perhaps it created some rounding issue? Perhaps, it is consumed by the 12v accessories and charging the 12v lead acid battery.

    At that moment, I may have released the accelerator a little bit. Despite the torque you feel from the electric motor at very low speed, it is pulling only a few hp from the battery - very deceiving! Seeing the realtime data really opened my eyes and put a different perspective on power, torque and speed.
     
  15. Mrn

    Mrn New Member

    Joined:
    May 8, 2010
    13
    1
    0
    Location:
    Europe
    Vehicle:
    2010 Prius
    Model:
    Three
    Thanks a lot for the feedback!
     
Loading...