Pulse & Glide vs Cruise Control - a study

Discussion in 'Fuel Economy' started by Doc Willie, Apr 14, 2008.

  1. Doc Willie

    Doc Willie Shuttlecraft Commander

    Here's my first attempt with preliminary data from seven runs. Details will be presented at
    HybridFest.

    Speed limit: 35-55 MPH 46.1 MPG 32 MPH Av

    P&G: 38-15 MPH 78.9 MPG 23 MPH Av
    CC: 23 MPH 67.1 MPG 20 MPH Av
    P&G: 40-25 MPH 66.2 MPG 27 MPH Av
    CC: 29 MPH 73.3 MPG 25 MPH Av
    P&G: 43-5 MPH 62.3 MPG 22 MPH Av
    CC: 25 MPH 76.6 MPG 22 MPH Av

    Summary: P&G in this range is more efficient only within a certain range (in my case determined by setting the pulse range by the IGN [ingition advance] readings on the ScanGuage.

    At the fifteen MPH and Max Pulses, cruise control is more efficient.

    This data is surprising to me, and I intend to make another set of runs.
     
  2. GreenVTEC

    GreenVTEC Well-Known Member

    I'm new to this but unless I'm wrong wouldn't the EV mode on the Toyota Prius be a factor in the lower results from a P&G run?

    If you are continually leaving electric only to speed up for a glide then by default wouldn't an electric only CC run give better results?

    I suppose it would depend on what distance travelled for the test!
     
    Last edited: Apr 15, 2008
  3. bestmapman

    bestmapman Fighting untruth and misinformation

    Hi Doc,

    What are the other factors in your experiment/observations. Such as temp, distance, legnth of pulse, etc.

    I seem to get the best results pulsing between 26 and 20. If I have a flat run and can P&G undisturbed, I can pull out over 100 MPG pretty easily.

    I'd like to see more of your results.

    DiamondLarry and Xcel may want to way in here as they have more experience and better results then I do.
     
  4. B.L.E.

    B.L.E. Well-Known Member

    I think your high peak speeds are killing your gas mileage. You are hitting high peak speed with a low average speed. Try doing it with a smaller speed delta, say pulse to 30 and glide to 25 and see how that compares with steady speed.
    Steady speed is actually the most efficient use of energy. The optimum results should be a tradeoff between the efficient use of energy that steady speed gives and the most efficient supply of energy that intermittant engine runs give.
     
  5. bomber991

    bomber991 Well-Known Member

    I'm gonna have to agree with this guy. Seems like based on the state of charge of the Prius battery it will determine how long you stay in ev mode, or how much assist is given. That would end up skewing these numbers.

    I just wonder what distance these tests were run over, and if your tires are inflated up to the max sidewall rating.
     
  6. desdemona

    desdemona Well-Known Member

    My impression as well. I also thought that maybe you are lettign the car get too slow. I don't know re: the Prius vs a regular non-hybrid. But I never let the low get lower than say around 25-27 mph. The car shifts down where it isn't as efficient. Perhaps i could be misreading your data??

    --des
     
  7. Doc Willie

    Doc Willie Shuttlecraft Commander

    Well, I don't want to pre-empt my HybribFest presentation, but since I started it, here goes.

    EV is not used at all.

    Course was a flat 5 miles, each run both ways, total 10.3 mi/run

    Temps were recorded. A gradual increase over 3 hours. Each PG run was followed by a CC run at attempted same speed.

    The second PG run was 40-25 mph. CC beat it.

    I will try narrower range on next set of runs. Hypothesis: this will begin to approach Cruise Control, which is the ultimate narrow range.
     
  8. JimboK

    JimboK One owner, low mileage

    FYI, I offered some comments in your PriusChat post. I'd link to it for the sake of others here, but PC is down at the moment.
     
  9. Skwyre7

    Skwyre7 Well-Known Member

    Again?
     
  10. Dan

    Dan KiloTanked in post 153451

    Doc, Cool stuff... I like the data, I could never quite find that CC sweet spot. Texas is big on overpasses, so there is always a 30 ft raise and fall the flips CC out.

    Things I'd suggest as a control for your experiments (not to nit-pic)
    1. Make sure you you gather your data on a round trip then average the legs (ie two legs, one out, one back).
    2. Record windspeed and direction at various points on the track.
    3. Force your Prius into S4 before each leg.
    4. To remove SoC as a possible culprit in skewing numbers, force charge the battery before each leg to an exact SG % reading. I suggest 55%. If you need to drop your SoC to 55% just turn on headlights and AC for a minute or two to drop it down to 55%. Just avoiding EV-ing is not enough since SoC will control your RPMs while running the ICE.
    5. Record your block temp at the start and end of each leg.
    6. Record your SoC at the end of each leg (using the SG percentage reading).
    7. Try to do the P&G -vs- CC test on the same day at the same barometric reading and humidity level (same day should keep this from skewing much).
    8. Perform all your glides in 'N' since regen or assist below 3 Amps won't display on the Energy screen.
    9. Try to keep all pulses in a certain RPM tolerance (ie 1500 - 1600).
    10. Experiment on different RPM tolerances (ie 2000-2100).
    11. Experiment with different starting SoC levels.
    Good luck and I look forward to your presentation.

    ref: http://privatenrg.com/#terrainfollowing

    11011011
     
  11. xcel

    xcel PZEV, there's nothing like it :) Staff Member

    Hi Doc:

    ___I am glad to see you are recording your tests in a detailed format but your real world results do not match any number of members when left to real world traffic and terrain conditions, over less onerous deltas or when able to let it loose in an FE Hypermiling Challenge of any kind?

    ___During the Hypermiling clinics in Plano, Texas last Sunday, the lowest achieved mpg from the Toyota supplied and loaded Prius-II Touring over a 5 mile RT roadway was 75 mpg, highest 85 mpg w/ a reset of the aFCD between each. Those results were achieved from brand new Prius drivers; in heavy 15 + mph side winds on both legs of the RT loop and in what some would call heavy traffic at times. Delta’s included maybe 15 mph on the bottom (besides the one or two, 0 mph starts from the stop lights) and maybe 35 mph on the high end.

    ___Then came a slew of standard drives over the next 18 hours … With yet another reset FCD and after an hour or two of cold soak for the Sunday night and Monday errands including a trip to the local movie theater (twice as I was way to early the first time there), a local Tex Mex place for dinner, a stop at a local gas station asking for directions to any local breakfast diner, a stop at a local iHOP, an excursion to the SouthFork Ranch, a trip into downtown Dallas to see the School Book depository (never made it due to traffic) and finally back to the airport. With all of the above including maybe 8 power button shutdowns with minutes to multiple hours between restarts and of course one from an over night cold soak in temps from 50 - 65 degrees F the following was achieved.

    [​IMG]

    ___This included starting out at the hotel in Plano, through a number of local small towns/suburbs to the South, North and West and than back to the airport in Dallas via the Interstate. Except for the segment from the SouthFork Ranch to the Dallas Airport, the individual segments were no longer than 7 miles w/ powered shutdowns and no lay of the land, traffic or local roads like anybody would when closer to home. I experienced some of the worst Interstate traffic (~ 5 miles over 35 minutes -- classic large city stop and crawl) and some of the best, some detours due to torn up local roads through the hundreds of stop lights and stop signs in the Dallas suburbs and of course some empty boulevards to do what I wanted ;) All in all, there was nothing steady state about it nor could there have been. Your conditioned test results do not match up to the real world results above let alone what the CleanMPG heavy hitters including Larry, Jerad, Marc, Dan and Jim achieve daily by any stretch???

    ___When the HF FE challenge drivers achieve 110 - 125 mpg the day before your presentation; expect a number of rolling eyes from the crowd. Hopefully on Friday evening or possibly Sunday afternoon while you are in Madison, there will be a number of highly qualified members to give you a Prius specific Hypermiling clinic and afterwards, you will see how all of this is accomplished.

    ___Good Luck

    ___Wayne
     
  12. diamondlarry

    diamondlarry Super MPG Man/god :D

    Today, on my way from my work to my wife's work(10.2 miles), with a cold ICE that sat just under 9 hours, I had 90.2 when I pulled up to her lot to pick her up. The only time EV mode was used was to FAS and to get up to 10-15 mph from stops(4). The temp was 58F and there was a SW wind at 5-10 mph. My route was primarily north. The return trip(6.3 miles) was at 85.9 mpg and the route was primarily south. As some have suggested, the speed delta may perhaps be too large. When I first started out, I was pulsing between 34 and ~27 mph since I wasn't at 155F on my coolant temp and had to use EV for FAS. Once I reached 155F+, I pulsed 40-~33 mph.
     
  13. mparrish

    mparrish Rosie the Riveter Redux

    Great job on all the testing Doc. Testing is always good. I can be counted on for any group testing as suggested by JimboK & Bestmapman.

    I must admit I'm having trouble understanding the tests. How would I ever be able to pull off CC at 23/29/25 MPH in city driving? How do you P&G 40-25 MPH, average 27 MPH, and only pull 66.2 MPG? How can you "G" at 43-5? Looks like 3mph of "WS" to me. ;)

    This may be worthy of a separate thread, but I'll go ahead and bring it up here. With all of the talk of Toyota raising the glide limit from 41 to a potential 62, I find myself wondering the following:

    What would one pull on the highway doing, say, a 55-62 P&G on the highway once that limit is raised by Toyota?

    I'm guessing 80mpg, but that's just a shot in the dark. Somebody go to Cal-Irvine, steal one of those babies for a weekend, and get back to me. :)
     
  14. JimboK

    JimboK One owner, low mileage

    OK, it's back up. Here's the link.
     
  15. Doc Willie

    Doc Willie Shuttlecraft Commander

    Thanks for the suggestions. I have controlled for most things as you suggested, and have the data on temps, etc. My ScanGuage does not report SOC. Can I use the battery indicator on the MFG? Maybe charge it to 6 blue bars before each run?

    Wind speed a little hard to come by. Hopefully compensated by round trip and paired runs.
     
  16. PaleMelanesian

    PaleMelanesian Beat the System Staff Member

  17. Dan

    Dan KiloTanked in post 153451

    This will tell you what you need to know. To get a good control point to start each experiment, you could let the battery drop to 5 bars then force charge up to 6. Shut down as soon as you cross from 5 to 6. That should give you a SoC of 57.5%. Remember going from 5-6 does not happen at the same spot (57.5%) as going from 6-5 (55%).

    I think the real problem will be in the fact that you don't know the SoC at the end of the run. If you clock a 65 MPG segment on P&G with a SoC delta of +15% and compare that to a 72 MPG CC segment with a SoC delta of -12%, the comparison is flawed at a very fundamental level.

    SoC has more than a marginal effect on MPG. It is the primary factor in my opinion. The problem arises in trying to control SoC while in CC. I could imagine that CC is not interfacing directly with the battery ECU and simply allows the battery ECU to panic once the SoC hits the shift threshold (somewhere near 52%). Whenever the battery crosses this threshold on the way up, RPM drops by 2k. It feels almost exactly like an automatic transmission shifting into high gear. Whenever you get below the shift threshold your FE drops off dramatically.

    I guess one way to illustrate the difference would be to start the experiment at 3 bars (P&G will win) then try again with SoC at 7 bars (CC will win). Those are extremes but it should show how SoC has an effect.

    Something that might work to control the end SoC would be the following:
    1. Reset your CURRENT trip gauge.
    2. Reset your TANK gauge (FILLUP>DONE>DONE).
    3. Start the runs at a specific SoC at a dead stop.
    4. Finish the runs by coasting down (no regen) to a dead stop in Neutral.
    5. Record the MPG and distance traveled (the coast to stop may give you +/- 0.1mi).
    6. Reset your CURRENT gauge.
    7. From the point where you coast ended, accelerate back up to 50 mph.
    8. Set CC at 50 mph for 10 miles.
    9. Shift to 'N' at 50 and coast down to a dead stop.
    10. Hit the 'P' switch and wait for the engine to cut off.
    11. Record the CURRENT gauge (mi and MPG).
    12. Record the TANK gauge (mi and MPG).
    Steps 8-12 with restore your SoC back up to 60%. So now you have a complete track with the a known and predictable delta on SoC. The purpose of step 9 is to prevent the SoC from changing from it's highway norm. Now what I'd expect is that the TANK gauge will favor the legs that started with P&G. I'm still suspecting that the favorable CC numbers you are seeing will be followed by very thirsty HW segments as SoC is returned to the pack. On the flip side of that, I suspect that the poor P&G segments will be followed by very lean HW segments because of a high SoC.

    Depending on your pulse rate SoC through a P&G segment will either stabilize at 4 or 5 bars. I usually pulse at 1500 RPM, use next to no brakes, and almost never see anything but 4 bars. If I change my pulse rate I can hold 5 bars. I use 3 windows depending on conditions.
    1. Highspeed with tailgaters - 50kph - 65kph
    2. Go with flow traffic with buffers - 45kph - 54kph (52kph - 54kph in EV)
    3. Open road no tailgaters - 30kph - 54kph (52kph - 54kph in EV).
    That usually keeps my SoC at 4 bars and yields 80-90 mpg depending on conditions.

    11011011
     
  18. JimboK

    JimboK One owner, low mileage

  19. bestmapman

    bestmapman Fighting untruth and misinformation

    The big problem with PC is that it is down a lot lately. It seems it is down as much as it is up.
     
  20. Dan

    Dan KiloTanked in post 153451

    Fine with me, but I used to get some pretty odd looks at PC and try to keep mileage discussions off of that board. I don't want to the the resident P&G loony, but sounds like you got my back, so in I go.

    11011011
     

Share This Page