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

Extreme hypermiling attempts

Discussion in 'Gen 3 Prius Fuel Economy' started by Ferls80, Apr 24, 2011.

  1. Ferls80

    Ferls80 Performing some hypermiling techniques.

    Joined:
    Oct 15, 2010
    66
    50
    0
    Location:
    Padova, Italy
    Vehicle:
    2010 Prius
    Model:
    N/A
    Hi all
    I just want to open this new thread to post some "hardcore" P&G tests I made so far with my 2010 Prius. New posts and suggestions are welcome.

    When I say "extreme" in the thread title I mean that I try to find out the best road conditions (no cars and no traffic lights), the best P&G and the best car setting to maximize fuel economy.
    During these tests I had to consider:
    • no gradient paths in favor (ex: start the test at 10m, end at 10m high).
    • overall consumption of traction battery = 0 (ex: beginning SOC=55%, ending SOC=55%, SGII is very useful here).
    • commutes should be far enough to find out a reliable average (I would say at least 20 km that is around 12 miles).
    • Beginning and ending speed must be the same (starting from 0 km/h, ending at 0 km/h for example).

    The first test I made was at the end of march 2011, one month ago. the P&G I used was the following: accelerate from 35 to 57 km/h (22 to 35 mph) with ICE running at 1360 rpm or so. In this way I could complete almost every P&G cycle with energy balance "from" and "towards" battery = 0.
    I tried to keep the SOC at 58%.
    The result is shown here:
    [​IMG]
    TripB: 62.3 km (38.5 miles) - Cons. 2.6 l/100km (90.5 mpg) - AVG 40 km/h (25 mph).
    Outside air temperature was 8°C (46°F). To keep ICE coolant between 85-87°C (185-190°F) I had to use lower grill blocking, covering about 50% of the surface.

    ****************************************************

    Since I have a target of 2.0 l/100km (117.6 mpg), this early morning coming back from a club, I tried some more tests, hoping that warmer temps could help me to do a better job.
    I made 3 different P&G tests:

    1. P&G from 35 to 50 km/h (22 to 31 mph) and SOC = 60%
      [​IMG]
      TripB: 18.1 km (11.3 miles) - Cons. 2.3 l/100km (102.3 mpg) - AVG 37 km/h (23 mph).
    2. P&G from 35 to 50 km/h (22 to 31 mph) and SOC = 45%
      [​IMG]
      TripB: 17.6 km (10.9 miles) - Cons. 2.4 l/100km (98 mpg) - AVG 38 km/h (24 mph).
    3. P&G from 35 to 68 km/h (22 to 42 mph) and SOC = 55%
      [​IMG]
      TripB: 21.6 km (13.4 miles) - Cons. 2.5 l/100km (94.1 mpg) - AVG 46 km/h (28.6 mph).

    In all tests I used to pulse at 1800 - 2000 rpm, to keep ICE at the most efficient speed: I guess you all know the 2ZR-FXE chart that Ken1984 showed us:
    [​IMG]
    where is clear that the most efficient 220 g/kWh area is between 1250 to 3000 rpm: Thanks again Ken.
    For these tests: 15°C (59°F) outside temp - with no wind - on the same flat roads - starting and ending at the same place, with the same speed and with the same SOC.
    In the second test where SOC = 45% I noticed that during the pulse phase more ICE energy went to the battery. I think that this was the reason why I had a worse fuel economy result than the first one.
    The car has more or less 19500 miles and has a solar roof.
    Tires (Bridgestone Eco Pia B250 195/65 15) were inflated 2.9 bar (42 psi) front and 2.8 bar (40.6 psi) rear.

    I know that someone got 2.0 l/100km (=50 km/l =118 mpg), so now I'm wondering what there's to improve on the things written above to reach such a great result.
    - Maybe a warmer outside temp can help? With no grill blocking ICE coolant was at 80°C (176°F) with 15°C outside, perhaps a 88-90°C coolant temp would give a better ICE efficiency.
    - Maybe a different P&G speed range?
    Suggestions are welcome.
    Bye
    PS: I wanted to share my driving experiences with you, but I apologize for my english, I know it's not good: I didn't have the chance to practice a lot.
     
    10 people like this.
  2. ItsNotAboutTheMoney

    ItsNotAboutTheMoney EditProfOptInfoCustomUser Title

    Joined:
    Jul 18, 2009
    2,287
    460
    0
    Location:
    Maine
    Vehicle:
    2010 Prius
    Model:
    II
    If conditions allow, a higher tire pressure should help more to extend the glide.
     
  3. F8L

    F8L Protecting Habitat & AG Lands

    Joined:
    Aug 14, 2006
    19,011
    4,080
    50
    Location:
    Grass Valley, CA.
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    *edit* Never mind. I missed it in your first post. Sorry

    What tires are you running and at what pressure?
    What mode are you running in when you do these tests?
    Are you using the A/C at all?

    Your numbers are very nice but it would help others to know more about your car. :)
     
  4. Ferls80

    Ferls80 Performing some hypermiling techniques.

    Joined:
    Oct 15, 2010
    66
    50
    0
    Location:
    Padova, Italy
    Vehicle:
    2010 Prius
    Model:
    N/A
    Tires (Bridgestone Eco Pia B250 195/65 15) were inflated 2.9 bar (42 psi) front and 2.8 bar (40.6 psi) rear.
    Running in ECO mode all the time, without using A/C , fan off.
     
  5. giora

    giora Senior Member

    Joined:
    Aug 28, 2010
    1,966
    730
    0
    Location:
    Herzliya, Israel. Car: Euro version GLI
    Vehicle:
    2013 Prius Plug-in
    Model:
    N/A
    Thanks for sharing.
    As far as I remember from Ken1984 reports on P&G by Japanese experts, best results are achieved when the average speed on the display converge to about 33 to 35 kph (Ken, correct me if I'm wrong). So try and lower your pulse top speed and glide bottom speed a little bit.
    Having said that, I must admit your numbers are impressive already!

    On the pulse part, if you switch to HSI, where about are you? and on the gide part?

    Giora.
     
    1 person likes this.
  6. Tideland Prius

    Tideland Prius Moderator of the North
    Staff Member

    Joined:
    Oct 2, 2004
    44,837
    16,073
    41
    Location:
    Canada
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    Well I'm impressed at your higher average speeds of up to 40km/h! Maybe because it's a short route, because my tank speed averages are usually in the low 30s.
     
  7. F8L

    F8L Protecting Habitat & AG Lands

    Joined:
    Aug 14, 2006
    19,011
    4,080
    50
    Location:
    Grass Valley, CA.
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    I wish the GenII had a tank speed average. I can't get my SGII to report this correctly. It always ends up reading 100+mph. :rolleyes:
     
  8. SageBrush

    SageBrush Senior Member

    Joined:
    Jun 4, 2008
    11,627
    2,530
    8
    Location:
    Southwest Colorado
    Vehicle:
    2012 Prius v wagon
    Model:
    Two
    I think this is the 5% you are missing. Wait for 25C weather, and start the test after the ICE coolant temp is nice and toasty.

    Were you really playing these games at 3 am ?
     
    1 person likes this.
  9. Tideland Prius

    Tideland Prius Moderator of the North
    Staff Member

    Joined:
    Oct 2, 2004
    44,837
    16,073
    41
    Location:
    Canada
    Vehicle:
    Other Non-Hybrid
    Model:
    N/A
    Sure F8L, blame the Scangauge for the high average speed reading ;) hahaha.
     
  10. Ferls80

    Ferls80 Performing some hypermiling techniques.

    Joined:
    Oct 15, 2010
    66
    50
    0
    Location:
    Padova, Italy
    Vehicle:
    2010 Prius
    Model:
    N/A
    Thanks Giora. About the HSI, I took a picture of it and I draw the zones where I was working, just to explain better. Here it is:
    [​IMG]
    About the pulse zone ICE was running around 1800-2000 rpm. In this phase some energy went to the traction battery (I see negative amps in my SGII, and a higher SOC at the end of the pulse).
    Typical SOC values before and after pulse were: 60.0% --->60.5%.
    About the glide zone, you see that is a very tight space. That's because every energy conversion has a efficiency less to 1, so there's always some waste of energy. I just glided to use the 0.5% SOC that I gained during pulse phase.

    Thanks SageBrush, I think the same: warm weather should help. But I doubt it will be enough to get the 2 l/100km I am aiming.
    Yes, it was 3 am, I was coming back from a club. Actually that's the only time when traffic allows this kind of games.
     
  11. SageBrush

    SageBrush Senior Member

    Joined:
    Jun 4, 2008
    11,627
    2,530
    8
    Location:
    Southwest Colorado
    Vehicle:
    2012 Prius v wagon
    Model:
    Two
    Hi Feris80,
    I was thinking more of 2.1l/100km.
    Did you read the Prius marathon reports that Wayne Gerdes and Dave Bassage et al participated in years ago using the G2 Prius ? I remember the tyres were reported *very* well worn, and they were inflated to high pressures.

    At the fringe of possible MPG, every percent matters.

    Anyway, have fun. Don't forget to get some sleep ;)
     
  12. Corwyn

    Corwyn Energy Curmudgeon

    Joined:
    Mar 6, 2011
    2,171
    659
    23
    Location:
    Maine
    Vehicle:
    2007 Prius
    Model:
    II
    Are those labels backwards?
     
  13. giora

    giora Senior Member

    Joined:
    Aug 28, 2010
    1,966
    730
    0
    Location:
    Herzliya, Israel. Car: Euro version GLI
    Vehicle:
    2013 Prius Plug-in
    Model:
    N/A
    As corwin pointed out, labels seem to be reversed.

    Also, I would try the pulse a bit further towards the power zone and the glide a bit away from charge zone.
    Only a bit of EV in glide helps and you are gaining it back on pulse.

    As sage said, outside temp will help a lot. Also, try during the day, no headlights, if this is possible (or are you doing it w/o lights at 3 am? lol)

    Giora.
     
  14. SageBrush

    SageBrush Senior Member

    Joined:
    Jun 4, 2008
    11,627
    2,530
    8
    Location:
    Southwest Colorado
    Vehicle:
    2012 Prius v wagon
    Model:
    Two
    I forgot!
    Why, that is 50 watts wasted. Since the car without lights is only using about 130 wh/mile, at 20 mph a mile takes 3 minutes, so the lights are adding close to two percent overhead.
     
  15. giora

    giora Senior Member

    Joined:
    Aug 28, 2010
    1,966
    730
    0
    Location:
    Herzliya, Israel. Car: Euro version GLI
    Vehicle:
    2013 Prius Plug-in
    Model:
    N/A
    I would say 130 Watts wasted...two headlights + tail lights + instrument...
     
  16. Ferls80

    Ferls80 Performing some hypermiling techniques.

    Joined:
    Oct 15, 2010
    66
    50
    0
    Location:
    Padova, Italy
    Vehicle:
    2010 Prius
    Model:
    N/A
    Oh, yes, thanks, just corrected.

    Ok, thanks for the advice, I'll do my best next time.
    But it will be hard for me to find a road that allows these tests during the daylight.
     
  17. sipnfuel

    sipnfuel New Member

    Joined:
    Jan 21, 2011
    1,080
    174
    0
    Location:
    So. Cal.
    Vehicle:
    2010 Prius
    Model:
    II
    Hi Ferls, thanks very much for your hypermiling trials and analysis.

    I use the same pulse & glide zones whenever possible.

    I employ SHM a lot since much of my commute is at freeway speed (65 mph speed limit, but you can go 75-80 mph without getting pulled over), and there are slight ascents and descents which favor pulsing & shm'ing, and the occasional glide. Gliding at higher speeds is not favorable as speed is lost so quickly, you spend more time pulsing, and much less time gliding.

    I notice you do your experiments in the dead of night when it is possible to do so. If I tried to travel at 23 mph, I would probably get murdered :).

    Perhaps you can try an experiment using micro-pulses and micro-glides, while keeping your speeds in a tighter range.
     
    1 person likes this.
  18. SageBrush

    SageBrush Senior Member

    Joined:
    Jun 4, 2008
    11,627
    2,530
    8
    Location:
    Southwest Colorado
    Vehicle:
    2012 Prius v wagon
    Model:
    Two
    Wayne Brown published energy/distance data at varying speeds with his G2 Prius. I put the data into a spreadsheet to ferret out the wh/mile at P&G speeds.
    [​IMG]

    The graphic was only to convince myself that Wayne's aero measurements followed the expected exponential increase with speed.
    Easy to see that non-aero energy consumption stays constant at about 110 wh/mile, and aero at 30 mph is around 25 wh/mile, for a total consumption of 135 wh/mile or 84 wh/km.

    Let's see what theoretical fuel economy looks like with Ken's best SFC of 220 grams/kwh:
    Petrol weighs 720 grams/liter, so one kwh at the driveshaft requires 305 ml. A kwh will push the car 7.41 miles (11.95 km), or 39.1 km/liter. This is quite a bit worse than experience shows, so either:

    My math is wrong,
    The SFC is incorrect,
    Waynes data is wrong (very doubtful),
    Or a mixture of the above.

    Addendum: This handbook says petrol at 20C weighs 802 grams a liter, so using that number the best fuel economy rises to 43.56 km/liter, or 2.30 liters/100 km. It seems fair to say that you are hypermiling really well, and 50km/L is quite unlikely. Could you take your Prius on rail tracks ? ;)
     

    Attached Files:

    3 people like this.
  19. sipnfuel

    sipnfuel New Member

    Joined:
    Jan 21, 2011
    1,080
    174
    0
    Location:
    So. Cal.
    Vehicle:
    2010 Prius
    Model:
    II
    ^^ Hi Sage, work/distance varies quadratically (the power of 2), not exponentially. Your log graph is characteristic of the graph y = 2 * log(x)
     
  20. sipnfuel

    sipnfuel New Member

    Joined:
    Jan 21, 2011
    1,080
    174
    0
    Location:
    So. Cal.
    Vehicle:
    2010 Prius
    Model:
    II
    Urgh, scratch this post until i fix it. bad math on my part.