Effect of tire pressure on mpg: collecting data

Discussion in 'Gen 3 Prius Fuel Economy' started by Robert Holt, Sep 29, 2019.

  1. Robert Holt

    Robert Holt Senior Member

    Joined:
    Aug 3, 2013
    1,279
    830
    0
    Vehicle:
    2012 Prius
    Model:
    Three
    Hypothesis: Increasing tire pressure from the recommended mid-30s psi range to 40 psi or over will increase overall fuel efficiency (higher mpg or lower L/100km).

    One result is:
    To answer this question objectively, I think it would be valuable to collect data from a variety of Gen III owners who are have recorded mpg (or L/100 km) data for the same basic driving route driven on tires with mid-30s psi and slightly higher psi.

    I’ll go first with the results from a standard loop from Fairfax to Manassas, Virginia, on I66 with a set loop of urban driving on the outskirts of Manassas and then a return to Fairfax.

    Tires at 34 psi: BB404543-EF26-4AA6-A7AD-93C0974A4C9F.jpeg

    Tires at 38-39 psi:
    07FBBA1E-67B5-4AA3-8770-D7F138E10809.jpeg

    The difference in mpg, 67.2 versus 68.5 mpg, was small but in the direction of higher tire pressures leading to slightly higher efficiency. However, the average speed was also slightly higher at 36 mph for the 67.2 mpg versus 34 mph for the 68.5 mpg, and given the known decrease in fuel efficiency with increased velocity, that average velocity difference could also account for the small observed difference in fuel efficiency. Thus, this is a positive result for the hypothesis but has a known confound that could account for it.
    I could use MORE SYSTEMATIC DATA from anyone willing to participate in this effort to confirm or disconfirm the relationship between tire pressures and overall fuel efficiency. Thanks in advance!
     
    TheeWood and Raytheeagle like this.
  2. Raytheeagle

    Raytheeagle Senior Member

    Joined:
    Jul 1, 2016
    9,399
    12,733
    0
    Location:
    Bay Area, California
    Vehicle:
    2019 Prius Prime
    Model:
    Prime Plus
    How will you measure the suspension components wearing out faster due to less energy being “absorbed” by the higher inflated tires;).

    I’m sure you’ll find some like @Grit that take hypermiling seriously participate:).

    Not bad mpg on your display though(y).
     
  3. fuzzy1

    fuzzy1 Senior Member

    Joined:
    Feb 26, 2009
    12,920
    6,450
    90
    Location:
    Western Washington
    Vehicle:
    2012 Prius
    Model:
    Three
    I don't believe most hypermilers would agree that there is that much difference. The rolling resistance vs pressure graphs posted long ago just didn't support that much improvement.
     
    #3 fuzzy1, Sep 29, 2019
    Last edited: Sep 29, 2019
    Vman455 likes this.
  4. bisco

    bisco cookie crumbler

    Joined:
    May 11, 2005
    84,563
    37,172
    0
    Location:
    boston
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    how many times do you have to repeat a test with few controls to be legitimate?
     
    ice9 likes this.
  5. Robert Holt

    Robert Holt Senior Member

    Joined:
    Aug 3, 2013
    1,279
    830
    0
    Vehicle:
    2012 Prius
    Model:
    Three
    Enough comparison pairs of data to achieve statical significance at a scientifically acceptable level. We used probability of < .05 in our work (less than 1 out of 20 chance of a false positive conclusion). Each data pair can be examined for confounding effects as I did, but that also requires some auxiliary information. The MFD includes the distance traveled (longer distances would give more stable mpg estimates), and average speed (higher speeds typically associated with lower mpg), but does not include ambient temperature (73-81 degrees Fahrenheit for both loops in my case), wind velocity (Near calm for both loops), or air conditioning use (not used in either loop). Finally, the exact composition of the gasoline should be the same for both data points as the energy content of gasoline varies by season (Summer versus Winter blend), alcohol content (E0 versus E10), and reformulated versus non-reformulated gasoline blends. Thus, I would simply recommend measuring both data points on the same tank of gasoline. With a large sample size, many of the random factors of driving each loop (such as stoplight sequences) will even out across the comparison pairs.
     
    RMB and bisco like this.
  6. William Redoubt

    William Redoubt Senior Member

    Joined:
    Feb 13, 2016
    1,195
    1,132
    1
    Location:
    Coronado Island, California
    Vehicle:
    2016 Prius
    Model:
    Two
    Suspension components will not wear out faster due to higher tire pressure. While it seems like an obvious result that the wear will increase, the statement is so vague as to be nonsense. In fact, I would bet that Ray the Beagle does not even know the units of measure for spring rate calculations, or the parts that may be affected by wear if the spring rates are changed. And whose specs are you going to use, anyway? The PSI on the door sticker? On the sidewall of the tire? In a writing posted to the world wide web?

    Come on Ray: Which parts will wear faster with higher tire pressure? Hint: All tires do not have the same spring rate at 35 PSI. And the spring rate of the tire is only one factor in rolling resistance, which is the number uno factor affected with increased tire pressure.

    By my calculations proven in real life trials, 5 pounds of increase in tire pressure from 35 PSI brings 5 mpg more efficiency, while increasing the spring rate by only two nanouggaduggas (2 ndg) -- the universal unit of measure for confined atmospheric induced change of spring rate. The calculation and units are very similar those used to understand the relative variance gravity effects of the turbo encabulator.
     
  7. Raytheeagle

    Raytheeagle Senior Member

    Joined:
    Jul 1, 2016
    9,399
    12,733
    0
    Location:
    Bay Area, California
    Vehicle:
    2019 Prius Prime
    Model:
    Prime Plus
    No sense name calling and pretending to know someone’s knowledge level;).

    I’m a degreed engineer whose worked with springs and related equipment in industrial settings:).

    I for one have noticed a ride difference when I “overinflated” our tires:cool:.

    And all you mentioned is one component of the suspension:whistle:.

    And a coworker of mine has a PIP and had a similar commute and he maxes out the pressure (something like 45-50 psi):eek:.

    He replaced the shocks at 120k as they were worn out. Mine were in the same worn out position at 180k miles:D.

    While your talking about springs specifically, I’m talking about the gas shocks, the rubber bushings and even some components that rattle more in the cabin due to the rougher ride:oops:.

    Keep it positive and think a bit about the total picture my pink Prius friend(y).
     
    Mendel Leisk and RMB like this.
  8. William Redoubt

    William Redoubt Senior Member

    Joined:
    Feb 13, 2016
    1,195
    1,132
    1
    Location:
    Coronado Island, California
    Vehicle:
    2016 Prius
    Model:
    Two
    Oh, the comfort afforded by cognitive bias. :) "Feeling" a ride difference does not have the sharp edge of definition that bespeaks a fair judgement about reality.

    Worn shocks on two cars owned by two people at different miles driven on different routes aren't the type of comparative data that has any value. At least for actual analysis. So, a sample of two, if tire pressure is THE factor in shock wear, says that 150,000 miles is the average life a shock absorber at a range of 35 to 40 PSI. Keeping in mind, of course, that a single pothole can destroy a shock at any time.
     
  9. Robert Holt

    Robert Holt Senior Member

    Joined:
    Aug 3, 2013
    1,279
    830
    0
    Vehicle:
    2012 Prius
    Model:
    Three
    Comparison data point 2 from our standard shopping loop out to Manassas and return:

    With tires set at 34 psi (cold) and ambient temperature 66 degrees Fahrenheit (lights on, no AC):
    1E6D89F7-6B0B-4FD5-996F-B00DD2022E08.jpeg

    With tires set at 41 psi (cold) and ambient temperature 67 degrees (lights on, no AC):
    7C01CA53-62A6-491C-A791-172B06BD690B.jpeg

    The observed distance was nearly identical (28.6 vs 28.5 miles) as were the ambient temperatures, and the average recorded speed of 36 mph was identical. The observed difference in mpg was 2.4 mpg. In percentage terms, he observed increase was 2.4/64.9= 3.7% increase in mpg. Since the difference in tire pressure was 7psi, the percentage gain in mpg for each 1 psi increase in tire pressure was .53% . [That is slightly more than the .31% gain in mpg per 1 psi increase in pressure that I was expecting based in the meta-analysis average of previous research on tire pressure effects on mpg in the study cited by fuzzy1, but it is still a fairly small effect size.]

    Anyone else willing to post their data on this issue?
     
    bisco and RMB like this.
  10. Vman455

    Vman455 Active Member

    Joined:
    Feb 8, 2014
    358
    337
    0
    Location:
    Champaign, IL
    Vehicle:
    2013 Prius
    Model:
    Two
    Since MPG is a function of several variables, % difference of MPG is also a function of the % difference of those variables; if you didn't hold everything else constant (and you didn't--wind, traffic, exact temperature, direction of sunlight [differential heating of tires on each side of the car], etc. all changed), then you need to know the % difference in each of those and calculate the partial derivatives with respect to each variable to find how much the tire pressure actually contributed. Not an easy task! I appreciate what you're going for here, but there are simply too many other factors to start ascribing changes in MPG solely to tire pressure.

    The best way to do this, I think, would be to rent a large hanger or other long, indoor space where you can let the car coast in neutral and measure the distance covered from an arbitrary speed to an arbitrary lower speed. That would remove the aerodynamic influence of wind and other cars, the gravitational force due to elevation grades, and temperature variations (assuming it's a temperature-controlled space). Next best thing would be a level test track outdoors, where the effect of wind could be calculated and temperature variations compensated for if measured at the time of testing.
     
    fuzzy1 and Salamander_King like this.
  11. Salamander_King

    Salamander_King Senior Member

    Joined:
    Nov 8, 2015
    4,906
    3,704
    0
    Location:
    New England
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Premium
    +1

    I completely agree with the above statement. It is just not possible to determine the real effect of tire pressure on mpg by just changing the tire pressure and driving around.
     
  12. frodoz737

    frodoz737 Top Wrench

    Joined:
    Aug 26, 2010
    4,189
    2,083
    33
    Location:
    Texas
    Vehicle:
    2015 Prius
    Model:
    Four
    Yes it is, and I've had 2 Prii long enough to know door jam vs 42f/40r the difference IS very noticeable long term. But as they say, YMMV. ;)
     
    StarCaller likes this.
  13. bisco

    bisco cookie crumbler

    Joined:
    May 11, 2005
    84,563
    37,172
    0
    Location:
    boston
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    while its not scientific, we appreciate your time and effort. with each test, i see a bit more 'proof' (y)
     
    RMB likes this.
  14. fuzzy1

    fuzzy1 Senior Member

    Joined:
    Feb 26, 2009
    12,920
    6,450
    90
    Location:
    Western Washington
    Vehicle:
    2012 Prius
    Model:
    Three
    We also have numerous pieces of actual scientific or research lab proof too. Not large improvements, but still noticeable.
     
    Robert Holt likes this.
  15. ice9

    ice9 Member

    Joined:
    May 3, 2019
    457
    221
    1
    Location:
    norfolk va
    Vehicle:
    2018 Prius
    Model:
    Four
    Why not rent a speedometer calibration stand and calculate mpg from that. If all you are really concerned with are the deltas for different tire pressures at different speeds, you can effectively establish controls on such a test stand. I would guess the most significant/difficult variable to control would be starting and ending tire temperature and pressure. But even that would be easier to control on a test stand than by driving a circuit.
     
    #15 ice9, Oct 10, 2019
    Last edited: Oct 10, 2019
  16. Dimitrij

    Dimitrij Active Member

    Joined:
    Feb 22, 2015
    377
    175
    0
    Location:
    Virginia
    Vehicle:
    2014 Prius c
    Model:
    Two
    If you are implying that in this situation personal opinion is powered by cognitive bias, what makes you think that?

    What makes you think that? Is there a law of physics that supports your opinion?
     
  17. Dimitrij

    Dimitrij Active Member

    Joined:
    Feb 22, 2015
    377
    175
    0
    Location:
    Virginia
    Vehicle:
    2014 Prius c
    Model:
    Two
    I have never been a dedicated hypermiler, but I did explore the effect of some driver-controlled factors on energy efficiency. After 160K+ miles in an hybrid and an EV, here is what I have come to believe that the speed and the driving technique affect the mpg and mi/kWh more significantly than other factors. Increasing the tire pressure from 35 psi to 40+ psi didn't seem to yield visible improvement in efficiency. I don't have a record of the facts and figures, though.
     
  18. ice9

    ice9 Member

    Joined:
    May 3, 2019
    457
    221
    1
    Location:
    norfolk va
    Vehicle:
    2018 Prius
    Model:
    Four

    Don't you find it odd that your use of the word "notice" was changed to "feeling" while it was being cited as a form of "cognitive bias"? :whistle:

    ...Just saying.
     
    #18 ice9, Oct 11, 2019
    Last edited: Oct 11, 2019
    Dimitrij likes this.
  19. William Redoubt

    William Redoubt Senior Member

    Joined:
    Feb 13, 2016
    1,195
    1,132
    1
    Location:
    Coronado Island, California
    Vehicle:
    2016 Prius
    Model:
    Two
    No. What's your point? I would appreciate it if you would let me know what the difference is. "I noticed," seems analogous to, "I felt," to me. Both are subjective, unless there is quantitative expression added to the statement. Or am I missing the nuance?

    Here is the first synonym that came up in Google: "Notice: to become conscious of someone or something by seeing, hearing, or feeling ..."

    I don't think Ray was saying, I" noticed, based on the difference in before and after data from vibration analysis using a Fluke 250 handheld vibration meter, that the ride was more harsh." I think he was speaking qualitatively. Using the seat of the pants meter. Which is subject to cognitive bias.
     
    #19 William Redoubt, Oct 11, 2019
    Last edited: Oct 11, 2019
  20. Grit

    Grit Senior Member

    Joined:
    Aug 22, 2017
    2,871
    2,426
    1
    Location:
    BOY
    Vehicle:
    2012 Prius
    Model:
    Four
    Guess I won't be able join with my mid 40s psi data over the past year. Oh well, this wine will make this frown turn upside down.
     
Loading...