(I'm opening a new thread instead of continuing to comment on a different posting in another forum from two days ago that was on a different topic but drifted into a discussion on mileage)

People have suggested I just try 93 octane and see what happens.

I'm happy to run an experiment with 93 octane but since mileage differs so much from tankful to tankful just by random, uncontrollable conditions (wind, air temperature, atmospheric pressure) it wouldn't be precise enough to me to be meaningful to just use different gas and guess if it made a difference. Plus I'm self-employed so it's not like I'm making the exact same drive to work everyday to compare it to. And even if I did, well, I'm near NYC where there's lots of traffic so even the same commute day-to-day wouldn't be under the same conditions.

I would like to run a controlled test, and a friend of mine also has a 2010 Fit AT, although unfortunately it's not a Sport so it doesn't have cruise control. I'll see if I can talk her into driving the same stretch of road as me (if someone near me has a Fit Sport AT and wants to do the test with me, let me know).

Here's what I figure to do, if she agrees:

Fill both tanks at the same station with 87 (which we're both using now, and probably from the same Getty station).

Drive to a nearby highway that's not highly-travelled. Possibly very late at night when there are barely any other cars on the road.

Agree that we'll reset our trip odometers at the exact same spot on the road, after driving next to each other at the same speed. If there's too much traffic I'll ask her to drive behind me at the same speed.

Match speed until we exit ten miles later. Check mileage computer.

Do the same thing on the way back.

Then wait until my tank's almost empty. I'll fill with 93, she'll fill with 87. Repeat test, compare mileage differences between cars.

But we have to wait until there's no snow on the road, plus my car has only 900 miles so we probably should also wait until it's more broken-in.

If anybody has any polite suggestions for how to improve this test to make it more scientific, please let me know. I'm happy to include a bit of acceleration in the test, maybe flooring it to 60 mph on the trip back.

-Shaun

 

This "test" will not tell you much if anything. I politely recommend you both buy scangauges and go rent a circuit if you are looking to yield anything approaching useful or conclusive.

Why don't you just run the experiment I told you to in the other thread. There are way too many parameters to control for that you have not and in some case cannot take into account.

Try the octane increment test I outlined for you and keep it to one car and one driver.

I don't think you actually know what you are looking for when comparing different fuels.

This is more than just the minor gain in combustion efficiency, this is about how the engine responds in low rpm/high load cells, part throttle tip in, cruise spark advance, etc.

If you see more than a 2-3mpg difference I would be shocked, and as I have noted in several other threads, which you didn't bother reading of course... for some conservative drivers they may find the difference is neglible or negated entirely by the difference in cost between grades at a specific location.

Long post short... this will not yield anything useful, especially since it is being championed by a comedian with a grudge and no understanding of what is involved.

If you really want an idea of the difference fuel can make, spend the $35-$40 a 5 gallon can of toluene costs, run your trials in ascending order from regular, to premium to premium plus toluene, then back to regular. Resetting the ECU between tanks and noting scantool data.

 

This shows the testing procedures the epa uses and shows all the variables. Detailed Test Information

The best way is to step up octane using the same station. After the 3-4 tank of premium step it back down and you will definitely notice the car acting differently and mpg falling. I notice after the 1 tank.



Here is the 2007 Fit mpg http://www.fueleconomy.gov/feg/calcu...umn=1&id=22861, the new Fit will get better mpg because of the maf sensor which operates over a broader range. My 2008 Fit averaged 36 plus lifetime and best of 42 mpg.


Your mpg will still vary http://www.fueleconomy.gov/feg/why_differ.shtml

 

First of all, what I do for a living has no bearing on the experiment, other than the fact that I'm not an automotive engineer. I could be a biologist, a teacher, a nurse, unemployed, retired... why do you bring this up?

Secondly, I read a LOT of threads on this site over the past month; don't tell me what I read or didn't read. I assume you're not standing behind me.


As I mentioned I want to measure something subjective-- differences in power, etc., unless measured at the track or on a drag strip, can't be accurately measured by feel. There's too much subjective bias- you say I won't believe it because I'm biased, and others who believe it will see a difference due to their bias/preconceived notions.

As I mentioned in another thread when I tested an oil additive I thought it worked because I expected it to, until scientists tested it and found it worthless.

Which is why I want to do a scientific test. And you can't do that by just changing fuels from one day to the next- even a minor amount of wind would throw that off. But using two similar vehicles at the same time and comparing differences takes that out of the equation.

And thanks, but I'm not spending $170 to buy a scangauge, especially not willing to connect something to a brand new car. I am willing to spend $20 on fuel.

To repeat, I want to be scientific about not, not to just put something in my car and see if I 'feel' the difference.

And a 2 mpg increase wouldn't pay for the extra cost of the fuel anyway, but if it were close it'd be worth it simply by reducing the number of fuel stops I'd be making over the life of the car.

 

First the only thing you can do is put it in and drive 5 tanks and then go back to regular. Nothing scientific about that. A scan Gauge would show you what the car is doing and you would learn to drive efficient and then notice that premium mpg is higher.

Second a 2 mpg will pay for the difference. The gas I use is 28 miles per tank which in a 10 gallons tank is 2.8 miles. The cost difference is 2.40 dollars and how far will 2.40 of regular take you? 20 miles? Premium is paid for and you saved 40 cents. The other benefit is a better running motor.

 

We have RFG gas plus ethanol so the mpg loss is typically 7 percent over 0 ethanol fuels and 1-3 percent from E10. So when I get 36 plus its like getting 39 mpg in other areas. Here are the other variables Many Factors Affect MPG for a possible loss of 8-9 percent loss in the fuel alone due to ethanol and season changes. So its a big deal when I get 40 mpg plus in mix driving. With conventional gas that would be 44 mpg plus.

 

1. Exactly- just 'trying' something and 'feeling' the difference is not scientific, especially with weather changes, differences in fuel season-to-season... and you can't compare two random trips because conditions vary. Maybe not on a windless plain but certainly where I live.

2. I don't quite understand your calculation. Here premium is about 10% more expensive than regular. 3 mpg is 10% of 30 mpg so it's a wash (although extra range means fewer gas-wasting fuel stops).

 

Ive already compared the gasolines in two exact cars, I am lucky then that premium costs 7 percent more than regular and the mpg gain is about 10 percent or higher. My commute is the same so its easy to duplicate the driving.

You just agreed that there is a difference and not understanding the difference or how the ecu works, Premium gas allows the ecu to run in close loop compared to open loop 11 to air fuel with loads over 70 percent. The engine loads at take off from a stop is 95 percent. Thats a difference of 25 percent when the ecu is in open loop. You wont get 25 percent better mpg but it averages out to 10percent or so because the ecu runs in close loop sometimes with regular. That why you need a scan gauge if you really want to see what I am talking about. Ive put out info to prove it but because you dont understand it doesn't mean I am wrong.

 

Exactly.

The scangauge gives you the direct sensor feedback. The ECU's numbers don't lie, but yours might Shaun, intentional or not.

The scan tool should even be able to give you how many pounds of fuel you burn per hour based on IDC's and fuel pressure.

The whole point is you are not an automotive engineer, or an engineer of any sort. You don't even have the basics down.

 

I've not said anybody's wrong, I've just asked for data and explanations.

How did you compare fuels in two different cars? Did you do the test I'm looking to do or did you just change fuels in one car and see a difference, then repeat with another car?

As I've stated, there is a lot of inaccuracy in making a drive, refilling the tank with a different fuel and then making the same (really only a similar) drive at a later date. Too many variables...

Question though-- the scangauge is $150. I see an ultragauge for about a third of the price. Anybody know the difference? Oh, and where's the OBD port?

 

One more thing- nobody here has met me nor read my resume. All you know is my current occupation and I believe in one thread I did say where I went to college.

I have a technical background, have studied engineering (electrical) and physics though I didn't major in either, and I have a few other qualifications that I don't think I should have to list.

So let's stick with science and leave the name-calling and other irrelevant stuff for the BMW forums.

Because I am talking about science and much of the response I get is either "You don't know/understand this stuff" or "I tried it and it worked better" based on unscientific, and flawed, calculations.

If someone here has gone back and forth for ten tankfuls, five each regular and premium, and has a very consistent commute, that'd be something. But "I switched to premium and my mileage went up and my car seems faster" doesn't satisfy me. I wish I had a steadier commute so I could play around with different fuels, but I don't.

That's why I want to compare identical cars, on the same run under the same conditions, then switch fuels and repeat the test.

 

I build and rebuild engines, then test and tune them in the cars and trucks they ultimately go in. My own vehicles I build and race out of my garage. I have gotten to play with various forms diesel, gasoline, VP110, C14, Q16, methanol, toluene, xylene, E85, propane etc. even little nitro RC cars.

This isn't something you are going to be able to master in a few months, I myself am constantly amazed at how much I have yet to learn as I start to play with more extreme projects. If anything this experiment of yours should be humbling and should pose you more questions than answers.

I and others have told you what you need to do to learn and participate in this discussion, let alone create and run experiments of your own in any effort to "prove" something one way or the other.

I am still willing to give you the tools, links and texts you need to get up to speed. But there is a lot more to it than you seem to have taken into consideration. It is not as black and white as either it is better or worse in a sweeping universal statement, and most of my comments on the matter reflect this.

Every single 1.5L GD or GE Fit, while built with identical parts, will respond differently in any given environment and driving situation. But an overall trend will show, that where available, a comparably more stable higher octane dose dinosaur juice will run for all practical purposes "better" under most circumstances than the minimum recommendation of US 87 octane.

You probably won't notice a whole lot going from 87 to 89 or 91, but there will be a distinct difference between 87 and 93 or greater. The explanation for this is quite simple, and it is merely based on what arbitrary "octane" number the ECU has assigned based on sensor feed back (usually 0-255 in most standalones and has nothing to do with actual octane content) This octane number helps the computer sort through which tables along the Min/Max v. Load/Rpm "octane" gradient are to be used at any given times.

This is where 3D Map tuning comes in. Because these Min/MaxOct tables are interpolated by the arbitrary "octane" assignment, the ECU may use a 3D cluster or diamond shaped chunk of cells across several different maps. This "octane number" is not necessarily fixed and will change in the presence of knock or changes in the necessary AFRs/Spark advance as things like IAT or coolant temps rise.

Things will also be skewed by any presence of ethanol or excess water (greater than 1-3% by volume) in any grade used. The problem here is unless you go to great lengths or have a friend with a small chemistry lab at your disposal that you won't be able to quantify any of these factors.

The closest you are going to get to yielding useful data is to provide a datalog from a scan tool or standalone unit along with the weather report for the duration of the test(s). Resetting the ECU between grades.

Even then, as the saying goes "your mileage may vary."

Under identical testing in a different car in a different region of the country, you could expect to see results tht would directly conflict with your data.

The best you will be able to do is provide a few unique statistics which in the end will mean little to nothing for the rest of us.

 

A scan gauge is a great tool because it offers many functions.. I generally monitor water temp, intake air temp, mass air pressure and voltage.. I use my electronic throttle controller to monitor the throttle position.... The throttle position readout is also on the scan gauge and I have seen it as useful in that it is possible see the effect of different fuels, wind, temperature, load, motor oil weights and brands and even between different tires... In 1957 a guy in junior high had me time him on his Whizzer motorbike from the corner to his drive way as he was determining whether the adjustments he was making to his carburetor was yielding more speed from a dead stop at that distance..... I still do that except from 10MPH and from the mailboxes to corner of a fence around a couple of neighbor's property lines while glancing at the speedometer as I pass it... I will slow to 10MPH in 2nd or 3rd gear when I turn on the access road to my house and start up a steep hill and see how much throttle it takes or at what point I have to down shift to climb it.... Those 2 things are a couple of ways to see if you are getting more 1st horsepower 2nd torque... When you have more torque it is possible to shift into a lower gear when the revs are lower and stay in a higher gear longer at slower speeds with out needing to down shift... More horsepower means you don't use as much throttle and both of these contribute to better fuel mileage as taught to me at an early age by a neighbor that was a car freak that had been a riding mechanic in the 1930s in Indianapolis 500 mile races and on the Blue Crown Spark Plug Specials that had 3 wins at Indy in the late 40s and 50s.... Just a couple of things you can do before and afters with that will show you there is a difference, or possibly not a difference but myself and others have been able to feel it.

 

I don't dispute any of this.

Luckily all I'm interested in is IF, not WHY. I'm sure there are a million factors which determine if, and how much, different fuels will change how my engine behaves. But for IF, I don't need to understand the engine, just the measurements.

I want to determine the effect on mileage (which is quantifiable). And if it seems to run better or have more power that's great but I can't determine that.

If my doctor gives me medicine that cures me, great! I don't care how it works, just if it works. And frankly there are a lot of drugs that work that the scientists don't know how or why. They just know it does. And they determine that by double-blind testing.

If you take an aspirin and your headache goes away, was it the pill? Possibly. If it comes back in 4 hours did the pill help? Maybe.

If you take an aspirin and the headache goes away and comes back in 4 hours and you repeat this and it keeps going away and coming back every 4 hours is it the pill?

Probably. But if during that 4 hours you're staring at the clock, I would have less faith that it's the pill, maybe it's your mind...

Back to the car:

To determine the effect I believe it is more scientifically valid to do the test my way rather than over a period of months which would entail different conditions.

Another way I could do this- wait until the tank's near empty. Fill a 5 gallon can with premium. Drive somewhere and back at a measured pace to when the tank's really almost empty. Dump the premium in, and repeat the drive.

 

Since you don't understand the "Why's" you won't know what your "If's" or "If Not's" are telling you.

This is not that straightforward. And again, these are results specific to your driving style, environment, specific vehicle and what fuel stations and grades are available. By no means does this prove anything for anyone but you.

Even you see benefit to premium cool, if not, it hasn't demonstrated anything conclusively.

So have fun wasting time and money to try and "prove" what would take months to execute if done properly and would lend us data other than your hand-calc mileage which requires us to take your word for it.

A datalog would tell me how fast you were cruising, at what TPS, IAT, MAP, Load, AFRs, Fuel Trims, IDCs etc. which would explain how and why you got that X mileage on Y grade fuel in Z conditions and whether these would be typical results for another random Fit owner. Then we could try and figure out the epsilon and delta were for a given data set.

What you are proposing tells us one small part of a giant piecewise function. Which if you remember basic Calculus (or even just the Algebra for some portions) is all but worthless if you dont have the rest of the "pieces." And if you aren't draining the tank between grades you are further skewing results.

 

Fuel quality too, plays a role. Both regular and premium have similar fuel air mixtures that they burn complete at 14.7A/F. VP c16 racing fuel is 14.8 so it would work in these cars if it didn't wreck the catalytic converter/O2 sensors, So dont use it. Unleaded racing fuel like GT100 from Sunoco has a 13.8 A/F ratio so that would get 7 percent less mpg in close loop and would run great at fuel throttle. Confused, May « 2009 « Total Performance Solutions Look at the links on the right by dates.

This explain the ecu basically How, EXACTLY, does the ECU work? - Toyota Tundra Forums : Tundra Solutions Forum Even though it a Toyota it the same principles. FlashPro Help need to read the whole thing this is for a Honda motors with maf sensors like the GE8. Theres people waiting for this to be for the Fits, I dont understand why they haven't come out yet for the Fits. Maybe because the Fit ecu is more advance and adjusts for more Parts and wont help. Flash Pro is for the Civics, but I think its possible to work with the Fits.

 

@BC: Thanks for offering to do this. It may be a valuable data point for figuring out the ifs of octane and Fits. If I may, I'd like to offer a few suggestions for the test.

First, it's good you have access to 2 Fits, as it essentially doubles the sample size of the test. However, when you do the drive, if you drive one in front of the other, the drafting effect will change the aerodynamics. The car behind will not be working as hard. You might want to take turns drafting, or just stay side-by-side as much as possible.

Also, since yours isn't broken in, you will see reduced mileage no matter what you do. You may still notice a difference in mileage and/or torque, but the overall numbers will probably be lower now than after break-in.

Another possible way to run the test: do the same A-B test with both cars. If you run a 93-vs-87 test on both cars instead of just one, it may help to eliminate the variables associated with the differences in the 2 cars, the driving styles of 2 different drivers, etc.

In fact, you could actually run 93 in one car and 87 in the other, drain the tanks, then run 87 in the first and 93 in the second.

I like the idea of testing both cars simultaneously.

Oh, and the OBD port is under the steering column, near the heater controls.

 

I don't think my friend much cares so I don't know if I'll be able to talk her into it, but I may buy a scangauge (or the other, less expensive, competitor, the ultragauge- I assume the data display will be equal). So if anybody in the NYC area wants to spend a day trying this out- reset computers, drive, refuel, reset computers, etc. let me know.

One thing I don't want to do, though, is drain the fuel tank unless it's something easy to do. Not ready yet to start taking apart my new car.

Either way, I'd keep the cars far enough apart to eliminate the drafting affects.

 

The scangauge is built better but the ultra displays more features. I have both and was gonna sell the SG but decided to keep it. KC and I will post the results. KC uses regular and the fuel trims add 10 percent fuel. I used premium and and see -9 pulling fuel.

 

Thanks. I mean they both just display numbers given from the car's computer so there's no difference in accuracy between the two devices.

Oh, since the weight of the fuel (approx 6 lbs per gallon) makes a difference, if you go from empty tank to full you might want to add a couple of 25 lb weights to the passenger compartment.