Mazda Skyactive octane/mpg

Status
Not open for further replies.
Joined
Jan 10, 2010
Messages
1,746
Location
Rochester, NY
This summer I got a new car, a 2016 Mazda3, stickshift, 2.0L. In the US the owner's manual recommends 87 octane and the engine has a 13:1 compression ratio, but in other countries it recommends higher octane and has a 14:1 compression ratio. I was interested in whether higher octane would have any benefit in the US version of the engine.

I didn't do a super scientific test, just did 2 fillups of 87 octane followed by 2 of 93, then 2 more of 87 and 2 more of 93.
Here's the raw data:
Code:


fuelup_date model miles gallons mpg octane

10/17/2016 Mazda 3 309.9 8.295 37.36 93

10/11/2016 Mazda 3 306.6 7.672 39.96 93

9/22/2016 Mazda 3 344.1 7.846 43.86 87

9/9/2016 Mazda 3 334.6 9.286 36.03 87

8/29/2016 Mazda 3 362.1 7.806 46.39 93

8/23/2016 Mazda 3 362.7 9.105 39.84 93

7/28/2016 Mazda 3 331.4 9.206 36 87

7/17/2016 Mazda 3 302.7 8.315 36.4 87


And here are the statistics. I tried to do a box-and-whisker plot, but I can't spare the time to figure out how to make Excel give me what I want!
Code:


87 octane mpg 93 octane mpg

median 36.215 39.9

q1 36.0225 39.22

min 36 37.36

max 43.86 46.39

q3 38.265 41.5675


So it seems I get a little better mpg statistically with 93 octane, but it's certainly not worth the extra cost!
 
I think this has been discussed before on Mazda forums and tends to come to a similar conclusion. Though a lot of them decide to either run midgrade or 93 because they claim it seems to have a little better acceleration. In theory this makes since as it is a high compression engine and higher octane fuel would help substantially with keeping this engine from pinging which would let the engine run at full power more often.
 
Its not all or nothing with octane. Its finding the sweet spot for optimum combustion timing. And sometime you might optimize a low rpm luggability to the detriment of high rpm. On my car with lowly 9.X comp ratio, it usually wants a couple to 3 gallons of 93 on top of the 87 tankfull in "dry" weather. In High humidity it does fine on swill.
Anymore octane and it gets sluggish. Also it appears to take more run time than you would expect to write over LT fuel trims and timing maps. I still am disappointed with typicall OBDII engine performance and mapping and snowballing/ detuning. Lambda control with safeguards aint what it could be. MAybe it is in UHP cars. I hear the monster corvette still has problems due to failsafe nannies where the engine choked at (possibly) bad fuel and would not build advertised boost/power. Think the was RodentRack?
 
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.
 
Originally Posted By: Bottom_Feeder
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.

The idea is that the engine may have to retard timing to get pinging in check, and retarded timing reduces efficiency.
 
The problem is gas mileage is dependent on conditions as well.

I always get better gas mileage in fall and spring.. or even with certain weather conditions.

Last night driving to Grove city I got 25.5mpg@75mph on the way there.

coming home I got 30.5mpg@80mph

Now If I had filled up @grove city would I say the gas did it?
but I didnt fill up it was merely atmospheric ie headwind.

Throw in A/C use traffic.. even stopping at a drivethrough.. and there is no direct comparison. In my Case it was a straight 90mile drive with no A/C use.

I do feel there might be a slight bump in MPG with 93 on my vehicle as well.. but its definitely not worth the 25% price difference for premium here
2.09 vs 2.59

Now If I had a turbo that is almost guaranteed to run better on 93 vs 87.
 
Originally Posted By: NateDN10
Originally Posted By: Bottom_Feeder
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.

The idea is that the engine may have to retard timing to get pinging in check, and retarded timing reduces efficiency.

I get that, hence my '... if the engine runs better on it' statement.
 
My parents GTDI Escape gets enough of a mileage increase going from 87 to 91 that the 91 ends up being something like $.50 cheaper per 1000 miles.

We calculated that a few years back when gas prices were higher and there was a larger price difference between 91 and 87. They went from 23 to 26/27 with it in daily commuting.
 
Originally Posted By: Bottom_Feeder
Originally Posted By: NateDN10
Originally Posted By: Bottom_Feeder
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.

The idea is that the engine may have to retard timing to get pinging in check, and retarded timing reduces efficiency.

I get that, hence my '... if the engine runs better on it' statement.


So what is it you do not understand? or were you just making a statement?

I would say most cars these days with anything approaching 10:1 compression and above would benefit mileage-wise and power by increasing the octane.

This is due to the higher efficiency attained by keeping the timing advanced closer to peak efficiency rather than the point where the knock sensor gets angry. (as has been pointed out already) I have proven this with my own vehicles as well.
 
Last edited:
Too many uncontrolled variables involved in any fuel economy testing described on BITOG. Also, Mazda states 87 octane is to be used for Skyactiv engines in the U.S. That is why the U.S. version has lower compression than other countries' versions. If the car would "run better" or get better fuel economy on premium gas what would be the drawback of Mazda saying so in the owner's manual? I have seen no convincing evidence that using higher octane improves anything. I ran a few tanks of premium in mine and realized that if there was a difference I would need some sort of precision equipment to tell because I didn't "feel" anything.
 
Originally Posted By: Miller88
My parents GTDI Escape gets enough of a mileage increase going from 87 to 91 that the 91 ends up being something like $.50 cheaper per 1000 miles.

We calculated that a few years back when gas prices were higher and there was a larger price difference between 91 and 87. They went from 23 to 26/27 with it in daily commuting.


Right that is a turbo car. Almost always its worth it to run at least 91 octane in those.

the FA20DIT in some subaru foresters has a 10.6:1 ratio and boost(psi) in the mid-high teens. giving it a very high effective compression ratio. .. it will run "crippled" on 87 but 91+ is really a must.

The FB25 NA has a 10:1 and I can tell with 93 some difference over 87 however my gas mileage would have to go up near 7mpg to cover the cost difference so unless I know I'll be towing in hilly terrain I usually roll 87 octane.

Now back when it was 10cents extra for 89 and another 10cents for 93 (20c total)
I would just run 93 because I hate pinging.

Right now the premium is more 50-60C total for 93 vs 87. or around 25%
 
In my '14 Mazda 6, I get better mileage on premium too.

The high-compression Skyactiv engines pull all sorts of tricks in order to avoid predetonation with lower octane gasoline. One particular trick I heard somewhere is that when you use regular gas, the engine runs slightly rich (maybe only under load) to keep the pistons cooler. Not sure if this is accurate, but given that it has been widely reported that Skyactiv engines have avoided fuel dilution by using premium gasoline, it stands to reason that the reason for the fuel dilution could be that the engine is running rich on 87.
 
Originally Posted By: danthaman1980
...it has been widely reported that Skyactiv engines have avoided fuel dilution by using premium gasoline, it stands to reason that the reason for the fuel dilution could be that the engine is running rich on 87.


I have seen it reported by one member on BITOG. Do you have other references to this phenomenon other than badtlc's posts? I ran 30,000 miles on the same oil using microGreen filters and had no signs or symptoms of fuel dilution in the UOA.
 
Originally Posted By: DBMaster
Originally Posted By: danthaman1980
...it has been widely reported that Skyactiv engines have avoided fuel dilution by using premium gasoline, it stands to reason that the reason for the fuel dilution could be that the engine is running rich on 87.


I have seen it reported by one member on BITOG. Do you have other references to this phenomenon other than badtlc's posts? I ran 30,000 miles on the same oil using microGreen filters and had no signs or symptoms of fuel dilution in the UOA.


I've seen similar discussions on mazdas247.com, mazda3revolution.com, and mazda6club.com - but I guess it is possible that all of those discussions were based on badtlc's BITOG posts
21.gif


Perhaps my explanation was unfounded, but my experience, however anecdotal my evidence might be, is similar to NateDN10's; marginally better mpgs on premium in the mazda. Certainly not worth the additional cost from a strict miles per dollar calculation, but if I decide to spend my disposable income on premium fuel instead of UOAs and microgreen filters, that's my prerogative.

Oddly, in my Chevy- after calculating fuel mileage almost every tank of gas between 2004 and 2014 (stopped calculating mileage in it a couple years ago) - or about 200,000 miles worth of fuel mileage calculations - I eventually found I got better mileage (marginally) on 87 octane, ethanol free. Actually there seemed to be a sweet spot at about 3% ethanol content (e.g., 1/3 tank of E10 and 2/3 tank of ethanol free). Why? I have no idea!
 
Good to know if the car will be used for a highway trip where gas stations are farther apart, or might be closed as you drive by. That extra 30-40 miles of range might be the difference between getting stranded or arriving safely.
 
Originally Posted By: Bottom_Feeder
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.


This isn't the dark ages. Modern engines adjust A:F in addition to timing. Higher octane in most modern engines will lean out the A:F. Plenty of UOAs on this forum showing less fuel dilution in UOAs when running premium is real world evidence in the event you can't go and view the timing and A:F on your own.
 
Originally Posted By: badtlc
Originally Posted By: Bottom_Feeder
Not sure where or why you're making a connection between octane and fuel economy. Higher octane is to fight pinging and pre-detonation, not for increasing fuel mileage although that may be the end result if the engine runs better on it (even if the manual says it doesn't 'require' it).

Running 87 octane in a 13:1 engine just does not seem right to me, but it's the 21st century so who knows.


This isn't the dark ages. Modern engines adjust A:F in addition to timing. Higher octane in most modern engines will lean out the A:F. Plenty of UOAs on this forum showing less fuel dilution in UOAs when running premium is real world evidence in the event you can't go and view the timing and A:F on your own.


Right. The ideal tune allows for stoich or lean combustion, and has to fatten up and/or retard timing for knock events. It doesn't need to ping to pull timing and/or add fuel. The ECU may be responding well before any audible knocking or pinging is heard.

One thing to additionally note: timing for direct-injection engines means not just ignition timing, but also injection timing. Fueling plays an important part in maintaining lower temperatures or injecting at a timing consistent with max pressure at a point too late to cause pre-ignition.
 
2 tanks alternating is not really enough time IMO to optimize some long term 'trims' ie atkinson mode 'virtual throttling' and cruising RPM range.
 
Status
Not open for further replies.
Back
Top