Turbo, Fuel Octane and Altitude

Wondering what people's opinions are with the turbo engine as far as octane choice as it pertains to altitude (and outside temp) to some degree. I'm taking a road trip in my GTR going from Sea Level with temps in the upper 70's to high desert (3,500-4,500 ft) and low 90's then eventually to mountains (6500ft and low 80's). Not towing anything. I'll be driving in Washington, Oregon, and Nevada. I've read the HP increase on the higher octane fuel is really over 4000 rpm, but that's important when passing. Think I should just fill up with Costco Premium along the way and not give it a second thought. It's so much cheaper than elsewhere even if I don't use the extra power much between the Costco price and probably slightly better gas mileage on a 1,500 mile total road trip I'm not going to net much savings being a cheapo with the 87.

Also curious if the turbo will compensate for the higher altitude or if it will be more limited by the max spindle speed than by the max pressure?
 
You need to go through about 2 tank of gas for the engine to learn and realise that yoi are now using premium. Turbo engines do compensate for higher altitude, to a point, meaning that the engine won’t be affected by altitude as much as a normally aspirated engine. Honestly i have driven fully loaded car with normally aspirated engines through the mountains, and the loss of power isn’t as big a deal as people make it seems.

use premium if you want the extra power to pass people above 4000 rpm, otherwise i wouldn’t worry about altitude at all and enjoy your trip !
 
The power loss in a NA engine is substantial - if you drive over passes of 8000 feet or more. The turbo will make a big difference at high altitudes. But at 5000 feet, not so much. Depends on the actual altitudes the OP will visit.

It will not take the PCM two tanks to “learn” the new fuel. The knock sensors will immediately see that the higher octane fuel doesn’t allow as much knock and will keep the timing advanced - direct feedback loop. I know that for my own driving (1500 to 9500 feet) I prefer 91 octane to 87. I don’t care about the small cost difference, I like performance when I want it. Higher octane fuel is a guaranteed 23 hp increase, a lot more than a low restriction intake or a cat-back exhaust costing over $1000.
 
I have a 2019 GT Reserve with the turbo engine. I have run nothing but 91 Octane since the first time I filled up the tank. I live at about 6000 ft altitude. Summer temps, like today are around 90 here and 100-105 down in the closest city that I go for groceries or other supplies a couple times a month. I'd run 93 Octane if it was available in California. The downside is fuel costs are outrageous here, like $4.79.9 in town and $4.69.9 down the mountain.

I now have around 11,000 miles on the car and the lifetime average is 24.9 MPG. I suppose if I were making daily freeway commutes where the car will get 27 MPG I may be tempted to try the 87 Octane, but I really like having at least 240 HP under my foot. I'm not driving very much now that I have retired so I'm not really impacted too much by the cost of gas now.

I can say that between the excellent gear ratios of the auto transmission and the abundant torque at low RPM's that the 2.5L Skyactiv turbo makes, I have no problem hauling ass back up the highway with a few hundred pounds of cargo in the back. Would I like 300 HP ? Heck yes. But I think the CX 5 does pretty well with around 250 HP.
 
Well luckily Costco is just about everywhere now so I think my gut that's it's a no brainer to just keep premium in it for the trip, get a little better MPG, enjoy they few extra horses. If I was doing a 4000 mile trip coast to coast I might think otherwise but for a few extra dollars having that little bit more oomph on passing is worth it.
 
Oh on the two tanks to "relearn" I think CarpeDiem is right, the engine can adjust the timing on the fly when you step on it. I do think the car does customize the shift points (non-Sport mode) based on learning your driving style. So if you are heavy footed, it shifts a bit earlier and vice versa.

Another interesting side note is I emailed Mazda one time about what the Sport mode actually does. They told me it changes the shift points AND alters the air/fuel mixture. That latter seems odd as that would mean to me you aren't getting all the full HP out of the engine unless you are in Sport Mode. So not sure if Mazda CS was correct on that answer but it was straight from Mazda. I thought it just changed shift points and slightly quickened the throttle response.
 
I live in Oregon. I prefer higher octane when not commuting and taking trips up higher elevations. Makes passing so much smoother
 
You need to go through about 2 tank of gas for the engine to learn and realise that yoi are now using premium.
Actually, the PCM does not recognize octane, Premium or the like, directly. What is does note as soon as higher octane fuel reaches the injectors (maybe 15 miles to make it from tank, through fuel lines, HP pump, etc) is that the knock point is further advanced for ignition timing, which has the effect of increased combustion pressure and thus torque at lower RPM's, and horsepower at higher RPM's. The PCM continuously monitors the knock sensors and contantly adjusts the timing just below knock point.
 
Last edited:
Actually, the PCM does not recognize octane, Premium or the like, directly. What is does note as soon as higher octane fuel reaches the injectors (maybe 15 miles to make it from tank, through fuel lines, HP pump, etc) is that the knock point is further advanced for ignition timing, which has teh effect of increased combustion pressure and thus torque at lower RPM's, and horsepower at higher RPM's. The PCM continuously monitors the knock sensors and adjusts the timing just below knock point on a continuous basis.

Not entirely correct. The benefit of the higher octane fuel starts at 4000 RPM for BOTH HP and Torque. According to this graph, the engine performes the same below 4000 RPM despite what octane fuel you are running. The torque just rolls off more gradually after 4000 RPM.

Update: I think you could be technically right as I forgot this graph is for the pre-2021 engine. Torque is up to 320 for 2021 with premium fuel and 310 with 87 so technically if the peak is 10 lbs higher at 2000 RPM, then yes, there is a very small torque increase below 4000 RPM.

TC.png
 
Last edited:
I've seen this graph before, and it appears idealized to support the marketing claims of octane. As the ignition timing can be (and if you monitored OBD PID's, it likely is) improved at lower RPM, and as expected, much more enhanced as boost pressure and RPM builds, which exacerbates knock.
 
Curious to know if anyone here as references to the fact that there is no adjustment period? I admit that when i said 2 tanks i was just going by what i had heard and what every 0-60 tester do as best practice. ( there is also the issue of octane dilution between the left over gas in the tank, etc.)

but i would be surprised that it adjust immediately. I feel like that would make the engine output very inconsistent. I agree that the knock sensor may cause the engine to immediately adjust to reduce knock and prevent damage when it happens, but i am not convinced that it reacts the same way in the absence of knock.

But i am curious and would be happy to hear if anyone as a viable source on this specific engine.
 
I've seen this graph before, and it appears idealized to support the marketing claims of octane. As the ignition timing can be (and if you monitored OBD PID's, it likely is) improved at lower RPM, and as expected, much more enhanced as boost pressure and RPM builds, which exacerbates knock.
The graph is from an online presentation given by one of Mazda’s engine designers. He stated that this is the actual dyno-tested difference in hp/tq between 87 and 93 octane, stressing that for most drivers 87 will work just fine since they seldom exceed 4000 rpm.

BTW - of course both hp and torque will increase, dynos only measure torque. Horsepower is calculated using the torque values.

HP = (Torque x rpm)/5252

HP and torque are always the same value at 5252 rpm (using SAE units).
 
Consider that the published curves are (almost certainly) WOT runs, where the mixture is rich (open loop) and ignition timing is heavily retarded at lower RPM's and high load. At partial throttle, where the PCM maintains closed loop operation (leaner than WOT), there is--for the naturally aspirated 2.5L at least, and I *suspect* similar for Turbo version-- a QUITE noticeable difference in throttle response, torque and ignition timing strategy, between 87, 89 and 91 octane. Probably 93 octane as well, if I were to seek that out in North TX.

While Mazda doesn't espouse it in their owners manual, others, like Honda in the Pilot manual, directly nod to the increased low and mid-range RPM bump in torque that reduces towing gear hunting, which higher octane fuel provides for vehicles with knock sensors. Toyota did a little marketing trick where the same drivetrain in a Highlander had slightly lower torque and HP published, than the RX-350, for 87 and 91 octane recommendations, respectively.
 
Lots of assumptions in the previous post, not supported by any data. I cannot dispute it on fact, but be advised it is pure conjecture. Since Mazda published the curves one could assume they are fact - unless one considers conspiracy theories in which case facts go out the window….
 
No conjecture, I have observed it with ODB data PIDS. Every vehicle I've owned that has knock sensors behaves in a similar way; using higher or lower octane fuel has a direct impact on ignition timing under a variety of load/rpm, which with a basic understanding of internal combustion engines means this directly impacts the efficiency of the power stroke, and the production of power, inferred as force on the piston, which is converted to twisting force on the crank, which we measure as torque, and which, by virtue of RPM can be calculated as the ability to do work, which we call horsepower. So let's agree the Mazda curves posted are fact, but also understand that this one curve set does not represent the engine under all conditions, but rather very specific conditions of throttle, fuel mixture, ambient temperature, etc. Under those conditions, no doubt the improvement of pre/self ignition are not significant until higher RPM.

Last, telling the average CUV or pedestrian sedan customer that they must use 91 or 93 octane fuel at a 50 cents or more premium per gallon for optimum performance, would be mostly lost on them, if not irritating. Telling a Miata owner to use premium (which Mazda does) for its NA Skyactiv, makes sense and would be embraced by the owner:

Snippet from 2020 MX-5 manual:

1626882266079.png
 
AVC is correct. As he said, the curve published is fact, but it is only a curve for Wide Open throttle under specific conditions. It is possible that the engine reacts differently under partial throttle when using 93 octane instead of 87. As he mentioned some cars do that specifically. That specific engine behaviour is what sometimes lead to claim of higher fuel efficiency when using premium fuel. It can when the engine can adapt and use the higher octane fuel to its advantage, but it isn't a rule for all engine. To the users there are not a lot of advantages to this, if the engine develops more torque at partial throttle with 93, then the driver usually adjust and give less throttle to match the acceleration they intended.

So has AVC said, partial throttle may be different on this engine as well when using 93 octane, but you will reach the same maximum torque at full throttle, and we don't have power curves to document the output difference. I also haven't seen any significants reports of better fuel economy when using 93 over 87 for this turbo engine.
 
I just ran a tank of 93 in the last week, only because it has the Shell extra skookum nitrogen stuff , and seemed useful to keep those fussy direct injectors clean. A quite noticeable uptick in throttle response at light to moderate throttle w/ the AC on. In theory there *should* be a slight improvement in mpg as the more advanced ignition timing improves power efficiency a tick. My tank avg was up 1mpg or so, which might be because of 93 octane, or just normal variance. If the former, 1mg improvement isn't worth the extra cost per gallon, per tank. IMHO.
 
Toyota did a little marketing trick where the same drivetrain in a Highlander had slightly lower torque and HP published, than the RX-350, for 87 and 91 octane recommendations, respectively.
Toyota tricks date back at least to 2004 Highlander, RX330 and Sienna which I recall from having checked these vehicles out at the Chicago auto show. Those vehicles shared the same 3.3L/5-speed drivetrain. Highlander and Sienna were regular fuel vehicles; the RX was premium "recommended." These vehicles had the same HP and torque specs except, oddly, the Highlander hit the 242 ft-lbs at 800 higher RPM as I compare them now.

Sienna was V6 only but with no AWD option at the time. The Highlander still had a 4 cyl. option. The RX was V6 only with the AWD option, inherently less efficient across trims sold compared to the others. Recommending premium on the RX was an evident attempt to get that model's mpgs across trims up a scootch, maybe 1+ mpg while RX buyers would be more inclined to pay up for fuel. I bought the Sienna.

On the subject of altitude driving, I was running 89 octane/10% ethanol in the 2004 Sienna because it happened to be cheaper than 87 octane ethanol free where I lived at the time. On a road trip while climbing from the flatlands toward Yellowstone it stated pinging under load, accelerating uphill. I switched to premium and the ping went away within a tank. The vehicle went another 125,000 miles of trouble free driving before trading it in.

Now, with my 2014 Sienna and the next gen drive train (3.5L/6-speed), I run 87 octane/10% ethanol and when climbing to 5500 ft. crossing the Appalachians twice per year I've never encountered a ping. Since 2014, Toyota had added direct injection, increased HP and had gone to an 8-speed trans until they went hybrid only in the Sienna.

The point of the story is that over 10 years and a next gen drive train, the Sienna, et. al. engine and software underwent further refinements to evidently include anti-ping controls at altitude. Despite hiccups with one maker's model or another, internal combustion technology across the industry has advanced since that pinging Sienna from 17 years ago.

If I were the OP driving to 6,500 ft., and I was accustomed to happy driving on regular fuel for economy's sake, I'd keep using regular fuel unless there is some specific evidence to the contrary in the Mazda turbo. If it does happen to ping or show an unsatisfying power loss, then I'd make the switch.
 
Last edited by a moderator:
Where are you getting E10 97 and 99 octane fuel?

As for ping, I've had opportunity to test this, but I suppose there is lower limit to ignition retardation--your 2004 would have had knock sensors-- that would make pinging pronounced; e.g the PCM detects the knock from the knock sensor, but can't make further timing retard, and thus only highly octane fuel would solve.
 
Back