Many of us have experimented with using regular gas despite the manual's warnings that OMG THE WORLD GONNA END, and found that it works just fine. I ran a 1,000 mile test and found the fuel economy was slightly higher with regular.

The motor should use the knock sensors to adapt on the fly to regular, retarding the timing whenever a knock is sensed. This should result in lower fuel economy, along with reduced power. So the results of my prolonged test were counter intuitive.

I think I figured out why I got the results I did. In the Variant Coding tool on my Autel MP808, I found that the ECU can be set to adapt to fuel quality. This would explain my results; the ECU had already adapted to regular, so running premium offered no benefit.

I found elsewhere a reading for the octane that the ECU had adapted to. But here are the switches available. One enables the adaptation:




And I guess this is the stored result of the adaptation:



The motor would probably be happier, and burn cleaner, with premium, but I question whether it is worth the 10% difference in fuel cost.

It would be interesting to run premium for a while to see if the adaptation resets Low Fuel Grade to "No" by itself.

Note this seems to imply that the system does not adjust timing on the fly, but rather over an extended interval.

I recall when I first switched to regular, I heard knocking in certain situations, like very light throttle and moderate rpm, in which one would not expect a knock in the first place - and certainly not the sort of situation in which a knock would be cause for alarm. Obviously not healthy, but not indicating panic either.

 

Do you have the 450 or 550? Wondering if the compression ratio difference between the two (if any) will make regular gasoline better digestible by one vs the other..

 

450

Same compression ratio.

Curiously: I assumed the bore was simply wider, but it turns out the stroke is longer also. That means the cylinder head's combustion chamber must be not only wider but taller, or the compression ratio would be higher.



​​​​​​​10.7:1 is a high compression ratio. I am surprised the motor burns regular as well as it does.

 

That's what I'm thinking too. Good to know that and you also have the empirical data.

 

Direct injection allows engines to run compression ratios that would have required race gas in the old days. We also have a 16 Mazda 6 in the house and it only requires 87 octane with a 13:1 compression ratio. The turbo version has 10.5:1 and can use either but Mazda claims ~25hp more with premium, but a turbocharged engine with a high static compression ratio is a severe octane test. Now if they could figure out a way to get ride of the noise from the DI pump at idle.

I'd be interested to see some datalogging on a GL to see what the difference in timing is.

 

Yeah, DI basically rewrites the script. The fuel goes in when you're ready for it to go bang anyway.

I'd be mighty interested to see some data logging too! I found no live readout of ignition timing anywhere. If you know of anything, please advise.

 

I'll put it on my list of things to do one of these days. I've got an app on my phone and a dongle I use on another car that should work.

 

Some people use octane booster with regular, I don't recommend it but interesting information OP, thanks for sharing.

 

Interesting - I have also come across this in DAS.
i wonder if it’s selectable so it doesn’t have to adapt.

 

Yes, the adaptation (#80, first picture) is a switch. I think it was Active / Disable.

 

My GL450 ran happily for 200k+ miles on regular gas. Went 200k before the first plug change. Didn't need oil between oil changes. I can say for a fact that it is OK to run during normal driving. If you tow or drive aggressively, you may want to run premium.

 

I think this is pretty much automatic… And that it senses knock against a given fuel map for a known octane rating and then trips the yes flag here. I have always ran premium in this but today I ran it dry and put in a quarter tank of regular and it had learned it.
It might be possible to do some damage were you to manually trip this in developer mode to NO and then run regular, in that it might ping and detonate in the time it would take to readapt itself to the low octane fuel flag.

 

There was actually a time when the EPA considered banning regular gasoline as the higher efficiency of high compression engines would create significant fuel savings. It was just a theoretical exercise and implementation obviously wouldn’t have been clear cut. As always, consumer cost is not important .. just consumption.

What remains, however, are basic principals … and despite reports “runs fine,” and “no problems,” engines with timing retarded to permit low octane fuel are running less efficiently and using more fuel when their existing compression is not being fully utilized.

The retard mechanisms are there solely for damage prevention.

The EPA might instead study whether price differences between low and high octane fuels are realistic based on actual production costs and because fuel consumption is still important. Perhaps high octane prices might be subsidized. The next rounds of CAFE (corp avg fuel economy) mandates will force more high compression engines without question. (Electrics off topic)

 

That is correct @lkchris , I am expecting some hp lost to retarted timing of running a lesser octane fuel than the engine is optimized for.

but the way I drive it won’t be missed.

 

Great info, and glad to see the confirmation from DAS - I am going to switch to premium and see how quickly it re-learns it. Thanks for sharing the info.

My family owned a 1971 Volvo. Thanks to cam timing or mixture or whatever, that thing would knock like nobody's business if you put less than 104 octane in it. My dad would insist on driving it in the highest gear possible for fuel economy, knocking horribly up hills. I was told somewhere around 150-200k miles the motor gave up. Preignition is bad, but it is no death sentence.

You might enjoy the experiment - like I said, when I switched to regular, I heard knocking in the oddest situations, part throttle at moderate rpm. Had it been knocking at full throttle, I would have been concerned.

I am sure the system does both a on-the-fly timing adjustment and a preset. I am curious how the on-the-fly feeds in to the preset, but there is almost no way to find out, short of a data logging of timing.

As far as the EPA is concerned: I doubt the occasional retardation of timing is less efficient than having a lower compression engine.

 

It is a good question. You lose some power, and some efficiency, but 10%? To justify the price difference?

On the other hand, if you don't drive very much, 10% of a small number is not a very big number. Does it also burn better and leave less carbon deposits?

If you get a chance, take a look at the mileage you get having switched to regular. Very curious to see what the results are. 10% would surely be a noticeable decrease in mileage.

 

Yah all are lucky. We only have one unknown grade of petrol in Nigeria.

Ferraris, Lambos, AMGs, are all running around here with unknown lifetimes.

 

I put in a full tank of 91 octane. I was surprised that after a 50 mile road trip the adaptation had not gone back. It might need several drive cycles, or it might not reset on its own, out of conservatism.

I set the flag to "NO" and also there is an adaptation for Ignition that seems to be the same thing; I set it to 93 (we are at some altitude here in Idaho; it was at 87, and has 91 and Base as options) and will see what happens. I didn't hear any knocks in the customary part throttle / low speed / up hill setting. My motor has enough problem with carbon buildup, I figure I should be as kind to it as possible.

Curiously, the one time I thought I heard pinging was right after starting up. Maybe the engine is still in open loop mode at that point.

One observation: The exhaust smells better. Recall I have gutted cats, and at first the stink was enough to make me want to replace the cats. Now it is more mild; I think the combustion is better. It will be interesting to see whether the spark plugs are less carboned.

 

I switched back to 87 octane several tanks ago. The ECU still has not set the flag to "low quality fuel". I had run the adaptation for 93 octane a while back, so presumably that is the map it is still using.

It is curious that the motor still has not adapted down. Winter driving? Boise is at 2800 feet, so that helps, but still ... Mileage is not great but I have generally observed that cold weather = worse mileage.

 

So my M273e46 runs happily on 87 and has for almost half a year now and 10k miles.

The family truckster’s GM 6.2; however, ingested one tankful of 87 and was rakkatakka kerrack krrkk krrrlling itself all over creation with not even the slightest attempt at adapting itself, until I could up-octane the fuel with half a tank of premium.

 

Odd that the GM ECU does not adapt to the gas better.

I wonder if my motor's adaptation to "bad gas" was due to the carboned intake valves. When carbon deposits cause a problem, it is through preignition - which I assume the motor responds to by mapping ignition to retarded settings. "Oh no. It's retarded."

I did not check the adaptation before the carbon became a problem.

If your motor quickly adapted to "bad gas", you might want to try resetting the value to No after a few tanks with a top end cleaner like Regane Complete in there. It could be you have carbon bothering the motor. A few bottles of Regane is a cheap experiment.

 

regardign fuel adaptations, you will see the quickest changes made when doing WOT pulls.
When we tested octane sensitivity and burn rates when i was doing my benchmarking job, the most aggressive and quickest adaptions were made along the top curve of the load map.
This was done by draining the tank, running the engine out of fuel so lines were clear, filling with new octane, then doing 3-5 back to back WOT pulls.
To fully populate the entire load map takes a signifigant amount of driving. 50 miles especially with having 4-5 gallons in reserve will not fully adapt.
I run regular in the winter, and premium in the summer. Heavy foot both times. But 95% of its time is spent cruising 60 mph back roads steady state. Noticeable power difference during WOT acceleration.

 

The majority of my driving now is 60 mile round trips which include going over a pass on a two lane road. Thus about once each direction I accelerate mightily.

It has been several tanks now, so I believe the 91 octane is long gone.

When I first switched to 87, long ago, I observed that preignition happened in the sorts of environments one would not expect. It was at altitude approximately 4000 feet, on warm days, with low part throttle. I could hear the rattling of light knock going at low speed around bends in mountain roads.

Now, the motor collapsed into carbon-based preignition around that point, so the carbon may have been causing the problem. This was the first of two incidents with carboned intake valves; a valve job was performed under warranty.

In other words, combustion chamber carbon is a variable that ought to be included. Perhaps if you always run Top Tier gas, or regularly use top end cleaners in the gas, it will not be an issue. I will continue to run 87 octane gas and will recheck whether the motor has adapted. I am tempted to also put a borescope back in the cylinders to look for carboned intake valves.

 

Update: The self adaptation is still set to "low quality fuel / no". I have been burning 87 octane the entire time. What's more, the manual adaptation I performed a while back was for 93 octane.

There is no audible knocking under any circumstances.

I have run several bottles of Gumout Regane Complete, which is chock full o' PEA, through the motor. Perhaps I now have nice clean intake valves.

Perhaps someone else, @Max Blast perhaps, who has a bidirectional tool and wants to experiment, can try running some cleaner and seeing if the adaptations still go to "low quality fuel / yes" afterward. I can't think of any other reason my motor should be this happy with 87 octane.

 

Yeah, maybe I can hook up the tool and see what mine still set at at… I’ve been driving regular for well over a year, about 12 K and I have never heard a ping out of the motor. Just a re-cage everybody’s brains, the day I switched to 87 octane on an almost empty tank; from the drive home from the station the ECU had sensed and automatically set the low octane flag so I didn’t have to do anything in star to make this happen.
So I would be very shocked, appalled and surprised if that flag were now set to no.
And no way in heck am I going to set it to yes with 87 in the system.

I am still a little fuzzy on how the actual adaptation occurs, whether it’s a set number of pings over a certain amount of time that automatically retards everything and sets the low octane flag. And then, conversely, in order for the ECU to set the low octane flag to no?