what does a car computer use to compute oil life remaining?

JohnnyG, No argument I was just trying to say what if the oil had a starting TBN of 5 or of 9, the OLM cant know which is in there, what if you have a bit more blow by than the factory accounted for or during the API change they dropped the concentration of an additive not speced in the manufacturers original testing.
 
deep neural network with three densely-connected convolutional layers, two pooling layers, adversarially trained with millions of labeled training samples from different brand oil meeting manufacturer specs at different mileage intervals and different driving conditions. they are state of the art regression machines.

just kidding! it’s a random number generator, pick a number between 5000 and 7500 🤣🤣🤣
 
I have tested my 2013 Ford Focus SE, by having my mechanic NOT reset the OLM. I do it manually when the notice comes on. I've done this 5 times, winter, summer in the hot Mojave Desert, etc. There's no algorithm, or anything complex. It's 9,950 miles. Period.
 
My 2018 F150 seems to simply use 10,000 miles.
I know at least on 2015 and newer Fords, they will count down the OLM about 2% a week (100% to 0% in a year) if the engine is never started. You probably drive your F150 all the time, so never see that part of the OLM working.
 
Before an oil life monitor could even begin to give an accurate estimate, it would first need to know what oil you used. Synthetic, a blend, conventional, quality and additive package, and more.
 
My 10 year old Honda ( 2012 but door sticker says made in spring 2011 ) MM OLM seems pretty smart and varies widely depending on type of driving. The Longest it will let me go with my type of driving is 12000 miles and around 7500 when mainly city and idling a lot.

It penalizes me when I drive faster. They raised the speed limit on the highway I use to go to work and I notice it’s less generous and says to change the oil more often. I trust it totally.

The OLM in mine certainly isn’t sacrificing engine life to lower the scheduled maintence to make Honda look like a cheaper car to own and operate.

Some people just don’t trust extended drains even when the manufacturer says they are safe when following the OLM. Not saying every one is accurate ( how would I know ) but an UOA would help them find that out.
I need to do a UOA on my two 2GR-FE cars and post it here (I have multiple 100+ mph for hours at a time), I just don't see how a well engineered engine at high cruising speeds even if you pushed it to 90 mph would take a toll on oil of today, I understand all the other parameters, just not highway cruising speeds, I could be wrong.
 
I need to do a UOA on my two 2GR-FE cars and post it here (I have multiple 100+ mph for hours at a time), I just don't see how a well engineered engine at high cruising speeds even if you pushed it to 90 mph would take a toll on oil of today, I understand all the other parameters, just not highway cruising speeds, I could be wrong.
My guess is that high-speed highway driving would be considered to be a similar usage regime to towing, though the loads on the powertrain would be less.

Engine RPM is higher, which can cause more rapid degradation of oil.

I would think that oils with less viscosity modifying polymer additive would degrade less, and more slowly. But, that should be the case in any usage regime.
 
Again, I think you're wishing for something (algorithm, multiple parameters) that simply isn't there.

I think all this car has is a rudimentary counter. Miles. Hours. Hours if you're lucky. Might just be miles and days.

No better than the counter on my Volvo, which adds up miles and days. When it hits 7,500 miles, or 365 days, it says, "Service Due". That's the sum total of sophisticated algorithmic parameter evaluation. Doubt GM put more thought into it than that when the car was designed over a decade ago.
OLM may be a good thing to have or not. Some may rely on it too much and not check the dipstick as often.
On short trips the Percent drops much sooner VS. highway runs.
 
Chevy Gen 4 engines used to go to zero in about 10,000 miles. Chevy Gen 5 engines count down to zero in about 7,000 miles, which is noteworthy in their fight against wearing out their DOD equipment. Yes, there are variations with the other items but doesn’t sound like it’s all that significant to me. I change at 5,000 miles anyway.
 
  • Like
Reactions: 4WD
Chevy Gen 4 engines used to go to zero in about 10,000 miles. Chevy Gen 5 engines count down to zero in about 7,000 miles, which is noteworthy in their fight against wearing out their DOD equipment. Yes, there are variations with the other items but doesn’t sound like it’s all that significant.
I reset mine … but once it’s below 20% it comes down to what day works for her and me …
 
I need to do a UOA on my two 2GR-FE cars and post it here (I have multiple 100+ mph for hours at a time), I just don't see how a well engineered engine at high cruising speeds even if you pushed it to 90 mph would take a toll on oil of today, I understand all the other parameters, just not highway cruising speeds, I could be wrong.
It’s not the speed of the car, it’s the “speed” of the engine ( RPM ). Engine RPM is definitely part of the programming.
 
Before an oil life monitor could even begin to give an accurate estimate, it would first need to know what oil you used. Synthetic, a blend, conventional, quality and additive package, and more.

The characteristics of degradation are known when running the recommended grade and rating.

This is one more reason why it makes sense to run the recommended oil.
 
Here are some references below:

Each auto maker has its own proprietary algorithm.

I think the best you will get is a "Meta-algorithm" which shows a general algorithm.

Shirley Schwartz cowrote this paper “Development of an Automatic Engine Oil-Change Indicator System” (SAE Paper No. 87043, 1987), and copatented the original OLM with Donald J. Smolenski.

Below is another but later GM patent:
 

Attachments

  • OIL LIFE ALGORITHM.pdf
    1 MB · Views: 24
  • Oil Life Monitor Patent.pdf
    429.9 KB · Views: 12
Last edited:
My wife's CRV gives every impression of using mileage (6,000) or time in the crankcase (1 year), which ever criteria trips first.
 
Or one year, whichever comes first.
As to the GM OLM, it was never a "rudimentary" device. Even back in 1999, the one on my STS was a very useful tool and would follow all of the parameters as described by slacktide, and possibly a few more. Yes, there had to be adjustments made to the algorithm over the years due to different engine designs and parts suppliers.
This subject has been written about on this site so many times, I'm really not going to go over it all again. I will say the my first reply on this subject was around 2005 when I let my Oil Life Monitor on my 2004 Chevy truck go to 12,000 miles and sent a sample to Blackstone. The report said that it had several miles of life left.
Just Google GM OLM and try to get the full story from GM. The woman engineer that devised the thing did a fantastic job. There is a patent on it of course, and none of the other car makers have been able to come up with a similar or better system of predicting YOUR oil change interval.
^^^This. I, too, have let my OLM go to 0% and done UOA (on ancient 2004 Buick). Oil analysis showed plenty of TBN left and useful oil. Of course, I run the World's Best Oil, Amsoil SS, YMMV, but that crude OLM on OLD GM products tracks more than mileage and RPM, and it is conservative. And I doubt the OP is going to access the formula on a patented device.
 
My 10 year old Honda ( 2012 but door sticker says made in spring 2011 ) MM OLM seems pretty smart and varies widely
We had an '05 Odyssey and our driving habits didn't change much, like most people, but the % when it told me an oil change was due always varied. It was always 6000+ miles up to 7500 miles.
 
Back
Top