Oil viscosity and fuel economy

[ekpolk]

Originally Posted By: fdcg27
I gave the '99 Accord an oil change last week, draining M1 0W-30 AFE and filling with M1 EP 15W-50.
The first tank (468 miles on 14.6 gallons) actually delivered better mileage than the last couple of tanks on the AFE.
The 15W-50 is a great deal thicker than the 0W-30 at any temperature, and has considerably higher HT/HS.
Also, the AFE would probably deliver better average fuel economy than the 15W-50, although the difference might be very small.
I used the same oil in my Accord last summer, and had a similar result.
The '97 Accord will get the same oil at its next change next month, while the van will get Tection 15W-40 next week, as it had last summer.
Those who claim that a thicker oil will kill fuel economy have seen diferent results than what I have observed.


I forgot to ask the most obvious question (and if I missed it, please forgive me...) -- why?

I'm not suggesting in any way that an xw-50 is going to instantly kill your elderly Honda engine, but OTOH, clearly, the engineers who built this engine never meant for it to be fed a 50 wt as it's primary, preferred lube. I'm not asking this in any disrespectful sense, but again, I'm just wondering what led you to give 15w-50 a go. I just filled all three of our cars yesterday with most of the rest of our stash of 12 cSt GC (including my hybrid Camry), so I'm obviously not allergic to thicker oils, but fer cryin' out loud, 50 wt in ANY Accord -- that seems extreme, even to me.

 

[ekpolk]

Originally Posted By: JRed
. . .
So yeah, oil weight vs. MPG is complete [censored] in my experiences.



OK, you were making sense until this last gem. I have concluded, and I make no representation that my conclusions are totally (or, well, even partially...) correct.

This said, it seems to me that ultimate fuel burn is a result of many factors. Internal engine drag/friction MUST be one of those factors, and oil vis is obviously related to that.

It seems to me that the REAL challenge here is to try to discern how, and to what extent, oil vis actually impacts the final outcome. Also critical is to ferret out the other important factors, and then determine how, and to what extent, they impact the final outcome.

The viscosity of an oil, and it's involvement in ultimate FE is not a factor to be dismissed out of hand. The real challenge is determining how viscosity, and all the other factors that are in play, influence the outcome.

 

[mechtech2]

Of course oil viscosity changes result in power and fuel economy differences.

 

[AEHaas]

API, SAE, ISLAC, and ASTM have goals, tests and standards. They have done for us the necessary tests to ensure that we protect our engines, increase efficiency and get the best fuel economy all at the same time. One thing they have shown us is that the thinner oils of today do give us protection, efficiency and the best fuel economy.

We know these things in racing as well. Formula 1 has made this a part of the regulations.

Oh and, thinner oils make for a better ring seal. All other thing being constant, the thinner oil will result in less engine drag, less wear on start up conditions, better fuel economy and more BHP. These are among the reasons thin oils are used in race cars.

Just because somebody smokes 4 packs of cigarets a day and lives to be 100 years old does not mean it is the best thing to do. When you combine the data 3 out of 5 people die from smoking related diseases. Seat belts kill a few people a year but save thousands.

I wear my seat belts and I use thin oils and I reap the benefits.

Some will always insist that smoking is good for you. They may even “back it up” with their own scientific studies. Follow them at your own risk.

aehaas

 

[fdcg27]

It wasn't that many years ago that Honda actually recommended oils as thick as an XW-50 for hot weather use in the US.
The Civics we had did many a mile on XW-50, and rarely saw anything thinner than 10W-40.
I am glancing through the car log for one of our '86 Civic Wagons, and I observe that even on M1 15W-50, the car would still average 40 mpg on the same commute I make today.
Why use such an oil in a VTEC Accord?
Why not?
I have long contended that oil grade made little measurable difference in fuel consumption.
I am trying to validate this contention.
Also, I have always though that the one grade for all climates, seasons and uses recommenations one finds in most owner's manuals is really just an effort to make things simpler for the average owner, not an optimal approach.
The average auto maker is not confident that the average owner will know that the 15W-50 has to come out before the onset of real cold.
The average owner's maintenance practices probably validate the auto makers' concerns.

 

[Art_Vandelay]

I just put Mobil 1 TDT 5w-40 into my Lincoln's 4.6.
Previously I ran Mobil 1 0w-20 in it.

After 20 minutes of driving I hit the reset button on my car's fuel economy computer. Then I tested TRUE city driving for about 1 hour. True city driving being defined as the car coming to a complete stop every 2- 3 minutes. Then averaging about 30 MPH while moving.

Yesterdays 5w-40 results were 12.8 MPG for that one hour.

The 0w-20 under the same circumstances was around 14.3 most of the time.

On the highway at a constant 60 MPH both seem to be too close to call. The car is always about 24-25 MPG.

It seems as though higher viscosity has the greatest impact on fuel economy from a dead stop. Once the car is rolling that deficit begins to drop.

Does that seem correct?

 

[Bill in Utah]

Computer in car = MPG GUESS = NO VALID DATA....

Thousands of miles vs accurate ODO vs how many gallons used = valid MPG data.

Bill

 

[Art_Vandelay]

Originally Posted By: Bill in Utah
Computer in car = MPG GUESS = NO VALID DATA....

Thousands of miles vs accurate ODO vs how many gallons used = valid MPG data.

Bill


You mean like when a scale is off by 10 pounds? Sure I agree with you.
The point you are missing is that such a scale can always distinguish the difference between a 100 pound object and a 110 pound object.

 

[ekpolk]

Originally Posted By: Bill in Utah
Computer in car = MPG GUESS = NO VALID DATA....

Thousands of miles vs accurate ODO vs how many gallons used = valid MPG data.

Bill


Bill:

Individual results may be off, but when you look at the data over the long haul, it's not usually that far off. I've found the computer to be within an mpg or so of the odo miles / gallons pumped method (and I've very careful to fill to the neck opening to keep things consiStent). 2/3 of the time, the computer is slightly optimistic, 1/3, it's pessimistic. Now, as for the rigorous validity determination, we can debate that for individual readings, but overall, the computer (mine anyway) seems pretty close.

But no, I would not bet anyone's life (especially not mine...) on the statistical validity of the computer's readings. Nevertheless, I've seen that can trust the readings to be very close to reality.

 

[Bill in Utah]

Not the ones I've used. (Many Toyotas, GM products, Subaru, and Scan Gauge (which you can make more accurate by filling into it's memory a few factors)

They are guessing via load, fuel flows, speed and such. Even a few tenths here and there is not really factual. The statement of I went this way and my digital readout said 22.1 for a week and after the oil change it now says 24.1 is not valid data IMB.

Go over a few tanks (even better a few thousand miles or a total OCI around the same time so weather/fuel mix are NOT in play) and total gallons into how many miles then I feel better about the data.

The trip computers are nice, but they are not as accurate as the real way to figure out MPG.

And stating that brand x oil or the type of oil gets 2 MPG better using the computer like many here do is not valid in MY Option.

Bill

 

[Art_Vandelay]

Bill I was only sharing an observation.

 

[fdcg27]

Bill, you are right.
Our Subaru's computer is consistently optimistic.
However, I would have to think that it is also consistent in its optimism.
Therefore, the instantaneous readout is probably a valid indication of actual consumption, even though the computer may not be valid in its calcualtion of fuel economy for an entire tank.