Others have offered several valid explanations as to why a specific engine might be more efficient at higher RPM. It is hard to design intakes and exhausts and fuel delivery systems to achieve optimal efficiency at all RPMs and loads. Another consideration for some cars is that an oversized spark ignition engine operating at low load is a big vacuum pump. As you increase throttle opening, you are spending a larger fraction of your energy doing actual work. That is why diesels are inherently more efficient than gasoline engines.

Those nice smooth curves of speed/resistance correspond to a cylinder in a wind tunnel, and when you look at the raw data, even the curves for a cylinder actually have irregularities in them as flow transitions. Those who have put real world objects in a wind tunnel know that the real curves are not nearly so well behaved.

In conclusion, if we had the time to do scientific tests under controlled conditions at 5 mph increments, most of us would find that the declining fuel economy curve is not a smooth exponential decay curve. A few might even find some bumps in the curve over certain speed ranges. All that being said, for the vast majority, the lowest speed that keeps you in high gear with the torque converter locked up is going to be your optimum speed, and there will be an exponential decay in fuel economy as you go faster.

There are almost too many factors to calculate between various vehicles, but yes, There are vehicles that will get better mileage at higher speeds than at lower speeds.

Using my diesel pickups as an example:

The oldest truck had mechanical injectors and an injector pump. It would idle for hours on end on vary little fuel. The newer trucks have electronic injectors and they burn fuel idling not much different than they do running. Part of the problem is all the new pollution garbage that requires the engine to rev itself up periodically to force the soot out which it didn't have to do previously, and couldn't do since it was mechancially controlled. Electronically, the computer does what it pleases with the engine and uses more fuel.

Road wise, the speed at which the engine gets the best mileage depends in a large part on the gearing in the trucks. As mentioned before, there is an optimum combination of horsepower and torque at particular rpms that will deliver the optimal mileage. For my truck, it's between 1800 and 2000 rpms. At that point, the turbo is delivering enough boost which in a sense substitutes for extra fuel to make power. The truck will get about 23 mpg if you keep it's engine running in that range which amounts to about 63 mph in 6th or high gear. Above 2200 rpms, or closer to 70 mph the truck gets about 19.9 mpg. The truck has a 3.73:1 gear in it now. If I changed to a 3.55:1 gear, I suspect it would move the best mpg speed up maybe closer to 70 mph.

This sort of relates to what the truckers are trying to say to the know nothing environmentalists around Knoxville, TN. They reduced the truck speed through the city last year citing the reason of wanting to reduce pollution from trucks. The problem is, most road tractors are geared to run optimally at 65 mph. They aren't geared to run optimally at 50 mph. Meaning, by slowing down, they are burning more fuel, having to shift more often, and creating more pollution than they would otherwise be doing.

Engines vary in the RPM range in which they are most efficient. My last 3 diesel pickups have all had 3.73:1 geared axles. They have however had various transmissions and engines. My 95 was a 7.3 Navistar Powerstroke with a 5 speed Borg Warner tranny. The 7.3 was pretty efficient running at 2500 to 2700 rpms and the final drive gear was slightly higher than the Dodge/Cummins/Aisin 6 speed tranny I now have. The 7.3 Navistar got around 18 mpg at anything from 60 mph to 85 mph and it really didn't make that much difference if it was loaded or empty. I had a 6.0 Navistar Powerstroke with a New Process 6 speed that didn't run worth a tinker's damn below 2000 rpms and was efficient at 2800 rpms or around 80 mph. It never did get more than 16 mpg empty. The Cummins absolutely drinks fuel above 2500 rpms. Mileage is down to 15 at that speed, but it sips along at 60. The final drive from the tranny on the Dodge is slightly slower than it was from the 6.0 L Ford and around close to the same as the 95 Ford.

Skipper

Just a bit more food for thought.
If vehicles are designed to get their best fuel economy at 48 MPH +/- then where does this scenario come in?
Assume xxxx model year Ford Ranger pickup with the V-6 and a 5-speed manual transmission. The EPA site will list mileage numbers for various models of Ford Rangers depending on engine/trans option.

My questions are - where is the axle ratio in all of this? Are you saying the mileage is going to be the same at 48 MPH even though there is a huge RPM difference between a 2:73 axle ratio and a 4:10?
I don't see the EPA providing mileage options based on gear ratios.

If one uses the logic that low speed/low RPMs provide better mileage then what will happen if one motors around town all of the time while shifting directly from 1st to 4th and never exceeding 30 MPH? If lugging is not an issue then mileage will not be affected according to most of the convential wisdom here.

While each powertrain combination has to be certified for emissions (including different axle ratios, IIRC), I don't know by what arcane process they figure the fuel economy for models with optional ratios. MAybe an average?

As far as shifting from first to fourth, apparently that works under light throttle with high torque, large displacement engines. This tactic was built into the shifters of Corvettes for a while: under very light acceleration, the transmission could only be shifted from first directly into fourth. Second and third were locked out. I guess this worked in the EPA testing, but I'll bet it was counterproductive in the real world. I'm sure 'Vette owners rarely accelerate that lightly or would deliberately hit the gas a litttle harder to bypass this goofiness. Unfortunately, I've never been in a position to test one of these cars. Anyone want to donate one, for purely scientific reasons?

I agree that most cars, and even almost all of the ones I've owned, get better fuel economy at slower speeds. Maybe my Lincoln Mark is the exception due to the engine (4.6 DOCH, 290 HP) and the 2:73 axle ratio. At an even 55 MPH on the open road that car is not a dog but it's barely running at about a 1000 RPM in Overdrive. At around 65-70 MPH it finally starts to breathe a little easier and it's a different car altogether.

Even then it's only turning about 1400 RPM. Since a lugging engine is a straining engine to some degree I've always wondered about the effects of things like MAP sensors on fuel economy.
I know the BMW I had used vacuum diaphragm carbs and that bike was also an utter pig at 55 MPH in 5th gear (the mandated limit back when I first bought it) and was just coming into the power band around 70.

Fuel mileage is an obsession with me and I pick it to death, especially on road trips. On Aug. 1st I'll be heading to CO for about an 1800 mile round trip, mostly on deserted 2 lane highways and Interstates. As usual, I'll be picking the mileage to death and will post back with the results.

It's pretty easily gotten around on a 3/4 ton or larger truck. They don't put mileage estimates on the sticker at all. There's way too much variation to begin with medium duty and above trucks. One may have a pickup bed and spend most of it's life empty, the next may have a tool bed and 9000 pounds of tools in it.

One thing I have noticed in trucks in particular is the larger the motor, a lot of times, the better it will do mileage wise in the real world, maybe not on a bench or test track or computer software.

I had 3 consecutive 300-6cyl F150's. The first was an 1986.5 that was an experimental model for fuel injection. It had a 4:10 LS axle in it and would get around 20 on the highway although it would top out at 80 (didn't have enough gear to run faster). In town, about 14 to 15. The next truck (88 model F150) had a 3.55 gear in it and got 14 in town and 17 on the highway. The 3rd had a 2.73 in it and got 13 in town/13 on the highway.

The first truck was geared for power and the best of the lot. I rented a tandem U-Haul once with it and the truck sat kind of high, high enough to bend the tongue and hold the surge brake down. Coming through Alabama I looked up in the rear view and I was absolutely smoking the axle on that trailer dragging it's brakes at 70 mph, but I had no problem pulling it at all. The 4:10 truck, you didn't have to keep it floored all the time to keep it going.

In contrast that 2.73 axled truck was the most no powered pos I ever owned. Worse than an old 78 Chevy 250 I-6 with burnt valves that I had once. You kept the heel to the steel at all times to keep it wound up and that's what burned the fuel like it was going out of style.

I had a buddy with a 3.51 F250 4wd. We went hunting a few times, and his larger truck would get 17 mpg on the highway v's 14 for my 6 cylinder 2wd truck. That 351 wasn't working near as hard.

I've experienced the same thing with Chevy's trucks. The older 350 powered trucks got better mileage on the highway than the newer smaller v8's.

Skipper

I don't know about your axle ratio question. I do know about one specific Ranger that fits your description. I have 7 years of mileage data (2001-2007) for a 1994 Ford Ranger Splash Supercab, 2x4, 3L V6, 5spd manual transmission, cruise control, and EPA rated 19/25.

Over 7 years of 30+ road trips over the same route to the same destination (98% highway, 544 miles RT), and using cruise control:

During the trips where I tried to go 90mph whenever possible, I averaged 23mpg.

On trips where I stuck to the posted speed limit at any given point (90% 65mph, 9% 75mph, 1% 45mph), I averaged 27mpg. 

These specific, meticulously documented, real-world mileage losses at higher speeds are consistent with the mileage losses generally predicted by Consumer Reports' Camry test.

Also of note: using cruise control and going the speed limit, I was able to consistently better the EPA highway prediction by 2mpg. This took patience, ameliorated by the fact that this truck was a smooth ride.

Not possible to increase MPG by increasing speed. Drag increases as the square of speed. Belief of better mpg at higher speed is the variable to be discounted.

Most of the people I have spoken to that get better gas mileage at 75 or even 90 than they do at 55 also seem to have owned one of those 160 mph bone stock Harley Davidson Sportsters back in their high school daze, killed a bunch of deer from 500 yards away with their dad's old 30-30 carbine, and catch at least one 20 pound bass every time they go fishing.

You got it...this is the longest discussion I have heard based upon a false assumption.

hehe! well, i'm not a harley guy; i don't hunt and i don't fish. i suppose that if i did, then i never would've come here to ask about it. ;)

What's even more amazing is that many automatically assume something based on CR, a theoretical formula that they apply to every vehicle made, a personal opinion of "that can't be right", etc.

And for those who think I'm BSing and want to apply the Sportster analogy I don't buy that either. A stock Sportster won't hit 160 but I did own one that would hit an honest buck and a quarter or 130. "Stock" it was not.
Never hit a deer at 500 yards either but my neighbor across the street has popped one at 400 with a .308.
Never caught a 20 pound anything in my life; a 6 pound bass is it.

One can believe what they want based on personal opinion and speculation but I'll say this. I used to make regular trips between my home and my daughter's home when she live in TX. It's 425 miles on way and she lived 1 stop sign off the Interstate. I live 20 miles off the Interstate with 1 stop sign between me and it.
I've made that run a dozen times at varying speeds and I know exactly what my car is capable of and exactly what it was doing. Of course, if you know more about my vehicle than me............

A few general comments:
1) Remember that the EPA tests are done on a dynamometer, so air resistance is totally ignored (I don't think this is done in a wind tunnel...).
2) Larger trucks may spend more of their power moving heavy weight, and so air resistance is a smaller factor in their mileage and may not noticeably "kick in" until higher speeds.
3) There are three kinds of loads:
a) air resistance/drag, which rises with the square of the speed, and at some point overwhelms everything else (and don't forget to consider windows down and A/C off, or vice-versa)
b) friction (internal and road), which is roughly linear with speed and thus can be ignored, as you double your load at twice the speed, but you drive half as long
c) speed-independent loads, such as air-conditioning or headlights, which are running for half the time if you drive twice as fast (i.e., a faster drive could help your mileage a bit)

I've had a 08 Highlander now for a couple months. This vehicle has a real-time fuel consumption graph and mpg display. To date the most economical speed range is 45-50 mph right when the transmission shifts up into 5th gear. The engine RPM is a low 1500 and if I was able to drive on flat road and not change speed the graph indicates that I could get over 30mpg. As I accelerate to 70mph the engine revs faster and there is more wind resistance. The fuel economy even with cruise on drops to the low 20s. I feel the only way that better economy is possible at higher speed is if the transmission has an overdrive that is most efficient at that speed and drops the RPM way down. I'm unaware of any stock transmissions with overdrives this tall.

No, it's not true. If you drove the exact same trip 20 MPH slower you would have improved your mileage by 20%.

The relationship is not always exacty linear, but it's true that if you slow down your mileage improves. I travel to the same city a number of times. My mileage ranges from 30 mpg at 75 mph with the A/C on to as much as 40 mpg when the going is slow (about 45-50 mph) due to weather condition and the A/C off. So, in my case I slow down 33% and get 33% better gas mileage.

I too have only anecdotal information but as I keep very good records of my fuel usage it may be useful. Here are the constants.

Every week day I drive from Huntington Beach, CA, to Corona, CA, in the morning and return in the evening. The commute is 37½ miles each way, 34 miles on three freeways. I drive a 1994 Nissan Sentra Sedan, 1.5L 4 cylinder, 3 Speed Automatic Transmission plus OD. I deliberately have chosen a commute time that is between the more popular times so I am before the "rush hour" to South Orange County. On weekends I drive only on local streets and rarely go more than 10 miles on both days combined.

I have not changed my driving habits for my morning commute but I have changed my driving for my evening commute. Two years ago it was my habit to set the cruise control at 65 mph and stay in the number 2 or number three lane, as appropriate. I only occasionally had to turn off the cruise control to slow for the impact of another drivers stupidity. I averaged 28 mpg.

After the first of this year I noticed that, perhaps due to the rise in gas prices, traffic was moving more quickly. I was able to keep the cruise control set at 70 or 75 mph and stay in the number two lane. Since I have done this I have averaged 33 mpg.


I suspect that the drive train of the car was designed for highway speeds. It performs poorly from a standing start, "jackrabbit" starts do not seem to be possible, perhaps due to the EFI that seems to say "Go ahead and floor it but I will only give the cylinders as much fuel as they can consume."

As I said above, this is another bit of anecdotal information but it is also slightly controlled data.

Peace

The problem appears to be that many do not understand lugging, its affect on an engine, and engine control items that are affected by lugging; MAP sensors, manifold vacuum, etc.

Consumer Reports is one of the last places I'd look to for expert automobile advice. CR is the same outfit who states (paraphrased here) "the owners manual is always the final arbiter as to maintenance on your car". CR is obviously wrong because there are a number of things that SHOULD be maintained more regularly than what any owners manual recites; trans fluid changes, fuel filter changes, valve adjustments on mechanical lifter engines, etc. In many cases air filters may need to be changed every 10k miles depending on the environment.
Constant short hop driving may lead to induction system cleaning more often.

Consumer Reports also investigated, AND backed, a lady who thought she was being ripped off by a dealer who recommended a timing belt change at 60k miles. According to CR anyway, there is no age limit on a timing belt and her 7 year old timing belt was just fine.
So owners of belt driven cars can sleep easy knowing that.

Wonder if CR is going to foot the bill for her if/when the belt gives out at 80k miles instead of the recommended 105k?

I hesitate to write this with so many naysayers writing in. However, I kept careful records from two trips from North Louisiana to Grand Canyon. AZ. On the first trip we traveled in a 2004 Toyota Highlander (4 cylinder)and on the second we traveled in a 2006 Toyota Highlander (6 cylinder). On the first trip we averaged about 24 mpg until we reached Colorado. At that point our milage varied from about 29 mpg to slightly over 30 mpg. We were traveling at higher speeds as we went further west (fewer towns, not as much traffic). We experienced the same thing on the next trip in a different car. We have always wondered if higher altitude was the reason. Most of our travel was on interstates and was never slow -- probably 65-75 mph until we reached Dallas and 75-85 after leaving Dallas. I notice that I failed to mention that our average mpg was lower over-all in the 6 cylinder vehicle.

Your basing your gas mileage estimates on TWO trips??? You need a lot more trips then that to get any valid results.