Experiment: What do I lose with AWD?
 

It's almost a given you're sacrificing a few MPG when driving an All Wheel Drive Vehicle. (AWD) I mean, that's 2 more wheels to push and a lot more weight right?

I've always heard it, but never saw any actual tests. I kinda lucked out when I realized I could manually force my car to drive in Front Wheel Drive (FWD) mode by way of a fuse. It disables the rear transfer clutch completely so the car acts like a regular FWD car (albeit, still toting around a rear end, drive shaft, hubs, and axles necessary to be AWD)

So being a good little ecomodder I set out on my test course tonight to see what I could come up with. Test were done by setting cruise control at 50 mph (speedometer and scangauge confirmed) and driving 6 runs, then putting the FWD fuse in, then 6 more runs, then back to normal mode for another 4 runs. Readings were taken with scangauge and after the testing I filled up at 13.9 gallons and the scangauge was within 1% accurate. Car was well warmed up as I had already been driving an hour when I decided to do this.

The course is 1.1 mile straight and level as can be around here, hardly any traffic in the middle of the night.

Temperature was 51 degrees, humidity was at 52% with no reported wind.

AWD:

A: 33.2
B: 33.3
A: 33.2
B: 33.2
A: 33.1
B: 33.2

AVERAGE: 33.2 MPG

FWD:

A: 32.9
B: 34.6
A: 32.9
B: 34.9
A: 32.9
B: 33.4

AVERAGE: 33.6 MPG

AWD 2:

A: 31.1
B: 32.0
A: 31.7
B: 32.1

AVERAGE: 31.725 MPG


This is some really odd data to me. A lot of it is consistent is the strangest part.

The first set of AWD runs I was really happy with how consistent they were. I usually don't get those consistent of numbers on the course.

Then the FWD runs were very consistent (all were identical even) in one direction, but not so consistent in the other. They were all a few degrees higher.

THEN the final AWD run the average dropped again to below the original test average.


Right now however, with that data, I can say that the FWD fuse on an automatic transmission Subaru doesn't do anything to really help gas mileage. Its all within the noise it seems.

Strictly speaking to test this, I should remove all the rear drive train components, but I'm not doing that for a simple test. This really is just to test the fuse itself.

ITS also worth mentioning here, that its my understanding that the transfer clutches give something like 5% to the rear wheels when cruising. I imagine this helps them overcome the drag when lugging themselves. So maybe that's more efficient than just letting them drag behind?

With a viscous coupled AWD or similar system, most of the time the AWD isn't doing much. It's only when there's wheelspin at either end that the VC (or electronic equivalent) locks and you get torque transferred.
So I'd say in normal driving the difference between having the rear end switched in or out will be minimal.
Physically removing the AWD parts will show the most gains. I did this on my little Nissan and there was more usable power and mpg, but to be honest nothing startling.
Landrover Freelander owners here in the UK do this too and call it 'mondo mode'.

Back in the day I read a comment from an Audi engineer saying the AWD doesn't hurt fe because a powered wheel has less rolling resistance than an unpowered wheel and that pretty much cancelled out the extra weight and mechanical transmission losses. They made it sound like an fe wash.

That said, I wouldn't build an ultimate fe car with AWD.

For your fuse thing to work I'd think you'd have to disconnect the rear axles too so all that junk isn't spinning.

Given the fact that I
1) Almost never go offroad
2) Live in a part of the country where snow is only something so see on TV
3) Took it to an independent mechanic

...makes me wonder if that fuse is something I should look for under the hood of my Forester. I took it to the place I bought it - independent used car dealer, sells mostly Subies and has his own tame wrench wrangler - when it started clunking in sharp turns. He said it needed the center diff - his words, not mine - fluid changed, and I simply took him at his word.

Now, I'm pretty sure that's really a viscous coupling in there, which acts as a sort of center diff while also turning some torque to the stern. And it seems to me that when I floor it on wet roads, only the front end spins. But maybe it's electronic? So hmm, maybe he's pulled a fast one? I'll have to check for that fuse.

Hey, I just went to some of the Scooby forums and read up a bit - it looks like activating the solenoid that is fired by that FWD fuse isn't a great thing to do - it's not designed for 100% duty cycle for extended periods.

I guess that's Subaru's way of saying that if you wanted 2WD you should've bought a motorcycle.

While not a "scientific" as using the scan gauge, I can tell you switching from 2WD on my Jeep to the Fulltime 4WD mode nets me at least a 1 MPG loss. But mine is a different system than the Subies...

I suspect the extra drop is the heat soak into the parts, you finally got your tires up to full temp or the drive train.

The ford Escape comes in 2wd and is way cheaper than most of the Subarus.

Dave

This is about the same results I had on my dsm.

You still have to consider that if you could remove the transfer case drive-line and all the AWD rear parts and and replace them with a lighter FWD rear end parts you would increase your FE by a lot. All these parts have more mass and parasitic drag, so by reducing their mass and parasitic drag you would increase you FE.

This is the same as when testing a car in awd on a chassis dyno verse fwd mode on a chassis dyno.

Too bad you only did A-B-A testing and not A-B-A-B. But I know how time consuming it can be, so thanks anyway:)

I'm with Mark that there shouldn't be noticeable difference while cruising. Better would be city driving, with lots of turns and stop-go, but this would be much harder to test.

Without the fuse, have you noticed any difference in handling?

The "clutch" that connects the front and rear wheels in an AWD car is typically a viscus coupler. The reason is that as you turn sharply the front and rear wheels go a different distance. with a solid connection like in a transfer case would give you binding and jerking motion as one or two tires slipped a bit to release pressure in the drive lines. It is likely that although you are no longer getting power to the rear wheels with the fuse pulled the rear wheels are still spinning a torque converter.

On a side note I gain about 1-2mpg when I have the Differential in 2wd and the hubs unlocked on my Suzuki Samurai. Unfortunately I hardly ever get to do that as we have snow/ice in the winter and mud in the summer.

1 wheel drive?

FYI, EcoModder member Xringer converted his Honda CRV from all wheel drive to front wheel drive by removing drivetrain components (axles, drive shaft).

He's since sold it. (No longer in his garage.)

I bugged him a few times to see if he'd do a writeup about his project, but he declined. If you ask him, maybe he'll share his experience/results.

Check it out: ROKON: TrailBreaker

There's sport bikes, and there's utility bikes. You don't see many other motorcycles equipped to pull a disc harrow.

Or a plow.

Hey, it floats too. If I had to have a motor, I'd hold out for a Rokon. I'm not in any hurry to get anywhere. But I do want to be sure I'll get there. Rokon is a strong insurance policy toward the latter.

I've also heard its a very bad idea to run around with that fuse in. But it doesen't seem to be making a worthwhile diference in economy. If any at all.

As far as handling, I waited till after this test to test it. I punched it up a hill and it peeled out. Was kinda cute actually. Normally the car never has any wheel spin in anything but snowy conditions. Uphill in rain I still can't get any wheels lose.

That's what I was thinking...

I thought of another tidbit on the testing.

The viscous clutch that connects the front and rear drive always has about 5% lockup. Your never truly in 2wd in one. So removing the fuse keeps it from going to full lock but does not give 100% disconnect.

Another thing to check, connect a meter to the viscous clutch energy lead, and watch when it turns on. I think that you'll find that even in 2wd mode (If you have that) it turns on when accelerating and when traction control activates.

This could be a bad thing if you converted it to 2wd and then drove in real icy conditions... It would be a careful experiment to perform. Depending on how the antilock and traction control interface is programmed....

Dave

The 1995 Subaru Legacy wagon was available with 2WD or AWD, and the EPA rates them at 21/29 and 19/26, respectively. AWD adds extra driveline friction and over 100lbs of mass, some of it rotational.

Good thing you did A-B-A, or we would have drawn unjustified conclusions. For example, if you had tested first with dimples and then without, we might have figured dimples were good for +5% MPG.

I take it you did one run heading north, then one run heading south? I see clear evidence of wind during your FWD testing, and a hint of wind in the AWD testing.

That just means it's under powered :p;)

I didnt think about wind. When I was driving and turning around between the markers, I rolled the window down and never really felt anything. But I think it has to be moving at like 8 mph before you can even feel it with your skin?

The weather station also reported 0 mph winds.

As for power, it would spin the wheels some while in FWD but in AWD. I think I need to drop a 350 small block in it to make sure I can squeel all 4 tires as I barrel towards the next stop light.

Related to the above posts.
Ford Australia produce the model called the Territory in both AWD and RWD variants.
The website lists the fuel consumption at:

Territory TX RWD Territory TX AWD

11.6 (with 5 seats),
12.0 (with 7 seat option) 12.5 (with 5 or 7 seats)

Engines are the same but the transmission ratios are different due to the FWD transfer case.

Cheers , Pete.

How does that work? I could buy a negligible difference, since a decent axle is at what, ~97-99% efficiency, and an extra ~50lbs won't do much, but I've never heard of a powered wheel having less RR.

I don't know how that works. That's what the dude said, back in the '80s and he didn't expound on it IIRC.

http://www.pistonheads.com/doc.asp?c=52&i=10065

Ahhh- here we are:

http://books.google.com/books?id=Ldk...stance&f=false

The only thing I can come up with is that by spitting power between two axles the axle losses are minimized over some operating range, so it may look like having a driven wheel minimizes RR, but in fact it's that axle efficiency isn't linear and can be optimized to some extent.

Edit- Jeez , I've had five beers and I'm talking about axle losses. I should just get a degree in ME and call it a day. :D

awd and hypermiling the redneck way
 

One thing that hypermiling can gain is eco friendly good habits but when you are picking a car to set up hypermiling it may be best to look at all of the aspects first , and dont fall into the old trap that any body can hypermile with any car , for example if you have a big van you may sacrifice any fuel gain due to the shear size of the vehicle . awd is also a factor in picking a vehicle as well ,why ? well take for example a subaru the awd is mounted in the rear and is always turning while the car is moving , the separation point is in the transfer case so that means that everything up to that point is turning , redneck thinking is this way k.i.s.s. keep it simple stupid (not calling anyone stupid) but by looking at all the factors of saving gas or diesel all factors are concidered, think of it this way..... can you get out and push your car down the street a ways? i know for a fact that a subaru is a tough push for one person , that said now you know the engine will still work hard even cruseing . so what im saying is that hypermiling can work but if you really want it than you have to think outside the box and disregard some of the rules . more later

I run AWD.. with a 3 main boxer. 90hp.
3 modes to transmission..
4wd hi == lsd rear,
4wd lo == full diff lock, and reduction
2wd == open front, no rear.

no difference at all bewteen 4hi and 2wd in gas consumption.
could stay there all day...in maine, I have many days.
there is even a momentum, after speed achieved..and no difference in throttle getting to a speed...but once at speed, it seems as though, I am even less toes on the throttle..a momentum is happening in the 4wd. I did learn this similar drivetrain from an inline four, and it was not the same.

But inline fours are notorious for completely dying to the word called "work", there is no torque, until it gets over itself. the boxer is just simply there at all times...every single last rpm...all the way to bogging like a diesel, and still keeps going.

vw AWD synchro van, and old subarus...similar results.

Extra weight is extra weight, and extra spinning weight is extra spinning weight.


I'd be surprised if there was any consistent difference in mpg whatsoever between an AWD car in AWD mode and an AWD car with all of its drivetrain bits spinning but only 2 wheels actually receiving torque from the engine. The wheels don't care if they are connected to the engine by a shaft or by a planet whose surface links the front wheels to the rear, the net result is the same parts spinning and causing drag.

As a side note, early (at least through 2001) 5-speed Foresters (and probably other soobs) directly transmit engine torque to the front wheels and only utilize their viscous coupling to link the front wheels to the rear. What this means is on those vehicles you can simply unbolt the rear diff & drive shafts and remove them entirely from the vehicle with no consequence other than loss of AWD. The output yoke on the transmission will just spin and the front wheels will continue to receive 100% engine torque.

In Subarus, cars with automatics have a 90/10 torque split while cruising to offset the dragging of the rear wheels (electronically controlled transmission control unit), but the manual models have a 50/50 (Center differential with a viscous coupling device).

Subaru's All Wheel Drive System - AutoWorld.Com

I was actually thinking about looking for an Impreza for my next project, since there's so many more aftermarket parts out there than there are for Tempos.

Sounds interesting, but I'm not familiar with any Audis that can flesh out that theory in terms of FE, at least according to the EPA (and my limited VAG experience). Of the models that I'm familiar with, available both in FWD and quattro (or haldex), the FWD always is rated at a better mpg. Maybe that wasn't the case with the early 80's models? Go figure.

I owned an 01 Forester with manual transmission and personally had it apart. The engine (through the transmission) drives the ring gear of the front wheels directly and a jack shaft drives the rear output flange to the rear axle through the viscous coupling. When I disconnected the rear driveshaft and attempted to verify this by crawling up a curb, and by dumping the clutch on flat dry pavement, it behaved like a front wheel drive car with 100% engagement.

Hypothetically if you removed the front axle shafts from the same vehicle and tried to drive it as a rear wheel drive, it would only be able to deliver half of its torque to the drive axle and would likely wear itself out by freewheeling the front flanges any time the engine supplied more torque than the coupling wanted to send the rear wheels (such as acceleration/climbing/etc)

I remember reading years ago about an AWD Can-Am race car that somebody was building and their conclusion was that since frictional losses were a percentage of the power being transmitted, splitting the power between the front and back axles cut the losses to each axle in half resulting in the total losses being about the same. This would explain why cutting power to the rear axle in an AWD vehicle but retaining all the moving driveline parts would reduce MPG because you would be using just as much power to move the vehicle and extra power to turn the extra driveline parts. A vehicle with locking hubs such as the Suizuki would be able to eliminate the movement of these parts and show some improvement.