[Tortfeasor STi]

O.K. The dccd is a system that has a variably locking center limited slip dif and mechanical front and rear. As you adjust the center differential you adjust the coefficient of friction in the center differential. All a limited slip differential is, whether mechanical or viscous, is a means of using friction to keep a certain percentage of the power going to an axle to be routed towards the wheel that has more grip. The center differential affects the lock up between front and rear axles. But it only applies friction when the drive shaft going to the front is moving at a different speed then the one going to the rear and vice versa. Adjusting the center differential only makes a difference when wheels move at different speeds, like in low or variable traction conditions or when turning. The only way this system can cause changes in gas mileage is when the friction of these differentials increases the rolling resistance of the drive train as a whole as experienced by the engine. More rolling resistance, less mileage. Now whether or not changing the variable center diff changed over all rolling resistance, I do not know. If the system is functioning properly with all 4 wheels have traction, then the center diff should not be a factor. However, as anyone who has studied physics knows, no system is perfect. So there very well might be some difference. Hence why I said I am not sure. Also, without verifying the testing conditions of the numbers quoted above, they are at best anecdotal evidence, just like the trip computer. After all, do any of you know how the ECU calculates mileage over time? What parameters it reads and what algorithms it uses? I certainly don't. So if anyone has access to the computer code, let us know. Otherwise, this conversation will continue to be pointless (i.e. it does not accomplish anything not already in another thread).

[warlord110aglia]

"All a differential is, whether mechanical or viscous, is a means of using friction to keep a certain percentage of the power going to an axle to be routed towards the wheel that has more grip."

You are incorrect. What you are describing is actually a limited-slip differential. A differential itself is simply an arrangement of gears that allowed separate control of each output shaft (allows the outer wheel to spin faster than the inside wheel when taking a turn). A differential will actually favor sending torque to the slipping wheel so that is why we have limited slips- using clutches, for example to correct this.

[Tortfeasor STi]

Originally Posted by warlord110aglia

"All a differential is, whether mechanical or viscous, is a means of using friction to keep a certain percentage of the power going to an axle to be routed towards the wheel that has more grip."

You are incorrect. What you are describing is actually a limited-slip differential. A differential itself is simply an arrangement of gears that allowed separate control of each output shaft (allows the outer wheel to spin faster than the inside wheel when taking a turn). A differential will actually favor sending torque to the slipping wheel so that is why we have limited slips- using clutches, for example to correct this.

You are correct I was describing the function of a limited slip differential. Like the three in the STi controlled by dccd that this thread is about. I will edit my above post for those hard of understanding.

[Crash1973]

Originally Posted by tshrey

I would like Crash to respond, but based on the numbers he gave I'm assuming his mpg was calculated from mileage and how much gas he filled up with, not the mpg display.

I have an '07, therefore no MPG display. My gas mileage was calculated using the distance traveled between fill ups divided by the amount of fuel put into the car. Even though I don't like to do this, I topped off my tank until I could see gas in the filler tube (yes, I have spilled it a few times too ). I also used the same gas station (the 76 at the corner of Rancho Vista and 10th St. West in Palmdale, CA) AND pump for every fill up................damn, I MUST have OCD.

Originally Posted by serendipity

If DCCD has a non-negligible impact on highway mileage, it's because the system is broken (whether by design or defect we can't say). DCCD in straight line cruise applies zero solenoid lockup. Look at the DCCD map.

Explain my mileage on the highway then. Remember, I was a non-believer until it happened to me (I lost fricken' $20 on this deal).

Finally, there is no such thing as "full rear" and I wish people would stop parroting this piece of misleading nomenclature. It is full OPEN, and in the 06+ cars, it's not even all the way open; it's as open as it goes but there is still a mechanical LSD in the diff which can effect a torque transfer if there is slip.

No need to be rude because you do not understand what I am talking about. J/K
"Full Rear" was referring to the fact that the adjustment wheel was rotated ALL THE WAY BACK...............hence, "FULL REAR"

Originally Posted by Tortfeasor STi

Adjusting the center differential only makes a difference when wheels move at different speeds, like in low or variable traction conditions or when turning. The only way this system can cause changes in gas mileage is when the friction of these differentials increases the rolling resistance of the drive train as a whole as experienced by the engine. More rolling resistance, less mileage.

This is EXACTLY how I thought BEFORE my experiment. Now, I am not so sure. I would LOVE for someone from Subaru to come on here and explain WTF about how the h311 I managed to get such a difference in mileage between the two settings.

Also, without verifying the testing conditions of the numbers quoted above, they are at best anecdotal evidence, just like the trip computer. After all, do any of you know how the ECU calculates mileage over time? What parameters it reads and what algorithms it uses? I certainly don't.

You don't know me so you don't really know how I am and how anal I am when it comes to running experiments. I am a big believer in the scientific method and am quite thorough in how I perform my experiments. I try to eliminate every variable that I can. Distances and speed are verified by using a Garmin GPS AND the Odometer (speedo is exactly 4.2 MPH too fast according to Garmin). Fuel is purchased at the same pump, same gas station etc. (and yes, I have waited with empty pumps available just to use the same pump). I have used the same routes to and from in order to get my numbers (that is one reason why I still have one more trip for my highway numbers. since I only use the one gas station for my fuel, any gas purchased DURING the trip, e.g. fuel purchased in Phoenix for the ride home, results in mileage that I did not count). I have even gone so far to throw out any numbers on days that did not fall in a certain temperature range. I have owned the car for 8 months and have just shy of 30,000 miles on it. I started tracking the mileage after my first fill up but did not start my experiment until I hit 8,000 miles. From 8,000 to 29,000 miles was when I tracked my numbers. I cannot explain why my car does this, nor can I say that ALL STI's do it. All I know it that my own numbers support the claim that on the highway, rolling it to the rear seems to help for some weird reason.

[HPH]

^^ Excellent! Thank you! This actually provides the sort of test that the gauges cannot.

Of course, the "why" is still something of a mystery, but I'm sure that there will continue to be all sorts of speculation. HPH

[Es Tee Aye]

Originally Posted by solovus

So then the trip computer has an input regarding the DCCD position? Can someone explain the "coincidence" since it cannot possibly be something that is actually occurring. What is the basis for calling it coincidental anyway? Is there anyone that can provide a technical basis for their position?

i cant provide any technical basis for my position but i do know that when i adjust my dccd all the way down (on the dash) i get better mileage on the highway. there is no proof behind this but only my experience... so for the people who says it is a coincidence then maybe your right...

[wyattroa1]

i would just think with less power going to the front wheels there would be less work for the engine to do..meaning less drivetrain loss... but then again im new to the sti so i really dont know ****..lol
robert

[HASTIhatch]

All I know is that I don't get how the F@#%@$^! you guys get into the 20s. I'm struggling to get past 16. On the freeway I've seen 18s and that's talking maximum. What the hell is wrong with me, aside from the obvious answer of laying of the lead foot...

[solovus]

Originally Posted by HASTIhatch

All I know is that I don't get how the F@#%@$^! you guys get into the 20s. I'm struggling to get past 16. On the freeway I've seen 18s and that's talking maximum. What the hell is wrong with me, aside from the obvious answer of laying of the lead foot...

I avg about 80mph on the highway and i'm getting about 20mpg according to gauge and verified at fill up. My car is completely stock. I think i've heard that the cobb maps actually improve mpg...can anyone verify this?

[Rootus]

Originally Posted by wyattroa1

i would just think with less power going to the front wheels

The DCCD doesn't change the amount of power going to the front wheels. The center diff is mechanically designed to split torque 41/59. In certain situations (low traction or you are turning the wheels) the LSD's in the center diff can reflect torque away from the direction the gearing sends it to dynamically alter what the front or rear sees. But when you drive in a straight line on anything other than snow & ice, the LSD's (all of 'em, including the two in the center diff) are 100% inert.

Which goes back to the point about mileage. If you are seeing a noticeable change in mileage, you ought to get your car looked at because it's malfunctioning in a big way -- actually two big ways. First you have something causing a pretty substantial difference in speed between the front and rear shafts, which is bad, and on top of that the solenoid clutch should be open when you're driving straight on a high traction surface. Both situations have to occur simultaneously. How long do you think those clutches in the center diff are going to last if you're constantly wearing on them 100% of the time? There's a good reason Subaru tells you to replace all four tires at once.

[Es Tee Aye]

Originally Posted by HASTIhatch

All I know is that I don't get how the F@#%@$^! you guys get into the 20s. I'm struggling to get past 16. On the freeway I've seen 18s and that's talking maximum. What the hell is wrong with me, aside from the obvious answer of laying of the lead foot...

i felt the same way the first couple weeks of owning my car, i found out that the way you drive impacts greatly how much mileage and if you do just city driving then say hello to 15 mpg... i can get 300 miles out of 13 gallons of gas from highway driving and 250 from city driving..

i get these numbers not from the auto guage thingy on the newer STi's but from setting the trip (A)or(B) to "00.0" at fill up and letting it keep track until the next time i am at the gas station

[fry_ed]

In a straight line (i.e. on the highway), DCCD AUTO will default to DCCD LOCK. There will be more friction (it's very easy to imagine this for an AWD car that uses a viscous type center diff to keep the fronts and rears rotating at the same speed), and therefore you will have to push the GO pedal a bit harder on the highway to keep up to speed, compared to having the DCCD in OPEN. You know, because friction = heat loss = more power loss.

In the city, there are lots of turns, and DCCD AUTO will default to DCCD OPEN. So it's not that much different than just setting it manually to OPEN. It seems very obvious to me that you should see a non-negligible mpg difference on the highway and a negligible mpg difference in the city, as CRASH1973 showed...

[toogood]

wow...at this thread...we should just submit this request to the "mythbusters" team

jk

while you all are worrying about that ill be enjoying the ride! lol

[Crash1973]

Originally Posted by fry_ed

In a straight line (i.e. on the highway), DCCD AUTO will default to DCCD LOCK. There will be more friction (it's very easy to imagine this for an AWD car that uses a viscous type center diff to keep the fronts and rears rotating at the same speed), and therefore you will have to push the GO pedal a bit harder on the highway to keep up to speed, compared to having the DCCD in OPEN. You know, because friction = heat loss = more power loss.

In the city, there are lots of turns, and DCCD AUTO will default to DCCD OPEN. So it's not that much different than just setting it manually to OPEN. It seems very obvious to me that you should see a non-negligible mpg difference on the highway and a negligible mpg difference in the city, as CRASH1973 showed...

I never thought of that. I guess that explains the mystery. I just couldn't get my head around the fact that in the city there was almost no difference but on the highway there was a HUGE difference. Thanks for explaining it.

[midnit]

Originally Posted by HASTIhatch

All I know is that I don't get how the F@#%@$^! you guys get into the 20s. I'm struggling to get past 16. On the freeway I've seen 18s and that's talking maximum. What the hell is wrong with me, aside from the obvious answer of laying of the lead foot...

That's what I would consistently get in my05 STI... Now that I've got the 08, I don't push the car at all like I did with the 05 every single time I drove it. I drive like a grandma right now in the 08, and I know that it's pissing off all my friends who like to go fast when we cruise together.

Once the break-in period is done though, I'm sure my mileage will go down below 20-21.