Subaru Automatic AWD isn't really, in most models

Andy Mason writes :

Thanks for this link, nice article. It points out something that occurred to me since my last attempt at reducing confusion.

In the case of Subaru's MPT automatic transmission, the front is 'hard wired' to the transmission, whereas the rear goes through the transfer clutch. Since the clutch is (almost ?) always slipping, there can be no transfer of power to the rear because the body of the car will pull the rear wheels into turning faster than the transfer clutch can, because it's slipping relative to the front. So under normal circumstances, the rear is actually applying drag to the system (so maybe the transfer ratio is 105% front / -5% rear :-) The rear can't be anything greater than 0%.

For the (slipping) clutch pack to act as a center differential, and transfer something to the rear it would have to be turning faster than the front, to make up for the slippage. This is mentioned in the above article. However, on Subaru MPT vehicles, the gearing is the same front and rear, and the transfer is 1:1 so this can't be happening.

So, although I've been trying to shine some light on the Subaru MPT automatics, it turns out they're even worse than I originally thought :-) I now can't see how the rear would ever be engaged at all unless a front wheel is slipping. So this business of the rear kicking in, in the middle of a turn, can't apply unless you're also spinning a front wheel (and thereby are at the limits of safety anyway). Not something that happens in everyday driving.

I just don't see the advantage of dragging around all this AWD stuff if it's only ever going to be used when you're stopped on ice, or when you floor it on a gravel road, or when you turn a corner so fast that you're sliding anyway. None of these hardly ever happen and none of them are related to safety (the last is actually kind of frightening to me). Might as well go with a FWD vehicle (and a set of chains) and save many thousands of dollars.

I have no problem with Subarus, only marketing bullshit. I have yet to meet a Subaru representative that knew how their systems work.

It's extremely likely that the Subaru engineers feel the same way and have so arranged things that the car doesn't operate the way you theorize. Sheesh.

Laughing my as* off. It's amazing how much people will try to think away something many engineers have spent many hours designing and putting together. Additionally, if that doesn't work for you think about this: all of that AWD 'stuff' cost money. Do you think managment is going to pay for 'stuff' just for the hell of it? To try and trick people into thinking it's a technological marvel? There are much cheaper ways to do such a thing. Simply because they can't understand it, it must be marketing bullsh*t! Subarus have the best vehicle control systems on the road. Anything else is just so much wishing. Just because a dealer or 'marketing person' can't explain it means nothing. When have you ever met a dealer or marketing person that could explain something technical?

John

Hey, if you know how the MPT automatic works please tell me. There's nothing I'd like more than fully understanding what's going on.

I'm a mechanic, have read most of the Impreza shop manuals, and scoured the Internet for weeks now trying to piece together some reliable understanding of the MPT system. I've presented what I've learned in as clear a fashion as I could in case the subject might be interesting to someone else. Is it really possible that :no one: is interested in the mechanical aspects of their car or the one they're about to buy ?

BTW, it took me about 5 minutes to understand how a Corolla AWD automatic works.

Like the other guy's MT theft derterant, i use the Stock radio theft derterant

Mikke

"John M."

You know, THAT is the first thing that i have heard that DOES make sence.

Mike

OTOH, have you ever met a GREAT engineer who could sell you on something extremely technical?? I've found that some of the greatest minds on the planet sometimes go around tripping over their own shoe laces ;)

As a GREAT engineering myself :-) my rule of thumb is that if I can't explain a good idea in five minutes or less, I don't understand it or it's not a good idea.

-- Dominic Richens | snipped-for-privacy@alumni.uottawa.ca "If you're not *outraged*, you're not paying attention!"

Absolutely true... that's why it takes all types to make the world go 'round and things are best when you are part of a great team.

John

I don't understand what you're trying to say. Do you mean that you can expain any good idea in 5 minutes *if* you understand it?

I think there is a vast quantity of ideas that can't be explained in 5 minutes whether you understand them or not. Fast fourier transforms, error correcting codes and single sideband come to mind off of the top of my head.

Jim I agree with you... there are so many things that can't be explained quickly. There are many things that most of us will never understand or be able to reverse engineer to our way of thinking. How many people understand Einstein and Theory of Relativity? If you don't quite understand the theory and you study it to try and understand it... and you are unsuccessful, do you then say the theory must have a fault? Just because someone can't understand how Subaru AWD systems work does not mean they do not work. It just means that that person is not qualified to make a judgement on what system is better; unless they have practical experience to guide them. John

Yeah, but you're talking wireless communications which is a black art. Any science that needs to take the square root of -1 to make it work is witchcraft! (It's Halloween).

vern....

this is from the NASIOC faq;

Q: What is the AWD difference between the AT and MT Impreza's?

A: Lets talk about MT cars first. If you think about the way a viscous limited slip works, it's obvious that "50/50" is a huge oversimplification. Put the car up on a lift and run the engine (not recommended). Now put a brake on the rear driveshaft (i.e., both rear wheels), and measure the torque there. Initially, there will be very little torque as the rear end slows down due to the braking force. 50% of the engine's torque is obviously not going there right now. Only after the rear shaft has slowed down, and the fluid heated up causing viscous drag, does torque begin to be delivered there. In other words, some amount of slip must occur before the torque "transfer" occurs.

Now let's take AT cars with the MPT (multi-plate transfer clutch) system. This is a very sophisticated system similar to the one used on the WRC (but built for MUCH lighter duty use of course). TCU, or Transmission Control Unit, does indeed "anticipate" slip by applying clutch lock-up depending on throttle, speed, and gear position. So when you accelerate from a stop at WOT, the TCU always sends torque to the rear, regardless of wheel slip condition. Note that the Haldex center diff used on VW's is mechanically identical to our MPT, however by not being connected to the TCU it is lacking this very powerful feature -- like the viscous coupling, the Haldex must sense slip before lock-up occurs.

So the "90/10" myth is exactly that. With the MPT clutch fully disengaged (low throttle or braking at highway speeds), maybe this is somewhat accurate, but at every other situation, it is not.

Note that you can force fully locked 4WD ("50/50") on the autotrans by putting the selector in "1".

fyi

Carl

Paul Pedersen wrote:

Hi, Sq root of -1 is an i. LOL. Tony

Under 5 minutes each:

(1) The special theory of relatively recognizes that the predominance of the velocity of light in the equations that describe electromagnetism is not a fluke. Instead, it is there because the equations are valid for any observer who ("in any frame that") is moving with a constant velocity. Therefore, this idea should be generalized to all of physics. Making the "other" physics of the time, classical dynamics, invariant against such frame transformations required Einstein to accept a then-known mathematical modification to space-time and apply it properly to the physics that describes the motion of bodies. This modification is not apparent in daily life; it only becomes important at very large velocities. An equivalent way of saying this is that the velocity of light is the largest possible velocity. Looking at the mass/energy of accelerated bodies, he quickly distilled E = mc^2 from that. In essence, this latter equation means that (because of the huge factor velocity-of-light-squared in it) even a small piece of mass contains huge energy, if you can set it free. From that derives the large energy of nuclear reactions.

(2) Einstein realized that the description of gravitation did not yet fit his new view. For one thing, Newton's gravitation acts instantaneously at a distance - a flaw even Newton recognized. In addition, the classical theory does not explain why we don?t feel a difference between acceleration (e.g., in an elevator or on a rocket) and gravitation. Einstein's general theory of relativity fixes this by generalizing space-time once more. This time, the generalization to accelerated frames means space-time must be intrinsically curved. Its local curvature is caused by the mass of bodies and at the same time forces other bodies (and even light) not to move in straight lines. This is a "geometrification" of the force of gravitation, a beautiful, powerful idea that has many consequences meanwhile observed in astronomical contexts (e.g., black holes, red shift of remote galaxies). Even Einstein?s cosmological constant, which he originally introduced to enable static universes (and then withdrew as his biggest blunder when confronted with observations of the expanding Universe), has earned a new place to describe the as-yet-little-understood repulsive energy that appears to accelerate the expansion of the Universe.

The above "geometrification" on the other hand has been the biggest problem in attempts to unify all forces of nature now known. Perhaps, string theory will resolve this issue. See current Nova programs on PBS for that.

- D.

You are right. The flaw in my analogy is that it involved one subject matter expert talking to another.

The typical context of the "5 minute rule" is two or more peer engineers working on the same problem, brain storming ideas. If after 5 minutes I still have a room full of blank stares, it's time to move on to the next idea.

-- Dominic Richens | snipped-for-privacy@alumni.uottawa.ca "If you're not *outraged*, you're not paying attention!"

Incorrect. The Haldex works in the same way as you describe for the MPT above, applying lock (transfering torque) depending on things like engine torque, gas pedal position and so on. No difference there.

/Staffan