Quattro vs. Symmetrical

[smoss]

Actually, I mis typed that. I had to get brakes at 35k, it wasn't 40k. Sorry abuot that. That still seems low to me. I also live 1 mile from work. So the service guy said they get a lot of crap built up on them in winter and that could be the reason. Of course my honda prelude got 50k miles on the brakes under same conditons. I was also harder on that car then the Audi, since it was more fun and had more power.

Thanks for all the info guys.

[Rollin Wes]

How about Cadillac's front-wheel drive Fleetwood Eldorado introduction in a 1967 model?

[Rollin Wes]

Oh, this is interesting too about who were the first ones to introcdue front-drive.

In 1965 Oldsmobile introduced the front-wheel drive Toronado, the first FWD car to be built and sold in the U.S. since the 1930s.

Sorry not trying to hijack the all wheel drive thread, just pointing out facts in response to those saying Japanese and Euro cars having front drive first.

[pdw]

The Mini preceeded the Toronado/El Dorado by many years. It was FWD but it wasn't really done for cost savings.

 

Oh, and interior space is another advantage of FWD (or can be if implemented correctly).

[IwannaSportSedan]

The thing about "just a few lines of code" is that a mechanical system is always present, and isn't just software, which depends on electronic inputs and systems, which are probably easier to fault.

 

A mechanical failure that would hamper a mechanical differential system would have to be catastrophic enough to effect the integrity of the drivetrain itself, and usually is much more rare, and less easy to cause.

 

I still wonder about a known wear item (brakes, more specifically pads and rotors) that is responsible for the stability and response of the drivetrain. If engineers, and even we know that the parts wear, why would we entrust vehicle stability, and AWD torque distribution to that other separate system? (other than it being cheaper...)

 

And if the "less expensive is better" philosophy carries out, then what happens if the brakes are undersized, or have cheap OEM pads, or cheap rubber lines, or any other manner of cost cutting that could compromize the brake system while it is being used almost constantly to manage the drivetrain, as well as decelerate the vehicle on demand. Heat Soak, brake fade, normal wear of the pads, rotors, and tires...

 

it is bad enough when those things happen to regular cars in the course of just the deceleration duties. What happens to an AWD car, when the brakes also handle power distribution via open diffs?

 

Also, a brake-dependent system has to be dependent on detected wheel slip (through the ABS sensors, usually), rather than a mechanical LSD which has some amount of preload, and in the Torsen/Quaiffe helical diff case, can in effect allocate torque before the wheel slip starts occuring.

 

The Miata we have has a torsen, and I am very well satisfied with it, and it is utterly smooth, because it is never technically disengaged. Some may call that lossy. It may sap a small amount of energy from the driveline. However, the engine continually provides that energy as long as it is running. With more powerful and efficient engines, driveline losses can be tolerated. AWD by it's very nature is "lossy."

Brake pads, on the other hand are wear items, and when EDL uses them more, they wear faster. When they are worn, they are less effective in two roles, rather than just their primary one.

 

No lunch is free, but I think I prefer the mechanical differential choice, even if it costs more than "a few lines of code" and a few more brake pads.

 

What I want to know is, how does Mercedes' 4matic system handle an almost frictionless surface when all of the tires have a very easy time spinning... does it just clamp down on ALL of the brakes? Not much forward progress with that one...

 

A subaru might have difficulties allocating torque, too, but at least the brakes wouldn't be trying to slow some wheels, while others try accelerate the car.

 

Helping a car move by using the brakes... seems somewhat counter-productive to me. Certainly counter-intuitive.

[LegacyTwin]

wow, didn't ever think this thread was going to get so long. but great info.....

[Scotty]

The use of computers is not due to bad engineering, btw I'm a mechanical engineer. Computers allow a engineer to think outside the box. The stability control systems are based off of fly by wire systems that are in most of the new commerical airplanes and pretty much all military airplanes developed since the late 70's/ early 80's. Next time you step on a Boeing 747, 777 or any Airbus planes, the pilot has no mechanical links to the plane. Most fighter jets today can't fly without a computer because they use an unstable design that is extremely manueverable, but difficult to keep straight.

 

Don't worry, a car won't become steer by wire any time soon. There are laws in most nations that require a mechanical link between the driver and the front wheels. That doesn't mean engineers are developing these systems.

 

Computer systems aren't cheaper to develop than mechanical systems. They cost significantly more since you need to have a feedback loop system in the electro-mechanical hardware. It also requires a lot of testing to develop the algorithms for most conditions encountered. Just imagine developing a "few lines of code" for traction control for an AWD car. 2WD is easier, compare the difference between the driven wheels with the non-driven wheels and don't let it exceed 15%, ABS systems work similiarly. To do traction control on an AWD car, it gets pretty complicated since you can spin all 4 wheels at the same speed, but still go nowhere. You have to add in accelerometers and a ton of code to compare the data from the accelerometers and wheel speed and figure out what is needed to get the car moving at the optimum rate based on the throttle input.

 

As for brake wear on systems that use brakes to control power distribution, well first part of engineering is to analyze how the vehicle will be used. More than 90% of the driving done in a car is nowhere at the limits of the car. On a car with traction control and stability control, more than 90% of the drivers will probably never see the light come on indicating the system is working for the life of the car. So when you have a system that might come on a few times during the life span of the car, a little brake wear is not of a concern. The brakes will wear more from everyday driving than the system engaging itself once in a while.

 

If you are worried about an electronic diff, well what about all the other electronics in the car? There are only two mechanical systems, not counting manual transmissions, left in a car, the steering and the brakes. Not even the throttle is mechanical. Only reason those two are still mechanical is there are laws that require them to be so.

[Beanboy]

Electronic systems don't have to be passive, in fact, I think that's one of their biggest strengths, especially over "slip has to happen first" system that regular manual transmission Subarus get.

[Deer Killer]

1. Computerized negative-feedback systems allow for a more fundametally unstable design, which means it _will_ happen (and you will not find a better fan of computer control than me)

2. Computer systems become incredibly cheap to manufacture in volume. That's because the real work is done in software, which besides popular corporate enforcement, requires nearly zero effort to duplicate. A mechanically complex and inherently stable design vs a mechanically simple and unstable design compensated by computer control, the latter is cheaper.

[IwannaSportSedan]

Interesting stuff, Scotty.

 

I am an IT professional by trade, so hardly an anti-electronic nut. But I also know how complex systems can get, and how complexity breeds fallibility. More things to go wrong. More feedback loops to fail.

 

When control systems fail on a fundamentally unstable system, it tends not to be "Fail-safe."

 

A simpler, more robust, stable system tends not to fail nearly as often, but unfortunately when they do fail, tend to fail big, and spectacularly. It also takes more energy to cause such a failure.

 

And besides, if 90% of drivers never use their traction control, and these other safety net systems, then why are they so common? Sales gimmick, or safety net for the one in a million chance, and do they really make that much difference in the event that the one-in-a-million chance happens? (maybe so. I am just posing questions, not trying to be difficult.)

[JonFo]

I found the Subaru system to be a little better than what I had in my ‘02 Audio Quattro A6 2.7T Sport. It’s AWD was good, but the ESP annoyed me no end.

I have some interesting switchbacks near my house in the mountains and I regularly got the ESP lights blinking (and power cuts) going up or down them.

 

In my Subaru’s (WRX and LGT) I can just blast up or down at higher speeds without any Nanny intrusions. Just clean power distribution.

 

Tires are critical, when I went from the RE92 to Michelin Pilot Sport A/S the handling and braking improved dramatically.

 

I’d like the mechanical system in the subies fine, but I do agree that some more electronics could also give it that extra edge. I’ve not seen discussions of the Subie VDC here. (there is a related thread at: http://www.legacygt.com/forums/showthread.php?t=7264&highlight=VDC+transmission).

 

(If they added VDC, I’d like on Off switch just like the ESP on my Audi’s, which I’d switch off when doing performance driving in the dry).

[apexjapan2]

Ah, but VDC, at least in its tune for Japan is more performance oriented. We tested auto Legacies with it on and off on Fuji Speedway, and the times were within a second of each other, with the VDC "on" car being more consistent (although we did set the left front brake pad on fire!!).

 

The VDC did not try to slow the car down dramatically, or turn the power off. Rather, it tried to make the car point where you wanted and kept things stable. Pretty impressive I thought, esp. since I am a hater of elxtronic nannies. Merc's various xxx-matic brake and throttle systems, and Porkers PSM of a few years back being the worst IMHO.

 

Personally I'd like a helical front and something a bit tighter at the rear of my B4

 

And yes, tyres are crucial - we are lucky to have got at least RE010s since the BG5B era std on our turbo Legacys, and now the RE050. Very good rubber, and surely helped the car get COTY in Japan.

 

Anyway, back to your normal programming

[fzanetti]

Guys,

 

I was doing some research on Subie's VDC and found this interesting article:

 

Subaru's Active All-Wheel Drive system is unique in its price range, and when coupled with Subaru's Vehicle Dynamics Control, or "VDC", we think it is without equal. Subaru's VDC system fixes errors made by the driver and automatically corrects them. The system monitors vehicle stability by continually measuring inputs from the steering angle, lateral-g, yaw rate, and individual wheel speed sensors. Using this data, VDC can detect understeer, oversteer and vehicle drift conditions and tell whether the car is going where the driver is steering it. Similar systems are found on Volvo.

Essentially, VDC monitors the vehicle's stability, and using the all-wheel drive system, traction-control, transmission, and brakes; and then reacts to correct understeer, oversteer, drift conditions and wheel spin. Amazing stuff, and it works like a dream.

We tested VDC on an Outback wagon, which we put into an intentional condition of understeer on an icy, snow packed rural road. Understeer occurs when the vehicle does not turn in the desired path from your steering input. Understeer is always caused by carrying too much speed.

To correct this understeer, Subaru's system applies braking force to the inside rear wheel. This counter-action pivots the car around the braked wheel and helps put it back on the drivers intended course. At the same time, VDC optimizes all-wheel drive control by decreasing transfer clutch engagement to reduce power at the front wheels. With a little finesse, we were able to hurl the Outback at a corner knowing full well we would otherwise be in big trouble without the VDC system.

A very normal driver response to understeer is to mash the brake pedal to the floor, and yank the steering wheel harder to the desired turn. As a result of the drivers input, excessive speed, loss of grip and other factors, the vehicle's normal response to these inputs is to spin like a top. VDC works well to keep this from happening. We also discovered the VDC system interacted with the antilock brakes, so that more braking was still applied to the inside rear wheel to correct safely.

Not all traction control systems and all-wheel drive system are created equal. If you are considering a vehicle with these features, we recommend one that will meet the challenge of your driving situation. You may only need occasional assistance with traction control in less severe climates. Alternatively, a full-fledged all-wheel drive system like the Subaru VDC system and those found on the Volvo Vehi-Cross and Audi All-Road may be in order for more wintry or wet conditions.

Overall, these systems can improve or impede safety. As a general application with correct tires and driving technique, they may offer an increased margin of safety. Drivers should always beware of overconfidence and a belief that the systems will somehow magically save them from danger.

 

 

 

Found at:

 

http://www.theweekenddrive.com/reviews/2001/reviews/allwheel.html

 

 

 

Happy 2006 for everyone!!

 

Flavio Zanetti

Boston, MA

[phoenix96]

I've found that my Legacy GT is somewhat worse in the rain than the other Subarus I've had (96 Outback, 02 Outback VDC, 92 SVX). I attribute it to the tires.

[IwannaSportSedan]

Thanks Flavio, for the interesting read...

 

I would love to drive a top-of-the-line subaru AWD system...

 

AWD 6MT with 65% rear-bias Variable Torque Distribution, electro-mechanical hybrid Driver-controlled Center Differential, Front and rear helical differentials...(torque sensing/biasing) with Vehicle Dynamics Control and a more sane final drive ratio. backing up four or six turbocharged, horizontally opposed pistons, please.

 

Most but not all of that is on the 06 WRX STI. I don' think it has the VDC bit, though, although most of the sensors are probably in place already with the VTD/DCCD and 4 channel ABS systems.

 

Wishes, wishes...

[SLegacy99]

Ok heres my 2 cents. My dad's '02 A6 handles no better than my Legacy. 16 inch tires on both. He has Pirelli I have Continentals and I prefer my car over his anyday.

[amason]

Does anyone have a link explaining exactly how the AWD system in the LGT 5MT works?

 

This is a great outline of different AWD types, and includes a bit about the Subaru system (only a bit, b/c it's really pretty simple (which is not to say "bad" by any means)).

 

http://home.comcast.net/~eliot_www/awd.html

 

Andy

[iyalla]

Nice article.

Its a shame the simple mechanical arrangement of the Subaru system is being tarnished by drivers who have gotten too used to traction control correcting their bad driving. I think the Subaru system is more durable due to its simplicity in the long run. But if former traction control drivers keep wrecking their rear biased Legacys or complaining then Subaru might rush the VDC into the lower models. Hopefully with an off switch

[dark_rex]

I have come to the opposite conclusion and I had a 00 S4 as well.

 

i didn't have one, but i've driven countless audis, and i think subaru's awd is FAR superior. predictable, responsive and efficient.

[RobY]

Electronic systems don't have to be passive, in fact, I think that's one of their biggest strengths, especially over "slip has to happen first" system that regular manual transmission Subarus get.

 

 

Not quite... power is always there just 50 percent of it to each axle. It still has a differential which is instantly acting cause its on all the time. If you want more than 50 percent to a particular axle then the viscous coupler comes into play.

 

No slip has to occur to have a 50/50 torque split. When slip occours say on uneven, non uniform traction pavement, the viscous coupler transfers more shared torque to the most tractive wheels. A 5mt subaru would still have 4wd if you removed all the viscous couplers It just wouldent be as effective.

 

Unlike haldex (a coupling) there is no response time because there is a center diffrential. Power is being delivered all the time with ZERO response time. The only situation where there will be a delay is if the front and rear set of wheels are on grossly diffrent tractive surfaces that requires somthing other than a 50/50 split of power. 99 percent of the time for most people the viscous couplings are along for the ride. Even in snow and ice... IF all the tires on relatively equally tractive surfaces the viscous coupling has no role to play.

 

The advantage of this system is that it will work if the wheels become unladen unlike a torsens which become completely useless when one wheel gets "stuck". You can disable an audi completely by jacking a single wheel up if it werent for the braking system.

 

The Viscous Coupling Diffrental Lock system is so simple reliable and effective that it is still the perferred system in Group N rally cars.

[Beanboy]

OMG! Subaru 5MT are 50/50? No way!

 

Do you at least agree the Subaru system is passive when it comes to wheels slipping? Unlike active electronic systems or the instant response of a torsen system?

 

Torsens DO provide some torque transfer, usually in a 3:1 ratio in the case of Audi. And they respond instantly, much better from the I perspective for on-road vehicles (not Group N rally cars) that rarely see gross traction differences like the Legacy GT.

[RobY]

A torsens is also passive and when it comes to slipping. Once it slips it will continue to slip. It takes a bit more to make it slip because of a bias ratio but if that is exceeded it is no better than an open diff.

 

The only diffrence between an open diff and a torsens is that an open diff has a 1:1 bias ratio and in audis case the bias ratio of the torsens is 2.3:1 both respond "instantly" but the threshold is just more in a torsens but in audi's case it dosent have a viscous coupler to stop slip.

 

2.3:1 isnt much if you get one wheel stuck in a snow bank and it takes 10lb/ft to spin up the wheels the torsens can transfer a MAXIMUM of 23 lb/ft to the other wheels. Probably not enough to move the car.

[Beanboy]

Passive in the instant sense, unlike a viscous coupling. More linear as well (obviously). As far as snowbank driving, the LGT is an on-road sporty car. Same goes with the WRX.

 

And as far as Haldex goes, do you agree that having a locked clutchpack BEFORE slip occurs can be a good thing? Especially since it can be easily combined with stability/traction control?

[Scotty]

The Viscous Coupling Diffrental Lock system is so simple reliable and effective that it is still the perferred system in Group N rally cars.

 

But active differentials are the preferred system for the World Rally Championship cars. For 2006 they did change the rules to go with a passive front and rear, but still retain the active center differential in order to reduce cost. Original specifications were to change to full passive, but the drivers and teams were not willing to go completely passive. From the racing point of view, the difference between active and passive is with active, you go fast by keeping the car neat through the corners and have the computer do the work. With passive systems, you have to get the car going sideways more to get around the corners, but at a slower speed than the active systems.

[RobY]

Passive in the instant sense, unlike a viscous coupling. More linear as well (obviously). As far as snowbank driving, the LGT is an on-road sporty car. Same goes with the WRX.

 

And as far as Haldex goes, do you agree that having a locked clutchpack BEFORE slip occurs can be a good thing? Especially since it can be easily combined with stability/traction control?

 

Jack the car up on all 4's then remove the viscous couplings all 4 tires wills till spin. The power is already there and handled by diffrerentals. The car dosent need a viscous coupling for AWD. Viscous couplings have NOTHING to do with the 50/50 split of torque the diffrentials do. It is full time AWD and 50 percent of available power is there all the time with no delay. It IS instant in a 1:1 ratio.

 

Viscous couplings are just there to adjust the proportion of power if one set of wheels needs more than the other. Its still awd without them.

 

In this case each wheel has a 50:50 split ALL THE TIME. There is not a time in the cars life not even a fraction of a nanosecond that any set of tires will get less than a 50 percent share of the torque, but one set of wheels can recieve MORE via the limited slip action of the coupler.

 

The bottom line is that there is a static distribution of torque at all times.

 

 

With a haldex coupling power is NOT already there. The coupling must engage before ANY power is sent to the wheels this is because there are NO differentals. It needs to have a locked clutchpack because thats the only way the power is going to make it to the wheels without a differental.

 

In this case the only time the rear wheels get ANY power is when the computer decides to give it to you. Since there are no differentals in this system there cannot have a static distribution of torque.

 

The system is passive because there is NO static torque distribution and must react to be AWD.

 

WRC differentals are diffrent that haldex. They are more along the lines with an STi or EVO because torque transfer is acomplished through DIFFRENTALS but the e-clutch takes the place of a viscous couplers and is the best of both worlds.

 

 

I hope that makes sense cause its hard to explain in words.

 

The bottom line is unlike haldex no action has to occur for power to be delivered to each set of drive wheels because the power is already there in a 50:50 proportion.

 

What i think some of us are doing is confusing a VCDL system with a viscous coupling system. In a viscous coupling system torque is transfered THROUGH the viscous fluid.

 

In a VCDL system (subaru) torque is transfered in a set proportion through DIFFRENTALS and slip is LIMITED by viscous pucks.