The technical literature of the TL SH-AWD states that it can transfer "up to 70%" of the torque to the real wheel during cornering.

As we all know, the TL doesn't really have a center differential, just a set of electrohydraulic clutches at the rear axle.

So the clutches thelselves cannot take more than 50% of the total available torque....50% is the theoretical limit when they are completely close (maximum clutching force)

So how the SH-AWD can send more the 50% of torque at the rear wheels in some conditions?? The only thing I can think of is that the torque transfer unit bolted to the front gearbox is already geared in a way that spit the power 30% to the front and 70% at the rear (through different gearing ratio, acting like a "fixed" center differential) and then the clutches at the rear modulate the actual power transfer reaching the real 70% rear power transfer when they are at their peak clutching action.

Am I correct?


Thanks


Regards

 

Popular Mechanics says the following:

I take this to mean that 70% of the 50% (or 35% of the overall torque) can be sent to the outside tire.

 

I quote the TL brochure:

"Up to 45% of power can be directed to the rear wheels during straight line acceleration"

"Up to 70% of power can be directed to the rear wheels when cruising through a corner"

"Take a corner and the system can send up to 100% of available rear wheel torque to the outside wheel and then spin that wheel slightly faster"

So I take that the system can send up to 70% of torque to the rear wheels and of that amount, 100% can be directed to the outside wheel....correct??

 

I don't know if this will answer your question, but I found the video below that describes SH-AWD. At around the 6:30 mark, it goes into the technical detail.

http://world.honda.com/HDTV/news/2004-4040401a/

 

I did already watch that video, thanks for the link anyway.
It idid not explain how they can get more than 50% of the torque at the rear wheels (it states that the system can send up to 70% of the available torque at the rear)
I suspect that it does through the torque tranfer unit bolted to the gearbox.

 

Perhaps the fact that the rear diff has the planetary gearing and a 1.3% (??) overdrive, means that when the wheels are all travelling at same speed, the input shaft to the rear diff is 'taking' more torque as it's technically turning slower.. Sorry I can't quite explain it, but I'm sure it's due to the overdriven speed thing..

Anyone else got a comment on that one?

My question is.. with the 6MT (the one I have) while doing a down shift, which wheels are providing braking traction? The SH-AWD display doesn't show me any TQ transfer when down shifting (not to accelerate, but to slow the vehicle).

 

That's how I'd read it.

 

It's all marketing-speak. You can't "transfer" torque anyhow, but no torque is applied if there is no resistance.

If only the left-front wheel has grip, locked-diffs (e.g. a 4WD truck) will "direct" 100% of the torque to that wheel, but the others will also spin... just at the same speed. Note that the diffs are locked, so there's no actual directing going on; it's just that the freespinning wheels take no torque as long as they're locked to the non-spinning wheel.

The SH-AWD difference is that the LSD is somewhat aware of the difference between stuck (the slower-rotating wheel is the one with the grip, so lock the faster one) vs turning (the slower-rotating wheel is the one on the inside, so ignore it and drive the faster one.) It doesn't really change that maximum "redistribution" is synonamous with "locked", but it does put a new coat of marketing-paint on it.

 

^ And you know this.. how? You haven't convinced me that the computer can't redirect power/torque and do what Acura says is does.

 

That is different from what I'm asking, I know that torque can be applied only where there is resistance (traction) and torque in a differential tend to "flow" to the path of less resistance (the wheel with less grip)

If you have traction on all wheels (resistance all around) you can distribute different amount of torque through various ways.....computer controlled clutch packs (Acura and BMW) with variable "biting force", different gearing or different meshing offset distance for the gears at the same rotational speed (an example of the latter is the Audi Crown gear set which distribute the torque 40% at the front and 60% at the rear with the two driven shafts rotating at the same speed)

 

What Technocat is suggesting is the case with certain conventional locked or locking AWD systems and under an instance of traction lose. When dealing with SH and it's rear differential, which is a similar concept used in the ZF differential that is available for some Audi and BMW models, it's very different.

Uniquely, it does not technically use any form of LSD nor does it have a center differential. Also, it allows for an active transfer of power to the rear wheels under any condition and there actually can be more torque applied to one rear wheel than the other whether only one wheel or both wheels have traction.

There is a difference between varying torque distribution and normal free spin variances. That's why they developed active differentials in the first place, which purposely drive one wheel or a set of wheels more than others vs a wheel or set of wheels having the same or static amount of torque applied but free spinning at different rates.

To Saturno's question, I'll guess that there is either an additional settings in the transfer unit or that the electromagnetic clutches don't allow for full engagement until a certain degree of steering angle is detected or something to this affect.

 

I think you are reading the literature wrong. SHAWD on the TL can put as much as 90% to the front with 10% to the rear or as much as 70% to the rear with 30% to the front.

When cornering it can send 100% of rear wheel power (which could be between 10-70% of total output) to the outside rear wheel and accelerate that wheel by 1.7% (1.7% faster than the inside rear wheel).

 

Now you're on the trolley

 

It doesn't take long driving the car hard to feel the awd at work; unlike a traditional helical type diff, you're not going to feel the power transfer using the high revs of a downshift to keep it engaged on the loaded wheel. I find that where I would normally downshift on a corner approach and slow to enter a good turn, if you keep in the power, even at WOT if you have the nerve, the AWD logic makes the car stick and accelerate all the way to apex. The worst thing you can do is lift on a corner entrance, the car goes into instant oversteer condition where you can feel the back almost floating away without any power keeping it on line. I'm extreme conditions, the shift will also induce understeer up front and make the car almost shimmy diagonally. You can recover quickly by getting on the power, but you're going to follow whatever line the nose is leading.

You'll also notice if you lift on a fast corner entrance, and then get back on the power, how long it takes the wad logic to calculate and recover. I'd say this is up to a second a half delay on my 09. Hoping they've sped this up on the '12.

 

Lift-throttle oversteer for the SH-AWD TL ?

Won't the VSA cut in to prevent any oversteering/understeering condition from happening ?

I thought all Honda/Acura vehicles, except the RWD S2000 and NSX, are OEM tuned with a mild understeer characteristic straight out from the factory, since understeering is much easy to recover than oversteering for the general public drivers.

 

Physics degree. Something Winslo lacks. Lots of other ways too, but nobody really cares about credentials here.

Acura is using "Marketing", which is a lot like "artistic license", but a bit divorced from technical reality.

 

It definitely gets loose; I can't recall if I've ever felt the VSA kick in without being under power.

 

Are you saying that Acura is making totally false statements in how the system works and that the car is basically a limited slip 4x4? And you know that Acura is using "artistic license" because you have a physics degree? I am not disputing your education or knowledge just that you haven't proven your point here.

Wouldn't Acura would be liable for falsifying these facts if they are as you say they are. I would think their rivals would be all over it as well if it was untrue , unless all other companies are also usin gthese "Marketing" techniques.

Credentials does not an expert make. Hard, proven facts make the expert. Otherwise, it's pure opinion/speculation.

but what do I know? ... I am just an electrical designer with no credentials....

 

So in what regards Acura is using "Marketing" versus technical reality concerning this particular discussion??

 

That is well and good and in terms of physics on whether torque can actually be transferred or not and arguing semantics on the term transfer, those may be great points but your understanding of basic automotive technology is apparently "lacking" and especially how that ties into the fuctioning of SH-AWD.

I am flattered that you chose to address me in response to someone else's post again and even despite that everyone else who posted on this thread doesn't fully agree with your assessment and how it relates to SH AWD. Also, good to see you assuming what other people's credentials may or may not be, again, as if that somehow makes you more qualified to speak on such matters. Keep thinking that.

Somehow you seemed to miss the point professor. I don't think anyone wanted credentials, maybe just an elaboration. There also is a question to the thread which you failed to answer or lend any opinion or perspective on. So much for those credentials. Since it's all marketing, exactly what is going on with SH instead? Not just what is supposedly not happening.

 

 

I didn't say it's basically an LSD; there are many varieties of LSD. In reality, the term LSD isn't sufficiently useful not just due to the variety but because there are two diffs (center and axle) between the engine and the wheel.

What I'm saying is that their system does not "direct" torque. And neither really does Audis or Subarus. It's not a ding against them, but it's not what they do. They selectively lock-and-gear, but that isn't the same as torque.

There's no liability if there's no harm really. To successfully sue (for money) you'd have to prove harm. And some other companies do use similarly incorrect terms. Acura isn't alone in it by a long shot. But it is misleading.

"Torque" is rotational force. If there's no resistance, you don't have any. Supposing three wheels on ice, locking all four wheels together would ensure that 100% of the "torque" is exercised through the fourth wheel, but it wasn't "directed"... and that's just a fully-locked 4WD system.

SH-AWD is seriously cool. I'd have liked a software update to tighten up slick-surface starts, but I haven't been in a more effective dry-road AWD system. It's just their words that are off.
Certainly more so than you, but I am flattered to have my own personal forum-stalker.
SH-AWD does some cool gearing... That's the crux of it. The rear wheels can be driven faster than the front and the left and right at different rates. That's what the "vectoring" is, and is way beyond anything else under $50K. So in dry conditions (when all wheels have grip), the rear right wheel can be the primary driver when the gearing kicks in for a turn.

But if the turn is steeper than the gearing, or either wheel has no grip, all the force (torque) still goes exactly where it does in any other car. What they really have is amazing All-Wheel Gearing.

 

I'll throw a wrench into this conversation; Lets say we turbo a 4G 6MT SH-AWD. How do you think the AWD will fare?

Right now, a 3G guy is turbo'd and is putting down 550 to the front wheels.

 

I said that before that if there is not resistance, you do not have torque...mechanics 101.....but if you have resistance you can "distribute" different amout of torque through different gearing, different mesh lock offset, or variable clutching engagement....that is also mechanic 101...you can vary the amount of torque sent to each wheel (again, if there is traction)

Thw SH-AWD, or the ZF sport diff on Audi or BMW works in a completely different way compared to a classic LSD unit...and there is no center diff on the SH-AWD...actually there is not rear diff either in the classic interpretation of the term.

An LSD locks the two wheel together when one spins (full locking or allowing some different rotational speed, depending on the specific implementation and LSD architecture) .....the SH-AWD "differential" just release its clutching engagement from the wheel with less traction....that is how you get a "practical" LSD effect on an SH-AWD system even if the conceptual mechanism is completely different...if you put one wheel on glass ice and the another on a regular surface, in principle the SH-AWD should not spin at all the wheel on ice (the relative clutch is completely open) due to the system capability of redirect 100% of the rear axle torque to one wheel...the clutches act as "switches" in the torque distribution "diagram".

 

I'm no expert, but I the TL is basically a FWD car and the SH-AWD is basically added on. If it is not transferring power to the rear, it mostly drives the front wheels. I believe the cars computer only opens up power to the rear when it decides that it needs it to push you forward or turn.

So I would think it is the front wheels that you feel providing the engine braking. At that point I think it is in 90 front wheel drive mode if you are slowing down. But since you are in a low gear if you give it gas and enter a turn the torque will all fly to the back.

 

It is definitely derived from a FWD design but it does transfer power at the rear all the time (in different amounts based on the situation)....as an example if you read your owner's manual under the "towing in case of emergency" section you will read a big warning about not towing the car only with the front wheel up as you will damage the transmission....you need to tow it with a flat bed truck (all wheels off the ground)

Instead you can tow an Haldex AWD based car with the front axle up with no problem since it is an "on demand" AWD system.

 

all I know is that the car does handle when pushed hard thru the turns

one of the reasons I replaced my 3G

 

http://www.honda.com/newsandviews/ar...spx?id=5933-en

Look towards bottom of webpage...it talks about SH-AWD

 

OK, I'll throw in my 2cents of speculation as a mechanical engineer who's read a couple things (sorry can't provide references) about SH-AWD:

1. The prop shaft is always spinning on SH-AWD.
2. The "rear diff" has two separate clutch packs, electronically controlled, one for engaging each rear wheel half shaft.
3. The rear diff, prop shaft, rear half shafts, universal joints, etc. are designed to handle a certain amount of torque (with design margins to account for shock to the system such as acceleration over railroad tracks).
4. At the engine end of the prop shaft, there is a transfer mechanism designed in a way that no more than 45% of maximum, wide open throttle, engine torque will ever be allowed on the prop shaft. This means that when the engine is NOT at max power (like when CRUISING), a greater percentage (up to 70%) of the *available* torque can end up applied on the prop shaft (but note that 70% of *available* torque when cruising will NEVER exceed 45% of maximum engine torque when at wide open throttle during max acceleration). I believe this is the crux of confusion and debate in this discussion.

For those planning on turbo-ing or supercharging SH-AWD, the real question is how that transfer mechanism works. It may be designed to not allow more than 45% of WOT torque of the 3.7 un-modified engine (ie it has a clutch pack designed to slip beyond a certain torque limit). This would be good if the SH-AWD mechanism design margins are tight. If it's 45% of max power regardless of power available, I'd be a little worried about failing U or CV joints in the rear drive train mechanics, or worse, a failed half shaft that was torqued over its ultimate strength, well beyond the design margin. Of course, this same concern has always existed for fwd only models as well....but the fwd half shafts are a lot beefier than the rear ones.

The clutch mechanism on the rear diff has me intrigued. How does it vary %age of rear wheel torque? When both clutches engaged, it's 50/50. When it's anything between 50/50 and 0/100, what's it doing? If it's allowing slippage, it seems this is a wear point of the mechanism. Or I wonder if the clutches are simply on/off solenoid activated and the engagement varies via frequency of the on/off signal. I suspect this is what the condition based maintenance algorithm is tracking for when to change rear diff fluid. The more in-between (ie the more slippage time, or number of on/off cycles there are), the more often you have to replace the rear diff fluid to flush out the worn clutch particles.

Also, with the rear brakes being a tad bigger than the front brakes, I wonder if there might not be a certain amount of "cheating" going on with application of rear brakes to vary rear wheel torque distribution rather than vary clutch slippage. Why else make the rear brakes bigger than the fronts?

Since max torque *ever* only occurs at WOT at a particular peak rpm on this car, there is a lot of opportunity for fuzzy math that can be used for clever marketing descriptions of torque variance. In this car's case, it's "up to 70% when cruising through a corner" that's got some of us thrown for a loop. It would certainly be interesting to see the SH-AWD computer algorithm. It might be simpler than we might think. It may only have two input variables: throttle position and steering wheel angle. Everything else is handled by ABS, traction control, and stability assist stand-alone algorithms, with maybe some tweaks to ensure the systems aren't "fighting" each other.

 

I can confirm that the rear clutch packs allow slippage (this is how you "vary" the torque applied at the real half shafts, varying degree of the clutches "biting" force)....works the same on the BMW xDrive, the Haldex based AWD applications and the new Ferrari FF....I know that on paper (I'm an engineer too) seems weak and inefficient but it seems to work....

Definitely the individual rear brakes are used by the VSA system but the rear "diff" (it is not a real mechanical differential in the classic interpretation of the term) has LSD capabilities on his own even if it works in a total different way compared to a classic limited slip diff...

I do not completely understand this point....the propshaft at the front end is bolted to the gearbox in a fixed torque transfer (hypoid gearing) unit......you cannot "modulate" torque transfer in various condition from that end...that is fixed...whatever goes in is split between front and rear at the fixed ratio (through different rotational speed I suppose)...however you raise a very good point about the up to 70% torque transfer sent to the rear only in "cruising" conditions and only a maximum of 45% in WOT conditions ....it may be true and if it is I bet is still accomplished only by the rear clutch packs electronic controls....

For example, if you completely "open" the rear clutch packs, the car becomes a FWD only but the rear propshaft and "T" shaped hypoid rear gear would keep being driven from the engine up the the rear clutch pack's input plates.

 

Highly unlikely.

Because if the SH-AWD driver is spending an hour or so, doing some spirited driving along some twisting mountain passes or going all-out on the race track, in which the rear wheel torque-vectoring transfer is working double-time; this "brake application cheating", if true, will end up with fading and smoking rear brakes.

The rear rotors are not vented like the fronts.

 

Hold on...

The Acura VSA still use individual braking to help the handling of the car in conjunction with the SH-AWD...is what Audi call "Torque Vectoring" in their own marketing material...some sort of "passive" TV....
For example, the LSD effect on the front diff of the TL (both FWD and SH-AWD) is obtained through individual brake application.

 

Yes, the VSA jabs the brake(s) momentarily to prevent over/under-steering situations, and the entire braking duration is only fractions of a second.

However, let's say the SH-AWD car is powering through a long sweep turn at high speed, and the torque-vectoring distribution is 90%/10% for the outside/inside rear wheels; then the entire braking duration on the inside rear wheel will last for a long time in comparison with the VSA duration.

 

ah no i already made a thread about a oversteer incident i had with my TL at 35mph in second gen thread. my tl is definitely on the oversteer side when i go into curves. suspension is completely stock. also after looking at my alignment specification the car is actually tuned for neutral-oversteer than understeer. thanks to heavier weight on the front tires they dont lose grip too easy but its light ass just goes around.

also with my sh-awd rdx the oversteer is even worst. if i take the same curve at over 40mph and i keep my foot on the throttle to keep speed, the tail starts going loose. if i dont take off my foot off the gas the car would definitely lose control. the bmw performs similar to the rdx but feels scarier.

Funny thing i can take this curve at 50mph with a corolla with no problems except excess swaying and body lean.

 

Were your rear tires getting bold compared to the fronts ?

I can never get my '02 TL-S to oversteer in the original stock form, nor even after I put in adjustable shocks with same stiffness setting on all four shocks. All I have got is understeer.

Only after I have adjusted the rear shocks to be stiffer than the fronts, then I'm able to oversteer the car.


Three ways to increase understeer :
(1) increase air pressure to rear tires while keeping fronts stock.
(2) install stiffer front sway while keeping rear stock.
(3) increase stiffness on front shocks while keeping rears stock.

Three ways to increase oversteer :
(1) increase air pressure to front tires while keeping rears stock.
(2) install stiffer rear sway while keeping front stock.
(3) increase stiffness on rear shocks while keeping fronts stock.

 

Definitely correct.

The VSA is not part of the main torque distribution process on the SH-AWD except for LSD effect on the front diff.
As you correctly mentioned, it just "pinch" for extremely short periods on the brakes to improve handling dynamics.

With the VSA, I heard that the FWD TL has excellent dynamics for being a front wheel drive car but I never tested one myself.

 

After having gone through the Honda material on it's own SH-AWD, from the link (post #28) as provided by SHAWD10, I now have more insight on how torque is distributed among the two rear wheels.

Each of the two rear wheels has it's own set of electromagnetic clutch for torque transfer, and the amount of torque transfer is achieved through modulating the pressure applied to each set of clutch pack inside the two clutches.

So, when the same (synchronized) pressure is applied to both clutches, both rear wheels will receive the same amount of torque => L/R = 50%/50%. By varying this synchronized pressure on the rear L/R clutches, torque is distributed in the front-to-rear basis.

In the (L/R = 100%/0) or (L/R = 0/100%) case, maximum pressure is applied to one clutch and no pressure is applied to the other. As a result, 100% torque is distributed to the wheel whose clutch is fully engaged, whereas the other wheel is receiving no torque transfer and is free wheeling since it's clutch has been disengaged.

By varying the individual pressure, and therefore clutch slipping, to the L and R electromagnetic clutches, different L/R torque distribution ratios can be achieved.

There is no need for any brake system intervention. Simply fully disengage the electromagnetic clutch and torque distributing to the respective rear wheel will be instantly cut off.

 

That is exactly how I read it.....the brake intervention for LSD effect work at the front wheels which, as far as I know, is a classic open diff....

My question is, if the clutches are completely "closed" (maximum biting force) what is the standard torque distribution from the front torque transfer case?? (the one that feeds the propshaft)..that is standard and doesn't vary....

 

The Honda.com quote that got me is "The clutches can be controlled as a pair to alter front/rear torque split".

With the "70% when cruising through corner" scenario, it's got me wondering if the rear clutches are never, ever completely engaged; and, if they were, could you actually get a "100% to the rear" scenario?

I'm beginning to picture the output shaft from the transmission first going to a bevel gear that joins with a gear at the end of the prop shaft, and *then* to the open front differential. In that scenario, when driving in a straight line, it seems it would be 50/50 front to rear split, but with the rear gearing designed to rotate the rear wheels a tad faster, maybe there's a 5% difference due to "torque following path of least resistance" principle (ie there's more resistance at the rear with it's different gearing than the front). I think that might be how Acura arrives at the 55/45 front/rear distribution in a straight line.

Also, in that design scenario, there is no slippage between transmission and rear wheels other than the rear clutches. I suspect the whole "70% when cruising through a corner" is not about "designed in variability in the software", but more in how torque is calculated in this particular design scenario. When you're going through a corner, the inside wheels turn less than the outside; and with an open front diff, I can vaguely see how one might arrive at a calculation that shows 70% of the torque might wind up at the rear of the car. I'll leave that to a "practicing ME". I haven't done real torque calculations in years...

 

As a funny side note.....if you engage in this level of technical discussion with the great technical minds that hang around the "Car Talk" section of the site, they would have smoke coming out of their ears...or reply with the standard line "Acura sucks...they were so much better 20 years ago..."