Wikipedia says this about the RL-
-------'A new RL debuted in 2005, this time with a 300-hp V6, improved styling, and Super Handling All-Wheel Drive (SH-AWD), a system capable of sending almost all of the RL's power to just one wheel in a turn.'------------

-How much power can go to one wheel??

from wardsautoworld-
------Acura's SH-AWD employs a collection of electromagnetic clutches and planetary gears in an elegant and complex rear differential unit that effectively handles both the front/rear torque split (70% front/30% rear is default) and the side-to-side apportioning of torque (either rear wheel can accept from 0 to 100%) that plays a critical role in achieving the torque-vectoring effect.

more from article in 2005-
The RL's torque-vectoring system has auto maker and supplier engineers buzzing. Some debate the technology's ultimate impact from a volume perspective, but most agree torque-vectoring is a formidable new advance in AWD development.

More telling, perhaps: Almost every auto maker and AWD-connected supplier is actively evaluating torque-vectoring AWD technology.

“There's a lot of interest around at the moment,” says Ben Reynolds, chief engineer, AWD driveline systems, Ricardo Driveline and Transmission Systems. Ricardo, he says, currently is developing torque-vectoring AWD for a vehicle that will be in the market “sooner than later,” and “several major Tier 1s are active in the field.”

For Audi, the big daddy of AWD in the premium market, torque vectoring plays a definite role in future quattro development programs.

 

And ZF is developing a torque vectoring driveline for BMW. You can google a press release on that.

Is there a question in this post? I don't really think so, but...

If the question is how much power can go to one wheel, most of the literature supports the 70/30 to 30/70 front to rear range of torque distribution. Of the 70% that can go to the rear, up to 100% of the 70% can be vectored to a rear wheel.

Virtually all of that information comes from Honda's own press releases currently, and in the archives at hondanews.com. There are some interviews of the designer, but almost anything of technical description can all be traced back to Honda press releases, in my research from before buying the car and since then.

It is especially interesting and entertaining to compare all the Honda documents and noting mistakes, such as specifically citing the PAX wheel size as 18 x 8, when in reality, it is a metric wheel size. In fact, the Honda Odyssey technical press releases at least get the Ody PAX wheel size correct, citing the size in mm vs inches. So much for Acura accuracy in its press releases. They should refer to the sticker and tables that Michelin produces to get the correct wheel size in mm.

 

Chas, the OP's question is exactly what you answered (and correctly at that).

 

My SH-AWD question is this:

Is the system sending some torque to one rear wheel even when the instrument panel display doesn't show it?

I'm not dissing SH-AWD, but anyone who has tried to invoke unequal rear wheel torque by taking curves fast knows you have to go seriously (and pretty much dangerously) fast to make the display show more torque to one rear wheel.

So I still feel that in normal-to-spirited cornering, you're not torque-splitting, unless the display just isn't showing it. Am I wrong?

.
.

 

That is a good question which I do not know the answer to.

Based on the technical press releasesI would say, yes, it is always torque splitting. Nothing I can find will say that directly, but from the technical press release quote below, you can infer that with constant inputs from various sensors, the computers are working to varying torque all the time.

"The logic and control of SH-AWD is integrated with the RL Engine Electronic Control Unit (ECU), and Vehicle Stability Assist (VSA) ECU. The Engine ECU provides engine rpm, intake manifold pressure, and transmission gear ratio data. The VSA ECU provides data on lateral g, yaw rate, wheel rotation speed and steering angle. The SH-AWD ECU monitors the status of the acceleration device and the right and left Direct Electromagnetic Clutch torque. Traction is calculated based on the information from the engine ECU. During an acceleration situation, lateral g and steering angle are used to set the torque split between the right and left rear wheels. At the same time, this data is used to control the acceleration device."

Just from observation, we also know that the SH-AWD works in deceleration situations also, adding torque to the rear wheels only until the brake is applied.

The question which remains unanswered is, what is the graph really showing you? Is it a linear graph, where each bar is 20%, or just some relative indication to something else? I do not find any specifics on that.

The manual offers the best info on the topic when discussing the MID function for SH-AWD. While there are many more pages in the 2008 manual vs the 2006 manual, this excerpt from the 2006 manual has not changed in the 2008 manual.

"Each torque indicator is displayed
as a bar graph divided into 5
segments. The number of
segments represents the amount
of torque distributed to each
wheel."

"When there is only a slight change in
torque distribution while driving,
such as cruising on level roads at the
same speed, the torque distribution
monitor may stop displaying the
amount of torque. This is not a
system problem. The monitor will
show the amount if the system
senses any change in torque
distribution."

This is all my educated guessing on the subject.

 

I think you would have to really press it on dry pavement to see it - and I may just do that if I can talk my wife letting me take it to an autocross - that is where you go through a course of traffic cones against the clock. I do it in my Porsche about once a month - and some members have brought out Subaru's - including wagons - that can kick all but the specially prepared cars butts.

Other then that - it is going to be in the snow or at least the rain - thing is - I wouldn't want to break it loose without a lot of room to recover - that is why I do autocross instead of the track - autocross you can push the car to 11/10'ths and all that happend is you crush some cones. On the track, you can hurt the car - or worse yet, yourself. Damage to the car is almost unheard of in autocross.

I washed off the car yesterday and took it out for a blow-dry. A Dodge Viper pulled out of a side road behind me then blasted by me (kicking up sand from a recent repaving project - thanks - A-- H--- !).

He then turned off the highway onto a side road that leads back to my house - a road I use to test changes in the suspension of the Porsche so I know it pretty well. I decided to follow him. He was moving pretty good - I don't know how far he was pushing it - but I was able to maintain the same distance behind him and maybe even gain a bit on the tight curves. I never felt it was near the limit on adheasion - but I did begin to experience brake fade - which I wasn't real pleased about. Hopefully, it was due to the pads not being fully seated and broken in (only about 800 miles on the car at the time). I didn't have enough familiaraity to use the manual shifting - that might of helped - so I was on the brakes pretty hard setting up for the turn-in.

 

Acura failed to mention that SHAWD only works while accelerating, and not while braking.

 

I think you're right about the braking part, but we've established it DOES work while coasting.

.
.

 

I wasn't expecting SHAWD to work on braking - but normally driving a stick shift - I expect some engine braking to happen - and it wasn't - so I found that it felt like I had to be on the bakes more then I would have if I had been using the paddle shifters - staying in the same gear that I just had accelerated in when I took my foot off the throttle inorder to start braking, rather then upshifting. Part of it is I am just not used to driving an automatic in a "spirited" fashion.

 

As we have previously posted before, the system is always active, in day to day normal driving situations not much if any vectoring (or over accelerating) is done or needed. The SHAWD display IS acurate. However, if driven more aggresively you will see side to side vectoring a lot. That is why most people fail to understand or use the system appropriately and then think the car is simply numb. To get the most of the system you have to give it gas while IN the curve which is exactly the opposite of what you do in any other car. However this aplication has to be judicious to avoid oversteer. The way to see what the system can do (it can do a lot) is in conjunction with the manual shifting:
1-Enter a corner in the lowest gear possible (so you will have the highest possible engine torque)
2- accelerate through the entire turn (judiciously as stated above) if done correctly gradual tire slippage from the rear will be your guide of when you are overdoing it.
3- Caution! big grins will invariably follow the maneuver!
The RL can easily pull over .9 G's if you do it correctly despite having only modest "touring" tires.

 

Yep, I think you're making my point. I've seen people saying they "feel" the SH-AWD all the time, but what I think they're feeling in day-to-day driving is just good steering geometry.

In short, the RL steers nimbly and capably because of its wheelbase, steering ratio, tires and wheels, and suspension tuning. SH-AWD will kick in on dry pavement when you're REALLY pushing it (aggressive cornering while on-throttle or coasting), somewhat like other cars' stability control systems. But it's not an "active steering" system that's always sending torque to this side and that side every time you turn a corner.

I personally view it mainly as a safety feature, but as an extra benefit it's a great all-weather traction system.

.
.

 

Huh??? It isn't the opposite of what I have always done in ANY car. You should be off the brakes and on the throttle ENTERING the turn - not when you are already in the turn and NEVER, EVER, lift off the throttle IN a turn or you will likely end up looking at where you just came from.

 

So.... wikipedia is wrong? or just semantics. I guess it's more single wheel power than most other cars.
'almost all power can go to one wheel in a turn'.

I guess I read that as implying almost 100% when in fact it is 70%.
that was my question.

 

That is my understanding 100% of the 70% (of the total) power that is going to the rear wheels.

It would be fun to rent a track for the day with a skid pad that can be wet down - last time I was at Willow Springs they were doing that at another section for the LA County Sheriff Dept.

 

According to what we've been told by Acura, the maximum torque sent to the rear is 70%. Now, 100% of that 70% can be sent to one rear wheel, but that still leaves 30% on th front wheels.

That, however, tells me the display is over-simplifying things. For example, when you see no "bars" on the front wheels and 2 bars on the rear wheels, that doesn't mean zero torque to the front (as the display would imply). So even though the display shows nothing, it can have 30% or more of the torque going to those wheels.

Maybe we should deduce that when it shows a bar on the display, it means MORE than 30%.

.
.

 

You CAN let-off the throttle mid-turn in the RL without a problem because the SH-AWD system will send the engine braking to the inside rear wheel which stabilizes the car. Very cool.

Back on topic, I've read on this forum at one point that the MDX or RDX (can't remember which) has a more "sensitive" SH-AWD display. The discussion was debating whether the system actually vectors torque more aggressively or if the display was calibrated to show you more than is really happening......

 

Yes, there was a thread implying the MDX was showing the SH-AWD in use ALL the time, even on regular turns (vs white knuckle turns), of course, you wonder if the higher center of gravity SUV has anything to do with it. Even as a front wheel drive car, I would think the RL is pretty stable, handling like an underpowered TL.

But since the RL was the showcase car for the technology, I wonder if Honda got a lot comments such as "this not working", and so they just tweaked the sensitivity a bit so you felt like you were getting more of what you paid for.

 

Chas!!! As the former owner of an '06 TL and current '06 RL owner, I feel compelled to say the RL handles nothing like a TL. To me, the TL seemed ponderous and truck-like compared to the RL. The TL was very obviously nose-heavy, and its heavier steering made that even more noticeable. The RL - at least IMO - has a far better balance of steering effort and turn-in, and that makes it feel much more nimble and controllable to me. The TL handles pretty well, but it requires a lot more effort and concentration to make it do so.

As for the MDX, I haven't noticed the display being more active. I'll watch that in the next few days to see if it seems like it's showing me more than the RL.

.
.

 

[QUOTE=Chas2]If the question is how much power can go to one wheel, most of the literature supports the 70/30 to 30/70 front to rear range of torque distribution. Of the 70% that can go to the rear, up to 100% of the 70% can be vectored to a rear wheel.

Virtually all of that information comes from Honda's own press releases currently, and in the archives at hondanews.com. There are some interviews of the designer, but almost anything of technical description can all be traced back to Honda press releases, in my research from before buying the car and since then.
[QUOTE]
This maximum torque distribution has changed as time has passed. 70:30 to 30:70 was correct in 06. The 2008 model is 80:20 to 30:70.

 

OMG! You're right about the MDX ... the SH-AWD display is much more active than the RL's.

I just made a Walmart run ( ) and I tried accelerating through a left turn from a dead stop. The display showed a couple of bars of torque vectoring. I don't think I could get the RL to vector at all under the same circumstances.

Either the dynamics of the MDX called for more vectoring, or Acura just decided to "liven up" the display like you suggested.

.
.

 

[QUOTE=black label][QUOTE=Chas2]If the question is how much power can go to one wheel, most of the literature supports the 70/30 to 30/70 front to rear range of torque distribution. Of the 70% that can go to the rear, up to 100% of the 70% can be vectored to a rear wheel.

Virtually all of that information comes from Honda's own press releases currently, and in the archives at hondanews.com. There are some interviews of the designer, but almost anything of technical description can all be traced back to Honda press releases, in my research from before buying the car and since then.
Good to know. What is the source? I would like to examine for additional changes. I would like to see the RL evolve and not just stay the same year over year.

 

On the RL you don't lift off you should give it gas through the entire turn, very similar to a RWD but with more enthusiasm, since the RL is much less prone to overesteer than a RWD car. Simple example, the same aplication of power to my s2000 will make it slide its tail freely.
The s2000 is a known fantastic handler but when you get down to it the RL has almost the same handling limits, however due to its suspension settings it doesn't "feel" at all the same, the s2000 gives you that buttoned down go kart sense, on the RL it still feels floaty in comparison but its like an invisible hand is keping you on track while you take the turn. I have a neighborhood road that has some very nice switchbacks, on both cars I can take them at almost the same speed (80- 85 mph) however your sense of control and roadfeel is overwhemingly greater on the S2000 (as is noise and all those sweet mechanical sounds that the RL filters).

 

Right - but you had said

and my point is that it isn't "EXACTLY THE OPPOSITE OF WHAT YOU DO IN ANY OTHER CAR" but EXACTLY WHAT YOU HAVE TO DO ANY CAR.

 

I think what you are trying to say is that you can give the RL more power in a turn (because of the SH-AWD) then you would with most other cars.

 

In most cars, the best way to handle curves is"fast-in/slow-out". In the RL it is "slow-in/fast-out".....

 

I can't think of a single car (including the 63 Impala Station Wagon my family had when I was learning how to drive ) I have ever driven where I would want to exit a turn slower then I went in.

Actually - I prefer fast in and faster out - steering with the throttle. Not drifting - that only waste speed. There is an optimum "slip angle" where everthing is in balance and the car travels through the turn at maximum speed. There are a lot of factors at play and having SH-AWD is going to be radically different from FWD or RWD and different again from RWD with the engine in the back (which is the configuration of what I do most my "agressive" driving in).

 

Unless the turn was decreasing in radius that is. That can be a very tricky situation.

 

you know its easy to say...but be carefull with your rl...i have experimented with my new car a lot and have been amazed by how well it grips the pavement around a turn...that being said, it really scares me when it breaks loose because the drift recovery takes a little longer (mainly because i am going considerably faster)

 

That is why I am going to take it out to a few autocross events to figure it out - where there is nothing but traffic cones and a big parking lot around. They don't call it "cone crushing" for nothing.

You want something scary when it breaks loose - try driving an early (pre '73 - or even better - pre 67) Porsche 911.

 

I agree ... race car drivers run up to a turn hard on the throttle, brake hard at entry, then power through and out of the turn. Letting off the throttle in the middle of a fast curve will generally get you in big trouble.

.
.

 

The mantra is "On the gas until you see God, then brake!" You are either on the gas or the brake - coasting is NOT going fast!

But the same is true for any car. You really don't want to let off the throttle in a turn. When you are in a turn, the angle of the tires and the angle of the turn are NOT the same - that is the "slip angle". The tires are turned more into the turn and slipping slightly sideways - pushing out of the turn. All stock street cars are set-up to understeer to some degree (push out of a turn). If you let off the gas in a turn, the weight shifts to the front, the rear tires loose grip, the front ones gain it, and you swap ends - what is called "snap oversteer" becuse the slip angle has decreased on the front wheels due to more traction so they are now effectivly turned more into the turn. Putting on the brakes only makes it worse as more weight shifts to the front, the tires stop turning (at least without ABS) and you now have even less grip on the rear. The car wants to keep rotating and continue sliding in whatever direction out of the turn it was going when you put on the brakes. You really want to turn out of the spin and get on the gas - but that is counter intuitive to most people.

That is why SH-AWD is supposed to be so great (at least as I understand it and one of the reasons I brought an RL) - the front wheels are pulling the car through the turn in the direction you have the front wheels pointed, and the rear is balancing the power on one side or the other to keep it going that way. If the car starts to understeer and try to go out of the turn - it puts more power to the outside rear wheel plus speeds it up to counter that. If it starts to oversteer and turn more into the direction of the turn, then the inside rear wheel gets more power and speeds up. Sort of the same as a limited slip differential or posi-traction on a rear wheel drive car. A normal differential has exactly the wrong effect as it puts more power to the rear wheel with the least traction.

The dynamics aren't any different on slippery roads then they are on dry - the same forces are in play in the same way. It is just that things happen at lower speeds becuase there is less grip to work with.

 

Hey Don, I must say a very lucid description of the mechanics in a turn. I actually learned a few things. All makes sense, just never gave it much thought.

 

There are two books that do a good job on the subject - the first is "How to Make Your Car Handle" by Fred Puhn. Another is "Secrets of Solo Racing" by Henry A Watts. Henry is local to me and he plus a couple of the drivers he refers to are "competitors" in the same events I participate in.

Autocross (in some places it is called Gymkhana) can really improve and sharpen your skills. It has helped my avoid problems on the highway. Any car that is safe to drive on the street is usable, and most events have loaner helmets.

It is the most fun I have ever had with my clothes on ( except perhaps for aerobatics in a high performance aircraft). I drive in events put on by the local Porsche Club of America chapter (I have two1969 Porsche 912's - one for "show" and the other for "go, man, go"), but PCA members can bring any car, so I will take the RL at some point (when I am sure I am out of the standings for a year end trophy - which is usually by the 3'rd or 4'th event of the year ). I would think the BMWCCA must hold events too. I am not really familiar with the SCCA but they probably hold introductory events as well.

 

correct

 

Don you have it all correct except for one thing, what determines the amount of "yaw" applied has nothing to do with slippage, although it is meant to reduce understeer, it is not reactive to it, it is proactive. The amount of rear overdriving force depends on steeering angle and engine torque, and is fixed for that particular angle and torque, the greater the angle and higher torque the more that outside wheel will be "overdriven". Remember not only is torque proportioned but the wheel is actually overdriven by an extra gear in front of the diff! The system is trully the snizzle. My point on the difference to other drive systems is simple in RWD your supposed to maintain power, not increase it since it will cause the car to oversteer, in a FWD car if you again accelerate the opposite will happen it will understeer. Thus no one is talking about lifting off but rather accelerating AKA increasing speed (DELTA V)rather than simply maintaining it to the limits of adhesion as you would in other types of drive systems.
In the RL SHAWD mainly serves to hide its nose heavy, inherently understeering nature. However, if driven very agressively it is truly SUPER! One ting that the RL neophyte expiriences often is understeer, precisely because they lift off when entering the turn. The trick is come in hot, break early and bury the accelerator jsut before you enter the turn

 

It does differ significantly from RWD, but as in rwd if you over do it it will slide the rear the main difference is that oversteer tendency is much lower than in a RWD vehicle, the other day I attempted an SHAWD style turn in my s2000, the rear whipped out faster than I could say ouch! It's fun no doubt but it isn't exactly the fastest way to go around a corner!

 

wikipedia on SH-AWD...
the system sends 70% of power to the front wheels and 30% to the rear under normal conditions but these ratios can reverse according if needed. The system can work because the clutches can slip under circumstances where the vehicle is turning, thus allowing the rear and front wheels to turn at different rates around a curve. SH-AWD in the Acura RL also features an "acceleration device" just ahead of the rear differential which, through the use of a planetary gear set, can overdrive the rear axles by as much as five percent. This is done only when cornering.

SH-AWD in the Acura RDX and Acura MDX use a default power split of 90% to the front and 10% to the rear. Like the unit in the RL, both can power the rear wheels with up to 70% of the engine torque. With this variant of SH-AWD, there is no acceleration device. Instead, the rear differential is constantly overdriven by 1.7 percent. During straight line performance, that difference is scrubbed off by the clutch packs. When cornering, the difference is used to overdrive the outside rear wheel.

 

I have discussed this before some months ago in another thread, but the other thing the RL has, that the RDX and MDX does not, is an acceleration device. Said device can overdrive the RL rear wheels up to 5%. Depending on your sources, that 5% may be additive to the same 1.7% inherent in the RDX and MDX design, so the RL is capable of a range of 1.7% to 5.7 or 6.7% overdrive to the rear wheels.

RL Power Train info says (italics added by me)

Acceleration Device

Positioned at the front of the RL rear drive unit, the Acceleration device typically passes torque rearward to the rear axle at very close to a one-to-one ratio (is this the 1.7% of the two SUVs or just 0.7%?). In cornering, however, the Acceleration device's output shaft spins faster than its input shaft.
The Acceleration assembly uses a compact planetary gearset to achieve its speed increase. Hydraulic actuators operate clutch packs that control the planetary gearset. When the input shaft is locked with the planetary gear carrier, there is no ratio change (this is the straight-line mode). During cornering, the carrier is coupled with the case, and the output shaft speed increases up to five percent. A speed sensor at the hypoid gear, downstream of the Acceleration device provides a feedback loop to the SH-AWD Electronic Control Unit to ensure that the system is working properly.

RDX Power Train info says (italics added by me)

SH-AWD™ System Layout

...The rear drive unit of the RL sedan has a clutch and planetary gear set, called the "acceleration device," that can increase the speed of the rear axle up to 5.7% (somewhere else, can't remember, I had found a source that said 6.7%). In contrast, the RDX's rear driveshaft is constantly overdriven by 1.7% - and the resulting overdrive effect is regulated by the left and right rear clutch packs. The difference in the chosen rear axle speed increase is due to the differing handling dynamics of the two vehicles. The RDX achieves the desired SH-AWD™ cornering characteristics with less outside rear wheel speed increase. A matched pair of direct electromagnetic clutch systems, one on each side, controls the power delivered to each rear wheel.

 

Acura Factory manuals as well as the RL brocures. They list it in the stats page. Unfortunately, I can't seem to locate one from every year the car has been out. I believe the torque split in 05 was different as well but I can't find the hard data to confirm.

The SHAWD on the RL is a more complex system than the RDX and MDX (just get underneath them and it becomes obvious). The RDX and MDX accelerate the outside wheel by 1.7% (faster than the inside rear wheel) where the RL can accelerate the outside wheel UP TO 5% faster (than the inside wheel). This is due to the planetary gear set that is used by the RL.

 

Slow-in/Fast-out.