I'm ordering a ITR Tranny tomorrow (obviously equipped with an LSD!), and I kind of understand the difference between the two types of diffs, but I just wanna know exactly what this LSD is actually gonna do for me, so if anyone could help me out with a brief explanation

 

ummm....brief explanation? ok, here goes. lsd drives both wheels simultaneously, whereas the open diff only sends power to the wheel with least resistance. bingo.

 

Thats kinda what I thought, I just wanted to get it straight. Thanx a lot!

 

Open diffs are what most cars have, take a turn, punch the gas, and the inside tires spins (GM J cars, civics, eclipse, etc.) the viscious LSD is fluid driven....(Nissan SE-R old ones) these types wear out after a while. My 1994 SE-R's LSD is just about worn out, and will be replaced with a Quaife Diff when I get the money....our R's is Helical type (new SE-R also) and this one takes the torque, and applies it to the inside wheel, pulling you closer to the inside of the turn instead of pushing you off the line!

I Might be way off and smoking something, but I am kinda sure that these are the different kinds of Diffs... open= Helical =

hope that hepls?
~Neeraj

 

i used to think the itr lsd was the bomb....until i drove a car in competition that had a clutch type lsd. the stock itr lsd doesn't even compare to it. oh, what's autocrossing?

 

Autocrossing is Parking lot racing, there are cones set up in a parking lot, and you race against the clock, not directly against anone else. Each cone you hit is a 2 second penelty. it is a lot of fun for $10-$30 per day.... check the scca site for local races in your area... http://www.scca.org

Yeah, I have heard that the clutch type is the bombtrack, but they are pretty noisy. that is what I have heard, never had the opertunity to drive one yet...

~Neeraj

 

Here you go: http://www.howstuffworks.com/differential.htm

 

One better, http://www.torsen.com they have several nice short descriptions with nice photos and one technical article. I strongly advise everyone to take a look...

Other things with an LSD: when cornering and you lift off of the throttle in a FF car with an LSD the car tends toward understeer which aids stability.

If you trail brake a FF car with an LSD it tends to promote understeer which aids stability.

Thru a corner at constant velocity it again promotes understeering and thus aiding stability.

Only when accelerating thru an apex and exiting a corner does it tend to "tuck the nose of the car" [i.e., oversteer] such that you can run a tighter line.

The concept of "one way" or "one and a half way"versus "two way" options is rubbish on a helical or Torque Sensing limited slip differential as the friction of running the gears varies with the direction on a true torsion and with a helical diff the same is true as is the frictional surfaces on either side of the housing as well as quite possibly the face of the side gears. So they are "two way" Limited-Slip Differentials. Actually, they aren't limited slip at all. They are actually Torque Biasing Units. Only the Clutch type are truly Limited-Slip Differentials.

Also, a viscous limited slip [speed sensitive like a clutch pack type] are the best for negotiating ice and what not, but that's not what our cars are designed for so this is merely anecdotal information.

There are several Society of Automotive Engineers Technical Paper Series articles on Limited-Slip Differential applications on FWD cars and the effects they have on vehicle handling behavior. I'll see how many copies I can make...

 

hmmm my experience was that the R LSD was ok up until it grenaded and popped a nice hole in the case
The R LSD was a little more transparent tho...meaning you didnt feel its effects as much
the Kaaz replaced it and felt way more aggresive (set up at 2way and not 1.5way)
but that coulda been the difference



[Modified by 555R, 12:16 AM 11/8/2001]

 

The "more aggressive" nature of the differential is much more than likely [I'd be willing to wager a substantial amount on this one] is due to the torque bias ratio. The greater the torque bias the greater the 'tug' on the steering wheel when traction varies between the front two wheels... all things being equal and in your case it was that. In other words, the same car, suspenson, etc with the only change being the diff.

 

hmmmmmm
trail braking
something I learned not to do as it upsets the balance of the car is a turn near the limit...
almost met the Blue Bushes that way

 

What kind of "competition" did you do?

 

How did you come to this conclusion? I'm interested to know how lifting off the throttle can cause understeer.

The understeer generated then lifting off the throttle and trail braking has nothing to do with the LSD, but with the transfer of weight towards the front in both cases. I don't believe an LSD can influence this behaviour.

Also, no matter how good your LSD is, an FF car isn't designed to power-on oversteer. Full acceleration almost always causes understeer in FF cars. This is why we can mash the throttle on oversteering situations to get the car back on track...

 

Trail braking is actually a very good way to get a fwd car (or any car that understeers a lot to turn-in). The Type R responds very well to this on a track.

The lsd in the stock type r in my opinion is very good. I've driven a CRX with a clutch pact type lsd and also one with a quaife. I do prefer the quaife unit the most.

 

I think you didn't notice rodney's sense of sarcasm. He is an avid road racer and autoxer...

 

i can see using trail braking in a autocross low speed application
but hell at 100plus mph its not such a good idea

 

You can still use trail braking at higher speeds, maybe not 100mph plus, but definately anything below 80mph. You just have to be really careful.

 

Trail braking is used at relatively heavy braking zones. If you're going 100 mph and the next corner is a 90 mph corner, you can not use trail braking. Your braking will consist of just a stab of the brakes, nothing more. Trail braking requires you to be on the brakes for a few seconds to transfer the weight to the front wheels and carry this balance into your turn in. The unloaded rear tires will have less traction this way, which is what causes the car to be able to rotate. In a short braking zone where you only need to get on the brakes for very little, you'll be done with the "slowing down" part very quickly and you won't have any braking left that you could do in your turn in. Since a type r can reach top speeds of about 120mph on most tracks, you can't really do any trail braking for a 100 mph turn since there isn't much braking to begin with. If you would be in a GT race car approaching that same turn from 150 mph, you could do it.

This technique is used mostly when driving a car that is more prone to understeer. A properly setup car does not require its use as much since it is supposed to be nuetral in the first place and should not have trouble turning in. In longer races when the tires go away towards the end and the car again starts to understeer, some drivers rely heavily on trail braking to compensate an understeering car and make it rotate better towards the end of a race. This is basically what I understand of trail braking, maybe someone can add to this. It does work, especially on a (stock) type r.

 

thanks ken, i guess i should have out a j/k in there....

 

--------------------------------------------------------------------------------
Other things with an LSD: when cornering and you lift off of the throttle in a FF car with an LSD the car tends toward understeer which aids stability.
If you trail brake a FF car with an LSD it tends to promote understeer which aids stability.


--------------------------------------------------------------------------------

How did you come to this conclusion? I'm interested to know how lifting off the throttle can cause understeer.

When you are applying torque to both wheels in a straight line you have a 50/50 torque split [right/left]. However, when you make a right turn you transfer weight to the left [i.e., outside] tire contact patch which unloads the inside tire contact patch. As the differential transmits torque to the outside wheel due to lack of adhesion of the inside contact patch it promotes OVERsteer. Correct? This allows one to power thru apexs and exits while keeping a tight line.

So what happens if you lift? The differential works "in reverse." The weight transfer causes the diff to transmit torque to the inside tire contact patch which resists the car from turning in. This can be measured by the degrees of steering wheel correction necessary to maintain the same radius in a curve. The larger the torque bias ratio the more pronounced the effect.

The understeer generated then lifting off the throttle and trail braking has nothing to do with the LSD, but with the transfer of weight towards the front in both cases. I don't believe an LSD can influence this behaviour.

Actually, it does play a non-insignificant role due to that very same weight transfer the Torque applied to the ground is a function of the normal force of the ground applied to the tire. It's true for coast or overrun as well as throttling the car.


quote:
--------------------------------------------------------------------------------
Only when accelerating thru an apex and exiting a corner does it tend to "tuck the nose of the car" [i.e., oversteer] such that you can run a tighter line.
--------------------------------------------------------------------------------

Also, no matter how good your LSD is, an FF car isn't designed to power-on oversteer. Full acceleration almost always causes understeer in FF cars. This is why we can mash the throttle on oversteering situations to get the car back on track...

Yes, but we're comparing an ITR with and without a torque-biasing limited-slip differential here. Or a Celica GT-S with and without the TRD-supplied helical limited-slip differential... The diff makes difference [pun intended] and if the bias ratio is large enough so will the difference be in these conditions.


These conclusions were reported in several SAE Technical Papers and Japanese SAE Technical Papers by Torsen and GKN. I also confirmed their results with several Ford Technical Specialists in Vehicle Dynamics as well a particularly technical individual at Zexel-Torsen and Visteon.

It appears that the LSD supplies more stable [predictable] dynamic behavior in limits of [or nearly so] adhesion handling situations. While it is initially counterintuitive [especially for constant velocity turn negotiation!!] a very good vehicle dynamics textbook will provide the fundamentals of applied forces for a vehicle negotiating a turn. I've also "played" in slippery conditions paying very close attention to what I'd have to do to the wheel. One thing that compounds the confusion is that if you lift in an ITR you get rear rotation. I was told time and time again that w/o the diff the chassis would be retuned to compensate for different behavior on the limit - and this was from a Ford of Europe Vehicle Dynamics engineer who works on the Focus Program.

Why would I "go off on a tangent" looking into LSDs? Because they initially confused the Hades out of me: Six months later and <<Voila!>> They're pretty simple, but I take my hat off to the brainy SOB who initially figured it all out and developed it. It's all too easy to follow those who've done their homework.

After studying LSDs for both work-related reasons and particularly their effect on FF vehicles I am firmly convinced that all vehicles should have them. So why wouldn't Honda offer one for the Civic Type R? And BMW for the M5 - there is one on the new Z3. And why are Porsche 911s open-diffed cars [except for the GT2 and maybe the GT3]? Believe me when I tell you that you've got something special and many here may open track and autocross much more than I do [I didn't do any events this season] so you're intimately aware of the finer points by experience. I shall be, too, but I'm trying to figure out what exactly did Honda do to make the car so damned nimble. If you have any further detailed questions please feel free to e-mail me: itrelise@aol.com

 

Actually, it doesn't promote oversteer but more appropriately reduces understeer. I see where you're going though and it sounds mechanically correct, but when trail braking, lifting off the throttle the weight dynamics mostly probably take over at that point and you don't really notice the LSD as much as you would while accelerating through a corner.

I must admit I haven't really researched how LSD work in such detail, but have had enough field experience with them. It seems like you've done some pretty thorough research on LSD's and how they work. VERY interesting information!

Someone did indeed do their homework. I'm sure all the information if pretty accurate, but it seems as though they only analysed the Torsen LSD charateristics alone and didn't analyse other things such as weight and chassis dynamics. That wasn't part of their search and probably didn't need to be.

I still believe that although a torque sensing LSD does cause the car to understeer more lifting off the throttle... But it actually plays a smaller role than we think.

Something I agree completely with. I always thought that every car should have an LSD. Manufactured in large quantities, I don't think they would cost even 1/3 of the price they are now. Not only for performance reasons, but also safety-wise in low traction situations.

I learnt a bit about LSDs as well today. Thanks for the information.

 

Actually, it doesn't promote oversteer but more appropriately reduces understeer.

You would be CORRECT. It mitgates or reduces the tendency of a FF vehicle to understeer. I OVERsteered.

... when trail braking, lifting off the throttle the weight dynamics mostly probably take over at that point and you don't really notice the LSD as much as you would while accelerating through a corner.

It's hard to say. I'm not an accomplished enough driver to separate such details while driving at MY limit. I know the mechanics and now need to practice. As it is I spend perhaps as much as 80 to 85% of my focus on negotiating with the rest analyzing what the car is doing - or more importantly what it is I am doing to compensate. More practice needed.

I have a suspiscion that if one were to remove the LSD that a GOOD Vehicle Dynamics test driver would instantly notice the differences - even if you trail braked or backed off of the throttle. I have a presentation with Ford VP Richard Parry-Jones shortly and if he has the time I'll ask him. He's a Vehicle Dynamics whiz and actually rally raced at one time in his career and is bloody quick on a motorcycle as well. It's rumored that when he/they/Ford visited Suzuki they offered to let him ride a GSX-R750. He was as quick as their test riders and they were impressed that he was able to negotiate one or two turns, in particular, quicker than they ever did. he then began to explain [in great technical detail] what the suspension and chassis were doing and what he'd like to have done to it and that he'd go quicker... Wiser, cooler heads prevailed and his test ride was over.

 

I agree with you in that every car should have a LSD just basically for Traction reasons... but NO!!! What we are doing, GM, is putting $$ into new technologys of Electronic traction control... all that money, and it does not do one bit of performance good!

~Neeraj