Torsen or LSD - Still more confusion

[ozz]

I know, this topic has been covered many times and even as recently as last week. I've looked at all that I can find on this and there is a lot of great information out there on this topic. BUT, while knowing the number of friction plates and friction material used in a factory LSD is helpful and interesting, I'd like to make the questions a little more specific and am hopeful to get the benefit from all of this group's experiences and spend (the limited budget) doing it the right way the first time. This one is long so - if you are frustrated easily - stop reading now

Like many on this board, I'm struggling with the decision between Torsen style and LSD diffs - likely a GT (Guard) product over factory or other aftermarket providers. The application is a 74 w/ 3.0 and 915 (few engine mods and typical suspension upgrades) that will be a multi-purpose car. Usage will be about 50% street, 25% AX, and 25% track use. I've read the rule of thumb answers about street/AX = torsen and Track = LSD but, think there may be a little more to it. To frame the questions a little more - assume dry conditions - no snow or rain driving in the mix. Given that there will be compromise somewhere, my questions are this.

1) Do the downsides of an LSD in AX usage outweigh the benefits gained in somewhat aggressive street driving and limited track use?

Seems to me the so-called "track" benefits (braking and handling benefits) with an LSD will carry over to street driving. I would think that in addition to benefits on the track, the lock-up on decel with the LSD for spirited street driving would be a good thing to maintain rather that the "free wheeling" on decel that will occur with the torsen style approach. Other than possibly very low speed/sharp radius parking lot issues, I can’t imagine downside to LSD for (dry) street use. So, is understeer with an LSD in an AX situation less of an issue (and safer) to resolve than oversteer in spirited street/track driving with a Torsen that unlocks on decel? The idea of weight shifting forward on decel, unloading the rear end while the (Torsen) diff is unlocking too doesn't seem like a good combination. This points me toward LSD instead of Torsen BUT...

2) In one of the posts, wear and maintenance of the LSD was raised as an issue ( clutch parts etc.) where there was relatively no maintenance (other than lube change) in the Torsen style units. Not too interested in down time and costs to have the diff pulled and clutch parts replaced regularly. Has anyone with the GT LSD products experienced any real maintenance issues or wear problems?

3) Then, one of the Grady's posts mentions the amount of lock-up on decel degrading over time due to component wear. It states **"On the track a LSD has some interesting characteristics when at the limit. A LSD set at 50% and with even slightly worn clutch parts will unlock during the transition from trailing throttle deceleration turning into a turn to applying the power just prior to the apex. As power is applied it will progressively re lock. This noticeably changes the handling of a 911 toward increased oversteer just when you don’t want it."** SO, this moves me back to thinking Torsen style since, if I understand what is being said, clutch parts will wear and, an LSD w/worn parts creates an (unpredictable) situation of unlocking on decel. If this unlocking is going to occur, I would think that as a driver I could set up the car and develop a feel for the "consistent" result of the unlocking of the torsen rather than what would appear to be a changing and unpredictable unlocking condition as the clutch parts wear in the LSD. Anyone have any experience with this?

Maybe I am missing something (very likely) or making more out of this decision than I need to. As many of you have already pointed out, either of these would be an improvment over an open diff. Also, my driving skills may never (at least not anytime soon) expose the shortcomings of one of these over the other but, any thoughts or experiences with the above points would be appreciated.

[Dantilla]

Okay, now my head is spinning. You already know more than I did.

You now have a decision to make. I think that whichever way you go, you will learn the driving characteristics, and be happy.

(oh- and I love my Quaife!)

[ozz]

Dantilla,

How long have you had the Quaife diff and do you AX or track the car in additon to street use?

[chrisp]

I do not have a locking diff but I am closely watching your thread because I will get one sometime in the short to medium term.

I think although you break out the useage 50/25/25 you should pick the priority and set it up for that. For me I put 3,000 miles a year on my car with about 10 track days included. So although I would say my car is 90% street driven, every decision I make is focused around what's best for the track while keeping it street driveable.

When I choose my locking diff I will glean some info from this and other threads but ultimately ask the guys who set up the fast track cars because that's what I am looking for. For example there are folks near me selling their Torsions to go to factory style. Again, track useage is the preminent concern.

My guess is that street driving isn't (or at least shouldn't be IMHO) aggressive enough to push the limits of either style diff. Of course if it's a very high horsepower car than a locking diff could be a necessity on the street.

[ozz]

Chrisp,

I agree with all of your points and look forward to hearing from the guys setting up the track cars as you do.

Re: street driving, to clarify, I agree that street driving will likely not push "limits" of the components nessecerily but, there will be cause/effect on the street as well. When its safe to do so, we've all taken the sweeping on or off ramps at 75 (or so) or the familiar twisty bits pretty quickly early on Sunday morning and I suspect the reaction of the diff on decel, braking, and handling in those cases will be similar to track conditions. That is my aras of concern more so than managing acceleration. Your point about high horsepower requiring as LSD is probably considering the acceleration side of the issue - on accell, it seems that the two diffs respond very similalry.

My point is only that there may be many similarites and results from the handling perspective between certain street driving situations and track conditions. So the question remains which approach (LSD or torsen) will create the least number of issues that will need to be accounted for via other suspension adjustments or driving style. In other words, knowing what the best option is to make the car easy and predictible to drive fast and require least amount of adjustment between the various uses we put our cars through.

[jluetjen]

In other words, knowing what the best option is to make the car easy and predictible to drive fast and require least amount of adjustment between the various uses we put our cars through.

Skip Barber (or similar) Driving School sounds like your best option. Guaranteed a couple of seconds per lap off of your DE times which will beat the impact of a Limited Slip (generally about 1 to 1.5 seconds per lap on a tight track) any day.

My point is only that there may be many similarites and results from the handling perspective between certain street driving situations and track conditions.

In my experience -- No. Any road feature that you take at let's say 30 MPH on the street will be a 50-55 MPH + experience on the track. Most entrance ramps would constitute a white knuckle 85-90 MPH on the track. As a "for example", Lime Rock has 180 degree corner (Big Bend) which is a 78-80 MPH (at it's slowest point) in 3rd gear corner taken (ideally) at a constant drift. There is just no way that you can take a corner on the street with that sort of commitment.

There is just no comparison between the experience of driving on the street and on the track in my book.

[ozz]

I've had the opportunity to attend 2 high-performance driving schools at Mid-Ohio, put on by Calvin Fish and his crew. Granted, these were front-drivers but, that was certainly a great learning experience. Just being faster on the track wasn't my point though, and I absolutley agree that you cannot (and should not) attempt to recreate track speeds on the hightway.

I'm probably not communicating my question very well. What I'm trying to get at is if with a Torsen I'm going to get oversteer on decel/braking from 60 MPH on the street or from 90 MPH on the track, I still have an oversteer condition to deal with. That is consistent. Let's assume that I can eliminate or at least reduce that with an LSD. If that is true, the LSD "helps" me on street and track. So then are the negatives that the LSD unit will induce in an AX setting for example be able to be overcome by suspension adjustment - maybe by reducing front sway-bar seetings or the like (just an example)?

The core of the question is with which unit will I pickup most benefit and least neagitves all things considered. IF the example above is accurate, I get benefit of the LSD on the street, on the track, and can adjust out some or most of the negative created in an AX situation.

IF this is correct, it would seem that the LSD would be the better choice over the torsen style. But then go back to my original post and see point 3 - THATs why this is such a puzzle.

[rdane]

I think you are putting too much thought into it.

AX can take advantage of the torsion type
better. Track and street will take better advantage of the LSD.

Just decide what your priority is, pick the best for that use, install it, and then go drive the car and figure out how to take advantage of either system.

[jluetjen]

What I'm trying to get at is if with a Torsen I'm going to get oversteer on decel/braking from 60 MPH on the street or from 90 MPH on the track,

As far as I know Torsens are open on overrun and so are no different when decelarating or braking then a normal open diff. They will reduce oversteer (increase understeer) when accelerating.

A properly configured LSD will reduce oversteer when decelerating. If you play around with the ramp angles (which will require custom machining if you are using a non-standard configuration) you can set up your diff to have little affect when you are on the gas. I don't know if this configuration is available, but if you were to run 85 degree acceleration ramps (virtually no clamping force) and 45 degree over-run ramps (a fairly high clamping force) you might be able to get to what you want. Reducing the acceleration ramp angle will provide increasing clamping force. Increasing the number of clutch faces will make the clamping "softer" while reducing them will make the clamping more abrupt. You can also increase the preload which affects the clamping force across the board. To be honest, since your running a 911, I doubt that you are going somewhere where others have not gone already.

Whatever you do, you will most likely need to adjust your sway bar and/or spring set-ups (if not camber and toe also) after you have installed the diff. If you didn't have to change those things, it would suggest that the diff didn't make any difference in the handling since all of the set-up paremeters are interrelated.

Maybe one of the gear-box experts will chime in.

[chrisp]

Your questions may be better suited for the manufacturers rather than folks on this board. I think whatever info you find out will be interesting and certainly worth adding to this thread.

You'll get a certain amount of salesmanship from all of the companies you speak with but with very specific questions like yours you should be able to work your way through that and get answers.

Try calling them and let us know what you find out.

[KTL]

Why not PM Geary and alert him to this topic?

Geary = Guard Transmission = someone who knows his stuff

[ozz]

Guys - this has been helpful - you have helped me to think thru the issues and break them down a little.

John, I agree that "A properly configured LSD will reduce oversteer when decelerating." and in fact I make that general point in the first part of my post. Thats the good news with the LSD. The bad news is that the LSD, unlike the Torsen, will almost certainly create oversteer in low-speed turns in an AX event. For what I'm trying to get to - this is all bolied down to these 2 questions.

1) LSD will be the ticket for me for street and track BUT - can I adjust out (most of the) oversteer created on low-speed turns with the LSD via a combination of suspension settings (sway bar etc.) when running an AX event?

AND THEN

2) See point 3 in my original post above - does the clutch wear in an LSD cause variing and unpredicatble unlock conditions on decel and braking (creating unpredictable oversteer)? If it does, this may be worse than predictable unlock on decel that we get with the Torsen style.

SO #1 is for you track guys and suspension gurus and I think #2 is for the Mfg as CHGRISP suggests.

At the end of the day, RDANE is probably right - I'm putting too much thought into it and just need to pick one a go drive it!!

Thanks again guys!!

[Geary]

I know nothing about how to properly tune suspensions to compensate for different differential types. As a matter of fact, I'm not much of a mechanic, period --- so please don't look to me for any definitive answer on this question. Both differential types are offered with the idea that whoever is setting up the car knows what he wants. However, I can pass along bits of info that I learn from a wide range of shops and teams that use the various components. Here are a few bits of info:

1) 2003's top ALMS team set their 50/80 LSD with so much preload (doubled-up Belleville washers) that I was surprised the LSD didn't explode. It was practically a spool. On the other hand, one of California's better-known builders replaces our heavy-duty Belleville washers with thin ones, yielding very little preload.

2) Autocross racers are adamant about the superiority of torque-biasing differentials over LSDs in Autocross. On the other hand, not a single ALMS GT3R has ever raced with a TBD, although two or three teams have tried them in test sessions with disappointing results. Two different types of racing, two different needs.

3) While the fibrous materialed friction discs found in the street 993 LSD wear out in short order, the friction discs found in the Motorsports or aftermarket version LSDs last a surprisingly long time in anything short of a track-only race car. I believe that Porsche recommended 30,000 miles for the 930 40% LSD (which had the good friction discs).

You can't have it all. How do you envision driving your street car? As though you were on an autocross course? Or as though you were passing Ferraris on the track? Hopefully the former.

"There is just no comparison between the experience of driving on the street and on the track in my book." -- Amen, John. Personally, I’d be choosing the torque-biasing differential for a street/autocross car.

[ozz]

Geary,

You've answered the question about wear in an aftermarket LSD and the related (and unpredicatble) unlocking on decel. Exactly what Paul Guard shared with me just earlier today. Given that, the only question that remains is can I compensate/adjust etc. for the oversteer the LSD may induce in the AX event.

I understand and agree with your point #2 Geary about differnt types of racing and differnt needs. Unfortuantly, I enjoy both types and driving on the steet but can afford only 1 car and 1 set up. So as I said, understanding that there are downsides to each approach, I am trying to find the one with the most benefits and least (or easiest to correct) set of downsides.

And once again, I agree with John and Geray both that "There is just no comparison between the experience of driving on the street and on the track in my book." They are two differnt worlds with two differnt approaches however, I don't agree that vehicle dynamics (weight transfer, over/understeer, etc) are not similar between the two which is what we are all effecting with varoious diffs, suspension mods, set-ups etc. I don't have to be driving (on the street) like am passing a Ferrari to notice and need to deal with these dynamics. With they be as sharp at street speeds - no, of course not. But, similar to what I pointed out above, whether I'm decelertaing from 60 mph to 40 mph and turing in to the Carosel at Mid-Ohio or from 60 mph to 40 mph and bending into an off ramp on Route 90, the vehicle dynamics are gong be the same and need to be dealt with.

Thanks eveyone for the info and ideas. I think I'm going to go with a Limited slip and see if I can't adjust out or compensate for oversteer when I go AX-ing. I'll post back here with how this works out.....