Talk:Limited-slip differential/Archive 1

howstuffworks.com
Has anyone gone to visit howstuffworks.com's differential section lately? It is quite clear. - Robert Turner 05:27, 9 September 2006 (UTC)

Fixed Differential - Spool
I removed this: " Compare this behavior to that of a fixed differential, where there is never a speed difference between the shafts". To my mind this is confusing on two counts. The linked article doesn't exist, so one cannot make the suggested comparison; and the description is odd, because it says that there is never a speed difference - if so, then how is it a differential? Perhaps the writer simply means a crown wheel and pinion gear(?) - this is not a differential. I also removed one of the section headings - it seems superfluous given that there are already numerous subheadings in what is quite a short article. Graham 00:50, 27 Jul 2004 (UTC)


 * It is somewhat common to call a simple (crown wheel and pinion, etc.) gearing a "fixed differential" in discussions of differentials. One would not normally use the term elsewhere, but it is proper here. AlbertCahalan 01:36, 20 May 2005 (UTC)


 * If it's a fixed crown wheel and pinion, it's not a differential at all, by definition. If some people call it that, then that's there ignorance - it shouldn't be propagated as a fact in an encyclopedia. Graham 03:03, 20 May 2005 (UTC)


 * Language is how people use it, so, yes it is a differential. Perhaps you'd prefer "null differential", but that's not the term people have settled on. In other news, "zero" is a number. AlbertCahalan 05:04, 20 May 2005 (UTC)


 * Not really - "differential" implies "difference" - in this case in the speed between its output shafts. If there is no difference (and there can't be here since they are joined rigidly together) it cannot be a differential. And yes, zero has always been a number, it just took man along time to realise it. I don't see the analogy. Graham 07:32, 20 May 2005 (UTC)


 * Ah, but it does allow a difference in speed! Of course, that difference is zero. AlbertCahalan 08:06, 20 May 2005 (UTC)

Although it is strictly incorrect to call a basic crown and pinion gearset a differential, that term has expanded to cover even this non-difference differential today. Back in the day (the early part of the century), "differential" always meant a gearset that allowed a speed differential, but this implication has pretty much gone by the wayside as the crown and pinion disappeared. --SFoskett 12:54, May 20, 2005 (UTC)

The word I know it by is "Spool". I wondered if it might be just an Antipodean word, but I checked Carroll Smith, and he uses it, so it's definitely North American as well, don't know about the UK. The traditional spool has a solid shaft connecting the rear wheels, i.e. no diff centre at all, and just a pinion/ring gear. Obviously that can only be used on live axles, but now there's spools where the solid shaft extends only to the uni joints for IRS. There's also the mini-spool which is the most popular in terms of sales because it's cheap and easy to fit (don't have to remove hubs/axles). The mini-spool keeps the existing axles, but just replaces the diff centre with a simple pinion/ring gear. I'm not sure whether a mention of the spool would be in order or not, because it's not a diff, and it's not limited slip, but it's used by similar people for similar purpose. RB30DE 10:07, 5 June 2006 (UTC)

A late-comer's followup: Removing "fixed differential" was a good edit. In the last month, I've talked face to face with three people in the racing industry who described a pinion turning a ring (crown wheel) fixed to a single solid axle as a "spool": a drag racer (who uses a spool), a sprint car owner (who uses a spool), and a custom welder. Two of them personally corrected me when I attempted to call a spool some sort of non-slipping or fixed differential (I don't remember how I said it exactly). The sample is small but it's 100% and from the group whose language use we should follow, in my opinion. -- Another Stickler (talk) 20:11, 7 October 2008 (UTC)

Changed explanation of dilatant fluids
Previous version described a dilatant fluid as "one whose viscosity varies as the inverse proportion of an applied force". This explanation is wrong. A dilatant fluid becomes MORE viscous as force is increased, not less. Also, this relationship is not linear (ie. not proportional). See Power-law fluid. Ryanrs 06:54, 10 November 2005 (UTC)

reversion of large edit
I reverted a big tract of added text, contributed by anon. There is some useful info in what was added, but on the whole it was more confusing than useful. IF such details are worth including (debatable, in an encyclopedia) then it needs to be much, much, much better written. As the contributor freely admitted, it was a cut and paste job from an online forum, where writing standards are different. In addition, it cut what was already a decent section of text clean in two. Can I suggest to the contributor in question that a little more experience of WP is gained before trying to put it back in, so you know what we are looking for. Graham 01:13, 22 December 2005 (UTC)

Needs Diagrams
If this page had some clear diagrams like the page on Differential (mechanics) I think it would have given me a much clearer picture of the device and how it worked.


 * Amen to that! Anon is right [and should sign his/her postings].  These phrases particularly beg illustration:
 * "A 1-way notch is cut like an upside down triangle"
 * "A 1.5-way notch is like an upside down triangle with a half triangle on top of it."
 * "A 2-way notch is shaped like a diamond."
 * --Nil0lab 17:02, 2 April 2006 (UTC)

plagarism
Parts of this article are almost literal cut and pastes from a few forums. I'm sorry to say, forum or not, the copyright is still held by someone. I'm going to be pulling this stuff promptly.-- Oni Ookami Alfador Talk 02:33, 2 June 2006 (UTC)
 * Proof: http://www.drifting.com/forums/showthread.php?t=6800 October 16 2004 was the date of the post on the forum.  This content was added after that.--  Oni Ookami Alfador Talk 02:41, 2 June 2006 (UTC)


 * I believe that the author of those forums re-used his own text here, so it is probably OK from the copyright point of view. However, I do feel that much of the added text is quite inappropriate and overdetailed for an encyclopedia article- so if you ditch it on that basis I for one would support that. Graham 05:18, 2 June 2006 (UTC)


 * I found nothing in the edit summaries that supports that, but I could have missed it. In any case, yes its not very encyclopedic, but it is good info and all pretty much correct to the best of my knowledge.  I'm rewriting a paraphrase version now that will work well in the encyclopdia and maintain all the key info with less of the manual/guide aspect.--  Oni Ookami Alfador Talk 04:33, 3 June 2006 (UTC)

Viscous LSD revert
I deleted the reference to changing the fluid in VLSD centres, which has since been reverted. This is incorrect. The Nissan R200 VLSD centre is sealed and not serviceable. The differential housing itself contains gear oil which can be changed, but the cartridge in the centre is quite separate from this. When they start open wheeling, the only way to fix it is to replace it, usually with an aftermarket clutch pack, because Nissan charge a ridiculous price for the insert. If you disassemble your diff you will see quite clearly what I mean. RB30DE 09:18, 4 June 2006 (UTC)


 * Sorry, not really reasonable of me to expect you to take the back off, I tried to find some images showing a viscous diff. Actually I found some photos labelled "viscous diff" that were not, so people don't necessarily know what is in their housing.  Anyway - Autozine Technical School - pretty decent site overall, new to me, also very good explanation of Torsen diff.  Visco-Kupplung - German page with excellent photo of cutaway diff. Physics Of Car Racing - another good photo of cutaway diff.  RB30DE 12:22, 4 June 2006 (UTC)


 * Oooh found a picture (in book) of a racing viscous diff which looks like it can be disassembled! I'm wrong! But still no passenger car VLSD that can be, hacksaws only to date. RB30DE 03:09, 6 June 2006 (UTC)


 * Fixed it. Why be distructive and remove info when all you have to do is clarify for the few people who're not going to get it?  ask any mechanic and they will tell you about Changing differential fluid/oil meaning the fluid in the housing.  Its just a generic speaking term but I do see where it might cause confusion.--  Oni Ookami Alfador Talk 03:50, 6 June 2006 (UTC)


 * Aiyo. Please do call my edit destructive, I'm trying to improve the article.  I think you've understood that the oil in the diff housing is not the silicone fluid in the chamber that forms the outer of the VLSD.  There are two entirely separate enclosures.  The chamber that contains the silicone fluid is inside the oil filled diff housing and is separated from it by seals.  But you haven't understood that you do not service a VLSD by changing the housing oil, you throw the VLSD away and start again.
 * This is the sentence we've got now - "However, they are less prone to breakdown as long as the housing fluid is changed regularly.". Whatever you do with the housing oil, if you run mashed bananas and sawdust in there, it has no effect on the VLSD centre.  Changing the housing oil does not extend the service life of the VLSD.  The VLSD will expire at its own convenience, and it will expire sooner than a mechanical LSD.  They are more prone to breakdown, not less, and they are unaffected by oil/fluid changes.  Why remove "info"?  Because it's false.  You keep reinserting these errors - is an incorrect statement better than no statement at all?  Why would you do that?  I'm baffled.  If I come up with a more encyclopaedic paragraph that covers servicing and service life, replacing the sentence that's there now, are you going to continue reverting?  210.0.100.153 03:17, 8 June 2006 (UTC)

Mechanical LSD Description
Current description is kind of a mish mash of characteristics of several different diff types, doesn't make sense. If talking about clutch type LSDs, it's not a difference in axle speed that activates the cam action which compresses the clutch packs, it's drive shaft input torque. The centrifugal rotor thing doesn't apply to clutch/cone/viscous/torsen diffs, the only design I'm aware of that it fits is the Detroit Locker / Lokka (tm) diff (possibly Weismann locker), which is an on/off design, either open or locked. Perhaps either the article should be a lot longer, or a lot shorter. Also if Detroit Lockers are going in, perhaps air lockers & hydraulic lockers come with them? There's a whole lot of other motorsport LSDs that passenger cars rarely see which could possibly get a sentence each down the bottom or something. RB30DE 03:09, 6 June 2006 (UTC)


 * Expanded the clutch LSD bit. Removed the part about centrifugal rotors etc.  The other little fly in the ointment is I've only just become aware there are (at least) two kinds of clutch LSD.  There's the type I've described, and there's a type where the clutches are compressed by springs(?).  In my part of the world, that kind of friction-surface-mated-by-springs LSD is a cone LSD, not familiar with the spring loaded clutch type.  I was going to write a cone LSD description, but that needs to encompass spring loaded clutches as well now I think. RB30DE 22:08, 21 June 2006 (UTC)


 * The article has changed since the above comments, but the article still needs some explanation about bias ratios and its effect on behaviour. That is only mentioned under Torque-sensitive LSDs, and the only figure mentioned there is 80% (and thus a bias ratio of 4 to 1). 121.218.77.253 (talk) 23:52, 28 October 2009 (UTC)

Mechanical LSD Left/Right Function
For an LSD the torques at the left and right wheels are different when the differential is "differentialling", and when it does the larger torque always goes to the slower wheel (at that moment). That means when turning a corner without significant wheelspin (at lower acceleration), the inner wheel is going slower so that the larger torque is there, and when that larger torque is sufficient (at increased acceleration) it will cause an increase in speed at that inner wheel until they rotate at the same speed and the LSD is "non-differentialling" (which means it is effectively "locked") and the distribution of torque depends on the "slip" at each tyre contact patch as they rotate at the same speed. At even larger acceleration it will cause more speed at the inner wheel so that it rotates faster so that the outer wheel is slower than it, and under those changed circumstances more torque goes to the outside wheel as is usually explained. 121.218.77.253 (talk) 22:45, 25 October 2009 (UTC)


 * A Locking differential has "lock up" where the differential has the ability to behave as one common shaft. This ability is the result of the manipulation of the differential mechanism. Regardless of traction, or even wheels for that matter. The 2 axles spin as one common shaft regardless of speed, torque, power, etc. Your definition of lock up, is not what anyone who sets up axles and/or goes off roading refers to as lock up.--Dana60Cummins (talk) 00:28, 27 October 2009 (UTC)


 * I was trying to explain the normal LSD function only, because someone did not understand with LSDs which of the wheels received the larger torque. I used the term "locked" in quotes to indicate the wheels were at the same speed without being actually locked such as by using a locker. The explanation can still be complete if I just leave out the part ""(which means it is effectively "locked")"".121.218.77.253 (talk) 22:15, 27 October 2009 (UTC)


 * I think you're assuming that the inside wheel's [angular velocity x torque] equals the outside wheels [angular velocity x torque]. This is not the case.  There is not equal power distribution (which would imply more torque to the inside wheel), there is either very close to equal torque or more torque to the outside wheel.  RB30DE (talk) 07:04, 4 November 2009 (UTC)


 * I am not assuming what you suggest, and what I wrote is based on the change in direction of the "differentialling" under the different conditions. Your first reference relates to an open differential and not to the LSD that I was discussing.  I accept your second reference.121.218.77.253 (talk) 04:36, 5 November 2009 (UTC)

I have changed it a little -
 * For an LSD the torques at the left and right wheels are different when the differential is "differentialling", and when it does the larger torque always goes to the slower wheel (at that moment) because at that side the differential output shaft is going "backwards". That means when turning a corner without significant wheelspin (at lower acceleration), the inner wheel is going slower so that the larger torque is there (by the bias ratio), and when that larger torque is sufficient (at increased acceleration) it will cause an increase in speed at that inner wheel until they rotate at the same speed and the LSD is "non-differentialling" and then the distribution of torque depends on the "slip" at each tyre contact patch as they rotate at the same speed. At even larger acceleration it will cause more speed at the inner wheel so that it rotates faster so that the outer wheel is slower than it, and under those changed circumstances more torque goes to the outside wheel (by the bias ratio).  121.218.77.253 (talk) 01:47, 10 November 2009 (UTC)


 * Explain what you mean by the differential output shaft spinning backwords I don't understand this at all. I've been working on axles for years and understand none of this.--Dana60Cummins (talk) 16:17, 10 November 2009 (UTC)


 * If there is drive torque giving acceleration, then the reaction torque coming from both axle shafts back onto the differential is effectively applied "backwards" compared to the direction of forward rotation. If one tests a differential by first applying equal torques in that "backwards" manner then no "differentialling" occurs.  If one torque is then increased (say shaft A) until "differentialling" occurs, then shaft A rotates in that "backwards" direction and shaft B in the "forwards" direction.  Thus the larger torque is in shaft A going "backwards".   If the test is repeated where the torque in shaft B is increased then the shafts rotate in the opposite direction, and the larger torque is in shaft B which is going "backwards".121.218.77.253 (talk) 22:45, 10 November 2009 (UTC)


 * I expected some further comment on this. Is there someone else out there who has also tested differentials? They might use their own way to describe this. 121.218.77.253 (talk) 22:45, 26 November 2009 (UTC)


 * You have a unique view on differentials.--Dana60Cummins (talk) 03:17, 27 November 2009 (UTC)


 * It just comes from testing differentials, including high-bias ones, for my own purposes, and I assumed that anyone else testing them would have essentially similar results, and so it would be "well-known", although I admit not actually "referenced". Surely there is someone out there? 121.218.77.253 (talk) 06:48, 27 November 2009 (UTC)


 * You have a extremely unique way of testing differentials then.--Dana60Cummins (talk) 19:38, 30 November 2009 (UTC)


 * I just used a very basic test. I had a frame to resist rotation of the housing, then fitted an output shaft to one output spline (say shaft B) with a lever on it with a weight on the end (lever length and weight used to determined a test torque for shaft B) but with the weight initially on the ground, and then applied sufficient torque to the other output spline (shaft A) to get the lever to lift the weight off the ground and thus apply the pre-determined torque to shaft B.  The bias ratio is that of the applied torque measured in shaft A to that in shaft B.  The total 'drive' torque simulated for the test is the total of those in shafts A and B.  Normal differentials gave ratios about 1.3:1 or so (as Carrol Smith says) and the high-bias ratio differentials that I was testing gave around 4:1 to 8:1.  I have done the same test without the weight by having someone else apply a 'fixed' torque to shaft B, and I have done the same test while a differential is still mounted in a vehicle.   121.218.77.253 (talk) 04:02, 1 December 2009 (UTC)


 * Drive torque is the amount of torque that the tire can turn into vehicle movement. You can't factor in drive torque w/o a vehicle.  Also:  Limited slips only transfer a certain amount of torque from the low traction side to the high traction side.  It does this by applying a certain amount of braking force to the low traction side.  This certain amount means the transfer of torque and braking force is limited.  This limit is surpassed when the braking force is overcome by the input torque from the ring gear.  This amount of torque is referred to as breakaway torque.  Nothing in your edits or your testing factor in any of this.  --Dana60Cummins (talk) 20:31, 2 December 2009 (UTC)


 * I understand your description and it is actually consistent with mine, and what you refer to as the 'certain amount' can be determined from my described test by taking half the difference between the torques at shaft A and shaft B (because it was subtracted from one side, and added to the other side), and the 'input torque from the ring gear' can be determined from my described test as the total of the torques at shaft A and shaft B. (Note - while looking at preview I note my IP address has changed from 121 218 77 253) 138.130.84.44 (talk) 03:01, 5 December 2009 (UTC)

Viscous LSD Description
Changed the mechanism description a bit. Moved the Gerotor paragraph out of that position, it didn't make sense there. I can't understand the Gerotor description too well, I don't know anything about them, can someone who does expand that a bit? Who uses them? RB30DE 00:10, 16 July 2006 (UTC)

"Electronic LSD"
I removed the entry for "Electronic LSD" as this is not an example of a LSD at all ( Note: LSD = Limited slip differential) this was an explantation of how traction control systems work and while some of the outcomes of both are the same they reach their goals in entirely different ways.
 * You are right, a system that uses brakes to limit slip, without affecting the differential, is not an LSD. But there are electonic differential locks, that electronically affect how the differential function, and we should cover it. For example, latest Skoda Octavia TDI PD 170 VRS, uses something that they market as "Electronic Differential Lock (EDL)", and we should probably cover it. Personally me don't know exactly how this Skoda's unit works. I've augmented the section about worm-gear-based LSDs, please review this section. Thanks: --Maxim Masiutin 11:54, 26 January 2007

It is a form of speed sensitive limited slip differential - look for a definition of differential. That is what it does. No doubt about it. The fact that it is using a brake on the end of the axel, not a clutch inside the differential gear housing is irrelevant - that is NOT an "entirely different way", it is pretty much exactly the same thing. That is how it is used and that is how it is marketed (see Toyota's ATRAC, Nissan's ALSB, etc.). Other articles (such as Jeep Wrangler) refer to this entry as that is what they sell on their vehicles. Leaving it out is confusing. I have reverted that omission fo rthis reason. 66.248.222.34 21:16, 20 June 2007 (UTC)(UTC)

Generic Name
The article says that LSD is commonly called "posi" (and the like), but I think this may only be true in the US. In England, at least, its commonly known as E-Diff, and this may be true for most other countries in Europe also. But, somebody more knowledgeable than I might want to check this out... Krea 21:00, 28 September 2007 (UTC)

There are three types of rear ends (that I know of)...

The single track or open end - where one wheel and one wheel only, turns. How lame is that.

The common slip or limited slip style diff that puts power to both wheels, but allows for that power to vary from side to side most commonly used in turning. It eliminates one of the tires from chirping or hopping in a turn.

Then there is what I call the True Positive Traction Diff. Not to be confused with the "posi-track" used by (GM) or "trac-lok” use by (Ford). This diff does not belong on the street. It is for race cars . It has no side to side slip. The gears are set, no clutches. And if the input turns "x" times both wheels turn "y" times. — Preceding unsigned comment added by 71.230.57.118 (talk) 13:04, 22 September 2011 (UTC)

Muscle Cars
Which kind of LSDs were used on muscle cars or super cars during the 1960s and early 1970s? LSDs like Positraction, Sure Grip, Safe T Track, Traction Lok and Twin Grip are already mentioned but there is no information how they work. Are those LSDs completely locked by a piece of metal or are they more sophisticated partial locking diffs. How often they had to change tires if they were permanent 100% lockers? —Preceding unsigned comment added by 87.188.144.195 (talk) 17:13, 14 August 2008 (UTC)

Posi- positive locking differential. Like (lock right). Actual physical meshing of metal pieces, using spider gears that are spring loaded. Which will override in turns and click to allow wheel speed difference, and fully lock and mesh on straight aways allowing full power to both wheels.

Lsd- limited slip differential. uses clutches and medal plates with springs, when wheel speed differs, spider gears will override and springs will expand allowing clutches to not rub on one another to smoothen out the power from one wheel to the other. When on straight away, springs will shut, and clutches lock with metal plates allowing power to both wheels. kinda like the clutch and pressure plate in a trani. that’s what 8.8 traction locks are.

there’s also worm gear set ups like the true trac, and wave trac. which are like Lsd, but uses worm gears to do the same thing. — Preceding unsigned comment added by 71.230.57.118 (talk) 13:03, 22 September 2011 (UTC)


 * What you are calling "Posi" is a Lunchbox locker. Posi is term coined for anything that likes to leave 2 black marks instead of one.  It's a broad term.  LSDs don't unlock and lock with turns, they simply allow the clutches move.  That's why you sometimes notice chatter when turning vehicles with tight LSDs.--Dana60Cummins (talk) 16:14, 22 September 2011 (UTC)

Cone Type LSD
They don't respond to input torque. The cones are compressed by springs, not by cammed ramps on the spider carrier. Once the static friction has been overcome, that's it, there's no further resistance offered, unlike a clutch type. RB30DE (talk) 06:09, 27 September 2009 (UTC)
 * Please stop adding cone LSDs to the torque sensitive section. The cone LSD spider carrier sits in a rigid bearing surface which fits the carrier exactly, is bolted up, and is immobile relative to the spider carrier.  This is in contrast to the clutch LSD where the spider carrier rests against a cammed surface that does not match the carrier shape, is not bolted up, and moves in response to spider carrier torque, compressing the clutches.  They're quite different.


 * The clutch LSD varies its resistance to differential action in proportion to input torque. In contrast, after the cone LSD has overcome the initial stiction, it applies a bit over 80NM of torque resisting the differential action.  This resistance comes purely from the springs pressing against the cones, and doesn't vary with input torque, hence it is not torque sensitive.  (It might be called age sensitive as it declines markedly with age!) RB30DE (talk) 07:27, 5 October 2009 (UTC)


 * Who says they are all compressed only by springs and without ramps? In any case, even if you test those that "only" have springs, you find that the reaction loads in the side gears provide forces which are a function of the torque, and thus even those are "torque-sensitive".124.183.92.167 (talk) 02:33, 6 October 2009 (UTC)


 * While it is easy to imagine as a thought exercise, I am not aware of any cone LSD that has been manufactured using ramps. But perhaps they exist.  Do you know of any?  "...the reaction loads in the side gears provide forces which are a function of the torque..." - this is misleading, and if you take it that far, there's no distinction between any LSD, even a viscous LSD would also be a torque sensitive LSD by your definition.  The point is that a clutch LSD binds the wheels together harder the more torque the engine applies, while a cone diff doesn't, the engine eventually overcomes the spring, then the diff open wheels.  IOW, clutch LSD two black lines, cone LSD one black line.   The BW78 cone LSD in the photo at the top of the page is not described in the article at all.  I have the service manual for this LSD, there is absolutely no mechanism for it to respond to input torque, the cones respond to difference in speed between output shafts.  The only cone LSDs I know of would fit in the speed sensitive section.  If there are indeed cone LSDs with ramps, they would fit in the torque sensitive section.


 * I am also wondering if it is pointless to make a separation between speed sensitive and torque sensitive LSDs, instead just list LSDs by mechanism. Either way cone LSDs should not be lumped in with clutch LSDs.  They have little in common.  I suggest you start a new section entitled Cone LSDs and write about them there. RB30DE (talk) 04:26, 10 October 2009 (UTC)


 * I consider there is a point in making a separation between speed sensitive and torque sensitive LSDs as their behaviour is different, rather than by physical type, which are just alternate means to the same end. In that case, the separate category of so-called "torque-sensitive" devices should be incorporated with torque sensitive cone and clutch LSDs.138.130.145.6 (talk) 02:04, 15 October 2009 (UTC)


 * Your assertion in the article/edit history that all cone/geared LSDs are torque sensitive is just wrong. The garden variety cone LSDs simply don't increase their resistance to differential action in response to input torque.  They are to some extent a mechanical analogue of a viscous LSD, although the analogy extends only so far. RB30DE (talk) 07:16, 4 November 2009 (UTC)


 * Are 138.130.145.6, 121.218.77.253, 121.217.134.129, 124.183.92.167 etc. all the same person? Anyway, it still doesn't make any sense to mix the clutch and cone LSDs in a single topic.  The first paragraph is now meaninglessly generalised.  The rest of the paragraphs describe only a clutch LSD, they don't apply to a cone LSD at all.  The mechanisms are quite different, the behaviours are quite different, thus the cone LSD should have its own section.  I have given up reverting the inclusion of cone as it keeps getting changed back.  Also most clutch LSDs do not have preloading springs, they are usually preloaded with shims.  I have seen one preloaded with a large spring washer.  They may well exist, but it's not the norm.  RB30DE (talk) 01:06, 21 October 2009 (UTC)


 * Yes they are all the same person. Revert to Locking differential for even more history on the user. --Dana60Cummins (talk) 06:32, 28 October 2009 (UTC)

Locking Differential
Although the Carrol Smith reference about LSDs is correct for larger torques when the limit of friction is reached, it does not apply at lower torques where the larger torques go to the slower wheel, not dis-similarly to that of the locking differential.121.218.77.253 (talk) 22:19, 19 October 2009 (UTC)


 * A: When a locker is unlocked, it applies 100% of the torque to the inside wheel. No other differential does that.


 * B: "the larger torques go to the slower wheel" I do not think this is correct. What is your source?  Even an open diff does not do this.  I have sources that disagree about the bias ratio of an open diff, but I have no sources that say more torque goes to the wheel with less traction.  They either say it's equal, or that more torque goes to the outside wheel. RB30DE (talk) 03:27, 21 October 2009 (UTC)


 * A: Agreed


 * B: That description was for an LSD and does not apply to an open diff when the torques are essentially "equal".  For an LSD the torques are different when the differential is "differentialling", and when it does the larger torque always goes to the slower wheel (at that moment).  That means when turning a corner without significant wheelspin (at lower acceleration), the inner wheel is going slower so that the larger torque is there, and when that larger torque is sufficient (at larger acceleration) it will cause wheelspin at that inner wheel and that rotates faster so that the outer wheel is slower than it, and under those changed circumstances more torque goes to the outside wheel as is usually explained.  121.218.77.253 (talk) 01:17, 22 October 2009 (UTC)


 * You've repeated yourself without explanation or source. "the larger torque always goes to the slower wheel" - I think you're assuming that the inside wheel's [angular velocity x torque] equals the outside wheels [angular velocity x torque].  This is not the case.  There is not equal power distribution (which would imply more torque to the inside wheel), there is either very close to equal torque or more torque to the outside wheel.  RB30DE (talk) 07:04, 4 November 2009 (UTC)


 * B: That added description at B above should not imply that there is no transition between those conditions.  During the transition the LSD is "non-differentialling" (which means it is effectively "locked" but probably should not use that word for an LSD) and the distribution of torque depends on the "slip" at each tyre contact patch as they rotate at the same speed. 121.218.77.253 (talk) 01:34, 23 October 2009 (UTC)

Article name
This article should really be at limited-slip differential with this pagename redirecting there. "Limited" and "slip" are used together to modify "differential", and conventional grammar dictates that compound adjectives are hyphenated. See non-governmental organization, non-profit organization, etc. 68.72.10.51 (talk) 22:49, 23 November 2009 (UTC)

General organisation, Speed-sensitive,
Why is traction control, and even stability control, mentioned under Speed Sensitive > Electronic? What has that got to do with LSD's?

Why are clutch pack differentials listed under Torque Sensitive? Torque Sensitive should be a sub sub heading here, covering only, strictly, what is under Geared LSD currently. Also, I'm not sure that Torque Sensitive differentials (geared) should be classified as LSD to begin with.

Finally I feel the page could be organized a lot better, it doesn't seem clearly set out to me. More focus should be made on the "true" LSD's like clutch & cone, the finer differences between and applications of one way, 1.5 way, 2 way etc.

Regarding the "true" torque sensitive diffs, (worm gear driven), I did not think they were even touched for drifting - I can't verify this statement at all. I know that in Rally they are avoided, most much prefer 2-way mechanical diffs. —Preceding unsigned comment added by 122.104.253.247 (talk) 01:40, 3 July 2010 (UTC)