I dont want to be 1WD anymore!

[JMortensen]

Those pics are GREAT Drax. I think it makes the whole thing easier to talk about. Thanks.

 

Also as best I can recall the side gears are free to move slightly, and would only constrained by the backlash on the helical gears (and the case itself) if not for the thrust bearings and washers.

 

So Jon you are saying the friction between case & helical gear is where the majority of the loading comes from... Fair enough, but I would think in that situation, the case would become a wear item. (it is hardened cast iron mind you) I can't see a lifetime guarentee given if the case (the main part of the Torsen diff) was a significant wear item. I know that if you want to replace the gears/bearings/shims in your torsen you can buy the parts off the shelf. If you want to replace the case, you need a new diff!

 

You've made my point beautifully. The case does wear, and the LSD does require adjustment because of it. The aggressiveness of the LSD is based on how hard the gears rub on the inside of the case. Nodular iron will wear out too, as you pointed out. As to the lifetime warranty, that wasn't offered on any of the gear driven types I used to sell (the only one I can recall off the top of my head was the 2 year warranty on Tru Traks). Maybe Quaife has come up with a different material, or maybe they just factor the occasional replacement cost into the original sale price. Maybe it's like warrantying a Tokico or KYB strut. Sure the warranty is there, but how many people are smart enough to figure out that they even have a blown strut? Even if it's worn it's still going to be better than an open diff for a LONG time. Or the car changes hands and the warranty is no longer valid, can't find the original receipt, etc.

[Drax240z]

So John C, any chance of you disassembling your Quiafe and taking some pics of the wear on the housing and the helical gears for us?

[Guest 2slo4u]

WooHooo. This thread is a long ride! I would think with the uses the original poster is considering for his zx, the 300zx lsd would be fine. I found a first gen. 300zx turbo in the salvage yard yesterday and I am wanting 2wd for my 81zxturbo. Do you guys think I can just bolt it on? The 300zx had the turbo, mas, and ecu taken already(and I needed some of those, imagine that!) Will it automatically be a limited slip if it is the turbo model? I saw no fins but it was bright aluminum and pretty clean.

 

 

 

81zxturborustybutruns

[JMortensen]

It's gotta be 87-89 turbo and not the SE model, which has a viscous LSD that won't work with your CVs. The SE models were pearl white with white wheels, so they should be easy to spot.

[Guest 2slo4u]

Thanks alot. I'll check the car when I go to get the distributor shaft from the guy who sold me the "fried" 85turbo ecu. Good thing the diff. is at a different place and I should get it for about $50. 8)

 

 

 

 

 

 

 

 

81zxturbo rustybutrunning

[Drax240z]

I was looking thought some old data I had and came across these, I thought I'd throw them into this discussion for completeness.

 

[tannji]

great thread.... but:

If you are wondering what I'd doing with the milling machine... well... don't ask.

 

Gotta ask....

[Drax240z]

Well lightening the center section, as well as drilling 3 holes through the entire section in order to mount it in a custom aluminum housing. Pretty rough going with that material, as it is surface hardened.

[260DET]

Great discussion but more gears in the Torsen/Quaife unit would explain why the oil runs hotter than in a conventional diff. The extra heat doesn't have to come from an extra high friction source such as clutches or high thrust surfaces.

[blueovalz]

I still do not understand completely the Gleason-Torsen differential action, but after reading this thread, I thought I'd throw this out. My impression of this type of action is similar to the everyday open differential with the exception in the way the gears (spider vs worm) interact with each other. In the normal, everyday, open differential, the spider gears provide the differential action, but because of their design, one gear spinning the adjacent gear offers no "real" mechanical advantage or disadvantage, thus they allow the axle to spin easily relative to each other, AND, with good traction, you have torque applied to both wheels. In the G/T differential though, my understanding is the slope of the worm gear's cut, and their interaction upon the element gears provides for enough of a mechanical "disadvantage" (under torque) that the two halves (axles) no longer spin easily relative to each other.

An example of this action is a rack gear in a typical steering rack. It is very easy to turn the pinion gear, and thus move the rack, because the ratio, and hence, the angle that the gear is cut, provides a mechanical advantage (or "gear ratio"). Now let's do the opposite. Move the rack gear in and out and you will immediately recognize it is harder to move the rack gear (and spin the pinion gear) because this previously mentioned mechanical advantage is now reversed and become a disadvantage. Kinda like a big gear turning a small gear verses a small gear turning a big gear. Only this time we are using worm gears which use the angle of the cut helical teeth to provide the advantage or disadvantage. Under load, this difference becomes more pronounced. Thus I contend (and this is the limit of my understanding) that there should be no more (or significant) wear caused by side loading than the OEM open diff has. It's basically a very complicated "open" differential using gear ratios to provide resistance instead of friction plates. The way you change the torque ratio is to change out the gears to a different angle (slope), which then changes the gears ratios, which change the mechanical "disadvanges" these gears have in turning the adjacent element gear. Being you have 6 helical cut gears, shearing the oil under high load, it's no wonder extra heat is produced. But I would think that wear on the internals should be a non-issue, at least in comparison to an open differential. BTW, I've got one in my car, and I love it. I had to change the way I drove out of the corners (set-up), but it was money well spent.

[JMortensen]

An example of this action is a rack gear in a typical steering rack. It is very easy to turn the pinion gear, and thus move the rack, because the ratio, and hence, the angle that the gear is cut, provides a mechanical advantage (or "gear ratio"). Now let's do the opposite. Move the rack gear in and out and you will immediately recognize it is harder to move the rack gear (and spin the pinion gear) because this previously mentioned mechanical advantage is now reversed and become a disadvantage. Kinda like a big gear turning a small gear verses a small gear turning a big gear. Only this time we are using worm gears which use the angle of the cut helical teeth to provide the advantage or disadvantage.

 

I had a cutaway working model of a Tru-Trak on my desk for about a month when I was selling diff parts. The funny thing about it is that there is almost NO resistance to the differential action in the helical gear diffs. For this very reason it is a very popular cheater diff with the roundy-round guys. If the rules required that they have an open diff, they would buy the Torsen because they could pass the usual inspection where they would jack up the back of the car and turn one wheel one way and the other would spin the opposite way.

 

I think I mentioned this before, but my ex-boss raced a Firebird in ITSS and he actually made a point of showing me how little resistance there was. We had the back of his car up on jackstands and he was spinning the tire with his pinky, and saying "I don't know how this thing works, but it does!"

 

When you have two helical gears sitting on each other and you start to turn one, the other will spin, but will also try to "walk" off the first one. If you held the one axle on the sample Tru-Trak I had on my desk and tried to spin the other one, you could see the gears inside "jump". They didn't want to turn on each other when one wheel had traction. They wanted to walk off each other and into the case. With no traction on either side, the gears just rotate around with very little resistance. That's why there is NO resistance to the differential action if you have no traction on either wheel. The more traction you get on both wheels, the harder it will jam those gears into the case.

 

EDIT--With a clutch type you still have to overcome the breakaway pressure even if one tire has no traction at all. This is also why my engineer friends thought that the Quaiffe was the reason that they kept breaking axles on their Midget. The one wheel would lift and instantly 100% of the power went to the tire that was off the ground, which would spin up to a pretty good speed, and when it came back down, BANG!!!

 

I understand Terry's theory, but I really don't believe that it is the correct explanation for this. I spent LOTS of time with one of these, probably the only one that was more fun to play with was the sample Detroit Locker. The gears inside are VERY easy to turn when they have no traction and when they have no preload.

 

I wish Drax's diagram said where the friction comes from. It shows that the worm gears resist turning and create the friction, but it doesn't say why...

[JMortensen]

One thing is bugging me about this now: It's that last diagram you showed Drax. Been gnawing at the back of my brain since I saw it yesterday: the Truetrac that I played with for so long has 6 long gears that are located sideways. Three on the left side, and three on the right side. They sit in between the side gears and the case, and there is no shaft holding each of the little helical gears in. You can access them from a plate on the end of the carrier.

 

The Torsens apparently have their helical gears front to back, in pairs, running on 6 "axles". I don't think that this changes anything with regards to how the thing functions as long as the axles that the helical gears are on have some slop in them. If the axles have no slop, then the gears can't move, which means they can't walk out into the case, which means that I've been describing the Truetrac and not the Torsen.

 

In doing a little research I found this Truetrac pdf:

http://www.tractech.com/docs/OEMTruetrac.pdf

 

This pdf from Tractech explains a couple things. First, I've been spelling it Tru-Trak and it is actually Truetrac. It also has a very clear picture of what goes on inside. More importantly though, here is an excerpt from Tractech's description of how it works:

 

The Truetrac can transfer up to 3.5 times more torque to the high traction wheel. This torque transfer ratio (called the bias ratio) is accomplished by using helical side gears and pinions. The bias ratio is the result of the pressure exerted by the side gears and pinions against the surface of the differential case. In certain applications, this normal gear engagement may produce temporary driver feedback.

 

I am still 95% sure that I have the right idea and I still think that this principle relates to the Quaiffe and Torsen as well, but the axles on the helical gears are making me start to doubt... damn it Terry and Drax!!! This is the kind of stupid crap that keeps me up at night!!!

 

So Drax, are the axles that the helical gears turn on "loose" so that the gears could get forced against the case? Does the case on the Torsens have a nice little hole for the worm gears to ride in and push on?

[Drax240z]

If you mean the "journal pin" for the last pics I posted, then I can't remember how they were held in. Yes the helical gears were free to slide axially on the journal pin though. The case does NOT have any kind of holes for the worm gear to ride in however.

[JMortensen]

OK, I think I've really got it figured out now. The newer Type II torsen is what Drax is showing pics of. The original Type I is different.

 

I am not an engineer, but maybe you engineers can take a look at this: http://www.sonic.net/garyg/zonc/TechnicalInformation/TorsenDifferential.html

 

What I'm getting out of this is that the helical gears slide on their axles and get driven into the case while being forced to turn with each other, which is what generates the friction in the Type II torsen diff.

 

The original Type I used a very similar design to the Truetrac, here is some info on that one: http://www.billzilla.org/diffs.htm.

 

Bottom line: They all use friction against the case to limit slip. The Type II looks to be a HUGE improvement over the Type I. The Quaife, Type I Torsen, and Truetrac all appear to function in the same basic manner from what I can tell.

 

And at the end of the day, I'd rather have a clutch style. 8)

[Sean73]

I will be using the viscous carrier out of a 240SX, 88 SS halfshafts, and the Modern Motorsports CV adapters. I'll swap the 240SX carrier into my existing open R200 (w/ 12 mm ring gear bolts).

 

The 240SX carrier can be purchased all day long on Ebay for $100(used)-$500(new). The 88 SS halfshafts snap right into the 240SX carrier and are $90 each rebuilt (Victoria British). The CV adapters are $215.

 

So, for about 600-700, I have viscous LSD with CV shafts (and new parts). I don't think that's a bad deal. I am not finished with the swap, but I did confirm on my own that the halfshafts will work in the carrier, so I take that as a green light to proceed with the install.

 

Sean

73 240ZT

[GabeRoc]

Well, i am an engineer... (not that this means I know anything in and of its self) and I do have some limited experience with gear-style LSD. The specific LSD that I worked with was torsens 'university special' a scaled version of their standard diff for application in formula SAE race cars. I was assigned the task of figuring out how it worked. After a lot of "how does this work" we came to a basic understanding and differences in different differential types.

 

So far blueovalz got closest to the understanding I have of the geared differentials. Basically it works off mechanical advantage, as levers do, but instead of giving advantage to a direct load (a force) it gives advantage to a torque (a force * distance).

 

In an open diff when a drive wheel is pushing a car down the road the torque (force*distance) of the wheel is being countered by the friction of the road against the tire times the distance from the road to the center of the wheel. So long as there is not going to be any slipping of the tire with respect to the road the torque being applied to each wheel will be basically the same. If there is slipping due to more torque than the friction of the tire can handle (ye 'ol burn out) or the friction force is reduced by lightening the load being carried by the tire (inner wheel lift in a corner) then because the torque being applied to the wheels must stay the same (in a slipping state, close to zero) and one wheel has almost zero torque being applied to it (the spinning one) the other wheel has almost no torque to drive. Summary: you get one wheel spinning and go nowhere.

 

Clutched LSD basically gives a fixed advantage to the non-free spinning wheel equal to the torque the clutches can withstand without slipping. So, for example lets say I have a clutched LSD where the clutches can hold 80ft*lbs before they slip. That means that if I lift a wheel off the ground the other wheel will have up to 80ft*lbs driving it forward. Similarly, if during a burnout the grabbing power of my tires does not differ by more than 80ft*lbs (or should I say the friction forces times the distance from the road to the center of the tire does not differ by more than 80ft*lbs) both wheels will spin at the same speed. Another way of saying this is it takes 80ft*lbs to spin one wheel with respect to the other. This also means that to some extent clutched LSD works like a solid axle in turns.

 

Now, in geared LSD you end up with a torque advantage. Some numbers I saw earlier were bias ratio (or as I have called it torque advantage) of 3.50 this mean that 3.50 *times* as much torque can be applied to one wheel as the other. This means that if you lift a wheel off the ground (a zero torque situation) the other wheel will have 3.50*0 ft*lbs of torque, 0ft*lbs, being applied to it. However, so long as you have *some* torque being applied to the less grabbing wheel the grabbier wheel will have up to 3.50 times that much torque being applied to it. However, unlike clutched LSD, there is not a torque required to spin one wheel with respect to the other. This means that the wheels can spin freely with respect to each other and still have more torque applied to one wheel than the other.

 

As for where the friction comes from for the geared LSD to create it?s bias, it isn?t a question of friction, it?s a question of mechanical advantage. Unlike clutched LSD there is no main friction surface, the normal gear friction is multiplied in the gear set to obtain the advantage.

 

As for which is better? it depends on application. If you are going to have a wheel coming off the ground under load (which means you have suspension troubles IMHO) then clutched LSD is the ticket. If you?re not then there are advantages to the open-diff like features of the geared LSD.

 

--gabe

[JMortensen]

So far blueovalz got closest to the understanding I have of the geared differentials. Basically it works off mechanical advantage, as levers do, but instead of giving advantage to a direct load (a force) it gives advantage to a torque (a force * distance).

 

I am definitely not an engineer, but I read the info from Tractech and Torsen below and I have good comprehension skills. I just don't think I'm wrong on this one. Especially since the manufacturers themselves seem to agree with me. Or am I missing something?

 

From Trachtech:

The Truetrac can transfer up to 3.5 times more torque to the high traction wheel. This torque transfer ratio (called the bias ratio) is accomplished by using helical side gears and pinions. The bias ratio is the result of the pressure exerted by the side gears and pinions against the surface of the differential case. In certain applications, this normal gear engagement may produce temporary driver feedback.

 

From Torsen:

The characteristic of torque bias is achieved in a very simple way. It is well known that frictional

forces are determined by the product of the coefficient of friction of a given surface and the normal force applied to that surface. Frictional torque, of course, is merely the application of that normal force at an effective frictional radius. All of the forces which are active within the differential are derivable from the torque which is being conveyed by the differential and the friction coefficients of surfaces within the differential. Therefore, all of the frictional forces which are generated within the differential, and all of the resulting resistant torques which oppose the transfer of torque between drive axles, are proportional to the torque being conveyed by the differential. Since the maximum difference in torque between drive axles which can be supported by friction is proportional to the combined torque of the drive axles, the maximum bias ratio remains constant with respect to changes in the combined drive axle torques.

 

In addition to providing a geared interconnection between drive axles which permits the usual

opposite relative rotation between the drive axles, the gearing also distributes forces which may be generated to resist differentiation over a large number of different surfaces within the differential. The surfaces over which the Invex gearing distributes forces are designed with different coefficients of friction and the Invex gearing is designed to distribute different loads between the surfaces. Collectively, the surfaces and the gearing are designed to distribute wear evenly over the surfaces and to control the overall amount of friction within the differential needed to achieve a desired bias ratio.

 

This section of the paper provides a mathematical representation of the basic frictional relationships within a Torsen differential which are responsible for achieving the bias characteristic between drive axles. The mathematical representation assumes that the direction of torque transfer through the differential is from a vehicle's engine to the drive axles. Figure 4 may be again referred to for identifying differential components mentioned in this section. However, specific forces and torques which are mentioned in this section are illustrated in Figure 5 in association with a schematic depiction of key differential components.

 

Engine torque

applied to the ring gear (Trg) is

substantially

equal in magnitude to the sum of

reaction torques

which are

developed at

each drive

axle (T1, T2).

The engine torque is

transferred to the drive axles through

(a) the differential housing which supports the ring gear and

 

 

( the Invex gearing which is carried within the housing and interconnects the drive axles for equal but opposite directions of relative rotation with respect to the housing.

 

Thus, in addition to transferring torque from the differential housing to the drive axles, the Invex

gearing also provides a rotational interconnection between drive axles which may be understood to function as a gear train for transferring torque between drive axles. Gear trains, of course, develop various reactions at gear meshes and mounting surfaces which generate friction opposing rotation of the train in proportion to the torque being carried by the train. Since all of the engine torque which is transferred to the drive axles is carried by the Invex gearing, reaction torque which opposes rotation of the Invex gearing is proportional to the engine torque which is transferred to the drive wheels. Thus, the transfer of torque between drive axles is also resisted in proportion to the transfer of torque between the engine and drive axles.

 

This feature enables the Torsen differential to support a torque imbalance between drive axles which contributes to the total amount of torque which can he transferred from the engine to the drive axles when the amount of torque which can be supported in one of the drive axles is limited by available traction. The major frictional interfaces which are responsible for supporting a torque difference between drive axles are listed below along with parenthetically enclosed symbols representing the coefficients of friction of the respective interfaces.

 

Side gear to element gear Invex gear meshes (µ1,)

Element gear faces to differential housing (µ2)

Side gear face to side gear face (µ3)

Side gear face to differential housing (µ4)

Typically, the largest reaction forces within the differential are side gear thrust forces (Fa1 , Fa2)

resulting from normal tooth loads (F1 , F2) acting at the side gear to element gear Invex tooth

meshes. The reaction forces are related to the normal tooth loads according to the following

equation:

 

Fa1 , Fa2 = (F1 , F2) x Cos ø x Cos ß

 

where 'ø' is the tooth normal pressure angle and 'ß' is the side gear helix angle.

 

These reaction forces are opposed by the respective frictional interfaces associated with the end

faces of the side gears and result in the generation frictional torques (Tf3 , Tf4) which oppose side gear rotation. These frictional torques are related to the reaction forces according to the following equation:

 

Tf3 , Tf4 = (Fa1 , Fa2) x (R3 , R4) x (µ3 , µ4)

 

 

where 'R3' and 'R4' are the effective friction radii at the respective side gear interfaces.

 

Frictional torques (Tf5 , Tf6) are developed at the respective interfaces between the element gears and differential housing. These frictional torques are also developed as a result of reaction forces at the side gear to element gear meshes, but the reaction forces being considered here are those which are directed along the respective axes of the element gears. These reaction forces (Fb1 , Fb2) are related to the normal tooth loads according to the following equation:

 

Fb1 , Fb2 = (F1 , F2) x Cos ø x Cos ß

 

The above-mentioned frictional torques (Tf5 , Tf6) are related to the reaction forces (Fb1 , Fb2)

according to the following equation:

 

Tf5 , Tf6 = (Fb1 , Fb2) x (R5 , R6) x (µ5 , µ6)

 

where 'R5' and 'R6' are the effective friction radii at the respective element gear to housing interfaces.

 

In addition to the frictional torques developed at the Invex gearing mounting surfaces, sliding contact between the side gears and element gears at the respective Invex gear meshes also produces frictional torques which contribute to supporting a torque division between drive axles. The respective friction forces at the Invex gear meshes may be represented as:

 

Fc , Fd = (F1 , F2) x (µ1)

 

The resulting frictional torques (Tf1 , Tf2) opposed to side gear rotation is related to mesh friction as follows:

 

Tf1 , Tf2 = (Fc , Fd) x R x Sin ø x Sin ß

 

where 'R' is the pitch radius of the respective side gears.

 

Therefore, the maximum difference torque which can be supported between drive axles is related to each of the above-equated frictional torques as follows:

 

T1 - T2 = Tf1 + Tf2 + Tf3 + Tf4 + (R / Rc) x (Tf5 + Tf6)

 

where 'Rc' is the pitch radius of the combination gears.

 

Letting the torque difference between drive axles (T1 - T2) be represented by 'Td', it follows that:

 

T1 = (Trg + Td) / 2, and

T2 = (Trg - Td) / 2

 

From this, the maximum ratio of torque which can be supported between drive axles (i.e., bias ratio) is expressed by the following proportion:

 

torque bias = T1 / T2 : 1

 

An alternative way of referring to drive axle torque distributions is by the term 'percent locking'. This term may be mathematically expressed as follows:

 

percent locking = (Td / Trg) x 100

 

Figure 6 shows the relationship between torque bias and percent locking over a range of comparable values of each.

 

S E CHOCHOLEK, BSME

Gleason Corporation, Rochester, New York, United States of America

[260DET]

The way I read it is that the Torsen's characteristics are derived from a combination of varying forces and friction involving all or some of the components, depending on the situation at any particular moment in time.

 

If friction involving the side gears against the case only was involved then this could be generated much more simply than the variety and complexity of the various components indicates.

[blueovalz]

As in the Phantom Grip?

[JMortensen]

The way I read it is that the Torsen's characteristics are derived from a combination of varying forces and friction involving all or some of the components, depending on the situation at any particular moment in time.

 

EXACTLY. This is why I said that I thought it was such a huge improvement over the type I.

 

What I'm getting out of this is that the helical gears slide on their axles and get driven into the case while being forced to turn with each other, which is what generates the friction in the Type II torsen diff.

 

HELP! PLEASE! ANYONE ELSE WHO UNDERSTANDS! Drax? Someone!

 

I'm going to try again. Torsen--Look at the yellow picture that Drax provided before while you go through my expalanation:

 

The invex gears slide on their axles. When the helical side gears turn on the invex gears, they slide until they hit the case. There is your "side gear face to differential housing". The side gear continues drive the invex gears, but now there is added friction. This is the "Side gear to element gear Invex gear meshes." Remember the bushing that Drax mentioned that sits between the side gears? That's there because the pressure of the invex gears pushing against the side of the case makes the side gears inside want to move the opposite direction, together. This is the "Side gear face to side gear face." In order to get a 1.5 way action on the diff, they've put needle bearings (which Drax also mentioned) on the outsides of the side gears, so that it won't make as much friction on decel. And that is the "Side gear face to differential housing."

 

Side gear to element gear Invex gear meshes (µ1,)

Element gear faces to differential housing (µ2)

Side gear face to side gear face (µ3)

Side gear face to differential housing (µ4)

 

The Truetrac is pretty self explanatory but if you don't get that one, read the pdf I linked to again.