LSD.....advice please

[Russ Kellermann]

Hey chaps,

my rig is a 4.2D Pickup '07, using a Limited slip diff. However, i seem to be having a problem "locking" it up at the point i break traction, leaving only one tyre churning away in the mess. To my knowledge the Lsd should lock the rear axle at about a 3/4 revolution of spin. is this correct? i may have a problem which Nissan must address.

[David]

3/4 turn sounds about right if I think what my Safari did with major cross-axles. You could feel there is suddenly traction.

Why don't you jack up one rear wheel while in 2-wheel drive and see if you can drive it off the jack? - If you can then the LSD is working.

[Gerrit Loubser]

Russ, you might be expecting a bit too much of the LSD, but of course I don't have the details of the situations...

The Nissan LSD is quite a good friction plate type LSD (much more effective than the LSD in the back axle of an 80 Series GXL Cruiser, for instance), but it can never do what a diff lock does. The Nissan LSD works as follows: It has conical spring plates to create preload on a pack of friction plates and clutch plates, that are splined (albeit indirectly) to the two side shafts in the differential respectively. This means that the two wheels will rotate together unless there is a difference in the torques transmitted by the two side shafts that exceeds the friction grip force generated by the spring preload. If you jack up the rear wheels on your vehicle, select 2WD, N and release the park brake, then you will notice that the rear wheels rotate in the same direction. You will probably find that the something like 100 Nm of torque needs to be applied to the each rear wheel in order to be able to rotate them in opposite directions. This is due to the preload created by the conical spring plates. As the friction plates wear, the preload is gradually reduced and one would find it increasingly easy to rotate the rear wheels in opposite directions in the test described above.

In a situation where one wheel on the axle has essentially no grip (such as when the wheel has lifted in a cross axle obstacle or if it has dropped into a hole in a muddy field) the LSD will apply some torque to the opposite wheel where the grip is better. This happens, because the friction created by the preload obviously fights relative rotation between the side shafts, but also because the torque transmitted by the axle forces the side gears away from the spider gears (this due to the pressure angle on the gear teeth), thus adding to the preload on the friction pack. This effect is directly proportional to the magnitude of the preload. Now the preload can not be made too high either, because then the LSD would interfere too much with vehicle handling on high grip surfaces (e.g. tar roads) and would lead to undesireable effects like tyre chirping. Unfortunately this means that the LSD is actually quite ineffective in terms of keeping the vehicle mobile in these situations. In a non ramp loading LSD, like the one in the Cruiser mentioned above, this is the end of the story.

Fortunately, in a ramp loading LSD, like in the Patrol, there is another mechanism that enhances the load on the friction pack to improve efficiency. The cross that carries the four spider gears in a ramp loading LSD is supported in V-shaped ramps machined into two components called the pressure rings, which fit inside the differential case (which in turn is bolted to the crown wheel). The nett effect is that any torque that is transferred by the axle forces the pressure rings appart (by virtue of the cross wedging in the V-shaped ramps). The pressure rings are arranged so that this then leads to additional pressure on the friction pack, thus improving the efficiency of the LSD way beyond what is possible on a non-ramp loading type. This is because the ramp angles can be set up with a view to creating this pressure on the friction pack and these angles are not limited to the gear pressure angle (determined by gear shape). The catch is that it still is not possible to transmit much torque to an axle where one wheel is in the air. An old trick to help in such a situation is to apply the park brake (if this only acts on the rear axle with the LSD) or the service brakes (partially) to load up the axle and make the LSD "bite". The park brake trick obviously won't work on vehicles with a park brake acting directly on the driveline.

It is a common misconception that an LSD will "lock up" after a certain amount of wheel spin. This is not the case. LSDs respond to torque transmitted and not to wheel rotation angle or speed.

The sad reality is that even the most aggressive type of friction type LSD will always have rather dismal performance compared to a locking differential. The locking effectivity of an LSD is expressed in terms of a locking %, which is calculated as follows:

Locking Percentage = LP = (Th - Tl)/(Th + Tl) * 100

where:

Th = Torque transmitted to the output shaft presenting greatest resistance to rotation (best grip)
Tl = Torque transmitted to the output shaft presenting least resistance to rotation (worst grip)

The locking percentage can range from around 15% to about 45% for typical ramp loading LSDs.

By re-arranging the formula above, one can see that:

TBR = (Th / Tl) = (1 + LP/100)/(1 - LP/100)

This ratio, called the torque bias ratio (TBR) states how much more torque can be transmitted via the wheel with good traction than the wheel with poor traction. If the LP = 15%, TBR = 1.35. If LP = 45%, TBR = 2.64.

In an open differential, the torque is nominally split equally to the two output (side) shafts (LP = 0%, TBR = 1). If the one wheel is in the air and is transmitting zero torque, the other will also be able to transmit zero torque.

In a 15% LSD the one side shaft can transmit 135% of the amount of torque transmitted by the other. Even if one wheel is in the air, the preload will ensure that a small amount of torque (perhaps 100 Nm) is generated between the output shafts within the LSD so that some torque can be transmitted by the other side shaft (135 Nm). Clearly, the greater the locking percentage, the better the efficiency in the wheel lift situation.

For interest sake, if a differential is locked mechanically, the locking percentage is 100% and the TBR is equal to infinity.

There are also other classes of LSDs, e.g. the gear type or Torsen LSD. I shall not go into these much, save to say that these can have TBRs of between 2 (LP = 33%) and 6 (LP = 71%) and are not subject to performance degradation through wear as they have no friction elements. When one wheel is lifted, however, they are totally ineffective, hence hybrid versions have been developed that add a viscous coupler to the Torsen LSD to load it up in these cases.

[Russ Kellermann]

Gerrit, ....You are my hero, i think only you and my Dad could possibly be this insightful , thank you.

My vehicle has done 93 000km , which i'm sure is still considered fairly unused as far as diff friction plates and clutch plates are concerned,. or am i disillusioned?

I have read elsewhere that one can add shims to decrease the amount of load needed to act on these plates, however, as you mentioned in your second paragraph,and i agree, this will most likely render the vehicle " twitchy" on sealed surfaces, and in my opinion, unsafe.

Putting load on the rear axle by means of park break is not an option as mine works on the drive train (2007 model).

I must say tho, that i was certainly under no illusion as to the effectiveness of a LSD when compared to a pneumatic,hydraulic or electrically lockable diff. I've had two of the pneumatic locking versions (gearmax) , one in my SFA Hilux and another in my 280DT isuzu , both worked unbelievably well, ...and i miss them so much.

I am much more enlightened now as to the operating conditions of my vehicle and armed with this knowledge i can better tackle my terrain as i will have the above facts in mind. what drove the nail in was your comments about it not locking under the condition of a lifted wheel,but,will deliver traction across both wheels (not equally) under circumstances when BOTH wheels are still on terra firma, and one of those wheels still propelling the vehicle .I immediately remembered a time about a month ago when i was on loose gravel and a tight bend at low speed, i sportingly gave it a bit of right foot and as the rear started to turn and break away, i felt a surge of traction and was shot out the bend perfectly straight, and i was wondering if that was the LSD which had come into play. ? My guess now is Yes.

I will most certainly one day put a driver activated diff lock in, and maybe in front too after time, but for now that cost is just way too high as i have worked for many years to simply be able to afford this vehicle, but did so knowing that it is a perfect platform to modify and improve on over time with the addition of quality gear that work.

Thanks again for your detailed response Gerrit, i really appreciate your effort here.

Best
Russ

[Gerrit Loubser]

Thanks Russ.

My vehicle has done 93 000km , which i'm sure is still considered fairly unused as far as diff friction plates and clutch plates are concerned,. or am i disillusioned?

Russ, a lot depends on what those km were like. If it was all highway use, LSD might be in very good shape still, but if it was used in town a lot with lots of sharp turns that would have constantly forced the friction pack to slip, it might be well worn by now.

I would suggest that you do a test as follows:
* chock wheels
* select 2wd
* select neutral

* jack one rear wheel up
* using a torque wrench on the wheel nuts, measure how much torque is needed to rotate the lifted wheel.

If you measure 100 Nm or so, the LSD is probably still OK.

You could also jack one wheel up using a bottle jack and try driving off the jack, as suggested earlier, but this does not quantify the condition of the LSD. It does not take much to drive off jack, so even a pretty worn LSD should do it. Perhaps you could place a brick in front of the wheel on the ground to add some resistance and see if the LSD allows you to drive off the jack while climbing over the brick?

Putting load on the rear axle by means of park break is not an option as mine works on the drive train (2007 model).

You could try using the service brakes, although this obviously brakes all wheels, not only the rears. Still, applying the service brakes partially and driving through them loads up the drivetrain and will het the LSD to work more effectively.

Actually, it will create a similar effect in the open front diff, albeit on a much smaller scale, becuuse open diffs are only theoretically 100% open due to internal friction.

I've had two of the pneumatic locking versions (gearmax) , one in my SFA Hilux and another in my 280DT isuzu , both worked unbelievably well, ...and i miss them so much.

Yes, there is no comparison. A vehicle like the Patrol will get very far without an axle diff lock, but there comes a time and an obstacle where a diff lock is essential. Then the IFS bakkies with their diff locks will keep going...

I immediately remembered a time about a month ago when i was on loose gravel and a tight bend at low speed, i sportingly gave it a bit of right foot and as the rear started to turn and break away, i felt a surge of traction and was shot out the bend perfectly straight, and i was wondering if that was the LSD which had come into play. ? My guess now is Yes.

Yip, that sounds like the LSD doing its thing. They are quite remarkably effective in these circumstances. Our Pajero has a rear Torsen LSD (lockable Torsen, actually) and one experiences a similar thing on wet roads,

[Russ Kellermann]

Hi Gerrit, ok cool, i'm gonna give that a try when i get my vehicle back, and ill post the results.

Thanks so much man,

Enjoy your weekend.

Best
Russ

[Engel]

nice post gerrit, just to mention 100nm is the amount to torque you can apply on a normal spanner, in 4x4 terms quite useless. Stick a arb locker in there and you will have a different vehicle offroad.

[Russ Kellermann]

Without doubt, but i've heard figures of R17k so will just have to wait...so until then and

[SRXy]

Russ,

My '98 Patrol 4.2D SUV has an LSD too....when I bought the truck it wasn't working....so I started researching getting new plates etc etc....I got a hint from one of the guys on the Aussie Patrol forum that I should check the LSD oil! The oil that came out was thin and pretty much normal diff oil, so I put in Castrol LSX (which is an LSD oil) and it made a MASSIVE difference. The LSD is aggressive and should "take" when you corner a bit fast in the wet....and trust me...you can't miss it when it kicks in.

Who serviced your truck? Find out what they put in when they last changed you rear diff oil....if they can't tell you then I'd strongly suggest puting some Castrol LSX in. Take it to an open parking lot and do some tight circle turns in both directions. I had to do this a few times before my LSD suddenly woke up....go0d luck! At R250 or so it's a hell of a lot cheaper than new friction plates!

[Russ Kellermann]

Nissan serviced my vehicle, and it was a 90k service so they did the diff oils too, i'll check up on the oil they used. If its not the same, i'm gonna put your theory to the test. As mentioned earlier, it does work, but maybe not as potent as it could be.
Very interesting, thanks