Balltec LSD

[NZeder]

I have a few questions about it's application:

Was it really a factory supplied diff in the JDM Silvias?

Not sure but most are pulled out of S13 and S14 JDM Silvias there is not pasts listing on a Nissan FAST system so after market is my best guess and also given they were sold in 95+ (from designers website in the company history section) that makes them out of the S13 for a factory install.

Do the splines match 280Z/280ZX inner axle stubs?

Yes 29 spline count so all good there

Does it fit into the 280Z long nose R200's?

Well given I purchased an OBX (the first one on this site to take the punt back then too - and I still don't have the car going) that was for a S13/S14 diff and it went into my long nose R200 with just some spacers made for the M10 to M12 mounting then I will say yes to that too.

[Flexicoker]

Craig and I having been staring at the pictures for a while. Best guess at the moment is that its going to function like a salisbury.

 

Is that a needle thrust bearing in between the 2 drums?

 

Anyone have one of these we can borrow and study for a bit? =)

[cygnusx1]

I just looked for the patent for it, but I only found their patent for the ball/plate differential, United States Patent US6463830 (also interesting). I didn't find the patent for this version. I'll keep looking.

[johnc]

There is also a rotational difference limit between the two halves. With the balls and plates moving side to side when the unit is differentiating that mass acceleration/deceleration and friction limits that difference almost like a lock ratio. A tight corner under power might have the inside wheel dragging a bit.

 

Or maybe I still don't get it...

[cygnusx1]

John, it sounds like you understand it. We are not quite 100% sure how it works either, but getting closer.

[JMortensen]

The balls are loaded in one direction from torque from the ring gear. Then as the wheels differentiate the balls ride up the ramps, and I think we've determined that the torque force acts on the shallow ramps on accel and the steep ramps on decel to give a 1.5 way effect. I still think this torque on the ramps is going to try and force the drums into the case, and it won't be constant as the balls have to travel down the steep side of the ramp on one side. Maybe that is counteracted by the ball going UP the shallow ramp on the other side. That seems like it would want to cause pressure on one side then the other of the case. That's the part I'm having a hard time with, is how do these balls which are captured and must remain equidistant remain in the tracks on the shallow and steep ramps at the same time. It does seem like it would oddly limit how the drums can move, so I think I see where John is going with the dragging wheels bit.

 

One thing I think I was missing was the speed at which the action happens. In retrospect, I don't think the differentiation action is going to be very fast. Maybe one of you math geniuses could figure out the length of a 180 degree arc on the inner and outer tracks of a Z's wheelbase with a say 50 ft radius and a 300 ft radius and we could determine the number of revolutions made by the drums.

[cygnusx1]

...

One thing I think I was missing was the speed at which the action happens. ...

 

Slow. lol, Unless a tire breaks completely loose and spins wildly.

[NZeder]

Is that a needle thrust bearing in between the 2 drums?

Nothing between the two drums on the unit I pulled apart - the other one however is setup different and has a large green (look at the pic I posted in this thread) spring - and I assume some belleville springs and clutches however I don't have a need to pull this one or a bolt long enough currently to hold the two drums together when I go to put it back together.

 

Anyone have one of these we can borrow and study for a bit? =)

Sorry shipping from NZ to US at any kind of speed will cost more than a diff so that is out of the question - hell are two on ebay currently for around $299 USD which is cheaper than shipping in both direction (it is approx $75 one way for the slow 6-8 weeks method)

[cygnusx1]

OK I was putting my 2yr old to nap and I was sitting fully relaxed in the rocker with my eyes shut and envisioning the Balltec differential working. I think I finally nailed it down. We all had bits and peices of it correct but I was able to transpose all of our ideas into one final answer.

 

OK:

When the input shaft applies a torque to the ring gear, the ring gear torques the housing, which passes the torque through the balls into the two inner drums via the tracks.

 

The balls force upon the angled ramps. These forces are applied evenly all the way around. Due to the angles of the grooves, some of the force tries to wedge or spread the left and right drums apart. On each outer sides of the drums are clutch plates that drag on the ring gear housing. The drums are forced outwards, with an additional force to the internal springs, into the clutchpacks to create breakaway friction.

 

The ramps are angled steeply in the forward direction so they apply more force on the clutches with more forward torque.

 

In the reverse torque direction (engine braking), the ramps are shallower so that a factor less of force goes into spreading apart the drums. Resulting in less pressure into the clutches.

 

During neutral input torque (neutral coasting), the breakaway torque comes only from the internal springs.

 

The left and right wheels can smoothly spin opposite from eachother and that has no effect on the breakaway torque.

 

 

Breakway Torque is the sum of the torque being input to the unit causing the drums to spread apart, plus the force from the internal springs.

 

So minimum breakaway torque is the force from the springs. Maximum breakaway torque happens when the motor is making maximum + or - torque.

[JMortensen]

The ramps are angled steeply in the forward direction so they apply more force on the clutches with more forward torque.

I think you have this part backwards. It's easier to produce a greater force on the case by riding up the shallow ramps than it is riding up the steep ones. The shallower the ramp the more aggressive the lockup will be. I'm pretty sure on this bit.

 

So minimum breakaway torque is the force from the springs. Maximum breakaway torque happens when the motor is making maximum + or - torque.

I also had a thought on the clutches. I don't think they're necessary for the action of the diff at all, but are probably there to provide a replaceable wear surface that the drums can rub against.

[NZeder]

I think that would work for the first version like those shown in the Japanese websites in bits - however how does that work on the other version ie the one I pulled apart that does not have any clutches, and no spring in the centre of the drums. It just the 2 large Belleville spings at each end?

 

I assume as the action of the diff is caused by the ball and grooves so as jmortensen as stated the clutches might not play a part in the workings and were only there to protect the surfaces of the drums and housing. So if this is the case then the version I pulled apart that only has the belleville springs I assume these are only there to apply some preload to the drums and to protect the drums and housing much like the other version shows. So the next question is would these behave like a 2 way or a 1.5 way? And is the break away torque set by the design/angle of the grooves.

[JMortensen]

So the next question is would these behave like a 2 way or a 1.5 way? And is the break away torque set by the design/angle of the grooves.

You could make it a 2 way by having grooves at the same angles on both slopes, or 1.5 way by varying the slopes. I would think 1 way would be theoretically possible with a straight vertical groove, but I don't think it would function well in practice.

[cygnusx1]

I might have to get one of these just out of curiosity. No clutches? That is bizzare.

 

 

This looks like a friction surface.

 

Therefore! Maybe the balls are PULLING the drums into eachother instead of pushing them apart. The friction (clutch) is between the two drum halves.

[Diomedes]

It's a diff very similar to a cam-and-pawl unit. We used to use diffs very much like this in our FSAE car. We'd get them from some kind of Honda ATV vehicle. It's already been pointed out how the balls in the carriers slide back and fourth, allowing the two outputs to rotate opposite to each other. When torque is applied to the input, and one wheel overspeeds, the unit jams up, and puts the torque to the other wheel. They are a torque biassing diff; they behave much like a torsen. I believe this is why there are clutch packs in the sides of this model; so there is some preload if one wheel jacks.

 

 

The good thing about cam and pawl diffs is they're light. As you can see, there's not much to them.

 

The bad thing about them is they don't last long. They're steel-on-steel. This design is better than most, as it uses ball bearings instead of "pineapple pieces," so there's more rolling and less sliding, but I still wouldn't expect it to last long. Also, when they wear, they allow more slip, which causes more wear, and the spiral continues, until you end up with an open diff. I could be wrong, though, these might last, who knows?

 

 

Here's a thread on another forum that shows the ATV cam and pawl, as well as some discussion:

 

http://forums.autosport.com/showthread.php?threadid=101634

[NZeder]

I might have to get one of these just out of curiosity. No clutches? That is bizzare.

 

 

This looks like a friction surface.

 

Therefore! Maybe the balls are PULLING the drums into eachother instead of pushing them apart. The friction (clutch) is between the two drum halves.

I have not closed the unit I took apart back up as I want to replace the cap screws that hold the housing together so I will take the drum out again and see if I can get some better pics of the clutch/friction surface on the drums for you all. Once I have this unit back together I will look at pulling the other one apart so we can see the difference between two setups.