V2 Diff rebuild + Gear Ratio Swap (4.10)

[sapote]

YYes, the V1 GKN unit by design allows the whole clutch stack and gears to "float" inside the carrier, so the there is preload from the free play shim, and the big bevel washer on the clutch stack, in the V2 unit the gears are locked in place, so the free play shim basically just jams the bottom spider and side gears together which will wear them sooner.

In a nutshell the factory the gears of both V1 and V2 are floating. When you say the gears are locked in place in V2, are you referring to factory setup or RD setup? Keep in mind that the clutch stack never puts any load on the gears.

[sapote]

I tried two of these kits. Each time, the diff locked up fine but then it would quickly start not locking. Still had preload at the wheels. Really odd. The V1 kits work great.

V2: Did you torque the ring to its original position (locked with a steel ball)? Doing the way that RD shows in their video is the wrong way, that they used the lock ring to preload the breakaway at 100Nm. The V2 factory used shims to achieve the spec preload, not spring washer and the lock ring.
Btw, watching RD video that uses the threaded ring to set the breakaway torque, which had the ring stopped before the it got to the locking ball, and this means the carrier right side bearing preload is much higher than factory when using the same stock bearing shim. Too many flaws in the way RD did.

[sapote]

Another data point, the original stack height is ~14.2mm, this is on a god knows what mileage, but at least I know is not completely shot as it still has 65NM of breakaway torque.

There are some shims on top of the clutch stack. Did you include the shims in your measurement?

[bigjae46]

Should last assuming the friction discs are the same as what I got with the 4 clutch ZF kit. It has been about 4 years and still works.

I did 1 V2 in 2022 and 1 in 2023 so not sure if I got the updated kit. Tightening the ring definitely increases the preload at the wheel. IIRC, we were on the edge of the backlash spec after tightening the ring. That's the PIA. Really need a jig to check the breakaway torque before putting the ring gear on.

Oh...removing the ring gear bolts isn't too bad with a big impact. Just have someone with strong wrists hold it and blast away.

[maupineda]

Yes, the V1 GKN unit by design allows the whole clutch stack and gears to "float" inside the carrier, so the there is preload from the free play shim, and the big bevel washer on the clutch stack, in the V2 unit the gears are locked in place, so the free play shim basically just jams the bottom spider and side gears together which will wear them sooner. also, the V2 unit has from the factory 8 clutches and 9 steel plates, RD kit initially was like OE, however, they redesign the kits and now both variants get 5 clutches and 6 steel plates, however, the V2 requires a taller stack. it will be cool for those that have opened a V1 to document the stack height. but for sure is less than V2s due to the lesser number of plates.

Once I open my unit I will be able to measure the stack height of the OE clutches and compare to the RD kit.

That said, my plan is to not use any of the bevel washers, I will measure the lash and if it is the recommended value (.15 to 0.2mm) from the factory, I will leave it, otherwise, I will shim it. for the clutches, I will replace only the clutches.

This is why I did not even think about using the top collar to adjust preload, as that changes the effective unit length, which would require new output shaft shims and adjust bearing preload, so all that I plan to keep to factory specs.

I only hope the quality of the friction material is close to OE, if so, then the rebuild should last long enough.

[bigjae46]

Mine coupe at 55k miles (90k kms) is very clunky. The one I am rebuilding is a spare of known condition so will do a full rebuild. May rebuild the OE one and sell it.

i also will document what I will do for clutches and spider gear lash.

The ring is left hand thread...found that out the hard way.

I tried two of these kits. Each time, the diff locked up fine but then it would quickly start not locking. Still had preload at the wheels. Really odd. The V1 kits work great.

[sapote]

The edge around the ball is tiny (1mm), you would need a specialty bit, and a proper setup and then how would you reinsert it in place? Either tap the same location, or around the barrel.

I see. A circular grinder bit (maybe custom made) would cut the 1mm lip. I will ask my friend machinist for ideas.

[0-60motorsports]

At this point im just going to get the wavetrac LSD:

https://www.ecstuning.com/b-wavetrac-parts/limited-slip-differential/30309165wkwvt/

[maupineda]

I have 5 V2 diff to rebuild this spring, all my diffs have worn LSD or whine...

Mine coupe at 55k miles (90k kms) is very clunky. The one I am rebuilding is a spare of known condition so will do a full rebuild. May rebuild the OE one and sell it.

i also will document what I will do for clutches and spider gear lash.

[maupineda]

Do you plan to replace the output flange bearings? For reuse I would tape off the bearings during drilling to keep the metal debris off.

On the locking ball drilling: they placed the ball in the hole then peen around. I wonder if it's easier to clean the "peen" lip and get the ball out vs drilling the whole ball.

Hi Sapote, no, I will replace all bearings, clutches, and all seals. So I did not care about debris, but yes, like you say, otherwise, taping and a proper cleaning will be a must.

i don’t think it would be DIY’able. The edge around the ball is tiny (1mm), you would need a specialty bit, and a proper setup and then how would you reinsert it in place? Either tap the same location, or around the barrel.

[Gt4]

I have 5 V2 diff to rebuild this spring, all my diffs have worn LSD or whine...

[sapote]

Do you plan to replace the output flange bearings? For reuse I would tape off the bearings during drilling to keep the metal debris off.

On the locking ball drilling: they placed the ball in the hole then peen around. I wonder if it's easier to clean the "peen" lip and get the ball out vs drilling the whole ball.

[maupineda]

After many months I finally started the rebuild. It will still take me a while since I do this on weekends and this being the first time, it does take a lot.

i started with drilling out the locking ball. It was very time consuming as I wanted to be as delicate as possible as I am not a fan of whacky jobs.

I first used a dremel with a small carbide bit to create a concave indentation, otherwise, the drill would just slip sideways



This is how I left it before taking it to a drill press



Once at the drill, I used a 5mm drill bit to complete the removal of the locking ball as well as drilling a hole and be able to tap it for an M6 fastener. I wanted to do this so I can close the diff after the rebuild. you can see the steel ball is solid, so it took a bit of time a patience.



Lastly, I tapped the hole, this allows me to install a M6 set screw. I drilled 10mm deep. The set screw is 6mm in length only as I wanted it to under flush to the diff carrier.




next, I will work on removing the ring gear and finally open the diff. This is resulting to be very tricky as the unit is heavy and I have not found a way to clamp the diff so I can undo the ring gear bolts and the locking collar. Will need to figure that out.

[sapote]

Another data point, the original stack height is ~14.2mm, this is on a god knows what mileage, but at least I know is not completely shot as it still has 65NM of breakaway torque.

Good piece of data. I wonder what is the thickness of the brand new stack - more than 15mm? The pump must have an internal spring to provide the breakaway torque on a used worn stack, as there is no spring washer used on the original diff.

[maupineda]

Another data point, the original stack height is ~14.2mm, this is on a god knows what mileage, but at least I know is not completely shot as it still has 65NM of breakaway torque.


I tried measuring the steel plates and clutches and seem the are 0.8 each