The Differential Thread

[Kgolf31]

Originally Posted by justpete

Ah, OK, gotcha. Makes perfect sense. I guess turn-in is more violent in autox than on the track then? Haven't seen this before I think but maybe I have and didn't know it.

Yea, typically it is quick transitions and flicking the car to make it upset and get it to rotate.

I'll eventually track the car (and want to do Time Trials with it) so 45/90 would still be okay for track applications too.

[justpete]

Originally Posted by Kgolf31

Yea, typically it is quick transitions and flicking the car to make it upset and get it to rotate.

I'll eventually track the car (and want to do Time Trials with it) so 45/90 would still be okay for track applications too.

Oh! Now i get it. It is difficult to rotate this car in a turn and I know the shop switched to an open diff in one car because of this LSD effect. Never looked at autox cuz all those raggedy ass orange stripes down the side would suck...

[Kgolf31]

Does anyone know if OSG has a 128i application? Everything I see says 135i

[Suprgnat]

Originally Posted by Kgolf31

Does anyone know if OSG has a 128i application? Everything I see says 135i

The diffsonline web site offers one...

http://diffsonline.com/bmw-128-diffe...l#.VIkN5zHF-G8

[Kgolf31]

Originally Posted by Suprgnat

The diffsonline web site offers one...

http://diffsonline.com/bmw-128-diffe...l#.VIkN5zHF-G8

But that doesn't correlate to OSGs site or anyone else for that matter so I'm.wondering what they're doing different.

I guess this begs a call

[Suprgnat]

Originally Posted by Kgolf31

I guess this begs a call

Yep

[skyscraper_g80]

I have an OSG in my 130i, standard 3.46 ratio which is the same as 128i isn't it?

[Kgolf31]

Originally Posted by evolution42

I have an OSG in my 130i, standard 3.46 ratio which is the same as 128i isn't it?

128i manuals are 3.23

[3002 tii]

Just did the install, forget the fancy AR tool or those $8 wrenches. $38 at HF gets you 6 jumbo metric wrenches and the 50mm one is the real deal. Felt like Fred Flinstone using a giant dinosaur bone LMAO.

[Bimmer-Bob]

I traded a few emails back and forth with Jim at PG and did a bunch of reading online a while back, because I'm planning on eventually ditching the open diff. The general consensus seemed to be that a Torsen-type differential was preferable for AX (because of the understeer issue with a clutch-type) but that a clutch-type was superior on a road course.

Kind of a bummer, because I really prefer to do track days, but end up at AX a lot because it's just so much more accessible.

OP, if you stumble upon a good dual-duty set up, I may be following your example. But part of me also says to just go for the Torsen, since this is after all a street car that has part-time fun pretending to be a race car, lol. The zero maintenance of a Torsen is attractive, but if the PG diffs only need a fluid change every 35K miles, then that's not really a big deal either. My 2010 128i doesn't even have that many miles on the odometer, lol!

[katsooba]

i think MFactory is comming with a plate type diff to the 1ers
should be aweomse as a 1.5 way diff! lets say,
40%\10% lock accel\decell should be awesome for everything

[MFactory]

Maintenance schedule on Plate type LSD's do not go by mileage, but by how hard you use it i.e The more the LSD will be used (a lot in Auto-X, due to constant slip/lock), the quicker the plates will wear.

[PeteA]

Given your experience in diffs MFactory, are you able to comment on diff bushings for this application?

I know there are two frontal small bushings and one larger one in the rear. Many of us want to upgrade these somewhat to avoid wheel hop and the likes. Are there any recommendations you can give which don't incur constant gear whine on the road?

I have some some extensive research and it seems the front small bushings on the later model M3's are stiffer than the 135i ones, however the rear one is identical.

Examples :
Standard: http://www.ecstuning.com/BMW-E82-135...ntial/ES56748/
M-specification: http://www.ecstuning.com/BMW-E82-135...ial/ES1844489/

[MFactory]

Yeah, the M3 ones are slightly stiffer, but are still relatively soft and disintegrate just like the regular ones because they are not fully solid bushings.

Because of the softness of the bushings though, they absorb most of the NVH that comes from the differential. The harder the bushings, the more NVH you "may" notice, especially so if you also install a Diff Brace.

The upside though is a stiffer rear-end with increased traction/less wheel-hop.

We're actually making our own diff bushings at the moment, which will be fully solid TPV bushings (not "solid" like Aluminium so less NVH, and won't crack/disintegrate like the oem). MSRP expected to be around $150 for the set of 3

[Kgolf31]

I traded a few emails back and forth with Jim at PG and did a bunch of reading online a while back, because I'm planning on eventually ditching the open diff. The general consensus seemed to be that a Torsen-type differential was preferable for AX (because of the understeer issue with a clutch-type) but that a clutch-type was superior on a road course.

Kind of a bummer, because I really prefer to do track days, but end up at AX a lot because it's just so much more accessible.

OP, if you stumble upon a good dual-duty set up, I may be following your example. But part of me also says to just go for the Torsen, since this is after all a street car that has part-time fun pretending to be a race car, lol. The zero maintenance of a Torsen is attractive, but if the PG diffs only need a fluid change every 35K miles, then that's not really a big deal either. My 2010 128i doesn't even have that many miles on the odometer, lol!

Here is the issue I have with Torsens.

If you create wheel slip in the rear you defeat the purpose and you go to an open diff. As well you can never achieve full lock.

For example, I lift half a tire in the rear, this is enough to start spinning the wheel and thus a Torsen would not help me here:



I talked with Dan at diffs online and he was in agreement that a 30/90 setup with low static lock would be ideal. Low lock on decel (essentially the static lock) would basically eliminate all potential push.

This is most likely what I'm going to get within the next month or so once I allocate my funds

[631twentyeighteye]

Originally Posted by Kgolf31

Here is the issue I have with Torsens.

If you create wheel slip in the rear you defeat the purpose and you go to an open diff. As well you can never achieve full lock.

For example, I lift half a tire in the rear, this is enough to start spinning the wheel and thus a Torsen would not help me here:



I talked with Dan at diffs online and he was in agreement that a 30/90 setup with low static lock would be ideal. Low lock on decel (essentially the static lock) would basically eliminate all potential push.

This is most likely what I'm going to get within the next month or so once I allocate my funds

@Kgolf, Doesn't the wavetrac style Torsen diff help counter that? I would like a clutch type but thought that wavetrac would be be a middle ground, so to speak.

[Kgolf31]

It should be. However on diffsonline they actually cost more than a 3 clutch custom ramp diff.

My car will be dedicated for autox and track next year, so I'd rather go the extreme route haha

[MFactory]

You drive a BMW with Traction Control, not a 20 year old car. The no-load issue has never really been an issue (wheel lifts up, traction control applies load to the spinning wheel, LSD functions again), and just something over-talked about. On a car with no traction control (or you decide to turn it off like a lot of serious race drivers do), the same can be achieved by balancing throttle in conjunction with left-foot braking. Over-talked, over-blown out of perspective by Plate LSD manufacturers and now a certain "hybrid" LSD manufacturer, and not something that would apply to 90% of users (i.e daily driver).

There are plenty of very fast cars (with good drivers) using a Helical LSD and are faster than most drivers on the field with a Plate type.

However, you are correct. For a dedicated track car, you would want a Plate type LSD, NOT because of said issue, but because of being fully adjustable for your particular track and application. If, however, you race on several different tracks and never intend to actually re-adjust your LSD (thus taking away the sole reason for getting one in the first place), you'd be surprised how well a high-bias Helical LSD will work (and there is only 1 company that offers a high-bias ratio LSD offering upto 75% torque transfer. The rest are all 66%)

[TheSt|G]

Originally Posted by MFactory

You drive a BMW with Traction Control, not a 20 year old car. The no-load issue has never really been an issue (wheel lifts up, traction control applies load to the spinning wheel, LSD functions again), and just something over-talked about. On a car with no traction control (or you decide to turn it off like a lot of serious race drivers do), the same can be achieved by balancing throttle in conjunction with left-foot braking. Over-talked, over-blown out of perspective by Plate LSD manufacturers and now a certain "hybrid" LSD manufacturer, and not something that would apply to 90% of users (i.e daily driver).

There are plenty of very fast cars (with good drivers) using a Helical LSD and are faster than most drivers on the field with a Plate type.

However, you are correct. For a dedicated track car, you would want a Plate type LSD, NOT because of said issue, but because of being fully adjustable for your particular track and application. If, however, you race on several different tracks and never intend to actually re-adjust your LSD (thus taking away the sole reason for getting one in the first place), you'd be surprised how well a high-bias Helical LSD will work (and there is only 1 company that offers a high-bias ratio LSD offering upto 75% torque transfer. The rest are all 66%)

Very nice post.

[Kgolf31]

Originally Posted by MFactory

You drive a BMW with Traction Control, not a 20 year old car. The no-load issue has never really been an issue (wheel lifts up, traction control applies load to the spinning wheel, LSD functions again), and just something over-talked about. On a car with no traction control (or you decide to turn it off like a lot of serious race drivers do), the same can be achieved by balancing throttle in conjunction with left-foot braking. Over-talked, over-blown out of perspective by Plate LSD manufacturers and now a certain "hybrid" LSD manufacturer, and not something that would apply to 90% of users (i.e daily driver).

There are plenty of very fast cars (with good drivers) using a Helical LSD and are faster than most drivers on the field with a Plate type.

However, you are correct. For a dedicated track car, you would want a Plate type LSD, NOT because of said issue, but because of being fully adjustable for your particular track and application. If, however, you race on several different tracks and never intend to actually re-adjust your LSD (thus taking away the sole reason for getting one in the first place), you'd be surprised how well a high-bias Helical LSD will work (and there is only 1 company that offers a high-bias ratio LSD offering upto 75% torque transfer. The rest are all 66%)

I run with TC completely off. Yes the E-Diff is still active and it is very much so a reactive E-Diff, not proactive. The response time on the diff is lacking, IMO. You're trying to control the braking of the rear calipers due to wheel speed. The car would need to measure, and adapt to the changes. From my experience on specific certain conditions (aka Lincoln Nebraska at SCCA Solo Nationals) have been very disappointing versus say, Toledo Airport where the concrete conditions are drastically different.

BMWs, in general, with suspension upgrades have limited droop travel, not to mention I run a rear bar...even with the BMW gods telling me I never should. I'm only complicating 1 wheel peel situations here.

If I could tune the E-Diff to work in correlation with a Torsen Diff, it would be a no brainer. However, unless within the past couple months someone has figured out how to tune that loop, it still isn't a viable solution for me.

A clutch diff, not only to adjusting the diff you can change the whole suspension around the diff. Front and rear bars now can change drastically as well as spring rates. Trying to up rear compression to get the rear end to come around is no longer an issue.

Originally Posted by TheSt|G

Very nice post.

Stig fell in love.

[TheSt|G]

Originally Posted by Kgolf31

Stig fell in love.

Quality posts are a rarity in these parts.

You can change eDiff profiles via coding.

[DietTab]

I have not experienced any rear unloading with the mfactory unit on corner exits. remember too when it's already started to apply power to the outside wheel, there is a mechanical momentum in the assembly that will keep the diff from unloading in a split second loss of traction, you can't really compare it to static behavior. I have nothing but good things to say about it.