Big LSD Shootout... OS Giken, Drexler, Titan, GKN, ZF...

[BanjoPaterson]

Thank you very much Driftflo for taking the time to answer my questions. My car's done about 50,000 miles (82,000km) and i had the diff oil changed earlier this year. So far so good, with only the very occasional clunk going around a corner.

But I will put aside some funds to change the diff when the car gets to 100,000-120000km. I may not use the car to its best potential, but I like to know that I could if I took the time to lean how.

[redpriest]

Originally Posted by driftflo

i guess you have never driven a proper clutch type lsd... but if you are happy then don't try it! saves you a lot of money

people who only drive a 335i are happy with it. the first time they try a m3 they will know what they missed so far.

i always thought a m3 is a fast car with good straight line pace (until i drove a c63), good cornering ability (until i drove a gt3), pretty emotional (until i drove and fell in love with a 488gtb).
you know what i want to say?! there is always something better out there (that costs even more money). if you like and enjoy your m3 in the current config, stay with it. if you want to improve it, a lsd-change is a good point to go for!

Oh no - don't take me the wrong way, I'm going to change the diff to a proper one as described in this thread as I gradually convert this car to full time track use, it's just that the one I have isn't malfunctioning yet.

[bigjae1976]

Hit the jackpot and found an E24 M6 diff...its a 210mm ZF with 3.91 gears.

My car is going to be a track car - mostly HPDE and some time trials. My driving style is early apex and lots of throttle steering. I had a 3 clutch 30deg ramp angle diff built my Metric Mechanic in my 330 which was just about perfect.

I want something that will be quick but smooth. So would a good setup be a 4 clutch ZF with 45/45 ramps and zero pre-load? Or should I want some preload to promote a smooth, quick lockup?

Thoughts?

[driftflo]

Originally Posted by bigjae1976

Hit the jackpot and found an E24 M6 diff...its a 210mm ZF with 3.91 gears.

My car is going to be a track car - mostly HPDE and some time trials. My driving style is early apex and lots of throttle steering. I had a 3 clutch 30deg ramp angle diff built my Metric Mechanic in my 330 which was just about perfect.

I want something that will be quick but smooth. So would a good setup be a 4 clutch ZF with 45/45 ramps and zero pre-load? Or should I want some preload to promote a smooth, quick lockup?

Thoughts?

if you already liked the 30° 3-cp version you will love the 45° 4-cp!
for even smoother engagement use the drilled clutches.
i would definitely add a thin cupspring for several reasons, especially to stretch the rebuild intervalls and get less extreme throttle on/off reactions.

[bigjae1976]

Originally Posted by driftflo

if you already liked the 30° 3-cp version you will love the 45° 4-cp!
for even smoother engagement use the drilled clutches.
i would definitely add a thin cupspring for several reasons, especially to stretch the rebuild intervalls and get less extreme throttle on/off reactions.

Thanks!

[Hujan]

driftflo

I sent you a PM about ordering an LSD from you. Please let me know if you did not receive it! Thanks!

[driftflo]

yes, i just answered all pending pn messages...

[Move Over]

Sent you a PM regarding diffs. Thanks for all your help.

[driftflo]

reply is out!

[Move Over]

BIG shoutout to Florian. He took 45 minutes out of his day this morning to talk to me on WhatsApp, he really helped me gain an understanding of how all of this works on the street and around a racetrack.

That really says a lot as some of these American diff companies just told me i couldn't afford what they were selling after i told them my budget.

Florian saved me about 800 dollars all said and done over an even lower performing LSD here in the states from a well known company. Won't say who but i'm sure you can use your imagination!

I do have an E46 m3, not e90 anymore, but i can be sure the unit will be coming with me if i ever go back to the E9x platform. Will do a track day review with the new unit once I install it! Thanks again Florian, really hope your US audience takes off, i will do what I can to help.

[dogbone]

Originally Posted by Move Over

BIG shoutout to Florian. He took 45 minutes out of his day this morning to talk to me on WhatsApp, he really helped me gain an understanding of how all of this works on the street and around a racetrack.

That really says a lot as some of these American diff companies just told me i couldn't afford what they were selling after i told them my budget.

Florian saved me about 800 dollars all said and done over an even lower performing LSD here in the states from a well known company. Won't say who but i'm sure you can use your imagination!

I do have an E46 m3, not e90 anymore, but i can be sure the unit will be coming with me if i ever go back to the E9x platform. Will do a track day review with the new unit once I install it! Thanks again Florian, really hope your US audience takes off, i will do what I can to help.

Florian is a super knowledgable guy. And very friendly. He started me down the path to understanding the world of diffs. When I started talking to him, I knew nothing about diffs.

I don't know anything about the E46 M3 platform, but if the CAN bus on the E46 M3 carries data of each individual wheel speed like it does on the E9x M3, an AIM Solo DL tied into the CAN bus can record the speed info from all 4 wheels individually. In the AIM software, you can overlay charts of the rear wheel speeds and see how their speeds compare. By looking at the wheel speed charts, you can get a sense of how the diff is acting. If you add charts of the gas and brake pedals, then you can really dig in and see how the diff performs based on your inputs. It's really nice to be able to see hard data on how a diff is performing.

I did a stupidly long 3-part write up analyzing diff behavior on my E90 M3 using my AIM Solo DL at the link below. I compared my stock diff with my aftermarket clutch-type LSD unit. It's a realllllly long read, but it has lots of pretty charts! If you want to learn more about diff behavior and what to look for, I think it's got some stuff in there to get you going. When it comes to diffs, people like Florian are rare. There wasn't much info readily available out there specifically regarding the E9x M3. So, I did that write up to put some basic info out there. (I credited Florian at the end of the write up.)

http://www.m3post.com/forums/showthr...1158445&page=8 (post #173, 176 and 177 (page 9))

[bigjae1976]

Originally Posted by dogbone

Florian is a super knowledgable guy. And very friendly. He started me down the path to understanding the world of diffs. When I started talking to him, I knew nothing about diffs.

I don't know anything about the E46 M3 platform, but if the CAN bus on the E46 M3 carries data of each individual wheel speed like it does on the E9x M3, an AIM Solo DL tied into the CAN bus can record the speed info from all 4 wheels individually. In the AIM software, you can overlay charts of the rear wheel speeds and see how their speeds compare. By looking at the wheel speed charts, you can get a sense of how the diff is acting. If you add charts of the gas and brake pedals, then you can really dig in and see how the diff performs based on your inputs. It's really nice to be able to see hard data on how a diff is performing.

I did a stupidly long 3-part write up analyzing diff behavior on my E90 M3 using my AIM Solo DL at the link below. I compared my stock diff with my aftermarket clutch-type LSD unit. It's a realllllly long read, but it has lots of pretty charts! If you want to learn more about diff behavior and what to look for, I think it's got some stuff in there to get you going. When it comes to diffs, people like Florian are rare. There wasn't much info readily available out there specifically regarding the E9x M3. So, I did that write up to put some basic info out there. (I credited Florian at the end of the write up.)

http://www.m3post.com/forums/showthr...445&page=8 (post #173, 176 and 177 (page 9))

You did a long post? I'm shocked!

Lol...that's good info. I'd be interested to see wheel speed data.

[Move Over]

BigJae bets me $20 bucks I?m gonna spin my first day out with the new diff lol

[dogbone]

Originally Posted by bigjae1976

You did a long post? I'm shocked!

Lol...that's good info. I'd be interested to see wheel speed data.

haha! You got me! That's ok. I know I'm long winded. I just hope there are some useful tidbits in my notes for people.

[CSBM5]

Originally Posted by dogbone

haha! You got me! That's ok. I know I'm long winded. I just hope there are some useful tidbits in my notes for people.

Yes, thanks for those posts as they were enlightening. I've never seen LSD diffs compared like that, so it sparked a good bit of pondering...

On the subject of the GKN Visco-Lok for second: Does it not potentially provide deceleration locking as a function of differing axle speeds? I always understood it did, and the documentation from GKN discusses bi-directional action. Of course given its design, it doesn't "seem likely" that it would often be significantly active on decel, but it appears it can be if rotation rates are enough different? I do know one thing though, it is really sweet having it open on low speed corner entry as it makes the M3 eager to turn-in followed by a very responsive tail to the throttle; in my E39 M5 with its OEM traditional BMW equal ramp angle Salisbury LSD, I sense it being too tight on decel in those conditions creating understeer and clumsy turn-in.

[dogbone]

Originally Posted by CSBM5

Yes, thanks for those posts as they were enlightening. I've never seen LSD diffs compared like that, so it sparked a good bit of pondering...

On the subject of the GKN Visco-Lok for second: Does it not potentially provide deceleration locking as a function of differing axle speeds? I always understood it did, and the documentation from GKN discusses bi-directional action. Of course given its design, it doesn't "seem likely" that it would often be significantly active on decel, but it appears it can be if rotation rates are enough different? I do know one thing though, it is really sweet having it open on low speed corner entry as it makes the M3 eager to turn-in followed by a very responsive tail to the throttle; in my E39 M5 with its OEM traditional BMW equal ramp angle Salisbury LSD, I sense it being too tight on decel in those conditions creating understeer and clumsy turn-in.

Yeah, there's not much hard data out there if you're trying to compare diff performance. Glad it got you pondering! That's exactly why I did that write up!

I'll leave it to other more knowledgable people to say what the stock diff can and cannot do. Sadly, there is a LOT of mystery that shrouds this topic. And a lot of plain-old wrong info. (I was always told the stock diff does not lock on decel....who knows if that's accurate info....not me. I didn't do any research into the stock unit.)

For sure, I can tell you that the OSG locks much faster on acceleration than the stock unit. But what is very clear in the charts, is that the diff is always open when entering turns. The inside wheels have no trouble moving freely on corner entry.

I said before, one of the most obvious initial changes in feel with the OSG 1.5 diff was initial braking stability. And now that I’m comfortable with the OSG unit, I virtually never send the car sideways accidentally. And if the car does go sideways, the combo of a smooth diff, solid bushings and a capable suspension makes for a predictable slide characteristic that’s easy to work with.

[driftflo]

Originally Posted by CSBM5

On the subject of the GKN Visco-Lok for second: Does it not potentially provide deceleration locking as a function of differing axle speeds? I always understood it did, and the documentation from GKN discusses bi-directional action. Of course given its design, it doesn't "seem likely" that it would often be significantly active on decel, but it appears it can be if rotation rates are enough different?

this is correct. the gkn viscolok is capable of locking on deceleration as well, it always locks when the wheels rotate at different speeds. but in what real driving situations would it be the case that on deceleration/braking one wheel turns significantly different to the other wheel over a certain amount of time. if you do have this situation the car is already lost and off track... so in real life driving conditions i cannot think of a situation where the lockup on decel can happen and will assist you in anything.

Originally Posted by CSBM5

I do know one thing though, it is really sweet having it open on low speed corner entry as it makes the M3 eager to turn-in followed by a very responsive tail to the throttle; in my E39 M5 with its OEM traditional BMW equal ramp angle Salisbury LSD, I sense it being too tight on decel in those conditions creating understeer and clumsy turn-in.

if a car does not turn in with a 25% stock lsd unit the problem is not the lockup value, the ramps or the 2-way setup but the preload that was silly high on the particular m5 e39 stock lsd setup!

[RomanK]

driftflo

It seems that you good know m3 diffs... I have a clunk from my rear subframe when I push/release the throttle on the high speed. It feels like a driveline backlash. I think it from the left flange of diff, because all others things of driveline I already changed/checked. Did you know how I can fix my clunk? Or I need to just change my gkn viscolok to any other?

[driftflo]

yes, it is the worn lsd and the short flange.
switch the lsd and flanges and the clunk should be gone!

[RomanK]

Originally Posted by driftflo

yes, it is the worn lsd and the short flange.
switch the lsd and flanges and the clunk should be gone!

Which probability that it can be on the almost new diffirential?
My differential almost new. It has 40k km... But I drove my car with this clunk already 20-30 k km as I remember...

Can you see these videos and tell me exactly that this clunk because of my differential:

Here on the 38sec the clunks what I am talking about:

I have solid subframe and differential bushings.

Thank you very much for your help...

[driftflo]

40k or less is not usual but also not completely uncommon.
plus solid bushings make things much worse!

i would still bet the root of the problem is the lsd! i had several customers with just this problem and in all cases the problem was gone after changing the lsd.

[RomanK]

Originally Posted by driftflo

40k or less is not usual but also not completely uncommon.
plus solid bushings make things much worse!

i would still bet the root of the problem is the lsd! i had several customers with just this problem and in all cases the problem was gone after changing the lsd.

Ok, thanks.

I think I will change my LSD in nearest future... So just another question...

I just want to ask about 2-way lsd... How it perform on the snow/ice road? I mean... i think there are will be a lot of unwanted drifting with released throttle? Or not?

For example if I choose your 45/45 with 50Nm preload...

I would like a 2-way lsd, but a little bit doubt.