FAQ: BMW eLSD - Not your dad's open diff!

[dailowill]

This might be a stupid question, but if the e-diff is coded out, would it affect normal DTC functionality? Does it even matter? I'm also curious as to why no one has tried this before... It would seem like every enthusiast with a real diff would want this.

[brakthru]

Originally Posted by dailowill

This might be a stupid question, but if the e-diff is coded out, would it affect normal DTC functionality? Does it even matter? I'm also curious as to why no one has tried this before... It would seem like ever enthusiast with a real diff would want this.

That's my question too. I'm waiting on a response from Alex to let me know for sure if this is possible. It would be nice to keep the ediff functionality as part of DSC/DTC on mode since ediff does let you throw the car around a little without losing composure. DSC/DTC off would let you drift or spin both wheels until your heart is content however, you wont have any kind of safety net since the car won't apply brakes or cut throttle at all which is nice for the intended purpose. I'll report back once I hear what Alex has to say about it.

[dailowill]

Originally Posted by brakthru

I'll report back once I hear what Alex has to say about it.

Any chance you heard back from Alex?

[brakthru]

Originally Posted by dailowill

Any chance you heard back from Alex?

I didn't get a direct answer on that point. He did say he can code out diff lock or power reduction when brakes are hot separately. I do believe diff lock would be coded out in all traction modes since it's greatest use is in dsc off mode. I haven't decided to pull the trigger on this yet.

[The Wind Breezes]

Originally Posted by paradoxical3

It worked. HUGE shout out to Alex at Alpine for doing the actual coding and staying late to help save my track day. There is also another value he coded out directly related to the brake disc too high, reduced power via DSC command.

Would you mind sharing what this value is, so we can all code it out?

[paradoxical3]

Originally Posted by The Wind Breezes

Would you mind sharing what this value is, so we can all code it out?

E84_FLR - Engine power Reduction (prevent brake disc overheating by reducing engine power) - nicht_aktiv

If you have a non xdrive DSC the value will still be there but it might not be a simple aktiv/nicht__aktiv. If it is a wert value you will want to set it to the one that corresponds with a hex value of zero.

I have tested this on the track and it works 100%.

To disable the diff lock you want to code out AX_REF_Diff_Lock or anything with a Diff_Lock in it (varies by chassis, but they all have it).

While you are in there, code off any variant of

E84_HBA - Hydraulic Brake Assist
E84_EVB - Brake Standby (reduce brake delay by lightly applying the brakes for when they might be required)
E84_HPS - Brake Fading Compensation (maintains pedal feel by gradually increasing braking pressure)

Ignore the e84, that is because I have an X1. Instead of e84 in those values, you will have your chassis code. This gives DRASTICALLY increased braking consistency and performance on track.

I have not noticed a difference in daily street driving but I am not liable if you code these off and hurt yourself. By coding those values off you accept the risk that something could go wrong and that you accept all responsibility.

Coding those braking parameters off gained me 1+Second on my laptimes at Mid Ohio because of braking consistency and pedal feel. We don't actually have real knockback on this platform, it is the stupid computers pre tensioning and un tensioning the brakes.

[brakthru]

Originally Posted by paradoxical3

E84_FLR - Engine power Reduction (prevent brake disc overheating by reducing engine power) - nicht_aktiv

If you have a non xdrive DSC the value will still be there but it might not be a simple aktiv/nicht__aktiv. If it is a wert value you will want to set it to the one that corresponds with a hex value of zero.

I have tested this on the track and it works 100%.

To disable the diff lock you want to code out AX_REF_Diff_Lock or anything with a Diff_Lock in it (varies by chassis, but they all have it).

While you are in there, code off any variant of

E84_HBA - Hydraulic Brake Assist
E84_EVB - Brake Standby (reduce brake delay by lightly applying the brakes for when they might be required)
E84_HPS - Brake Fading Compensation (maintains pedal feel by gradually increasing braking pressure)

Ignore the e84, that is because I have an X1. Instead of e84 in those values, you will have your chassis code. This gives DRASTICALLY increased braking consistency and performance on track.

I have not noticed a difference in daily street driving but I am not liable if you code these off and hurt yourself. By coding those values off you accept the risk that something could go wrong and that you accept all responsibility.

Coding those braking parameters off gained me 1+Second on my laptimes at Mid Ohio because of braking consistency and pedal feel. We don't actually have real knockback on this platform, it is the stupid computers pre tensioning and un tensioning the brakes.

Hey,
1. After coding these features off, did you notice any traction control intervention, braking, or throttle cuts at all when at the limits of adhesion?
2.Also, are you able to spin your wheels a good deal now when launching off the line?
3. Does drifting feel more normal now with both rear wheels trying to spin equally?
Thanks.

[brakthru]

Originally Posted by brakthru

Hey,
1. After coding these features off, did you notice any traction control intervention, braking, or throttle cuts at all when at the limits of adhesion?
2.Also, are you able to spin your wheels a good deal now when launching off the line?
3. Does drifting feel more normal now with both rear wheels trying to spin equally?
Thanks.

I saw your answer on a different post from 10/31/15. Good drifting. Good to hear.

[brakthru]

Update:
After much heart ache and pain, i finally figured out how to truly code off diff lock. I found out at least in my case it was impossible to code out diff lock with the DSC software update latest file from June 2015, SP daten v56. Even though i had coded out the diff lock function in NCS Expert, there was still a subfunction of DSC still operating and not totally unlocking the wheel. What a nightmare. I did some searching and found SP daten files v41 which are from 2011. 2011 was the last software update I had where DSC worked normally. I had multiple dealer updates in 2014 and 2015 in an attempt to fix problem. After re-flashing DSC with Winkfp and encoding with NCS Expert the 2011 files my car was back! Tons of power, wheel spin and drifting on tap now. No power cuts at all and no boost reduction. No feeling of the brake holding the inside wheel! The car feels unleashed, free and normal again. I have an LSD installed as well. Also, i had to recalibrate DSC/ABS pump valves with ISTA-D. I truly hopes this help someone. It took me over a year of to figure this out. Not even BMW knew what was going on. None of the dealers understood why my car was managing power and traction even with DSC completely off.

[Bullitt]

Originally Posted by brakthru

Update:
After much heart ache and pain, i finally figured out how to truly code off diff lock. I found out at least in my case it was impossible to code out diff lock with the DSC software update latest file from June 2015, SP daten v56. Even though i had coded out the diff lock function in NCS Expert, there was still a subfunction of DSC still operating and not totally unlocking the wheel. What a nightmare. I did some searching and found SP daten files v41 which are from 2011. 2011 was the last software update I had where DSC worked normally. I had multiple dealer updates in 2014 and 2015 in an attempt to fix problem. After re-flashing DSC with Winkfp and encoding with NCS Expert the 2011 files my car was back! Tons of power, wheel spin and drifting on tap now. No power cuts at all and no boost reduction. No feeling of the brake holding the inside wheel! The car feels unleashed, free and normal again. I have an LSD installed as well. Also, i had to recalibrate DSC/ABS pump valves with ISTA-D. I truly hopes this help someone. It took me over a year of to figure this out. Not even BMW knew what was going on. None of the dealers understood why my car was managing power and traction even with DSC completely off.

Thanks for the update! V41 seems most mentioned when users report successful coding and your post confirms that.

[Motohip]

So does this apply in low traction conditions? When it was snowy and icy, I was able to get around with summer tires on. It was actually pretty amazing. But when the passenger side was stuck in the mud I had to put it and drive and reverse over and over to get out. Is this active in those situations?

[imported_mega]

elsd is friggin terrible on the track, overheats the brakes then it starts to cut your power

even worse at autocross it does not react fast enough when the elements are so short and close together

[imported_mega]

it's worse than nothing, at least ax rules now allow it to be all coded out

the ecu starts to limit torque because it knows the brakes overheat due to the ediff bullshit

search the forum you'll find threads about it

[brakthru]

I know it's been a while since my last update. About a month after i thought i conquered the ediff/adb/traction control issue, the problem returned. At the time, I was experiencing misfire issues. After replacing plugs and coils, I decided to reset adaptions and variants. For some strange reason, my car reverted back to its previous state of traction control preventing the wheels from putting the requested power down. You can only imagine how I felt having brakes intervene again and power cuts at the limit with dsc completely off. The ecu will brake any wheel necessary front or back to keep the car going in the direction of your steering inputs. Ecu controlled brake pressure is seamless sometimes but, other times could be abrupt. Either way, the vehicle is slowed down.

I back tracked and redid everything. I reflashed DSC90 data file 6776067 in winkfp and recorded with ncsexpert. I used the v41 Daten files. I repeated this process over the months also using v54 files with no success. Be cautious when updating a module with winkfp. You don't want to brick an ecu like I did. Winkfp wasn't the problem. My laptop gave me an issue right in the middle of the flash. So, I have a new dsc unit installed.

I officially don't know how to stop traction control from braking a wheel or power cuts. It seems to me the brakes are already pretensioned even when just driving down the road ready to react to a possible loss of traction. I had new stock turbos installed which gives me some wheel spin from a 1st gear launch without a problem once 15 psi of boost kicks in at 3500 rpms. Thats only because i have a jb4 and mhd. No wheel spin on stock boost. The rear brakes still limit power with braking pressure while the wheels are spinning.

The good news is I have learned how to code a little more fun in the drive. On the MK60_87.COA module, it's COA_SDR_CTC or just CTC for cornering traction control. Theres only 2 values for this. Ncsexpert shows wert 01 data 00 selected by default. I decided to add custom data value 01 to see what happens. Well, it basically turns corner traction control off. Going around corners is a lot of fun now. It allows just enough drift before you break to much traction and get into trouble. Nice controlled drifts. The car will hold the drift with braking pressure. If you push the accelerater while in the corner or drift at the limit, the car will stop applying power until your back on a safe path. Everything is happening so fast, you don't really realize to a huge degree that power was gradually reduced to assist with getting out of the corner safely.

Although the computer will help you, be careful. Even the M2 has safety features while dsc is completely off. You can through that car into corners at over 100 MPH and the ecu controls the car perfectly. Letting you have fun and making you feel awesome. BMW has done this intentionally from my research. Safe and fun.

[brakthru]

I forgot add that disabling CTC will give you a little more freedom even with DSC on.

[dzid_]

I actually read the whole thread. Ufff
Amazingly no one gave simple physical explenation why eLsd is losing more energy.

It is related to that the eLSD controls slip "to frame" whereas LSD does this to the other axle.
Energy dissipated = torque x slip.
And since there is greater slip between the axle and frame instead of between two axles, the energy dissipation in eLsd is greater.

This is why brakes will heat up more than lsd.


Also, I wonder if anyone knows, how different characteristics of the brake pads influence the eLsd ?

[jlc45]

Congratulations to the OP for maintaining such a good attitude.
I for one do not remember "sucks balls" or similar terms as being defined in any of the several physics courses I have taken! LOL.

[juld0zer]

Originally Posted by brakthru

I know it's been a while since my last update. About a month after i thought i conquered the ediff/adb/traction control issue, the problem returned. At the time, I was experiencing misfire issues. After replacing plugs and coils, I decided to reset adaptions and variants. For some strange reason, my car reverted back to its previous state of traction control preventing the wheels from putting the requested power down. You can only imagine how I felt having brakes intervene again and power cuts at the limit with dsc completely off. The ecu will brake any wheel necessary front or back to keep the car going in the direction of your steering inputs. Ecu controlled brake pressure is seamless sometimes but, other times could be abrupt. Either way, the vehicle is slowed down.

I back tracked and redid everything. I reflashed DSC90 data file 6776067 in winkfp and recorded with ncsexpert. I used the v41 Daten files. I repeated this process over the months also using v54 files with no success. Be cautious when updating a module with winkfp. You don't want to brick an ecu like I did. Winkfp wasn't the problem. My laptop gave me an issue right in the middle of the flash. So, I have a new dsc unit installed.

I officially don't know how to stop traction control from braking a wheel or power cuts. It seems to me the brakes are already pretensioned even when just driving down the road ready to react to a possible loss of traction. I had new stock turbos installed which gives me some wheel spin from a 1st gear launch without a problem once 15 psi of boost kicks in at 3500 rpms. Thats only because i have a jb4 and mhd. No wheel spin on stock boost. The rear brakes still limit power with braking pressure while the wheels are spinning.

The good news is I have learned how to code a little more fun in the drive. On the MK60_87.COA module, it's COA_SDR_CTC or just CTC for cornering traction control. Theres only 2 values for this. Ncsexpert shows wert 01 data 00 selected by default. I decided to add custom data value 01 to see what happens. Well, it basically turns corner traction control off. Going around corners is a lot of fun now. It allows just enough drift before you break to much traction and get into trouble. Nice controlled drifts. The car will hold the drift with braking pressure. If you push the accelerater while in the corner or drift at the limit, the car will stop applying power until your back on a safe path. Everything is happening so fast, you don't really realize to a huge degree that power was gradually reduced to assist with getting out of the corner safely.

Although the computer will help you, be careful. Even the M2 has safety features while dsc is completely off. You can through that car into corners at over 100 MPH and the ecu controls the car perfectly. Letting you have fun and making you feel awesome. BMW has done this intentionally from my research. Safe and fun.

I know this is an old thread but no point starting a new one.

Brakthru - Any updates on your issue? Did you eventually solve it once and for all?

I just downgraded my DSC software from 6862873 to 6789303 (hardware number 6789303 on the control unit, 6789304 on the pump motor casing) using v41 daten files. Based on 808AWD325xi's posts, it seems plausible to conclude that the base version of the software applicable to a DSC module can be determined by the numbers on its control unit casing. Correct me if i am wrong though. My car has all the latest software from BMW prior to this point.

It was a bit nervous flashing the DSC module as there isn't any specific guide but if you can flash Alpina or DCT software, the gist is the same. You have to code the DSC module to your car again, and then recalibrate it. The ignition will turn itself off about halfway thru the second stage of the flashing. I held my breath at this point.

The things i have noticed so far:
- DSC light seems to flash more frequently
- Brake is stronger and engages with less pedal travel

In the past, any changes noticed seem to wear off after about a week so i will report back with more info as it comes.

[brakthru]

Originally Posted by juld0zer

I know this is an old thread but no point starting a new one.

Brakthru - Any updates on your issue? Did you eventually solve it once and for all?

I just downgraded my DSC software from 6862873 to 6789303 (hardware number 6789303 on the control unit, 6789304 on the pump motor casing) using v41 daten files. Based on 808AWD325xi's posts, it seems plausible to conclude that the base version of the software applicable to a DSC module can be determined by the numbers on its control unit casing. Correct me if i am wrong though. My car has all the latest software from BMW prior to this point.

It was a bit nervous flashing the DSC module as there isn't any specific guide but if you can flash Alpina or DCT software, the gist is the same. You have to code the DSC module to your car again, and then recalibrate it. The ignition will turn itself off about halfway thru the second stage of the flashing. I held my breath at this point.

The things i have noticed so far:
- DSC light seems to flash more frequently
- Brake is stronger and engages with less pedal travel

In the past, any changes noticed seem to wear off after about a week so i will report back with more info as it comes.

No updates as far as removing ediff or automatic differential braking.

The car still applies some braking when it senses a possible traction loss cornering and straight ahead with DSC full off. In extreme cases ignition or timing cut. I have a tune and do drive aggressively at times. I've grown to appreciate it in most cornering situations since I can corner much faster with electronic aides than without them even with my LSD installed. My tune is strong enough to let me have some fun with wheelspin in 1st, 2nd and attempts in 3rd but braking slows it down. Ediff seems to be hard coded to ABS. It's a brilliant system but sometimes it gets in the way.

I even feel the front breaks apply to help the car follow the steering or safe path.
I also noticed a firmer brake pedal and less pedal effort to brake the car.

What I've noticed is the car ediff or traction control system adapts to the type of road surface I'm on. If it's dry, the system allows more power down but tries to adjust to what the tires can handle. If it's wet out or raining, the system will adjust power to the wheels via braking or torque reduction based on the coefficient of friction. This allows the maximum amount of power down and forward motion in contact with the road surface.

I've actually proceeded full throttle in 2nd and 3rd gear in the wet before with DSC fully disengage and not lose traction. I've turned corners which should have caused the car to drift but instead it stuck to the pavement and power was adjusted perfectly. I can intentionally lose traction if I want but the system makes it very difficult which is good on the street. I don't track my car so handling would be different there of course at the higher speeds.

On the road, the car mostly brakes the inside wheel and once the turn is complete, the wheel straightened out, there's a sudden surge of power boost which can cause loss of traction. I don't know if 'all this is unique to my car only.

Keep us posted with your findings.

[kivyee]

Originally Posted by dzid_

I actually read the whole thread. Ufff
Amazingly no one gave simple physical explenation why eLsd is losing more energy.

Heh it was almost unbearable for me reading the discussions and knowing it was almost 6 years ago, and the original posters were not there for me to shed some light on the topic .

I think your explanation about covers it - and to illustrate, imagine the following scenario where one rear wheel is on ice:
1. Ediff - brakes are applied on the wheel on ice, to transfer torque to the other wheel - lots of braking and energy dissipation
2. LSD - clutch type for simplicity sake - clutch locks to transfer torque to the other wheel - no slip between the half shafts - no energy loss.

In one case you are trying to slow the wheel that's slipping with the brakes to "allow" the other wheel to catch up, in the other case you are getting the wheel that's slipping to "spin" the other wheel to speed the other wheel up. I think it's intuitively obvious that there would be less energy loss in the latter case.

[juld0zer]

Originally Posted by brakthru

I even feel the front breaks apply to help the car follow the steering or safe path.

On the road, the car mostly brakes the inside wheel and once the turn is complete, the wheel straightened out, there's a sudden surge of power boost which can cause loss of traction. I don't know if 'all this is unique to my car only.

Keep us posted with your findings.

After downgrading the DSC software, I also now notice more DSC intervention during understeer. It is actually noticeable now when it brakes the wheels, but the light also comes on a lot more than it used to. I am not sure if the earlier software had a more strict threshould or if the system works better with the old software - perhaps the intervention before was a lot more subtle? Or none at all in the same scenarios? The car as a whole feels so much better and predictable when chucking it around.

I also agree with your observation. If the steering wheel is anywhere further off centre than 1 o'clock or 11 o'clock and you are accelerating, it feels like the throttle is being restricted and exactly like you describe as you return the steering wheel to centre, the normal acceleration is restored. It is slightly annoying and distinctly noticeable from before. I am on stock power so losing traction isn't as easy in mine.

[Poochie]

Originally Posted by juld0zer

After downgrading the DSC software, I also now notice more DSC intervention during understeer. It is actually noticeable now when it brakes the wheels, but the light also comes on a lot more than it used to. I am not sure if the earlier software had a more strict threshould or if the system works better with the old software - perhaps the intervention before was a lot more subtle? Or none at all in the same scenarios? The car as a whole feels so much better and predictable when chucking it around.

I also agree with your observation. If the steering wheel is anywhere further off centre than 1 o'clock or 11 o'clock and you are accelerating, it feels like the throttle is being restricted and exactly like you describe as you return the steering wheel to centre, the normal acceleration is restored. It is slightly annoying and distinctly noticeable from before. I am on stock power so losing traction isn't as easy in mine.

I have a 235i now, I had a 135i two years ago.. There is a noticeable difference in how much less the traction control (eLSD?) bogs down the power around turns in the 135i, as oppose to the 235i. So they did change the threshold of activation.

Also, I did a couple of burn-outs before I gave back my 135i to the leasing company and all of them left only ONE rear tire mark; I almost threw up when I saw that.. Yuck and f-u e-diff!!

My 235 has an LSD now, even though it cost 3k for the privilege..