[kdiggz]

i don't have that tune and my car will probably be long gone by the time this gets figured out. hopefully whatever the solution is will work on the f30's system as well

[ajsalida]

Originally Posted by DannyZRC

An open diff ALWAYS, ALWAYS, ALWAYS, gives equal TORQUE to both outputs. ALWAYS. Torque is one thing, Power is another.

An open diff is a path of least resistance for power, and an equal distributor for torque.

the example of the spinning wheel on ice, it takes very little torque to spin the wheel on the ice. Because the open diff ALWAYS, ALWAYS, ALWAYS provides equal torque to both outputs, the wheel that has traction is receiving the same torque as the wheel spinning on ice, which is very little. However, the wheel that is not spinning is receiving now power, while the wheel on ice is receiving lots of power.

When an eLSD brakes a spinning wheel, it is increasing the torque required to turn it, and thus also increasing the torque provided to the non-slipping wheel. This reduces the power flowing to the spinning wheel, and increases the power flowing to the gripping wheel.

Torque, Power, 2 different things.

There is nothing disingenuous about stating that a locked transfer clutch can transfer 100% of the torque if it can do so. Some transfer clutches don't have the torque capacity to do this, and will slip under such high loads, others won't. I don't know the torque capacity of the clutch used in the xdrive systems in question. Don't lose sight of the reason for the system in the first place, which is to get the power to the ground effectively. If the rear is not slipping, why drive the front? If the rear IS slipping, then the front can get up to 100% of the available power.

As to saying it doesn't have a center diff, this has both up and downsides. A center diff allows the front and rear to both receive torque constantly, irrespective of wheelspeed difference, but this also means that now instead of a diff spilling power on a wheel on an axle, you can have a whole axle slipping. an AWD system with 3 open diffs and no slip limiting would lose forward drive if ANY of the 4 wheels lost traction. xdrive can't have this happen, since the nature of the clutch coupling in the transfer case is to limit slip. Even without eLSD, xdrive with it's open diffs would need to lose traction on a minimum of one wheel on each axle for it to lose forward drive, if either axle has both wheels driving the car will have drive.

I accept your explanation of TQ vs power for an open diff. The point however is an open diff spins the wheel with least traction. We don't want that to happen in our lame xDrive open diffs. We'd rather have a real LSD instead of e-LSD, but for now we are stuck with it. Unless we swap an qaiffe or wavetrac, and there is some question as to how the electronics will interpret that.

The TC cannot partially or totally disengage the rear drive shaft. Ever. Period. End of story. If you wish to spin this into 100% TQ to the front under very limited special conditions (perhaps like a welded center diff would, which is why I brought that up) be my guest. It cannot do so under any conditions performance driving sees, and was never intended to so so. How strong the TC clutch is has nothing to do with that fact. Even BMW states in fine print it is only "instantaneously" when both rear wheels have zero traction (glare ice) and fronts have some. Marketing has picked this tidbit up and has tried to camouflage the lack of a center diff, or spin it into a "feature" as you are doing.

It is clear you are very knowledgeable about these systems. Please continue and tell us how to tweak this system as is. Not how great it may already be, because it isn't. It is an adequate, well-done but simple, add-on system on an already great car. The subject of 100% TQ or not to front wheels has been beaten to death in these forums. I am sure everyone here is as tired of it as I am.

Now please, let us return to the subject TUNING the xDrive AWD system, which is where I thought this thread was going.

[Pseudo Nim]

Originally Posted by ajsalida

<snip>

Now please, let us return to the subject TUNING the xDrive AWD system, which is where I thought this thread was going.

hold on... i thought we more or less determined there's a great deal of wishful thinking involved here, and then we went on to talk about dogs

[ajsalida]

Originally Posted by Pseudo Nim

hold on... i thought we more or less determined there's a great deal of wishful thinking involved here, and then we went on to talk about dogs

Ha ha yeah. I for one would like to know if it is possible to do any of the following:

1) Box or module to disable AWD in favor of RWD only while keeping all 4 wheels' ABS and rear only e-LSD.

2) Box or module to TURN OFF e-LSD when a mechanical LSD was added.

3) Box or module to tweak parameters of existing xDrive (i.e when it engages e-LSD and/or TC logic) like say for different diameter tires F/R or more performance oriented behavior.

4) Box or module to lock center TC clutches (for snow ice conditions).

Unfortunately my guess is that the xDrive/ABS logic is all deep into the programming and hard to get to. But I don't know and that is what I hoped would be discussed here.

[DannyZRC]

4 paw drive, accept no substitutes.

Originally Posted by ajsalida

I accept your explanation of TQ vs power for an open diff. The point however is an open diff spins the wheel with least traction. We don't want that to happen in our lame xDrive open diffs. We'd rather have a real LSD instead of e-LSD, but for now we are stuck with it. Unless we swap an qaiffe or wavetrac, and there is some question as to how the electronics will interpret that.

The TC cannot partially or totally disengage the rear drive shaft. Ever. Period. End of story. If you wish to spin this into 100% TQ to the front under very limited special conditions (perhaps like a welded center diff would, which is why I brought that up) be my guest. It cannot do so under any conditions performance driving sees, and was never intended to so so. How strong the TC clutch is has nothing to do with that fact. Even BMW states in fine print it is only "instantaneously" when both rear wheels have zero traction (glare ice) and fronts have some. Marketing has picked this tidbit up and has tried to camouflage the lack of a center diff, or spin it into a "feature" as you are doing.

It is clear you are very knowledgeable about these systems. Please continue and tell us how to tweak this system as is. Not how great it may already be, because it isn't. It is an adequate, well-done but simple, add-on system on an already great car. The subject of 100% TQ or not to front wheels has been beaten to death in these forums. I am sure everyone here is as tired of it as I am.

Now please, let us return to the subject TUNING the xDrive AWD system, which is where I thought this thread was going.

Regarding the 100% front topic, I thought of a good analogy that hopefully will help is to paint the power as food, and that the rears get to eat first. This isn't 100% accurate, but "if the rears aren't hungry, the fronts will get all the food."

remember though also that there are many other things the xdrive can't do, like poach eggs or apply nail polish, which are equally useful as powering the front wheels 100% while the rears have traction.

About eLSD interacting with an aftermarket mechanical LSD, let's just consider how the systems involved interact.

The transmission powers the diffs, the diffs provide equal torque to the axles but allow the power to flow freely however it will.

When a wheel slips, it's because the applied torque overcame the available traction. When this slip occurs, the wheel speed difference across the axle rises, and all of the power is flowing to the slipping wheel, meaning it no longer powers the car's acceleration and is instead wasted. The computer detects this and applies a braking torque to the slipping wheel, this arrests the outflow of power and allows the wheel across the axle with traction to receive torque and power and resume accelerating the car.

If you replace the open diff in this scenario with a quaife or a wavetrac, what will happen? Both are torque sensitive automatic torque biasing differentials. (the wavetrac is also speed sensitive, I believe. I can't say I fully understand it)

Well, firstly the tires are less likely to lose traction from drive forces, because the torque distribution is now unequal between the axles in favor of the axle with more traction at a ratio equal to the torque biasing ratio. So, the likelihood of an eLSD intervention event falls.

Second, the brake corrections would be more effective due to acting on a limited slip diff instead of an open diff, so should be shorter duration and less severe.

There isn't really a lot of opportunity for excessively strange interaction, especially since the corrections generally seem to be very conservative as delivered.

(saw new posts while writing this):

Originally Posted by ajsalida

Ha ha yeah. I for one would like to know if it is possible to do any of the following:

1) Box or module to disable AWD in favor of RWD only while keeping all 4 wheels' ABS and rear only e-LSD.

2) Box or module to TURN OFF e-LSD when a mechanical LSD was added.

3) Box or module to tweak parameters of existing xDrive (i.e when it engages e-LSD and/or TC logic) like say for different diameter tires F/R or more performance oriented behavior.

4) Box or module to lock center TC clutches (for snow ice conditions).

Unfortunately my guess is that the xDrive/ABS logic is all deep into the programming and hard to get to. But I don't know and that is what I hoped would be discussed here.

#1&4 are more attainable, #2&3 are tougher. the rear only eLSD component of #1 is pointless because without drive it'll never activate anyway. Also, #3, running square tires would remain highly recommended. Remember that fat rears are just promoting understeer, and nobody wants to cook the clutches in their transfer case with excess slip under load.

[ajsalida]

Danny,
Actually you raised a good question earlier about how much power the TC could transfer, I wonder if better/stronger clutches would have an effect on higher power applications, or if they'd also require new control logic to exploit.

The non square thing was more for certain folks here who insist on running beyond the 1% roling diameter F/R threshold, and it might save their TC's. The rear only e-LSD Q was for after the front diff has been disconnected (as opposed to disabling all traction control), not sure what your response was to.

These are good discussions BTW.

[DannyZRC]

People running wrong sized tires are just doing the wrong thing, period and stuff.

my response regarding disabling front eLSD, expanded : If you run the car in RWD mode, where the TC clutch is never engaging, then there will be no drive forces on the front axle to cause slip, which means that the eLSD will never fire because it will never have any slip to limit.

[ajsalida]

OK so let me refine some of this from a product wish list standpoint. Like various stages of tunes, but for the xDrive system

Stage 1: all electronics, features could be

a) more aggressive performance-oriented parameters
b) locking TC case capability (within mechanical tolerances)
c) full open TC (basically RWD) capability with rear e-LSD & RWD DTC/DSC & all ABS still intact.

Stage 2

add true mechanical LSD
a) rear
b) front?
c) electronics to reflect/optimize

Stage 3
Upgraded TC components and logic to exploit.

As I said, wishful thinking while I am waiting for my ETS intercooler to arrive. Once the engine tuning side (finally) gets into adding fuel rails and over 600 HP, I am sure the xDrive crowd will be looking into some of this.

[ajsalida]

Originally Posted by DannyZRC

People running wrong sized tires are just doing the wrong thing, period and stuff.

Agreed.

my response regarding disabling front eLSD, expanded : If you run the car in RWD mode, where the TC clutch is never engaging, then there will be no drive forces on the front axle to cause slip, which means that the eLSD will never fire because it will never have any slip to limit.

No I meant e-LSD on the rear axle. Right now people just pull a fuse and disable everything. Would be nice to be able to have fully functional RWD car, with ABS, eLSD, DTC/DSC.

edit: and of course no codes or flashing warnings.

[DannyZRC]

I think the stage system may be a little misplaced, but I'll go with it.

Stage 1: add on box that allows RWD or LOCK, intercepts TC Clutch control and actuates it directly, would add a 3 pos switch into the cabin for RWD/AWD/LOCK, and probably make the LOCK button a momentary push, lights up an LED for lock, and then disengages above some speed like 15-20mph and reverts to AWD.

Stage 2: Box + a rear LSD

Stage 3 (by far the most work): Retuned AWD implementation either by recoding factory electronics or by piggyback computation, possibly still with inclusion of user input devices to select behavior of AWD system, along with tailored algorithms to account for presence/absense of LSD on either axle.

[DannyZRC]

Originally Posted by ajsalida

Agreed.



No I meant e-LSD on the rear axle. Right now people just pull a fuse and disable everything. Would be nice to be able to have fully functional RWD car, with ABS, eLSD, DTC/DSC.

edit: and of course no codes or flashing warnings.

If you set the TC clutch functioning to 0 (with no codes etc), the front and rear eLSD algorithms will be functional but the eLSD on the front will never fire, the rear will continue to act as normal (and probably fire more frequently due to more slippage being generated. This operating mode may prove undesirable depending on the factory eLSD logic).

[ajsalida]

Originally Posted by DannyZRC

I think the stage system may be a little misplaced, but I'll go with it.

Stage 1: add on box that allows RWD or LOCK, intercepts TC Clutch control and actuates it directly, would add a 3 pos switch into the cabin for RWD/AWD/LOCK, and probably make the LOCK button a momentary push, lights up an LED for lock, and then disengages above some speed like 15-20mph and reverts to AWD.

Stage 2: Box + a rear LSD

Stage 3 (by far the most work): Retuned AWD implementation either by recoding factory electronics or by piggyback computation, possibly still with inclusion of user input devices to select behavior of AWD system, along with tailored algorithms to account for presence/absense of LSD on either axle.

That sounds absolutely awesome!

[Doyle]

I agree and I feel like Danny's final "stage" list is the most probable.

I asked Harold @ HPA if he did anything special with the eLSD and it's interaction with the Quaife they have in their car. He said they got no weird issues, and it didn't sound like he though it would be an issue for us.

Additionally, I attempted to see if the rear wheel only work with the parking brake on, but all of the surfaces I tried provided too much traction. If someone has jack stands, they might be able to elevate the rear and try it out...

[Pseudo Nim]

Originally Posted by Doyle

<snip>

Additionally, I attempted to see if the rear wheel only work with the parking break on, but all of the surfaces I tried provided too much traction. If someone has jack stands, they might be able to elevate the rear and try it out...

ok come on, you KNOW i have to link this.

[Doyle]

Hahaha, true, true.

[ajsalida]

If a vendor came out with a product like Danny has laid out, they'd probably sell more than they could produce. Even soccer moms in X3's would want stage one.

[Doyle]

Yup. That is why I'm trying to get some vendors in on this.

I was doing some thinking back to my Sophmore year engineering classes, and I think that our way in is going to be through the servo motor (part no 13?):

http://bmwfans.info/parts/catalog/E9..._case_atc_300/

If it is a basic servo, then it simply uses an electric current (from part no 19?) through a potentiometer to set the % of clutch lockup. If we could bypass this loop and throw our own potentiometer inbetween the signal originator and the potentiometer, we might be able to switch from anywhere between open circuit (100% RWD) and closed circuit (total lockup). This would work just like an electric guitar's tone/volume knob. You could also have it as a push/pull pot, where pushed in is normal loop/bypass (normal AWD performance) and pulled out allows gives you full control over the lockup. It would be really trick to include a digital display that reads the current percent (.00-1.00).

Thoughts?

[Doyle]

Brilliant!

[DannyZRC]

it might be PPM or PWM instead of just a voltage, but you're right that controlling the clutch is the easy part.

It's doing it and making sure the car's various electronic systems don't freak out that's potentially challenging. How potentially challenging? won't know till someone tries.

[Doyle]

Did some more looking into the servo.

Here is an interesting read about the e83 servo. This is similar to ours, but due to packaging, it is different. However, it looks like there is a single plug that receives voltage that determines the position of the gears.

http://www.bimmerfest.com/forums/sho...d.php?t=566880

Also, here is a picture of our servo motor and a good BS-free read about how the transfer case and servo motor operate.

Danny, help me out with this one. The TCU receives signal from wheel speed sensors, steering angle, ABS, and yaw/lateral accel sensors. The TCU then communicates it's "decision" via metered voltage from a power source into the servo. So, if we were to intercept this signal surreptiously before it gets to the servo there should be no codes thrown, correct? For example, in our controller circuit, there would be one path that the TCU "sees" that acts as a dummy servo. The other path would be our potentiometer that truly controls the position of the servo. Would it matter whether it is PPM or PWM? I admit that I am way out of my element with those.

[DannyZRC]

no, voltage/pwm/ppm is just a detail of implementation.

the document linked is for the X3, but it's marketing pablum so who knows.

Example : my trooper's SOTF system wasn't working, but it was broken in such a way as to be invisible to the fault-finding apparatus, however when I engaged 4wd (which was only driving the rear wheels), and I spun the rears, it would error-code and disable the ABS system, since there was disagreement in the wheel speed sensors vs the transfer case output speed sensors.

So, until people start playing with it, nobody will know how BMW is accomplishing fault finding in the system, and whether or not it will be a surmountable obstacle or not.

For example, if you operate in RWD mode and you overpower the rears in a burnout or drift or other such goofoff, and the computer tries to send power to the front but the signal never gets there since the TCase is being manually controlled, then it may progressively increase the signal sent to the front until it thinks it's got maximum lockup in the TCase, and there will still be wheel speed disagreement with what it thinks should be happening.

It may just shrug it's shoulders, or it may throw a fit. Can't be known.

[DannyZRC]

also, extremely disappointing that they would use a plastic gear in this application. It's bad enough that pop up headlights used to have plastic gears to wear out (which you would then flip around and wear down again), why on earth would you stick one into the drivetrain. bad bad bad.