electronic locking differential

[Norm Peterson]

Not the same. At least ABS (by itself) will stay out of the way on corner exit.

I'd hate to think that stability control would take any hints from e-diff lockup % and start shutting down power or interfering with directional control if it didn't like the %. I hope you don't think that an ESC calibration developed for (say) P-Zeros on the street would be appropriate for R-comps or even some of the better Extreme Performance level tires on the track.


Very few average drivers know that the biggest advantage of ABS comes from being able to steer while they're stomping on the brakes for all they're worth. I'm not even sure they really know enough - on a gut level as opposed to vaguely remembering that somebody mentioned it once - to keep standing on the brake pedal even though it's pulsing back at them.


Norm

Sponsored

 

[EXP Jawa]

This is the part about an e-diff that I don't like, as it would tend to teach a certain amount of sloppiness - that by its intervention you can get away with being less "smooth" with the throttle than you should be. Somehow, I just can't make that sound like a good thing.


Norm

True, and this is the case with any of the electronic chassis controls on cars. However, if you look past that and consider the possibilities that a well-thought-out system opens up for a performance car, the advantages it can give, it becomes hard to ignore. The real advantage is in being being tied to the chassis controls - the level of integration that can be achieved between traction control, ESP and eLSD can produce seamless results.

The same could be said for a lot of things - DSG transmissions, for example. People said the same thing to Jim Hall when he first started putting downforce aids on Chaparral race cars in the '60s, but now optimizing a car's aerodynamics is standard fair... I just want to point that out. I work on mechanical LSDs, so its not in my interest to have eLSDs become popular, but I can't deny what potential they offer.

 

[EXP Jawa]

Thanks for the great description of LSD types and their effects, it's posts like these that keep me coming back here....

Your description makes it clear how the M Performance LSD (a dealer/port install option for the 3 series and the M235i) does its variable lockup magic--it's purely mechanical, but it's been tuned to give a low degree of lock on corner entry and about 3X the lock on power-on after corner exit (although nowhere near 100%). Impressive engineering voodoo!

Variable lock-up, or at least different levels of lock-up between corner entry and exit, are pretty readily achievable on mLSD (by mLSD, I mean mechanical LSD, not a BMW "M" LSD) products. The overall lock-up varies anyway with torque, typically, but maintains a given ratio of tire traction from left to right before allowing slip (torque bias ratio). What you're really seeing from corner entry to exit is the difference in lock-up from the "drive" mode (on-throttle) to "coast" mode (compression braking). If coast has low lock-up, the car will rotate readily at turn in. Drive mode is tuned to give enough lock-up to prevent inside wheelspin, but not so much as to induce understeer. As I said before, the amount needed to do this can vary quite a bit from one chassis to another or from one type of track to another.

Salisbury-type clutch plate diffs could achieve this by adjusting the ramp angles on the plate carriers; 1-way, 1.5-way & 2-way refer to exactly this - and describe the relative drive vs. coast bias levels. The naming convention isn't real precise, but the product is well understood. Helical gear differentials like Quaife & Torsen were traditionally tougher to fine tune. Once the gear geometry was set, there was little that could be done to change it, and typical drive and coast bias levels were slaved together. We've made some design improvements in the last few years that allow relatively infinite tuning options, and have decoupled the drive from the coast mode. That allows a customer like Ford to evaluate a matrix of combinations and determine what suits the chassis tuning best for the intended use. We're actually at risk of offering too many choices and muddying the waters...

But anyway, I didn't mean this to become a lecture on mLSD characteristics, however, since I wrote it already, I'll post it. But to address your comments earlier, yes, mLSD products can also offer a lot of lock-up levels and tuning options. Its probably fair to say that they can do most of what an eLSD system can with a lot less complexity, and if the chassis controls are calibrated to be optimized around the mLSD, the net result can be very similar.

 

[Norm Peterson]

True, and this is the case with any of the electronic chassis controls on cars. However, if you look past that and consider the possibilities that a well-thought-out system opens up for a performance car, the advantages it can give, it becomes hard to ignore.

I'm really glad that development still continues with mechanical solutions. Too many people think that getting a computer involved can only make things better.

I really do grasp the concept that working individually with each tire's contribution can result in better performance, at least in terms of measured numerical stats.

What I'm not at all ready to do is grant "co-driver" status to electronics that are programmed with somebody else's preferences, prejudices, and rather conservative assumptions regarding driver skill and capabilities. I need to have some tactile "feel" of what's happening, and if that "feel" is artificially altered it's like my car is lying to me. I don't like my car (or my passengers) suddenly throwing surprises at me when I'm driving, so no thanks. FWIW, I can't reliably keep any car in an electronic game on the track even down the straights, because the feedback I'm looking for just isn't there.

Stuff I can't possibly keep up with like fueling and spark control are the ideal things to leave to the computers. For things that I can keep up with in real time, computers aren't nearly as necessary or even as desirable.


Norm

 

[EXP Jawa]

Oh, I'm with you Norm, but as I said, my stance on the matter is somewhat "biased"...

I've got a vested interest in continuing the development of mLSD systems. These days, there are a certain portion of the market that feel doing so is akin to continuing to develop the buggy whip, but there really is demand for the product. Like you say, sometimes putting a computer in charge can yield improvement on paper, but a lot of people that are the target buyers like tried and true mechanical systems. Obviously, cars like the C7 seem to want to demand state-of-the-art tech, but at the same time, GM still came to us for the diff in the Z/28, just as Ford had done with the Boss 302. And that came from known success with the T-2R product in the aftermarket over the last decade.

I know that at one point in time, an eLSD was on the table for the GT500, and even looked likely. However, the cost and complexity of integrating and calibrating drove them to use the helical diff from the Boss instead. Now, I doubt that eLSD would find it easy to get back in since the current models have done well.

 

[Norm Peterson]

These days, there are a certain portion of the market that feel doing so is akin to continuing to develop the buggy whip

I know, mainly digitally-oriented types who'd rather forget that nature, engineering, and mechanical car parts and handling behavior are really analog. Just because a task can be accomplished with electronic involvement does not inherently mean that it should be accomplished that way, and I guess they missed the memo.


Norm

 

[lostwon]

Sport mode....

I hope you don't think that an ESC calibration developed for (say) P-Zeros on the street would be appropriate for R-comps or even some of the better Extreme Performance level tires on the track. Norm

I do think, from experience, that Hoosier A6s work pretty well off the corners in Sport mode on my very stock 2012 Mustang GT in auto-x, Norm; for what it's worth. I've tried it both with the nanny on and with it off, and (as I'm not much more than an average driver) I feel it's better (faster) than me trying not to overpower the tires. Definitely feel the throttle intervention, but timing slips show the nanny is consistently faster. Even with fresh 295/30-19 A6s, and fresh diff clutch pack, the Coyote will easily break the tires loose, especially around tight turns like you see in auto-x. I have no recent experience in tracking the Mustang, so that would, I assume, be a whole different ball of wax...

 

[Norm Peterson]

I do think, from experience, that Hoosier A6s work pretty well off the corners in Sport mode on my very stock 2012 Mustang GT in auto-x,

Key words boldfaced. I'm assuming that means you still have the OE tune (virtually all of the off-the-shelf aftermarket tunes are even worse about this). Know that that has been problematic as far as fine throttle modulation is concerned. In that case, the nannies are covering more for unacceptably high throttle tip-in response rather than for your driving.

I used to autocross a fair amount at the regional level, and for one car I had to dial back the early throttle opening rate on a completely mechanical system to solve essentially the same issue. A pre-nanny car that you couldn't Band-Aid by leaving any of them on.


Norm

 

[EXP Jawa]

I assume, be a whole different ball of wax...

A whole differential ball of wax... Sorry, bad driveline humor.

But this also does point out that no differential is better than the tires. If in this situation the electronic throttle makes it hard to modulate at tip in, and the tires (even R-compound) are easily overwhelmed, then there is relatively little an eLSD can do to help. It might be able to be locked on command from the traction control, but if grip on both drive wheels has already been lost, then it doesn't help much if the diff is playing catch-up.

With a mechanical biasing diff, there's a chance that it'll be able to progressively load the tires, and potentially prevent break away by putting more load to the outside as the inside tire starts to loose traction. But if, as Norm says, the throttle isn't fine enough to prevent an abrupt torque application, then even that will overwhelm the tires.

 

[lostwon]

New Fangled Diffs

Key words boldfaced. I'm assuming that means you still have the OE tune (virtually all of the off-the-shelf aftermarket tunes are even worse about this). Know that that has been problematic as far as fine throttle modulation is concerned. In that case, the nannies are covering more for unacceptably high throttle tip-in response rather than for your driving.

I used to autocross a fair amount at the regional level, and for one car I had to dial back the early throttle opening rate on a completely mechanical system to solve essentially the same issue. A pre-nanny car that you couldn't Band-Aid by leaving any of them on.


Norm

Yup, OE 'tune'. "Fine throttle modulation" not a part of my Mustang's auto-x arsenal, unless the nanny's on. Just assumed it was mostly me that was the problem. Never have been a big fan of the "fly by wire" throttle and steering assist in cars. Spoiled, I guess. Works pretty good in late model fighter jets, but I suspect Ford didn't spend quite as much money on car systems. Kind of like HUDs; they need to be focused at infinity, and in your line of sight to be effective, not at the base of the windshield. Legal liability problems, I guess. Don't think any of the F-18 Hornet drivers ever sued the government....

As usual, many good points made Norm. Thanks.

John

Sponsored