The reference to the AYC removal wan't for USDM but for everybody else in the world. The only reason I brought that up is because there can be good gains from tuning the AYC map itself.

You make really good points about the motec- it is a fine piece of equipment for sure. What we are trying to do is offer a cheaper solution that can match up to it. There are some features that the stock can't match up to - like your limp mode scenario, but it seems to match up everywhere else, minus one switchable map but it does retain AYC for folks who have it.

Motec= 4 switchable maps, limp mode, no ayc control (not a problem for usdm evo 8/9)
$1700
Stock= 3 switchable maps, no limp mode, ayc control
$500 (for 3 maps)

You are also correct about making the adjustments with instrumentation. That's what we have done to make the adjustments we have made to our maps. Yes, tail out is fun, but not always fastest. More grip and rotation is what equals speed through turns.

Matt

 

Yeah, the AYC is even a bigger thing with the X. The AYC brake (Auto brake differential) also has some parameters we can only guess at this point. It allows the AYC to artificially induce oversteer, which is a huge gain in certain situations.

Motec also has no OBD2 capability, so it's technically illegal for any emissions.

On the general subject of ACD tuning, there was a theory posted a few months ago on another thread where Malocas proposed that if you take the G force lockup reducer tables and subtract by a factor of 60 you get numbers than mean something when compared to mechanical LSD ramp angles. The ACD is emulating a conventional LSD's properties, in other words.

That might be the key to convincing more people to understand how the ACD works. Lots of people can understand how a 60 degree ramp angle effects handling for example, it's such an old technology there is tons info written about it.

60 would be aggressive and 0 would be very slow acting, it seems to make sense, thoughts?

 

If you are referring to the "A" maps, those numbers are referring to pump pressure...

 

Not to turn this about AYC, but at the very least my assumption was that the AYC is a controller, just like the ACD, which takes in parameters to tell the rear differential how to behave.. Mitsubishi's engineers programmed how the AYC reads each input variable and translates it to how and when the rear differential locks up.. the parameters for us may be unknown but it's not exactly a magical mystery.. Improvement of the AYC programming will take an ungodly amount of man hours as well as probably a PhD in physics. ie knowing that input from steering brake, yaw blah blah means the car is in a certain condition, and by adding 5% more lock up at x time will give the car a benefit if y% exit speed or blah blah.. changing the AYC programming could result in the differential doing some very strange things, good or bad. More oversteer is not necessarily a good thing, for every condition there is probably an optimal torque split between the rear wheels but good luck figuring out what that optimal is, especially considering that the condition is changing over continuous time... There's a reason there hasn't been any proliferate development of aftermarket AYC tuning.

 

The ayc map has an axis for speed and an axis for steering angle degrees. The numbers in the map that are being adjusted are labeled as 'outside wheel tq%'. There are very good gains from tuning this map as well, although you must be careful because like you said- weird things happen. This is the only map that actually needs to be tweaked rather than completely made over. The ayc just really compliments everything else very nicely.

Matt

 

The ACD computer has nothing to do with emissions capability. You can run a MoTeC MDC with a stock ECU, AEM, MoTeC M800, etc, its on a completely different set of wires.

Dave

 

Agreed.

 

AYC has tables and has been tune-able for some time, this is not my point. Your whole argument is strange considering the ACD tuning yields great results, why is the AYC any different? Should we stop tuning the ACD as well according to this logic?

The issue is the AYC in the Evo X has added functionality that does not entirely reside on the Evo X S-AWC. It looks like a special set of tables on the ASC (ABS computer) side is governing the AYC brake, which is a kind of hyper-torque vectoring aid. The old Evo's with AYC did not have this functionality. Even if we do change values for outside wheel torque % in the AYC tables we would be ignoring the AYC brake params. The question is where the tables controlling it reside, there is an outside chance the CAN between the ASC ECU and the SAWC ECU is one way and the tables are just on the SAWC, but I doubt it.

Thanks for the info on the Motec, did not know their SAWC computer worked with stock ECU.

 

I'm not sure what you mean. Are you disagreeing that it's more complex? Are you disagreeing that it's not as proliferate, no matter what country you reside in? I'm not saying there aren't gains to be had, but I will say they are not as large as that can be had in the ACD.

How much MORE torque can you push to the outside wheel when you already have enough to do a power on oversteer in? So yes, I would say there is a difference. If I were considering the GSC 40:60 center diff (which isn't available anymore), I would think similarly.

 

Right, it's 0-4v to the ACD proportioning valve. AYC has a similar proportioning valve with similar values, but with a directional valve added to switch pressure to RH or LH clutches.

The A maps are emulating a mechanical limited slip's ramp angles. That's a very good theory. If you read about LSD ramp angles you find that racers have been tuning 1, 1.5 and 2 way diffs for over a century now, and it has tons of info. It's like having a mechanical accel and decel map, really. The ACD is much more complex and much more functionality and sensor input beyond even a 1.5 way diff, but there are similarities.

Converted proportioning valve voltages for the % lockup reduction tables read:

60-60-60-60-60-58-55-54-40-40

with the -60 math trick applied. That right there explains a accel side of a limited Slip diff very nicely, albeit the ACD can change it's variables. So at low speed it acts like a 60 degree ramp angle (aggressive) and a 40 degree at higher speeds and G-forces it tapers off to less aggressive.

The pump voltage is relative to G forces:

Pump voltage 1.0-1.5-2.0-2.5-3.0-3.5-4.0
Acceleration 1.5-1.0-0.5-0.0-0.5-1.0-1.5 G force

Again, it's a theory to describe a basic set of rules for the ACD to follow. May be right or wrong, but if it is right it explains to people how it functions. Other than this theory, how do we even know where to start in terms of tuning except for radical trial and error?

 

Matt just hinted that the table is more easily tweak-able than the ACD maps. the physics side would be my strong point, I have read all JSAE papers on the subject and think I have a good understanding how the system works on a physics level. Artificially generating a polar moment is simple badass, and that's the technical term.

I am not sure if it's as effective as tuning the ACD, but I am sure there are gains to be had. Even more potential gains with the AYC brake control.

 

I meant complex in terms of the physics, but I see you've got that covered (no sarcasm). I was just thinking there could be a lot of circumstances when you make it perform worse instead of better, especially considering moment by moment changing conditions. I assumed the physics involved in improving function by a center differential, especially one that has a tune nowhere near its potential, would be a bit more straightforward.

ie, how do you know, in a certain situation, with all the various inputs, that, say giving 5% more torque to the outside wheel is better (and by better, resulting in higher exit speed) than the stock's tune? How do you know it won't result in too much oversteer? etc.. for a center differential, it seems more straightforward, which was the intent of my original comments.

 

You have a good point in that the reading of the physics in the papers raises some simple points about how precarious the torque vectoring effect is on the "positive" side of the handling. They imply that it can easily work against you.

The physics are more understandable for me on the ACD, since influencing the handling of the car through the center diff seems more straightforward to me too, the parameters are broad. Torque vectoring seems to have a very narrow sweet spot.

 

Loving this thread- I am going Motec but, my goals are Time attack and circuit. Will be a few hillclimbs and eventually ship to the states in 2013 to compete in pikes peak.

Yes weekend warrior car, serious goals I will achieve and in such an environment I think the Motec is a better choice.

 

It would be interesting to see how much performance benefit the MoTeC can result in the end over a flash like ACD-Tuning for the cost. I think the biggest benefit will have to be the custom map for a specific track and driver - otherwise everything that can be tuned by the MoTeC should be able to be tuned by ACD-Tuning, correct?