sba

Yes we can do this. And we can set it up to 13 bars of oil pressure.

EVOlutionary

iTrader: (38)

The full lock thing could be good for drag racing too. Full lock until over a certain MPH then go open to reduce parasitic drag.

lan_evo_mr9

iTrader: (10)

Here is a mini-review and some results from a member down in Florida. He will fill everybody in on how the car feels and the exact details of the runs as soon as his acct is reinstated.

https://www.evolutionm.net/forums/ev...-review-2.html

BluEvo210

Subscribed! I haven't been through every page of the thread yet, but I imagine someone must be doing this for the Evolution X. I'm really a RWD driver. One reason the Skyline GT-R was my favorite unattainable car was that the AWD is really RWD by default, and sends power to the front as necessary.

I can see where there's a lot to be gained, in both speed and tire life, by adjusting the center differential.

kyoo

iTrader: (29)

car will always be 50:50 though, no matter what adjustment in tuning you make to the center diff. the only way to get a different distribution is to use a different center diff. this will however, change when and how much torque is distributed

discogodfather

iTrader: (2)

Center diff is always an open 50:50 ratio, but it does change quite a bit based on the mode. This isn't because the ACD is proportioning torque, rather it is a brake and is limiting slip, but since it operates on different maps and different locking delays the center diff does vary the torque.

Problem is open diffs send power to the wheel that is slipping. Take a look at this dyno of a tarmac mode pull, the vehicle is basically 60:40 with a strong bias towards the front when accelerating. This is because the Tarmac mode is the most liberal slip governer. Looking at the same pull in Gravel and Snow it would be closer to 50:50.

Decel is going to be mostly 50:50 because of aggressive lock, and turning is going to be open so any ratio, but mostly biased towards the front (with understeer). AYC counteracts the understeer by pushing power to the outside rear wheel, generating an artificial polar moment and inducing oversteer. The AYC brake, or ABD, also does this and adds to the oversteer effect.

kyoo

iTrader: (29)

is this something you're absolutely sure about? my understanding is that it can be open, and it can be everything in between open and fully locked, can be open when steering and turn in etc, and lock up more on exit.
if the ACD is acting as a brake and just slowing down the front wheels for understeer then there wouldnt be anything gained unless there was an actual torque transfer to the rear. acting as a brake to limit understeer just helps drivers with heavy feet trying to get on the gas too early. i can see how it will help with ayc/aggressive rear diff, as front will "braked" while outside rear wheel is sent torque.

still, i'd like some clarification, you seem to be suggesting that the ACD is always open and simply limits the front wheel speed based on input

discogodfather

iTrader: (2)

The ACD is a brake that actuates on the intermediary torque transfer gears / tubes inside the T-Case. When it actuates, it clamps equally on the front diff and rear diff sides, reducing shaft speed differences.

Center diff is just open until the computer decides to lock the ACD. Open diffs send power to the wheels with the most slip. On a hard accel in Tarmac mode, the front wheels lift up and the rear wheels squat down. Power rushes to the front wheels, but the ACD map controlling the Tarmac mode allows lots of wheel slip. It doesn't actuate until the WSS difference is approaching 20%. That's why we see a 60:40 spilt in the dyno. Gravel and Snow allow less slip. This is the Accel map.

Since you can't turn with the locked diff, the ACD opens during hard cornering. It locks on decel to help braking, but it also needs to maintain slip during cornering. This is the decel and C maps.

An open diff can be anywhere, it could go 100:0 or 0:100, but in reality it never gets to this level because the physics are extreme to get it to do so. I had experience back in the day with the Original Audi Quattro (UrQ), which was a totally open system with no locking capability in the center other than a manual lock (think 4wd on a truck). It had problems on low traction surfaces but it would be nearly impossible to get the rear or front to get all the power. Most of the time, the difference in front wheel wheel speeds is not great when you start with the 50:50 ratio. It's going to be within 20% almost all the time.

If your in Tarmac mode and hitting the track with an Evo, when the car turns in it will normally understeer. Thats the feeling I get on an older USDM 8/9 and a current RA, they all have mechanical rear diffs. The X oversteers because the AYC is mitigating the power going to the front wheels by pushing power to the rear outside wheel. This has the effect of creating the yaw moment. The AYC is really a diff designed to overcome the inherent problems of the open center diff, even with the ACD, at an open 50:50 ratio. That is Mitsu's design philosophy.

Other AWD manufacturers solve this problem by torque biasing (like Subaru), varying the initial torque split to more RWD in ratios like 45 : 65. This helps because you want less power to the front in most situations including accel and turning. 50:50 is better for braking.

The ACD is a brake, the AYC is a torque vectoring diff (it has two brakes, two clutch paks) so it can send power to where it wants, not just limit slip. That's the difference between an LSD and a "Torque vectoring" diff.

It would be nice if the ACD had two clutches, but everything I have read says you don't need torque vectoring on the center diff.

kyoo

iTrader: (29)

i dont see how power "rushes" to the front wheels, since power is always 50:50 - if anything else then are you suggesting torque is sent to the front, or at least more efficiency/driveshaft deficiency?

so you're saying the center diff will slow down both sides when operating rather than actually sending torque to the rear (typically rear, since front is traveling faster). Basically what you're saying is that the ACD slows down the faster side rather than send torque to the slower side, correct?

You also seem to be suggesting that no lsd actually sends more torque to the other side, but isnt that what torsen differentials do? send torque to the wheel with more grip? Wouldnt any 1.5 way clutchtype/torque sensing limited slip differential will help the ACD function? On accel or decel torque will be sent to the wheel with more grip? With AYCs relying on programming and mechanicals relying on physics? It seems like you're suggesting that mechanical LSDs have no merit outside of limiting slip? Slowing down an inside spinning wheel rather than sending the extra torque to the outside?

(ps, goes without saying but i'm not meaning any offense in what im saying, if it was offensive to you at all - just trying to understand the car better)

I did hear about the AYC/active rear diffs overcoming an inherent weakness, but I thought what they said in regards to a front heavy car, not because of an open center diff.

discogodfather

iTrader: (2)

Power cannot be 50:50 always, that is the starting point. As soon as the front wheels turn at a larger radius than the rear (i.e. a corner) then that distribution needs to change. If it's locked, well, you chatter around the corner. Try it in a 4wd truck, it hops along because the lock doesn't allow any front to rear speed differences.

ACD does not slow down either the front or rear, it clamps on both at the same time. Look at the tech diagram of the ACD, it becomes clear when you see it. It's like a governor that, when fully clamped, or fully locked, is at 50:50. Anything inbetween limits slip between the two, but does not negate it.

So when the Evo is humming along without anything happening it's 50:50. When it turns, it throws the ratio off. When it accelerates, it throws the ratio off due to weight transfer. .

If you have a torsen then yes, it is the only mechanical diff that can theoretically torque vector, but it's highly limited by the fact that it simply measuring the resistance to turning. It's still a dumb mechanical diff in that respect.

The only plate clutch LSD I have seen that can push power to the other wheel is on a current GTR, where there are two sets of clutches (one right and one left that are disconnected) but it beats me how that works. At any rate, it ain't active so it's just physics and mechanics. The vast majority of LSD's only limit slip. 1, 1.5, or 2 way refers to their coupling characteristics under load. I've often described the Evo ACD / AYC as 3 way diffs, since it acts completely differently under any load condition and any turning, accel, decel situation (and unlocks with a handbrake, so maybe 4+ way, lol).

Power "rushes" to where ever there is slip, this is how an open diff works. Your still thinking in terms of the ACD proportioning or sending power. It does nothing of the sort, it just clamps inbetween the front and rear diff and minimizes shaft speed differences. It's just a glorified mechanical brake, and the physics are limited to when it clamps, how fast it clamps, and by how much it clamps.

The AYC can only "send" power to where is wants by braking one clutch at a time, and since the power needs to go somewhere it goes to the other side.

kyoo

iTrader: (29)

by power i mean the power sent by the engine to the wheels, not the torque that is distributed by the center diff. The power sent to the wheels will always be a 50:50 split, while the ACD will manage the torque based on conditions, and by what you say it's just acting as a brake.

Isn't that different from clamping to minimize speed difference? when it clamps wont the energy from the front, assuming front is going faster, be transferred to the rear which is going slower? Physically, that has to be the case - both wheels arent gonna go at the fast speed, arent gonna go at the slow speed, so they'll have to be somewhere in the middle, so in effect front will be going slower than what it was going and rear will be faster than what it was, effectively meaning that torque was sent to the back? In any case, "speed" will be sent to the back, correct?

In terms of torsen LSDs, doesn't the power go to where there IS grip? I mentioned the 1.5 way going along with what you were saying, in that torque will always be sent to the outside wheel (depending on the LSD I guess). This is the one I have http://www.cusco.co.jp/en/parts_prod...rs_typemz.html, judging by its ability to send me completely sideways it feels like its doing its job anyway. If you were referring to the USDM versions of the older 8/9s, their rear differentials weren't assembled properly, assuming you drove one that didn't have a TRE upgrade - so those cars have absolutely nothing going on in terms of rear differential function - not sure about RAs but I assume its the same. They werent the 1.5 way RS diffs that the JDM Evo RS's received

discogodfather

iTrader: (2)

It's all based on speed. The entire SAWC is a speed based system. It uses speed measurment rather than power or torque to make it's decisions. If I turn in hard and overcome the tire 100% rule, the front tires are going to slip with an AWD vehicle. That's why we always refer to RWD as having sharper turn in, because 100% of tire traction is available to the front tire. In AWD, especially 50:50, only half the front tires traction is available to develop a slip angle and turn the car.

So the front wheels loose traction, car understeers, and front wheels are effectively spinning faster than the rears. Some of that slip is what we want, because we need a front rear speed difference in order for the car to turn. But that's a tiny percentage compared to what we are loosing to wheel spin.

The ACD clamps and locks the front and rear driveshafts together (to the extent that the actual ACD clutches can hold) and does exactly what you described. If the front is spinning faster (slipping) and the rear is turning slower, then the ACD equalizes the difference until it locks at a full 50:50 no slip situation. The ACD cannot clamp 100% on turning or else we would not be turning very well. It does lock 100% in braking and accel conditions, but only in a straight line.

Think of it this way, the ACD is a brake and is by definition parasitic. When it clamps, it takes shaft speed differences and converts them into heat energy (inside the clutches). In exchange for that parasitic load, it creates more equality between front and rear wheel speeds. Parasitic load can reach up to 10% from some papers I have read.

The AYC is not parasitic in the sense that it allows power to go in the other direction. Sure, some power is converted into heat, but it has a path go somewhere instead of being a cancellation point.

The diagram is hard to imagine, but effectively the ACD is in between the front and rear. If you took two shafts and butted their ends together, then had a disc attached to both the ends at the point where they butt together, this is basically the layout of the ACD clutches. If we then spun both shafts and had them moving at different speeds, then clamped our fingers on the discs forcing them together, we would equalize the shafts speeds relative to the pressure we apply to the discs. Power and torque would be theoretically going from the faster one to the slower one.

Check out the diagram again, notice the arrows and the direction of power flow, the ACD is a cancellation point when locked (by that I mean power and torque have no where to go but into the clutches until speed equalization is achieved).

Torsen diffs are amazing, and they do the same thing as the AYC, but they cannot act in as sophisticated a way as the AYC or an active diff. They are limited in the torque ratio, meaning how much power they can push to either side, among other limitations based on their 1,1.5,or 2 way design. They also need power to be applied (in the form of torque) in order to even operate, since they do not have an ability to pre-bias the power going through them, they are always 50:50 as well as a starting point.

The real brain bender is that an active diff can proportion power before traction conditions are changed, meaning it can do something before problems start based on sensor data and predictive algorithms. It's like a preemptive diff, where as any mechanical diff can only operate after something has happened.

kyoo

iTrader: (29)

Right, I think that bolded part was pretty much what I meant. No doubt the greater abilities of AYC over a mechanical, but I do think the driver needs to be aware of at least a general idea of those algorithms, so that the driver can utilize the part better - ie he may do something that signals to the AYC to do one thing but he really means to do another (ie slaloms). I would almost thing the AYC programming would need to be tailored to an individual driver, but that's a different talk again.

fantrman

iTrader: (27)

Subscribed

kyoo

iTrader: (29)

bump! can't wait til the season starts - I did some light driving in the autocross mode, definitely noticed more aggressive lockup compared to the street map that I've typically been running on