Thank you for your explanation. It seems counterintuitive but like you said, there must be a reason for it to allow some wheel spin. Watching your tests I have also seen that it took some time and wheel spin in some cases to pass the obstacles.
I give credit what you have said but forgive me for not trusting Fast Lane Car. Partly due to lack of expertise, partly due to their grudge against Mitsubishi. They post videos daily and this factory-type publishing cannot be coupled with understanding the facts thoroughly. These guys can barely read even the technical specs about AWD systems from their sheets. Please, see this video for proof (Best winter cars, Part 1):
https://www.youtube.com/watch?v=NBcrFy9TFOg
If you do not want to watch the entire video just go to 10:10 (Range Rover) and 11:45 (Outlander Sport). The tall guy also said (at 12:15) that the OS/RVR has a "crappy ground clearance". Here are the ground clearances he must have compared our car with as they appeared in the same 2 episodes:
Nissan Rogue: 7.4"
Buick Regal: 5.5"
Ford Explorer Sport: 8.3"
Volvo XC60: 9.1"
Range Rover Evoque: 8.5"
Mitsubishi Outlander Sport: 8.5"
Subaru XV Crosstrek: 8.7"
Why 8.5" is crappy?
You may watch part 2 of the best winter cars from them:
https://www.youtube.com/watch?v=ipv1FFQrhr0

The same guy in the video below at 7:30 labels the 7.9" ground clearance of the Volvo V60 Cross Country "good".
https://www.youtube.com/watch?v=pkXngkzWZ1g
I do not know about you but if someone knowingly gives away false information that person has simply lost his credibility. In web publishing credibility is your #1 guide whether you take an information or reject it. These guys are very successful because people watch their videos for entertainment. Automakers support them. There is a lot of factual error in their reports though.

Thanks again for the detailed explanation.

 

I agree, they do fun tests, not real tests, but in this case this one sentence about wheel spin and moving is exactly the same with my opinion after 5 years with this car.

When it comes to time of reaction shown in my videos - that is a little bit of human factor, I did not push acceleration pedal rapidly to reach 15 km/h, so I think we should look at the time from the moment of first TC flash.
In my private opinion - it reacts faster in Auto mode, Lock allows more wheel spin.

 

This is a valuable detail, thanks for sharing it.

 

Fast Lane Car is a video version of Car and Driver. There's a reason the latter is referred to as Car and Drivel.

 

Comparison with Quattro shows that time is almost the same in Audi, but speed of wheels is different - roller test one side is the best way to see it:
https://youtu.be/j33wMZO6zI0?t=1m4s

 

Yes, your comparison with Audi shows that the AWC (not even the S-AWC in the EVO or AWD in the Ralliart) is competitive with Audi's Torsen system (just a remark for some people, who do not follow the latest changes with Audi: smaller models, that have transverse engine layout (A3/Q3/S3 and thus, Tiguan from VW), are equipped with the same Haldex system, what Volvo is using; not all Audis are equipped with the Torsen system (only those with longitudinal engine layout) , yet they are all called "quattro" by Audi's nomenclature).

BTW. You have nice custom wheels. Why don"t you take a picture and post it in the "Show your custom wheels" thread? If you could, adding specs would be nice.

How it works:
Torsen:
https://www.youtube.com/watch?v=JEiSTzK-A2A
This demonstration is cool with LEGO:
https://www.youtube.com/watch?v=osKLMzYxyyw
Haldex 6 (latest generation):
https://www.youtube.com/watch?v=CDRVTjMPK9Q

 

Done


Here is how our EMCD works:
https://www.youtube.com/watch?v=5bZoYQFL_Qg
It is not as strong as Haldex (especially IV), but much much easier to control and much simplier in construction.
BTW - your last link is about V generation, I think there is no VI yet...

 

Yes, it was a typo from my part. Sorry, the link describes the fifth generation.

Here is another video (perhaps even better demonstration), how Audi's Torsen system works (again the video says "quattro" but it only applies to longitudinally mounted engines):
https://www.youtube.com/watch?v=MjaCbpw84kw

 

No, it is not Torsen (which in fact is almost 100 years old...)
This video is about another Quattro system - crown gear center differential.

 

The more one reads the published descriptions of the different systems the more dubious they appear. Some of the statements in these description videos must be incorrect. For example, in the Torsen video, at 4:24 the narrator says that the differential locks up if slip is detected.
https://www.youtube.com/watch?v=JEiSTzK-A2A (PROPER LINK INSERTED NOW)
If this was the case we would not be able to see wheel spin in vavoom's comparison video:
https://www.youtube.com/watch?v=j33w...outu.be&t=1m4s
The way I understand that Torsen must be more like one kind of limited slip differential with a smart >1 torque bias ratio (due to the angled teeth on the gears). Understandably manufacturers do not give away all details, so it is very hard to get accurate information on how the systems work in reality.
Similarly, in our AWC system the front and rear differentials should not be a simple open diffs. If it was so the car should not pass the roller test when the rollers are on the same side. According to Mitsubishi only the S-AWC Outlander has active front differential that would feed the torque side-to side (besides EVO and Ralliart of course). Can anyone chip in with some additional info on this?

 

There is no word about Torsen, it must be wrong link

There would be no wheel spin only in case of fully locked diff, not self-locking diffs which requires time and wheel spin to lock.

I do not understand your state, our AWC would not pass roller test with Traction Control OFF, doesn't matter if rollers were on the same side or diagonally.

 

Yes, here is the correct link:
https://www.youtube.com/watch?v=JEiSTzK-A2A Now look at 4:24. If what the explanation says within one rotation of the axle the differential should lock up. It did not happen in your test.

It is not what the above video says. Viscous coupling and limited slip differentials need wheel spin, I understand. That is why I mentioned that Torsen must be a kind of limited slip differential and it does not lock up fully as the video states at 4:24.


The Torsen, Haldex, and our electromagnetically controlled clutch plate (EMCD) distributes the power between front and rear not side to side (as far as I know). The diagonal or same side roller tests should also apport torque from side to side in order to pass the test. Roller means nearly zero torque is required to spin the wheel on that axle. That also means that the same amount of torque (nearly zero) is going to the other side wheel on the same axle. You may send any torque between back and front (even with Torsen's torque bias ratio: 3 x 0=0 still) the torque without intervention will be nearly zero on the wheels that have traction (if it is an open differential, OS does not have active front or rear differentials) and the car cannot move forward. Why do we need more time for then? My suggestion is this. When you turn off ESP (watch my word, not the traction control, the electronic stability control) you actually do not turn off the traction control. This is the only way one can pass the roller tests (diagonal or the same side). In this case the traction control applies the brake on the spinning wheel(s); now more torque is needed to turn this wheel, so the same, now higher torque is going to go to the other wheel on the same axle and the car moves forward. Just a reference that the traction control does not get turned off in our car even if you deactivate the ESP (check this video at 3:44)
https://www.youtube.com/watch?v=U-ICb6Slgtw
This is the only way I can see that our car passes the diagonal and same side roller tests. Again we have no active or self locking differentials at front or rear that would be able to distribute torque from left to right (the ABS intervention due to the traction control can only let this happen)
This is why I suggested to the testers, who did the Diagonal test (see post #1 in this thread) that they should have tried for more than 5 seconds to allow the traction control to do its job. Interestingly in their test for CX-3 and CX-5 GT they kept pushing the accelerator for 16 and 14 s, respectively. See the proof yourself in the following videos between 1:46 and 2:02, and from 0:18 to 0:32, respectively.
https://www.youtube.com/watch?v=n-uAtG9gCvM
https://www.youtube.com/watch?v=hbFaox-jZ5Y
I made a respectful comment but my posts are censored and do not appear any more.
Someone, who did not pay attention would conclude that Mitsubishi failed the test and the Mazdas passed. Was this intentional or accidental by the testers, I do not wish to go into that.

 

Yes, that is wrong information in the video, Torsen never locks. In some conditions, car may be even one wheel drive...
https://www.youtube.com/watch?v=pnuOJTj9Fxg

When it comes to our AWC - my tests were always with ESP ON, in fact we don't know how TC works then. The only thing for sure is that it needs some wheel spin (~ 15 km/h).
In January I will try to do some roller tests with ESP OFF, we will see if it really goes fully on after 5 seconds, however it will not make any difference for its final result.

This kind of tests (rollers, ramps) are easy to manipulate - Mazda passes, Mitsubishi fails, X-Drive passes, Quattro fails, there are tests were Mazda wins with Subaru, total fail of Honda etc. That is business.

 

Now, finally we are on the same page again. Thanks for your input.
I would rather say that we do not know how the ESP is connected to the traction control and most likely turning ESP off does not turn the traction control off. I agree with you, it may not make any difference whether ESP is on or not. I would want to refer for this reason to your original test on ice (some four years ago). Was any difference in climbing up with ESP on or off? Watching your video again it did not seem to be.
https://www.youtube.com/watch?v=KYws-l-t0ZQ

 

There was some difference in 2WD mode, but in my opinion - this difference was made not by the brakes (I don't even know if they were braking the wheels, no audible sounds), but by engine intervention - reduce of power, rpm.
4WD mode - no difference, but as I found out years later - in this mode there is much more wheel spin allowed by the system than in 2WD mode.
LOCK mode - almost no wheel spin, so no intervention of TC.