AWD Testing and How it Works

One of the most nebulous topics for the MS6 since it came out was how the AWD system worked. How it transfers power to the rear wheels, how much it can transfer, and when it transfers are mostly unknown to the masses. Let's try to clear some of that up, and let's do some testing to see how much we can really transfer to the rear.

I'll just say it up front - The MS6 can go 100% rear drive.


The general layout of the drivetrain looks like this. As you can see, there's only two points of potential slip in the system.

1) The clutch - operated by your left foot. This effectively becomes direct drive once you engage it.
2) The rear diff coupler - full of black magic.



Note that the PTO/Transfer Case is NOT a place of slip (until it explodes). The tcase is splined directly to the front differential case through a third set of splines:





This is what it looks like with everything pulled from the trans:




Now that you know the transfer case is splined directly to the front diff housing, you should realize that you can remove the front axles entirely, and it could be 100% rear drive. That also means that if you hit an ice patch with just the front wheels (0 traction, let's say), that the rear could get 100% of power being produced if they had traction.

In the next post, we'll talk about the rear diff coupler.

 

Diff coupler is a hot topic, but let's take a look at the system from transfer case to rear diff output first.

We know the transfer case is all direct drive through gears. It's important to understand there's a gear ratio in the tcase, and that it's a 1:3 (2.93, actually) ratio of input to output. That means the driveshaft gets spun at 3x the speed of the input shaft. That also means we have a torque reduction by a factor of 3...



At the rear diff, though (after the coupler, we'll come back to that) we have a gear reduction back in order to match the front wheel speed - That same 3:1 ratio, just in reverse. We've also just multiplied our torque by 3.


So I really wanted to know how much torque the rear diff coupler could withstand, because there's a lot of broken drivetrain parts out there, and I think the coupler's operation has a lot to do with it. I pulled one of my extras apart and extracted the diff coupler, input flange, and pinion (splines to diff coupler output).

The diff coupler is really just like a motorcycle wet clutch (multi disc) that is squeezed be an electromagnetic solenoid when it is energized. Here's a cutaway, but I don't think it's a great representation:



The idea is that the solenoid will squish the clutch plates together when you apply voltage to it. Pretty simple. The factory BCM does this by PWM of a 12V signal to the solenoid. Fine, but let's see how much this thing can transmit:






My testing method skips the PWM, and applies static voltage to the coupler. I'm using a variable DC power supply, and I started at 0, working my way up until it slipped or I broke something. I went up in 0.25V increments to see the behavior. This went through a couple modifications because it took a ridiculous amount of torque. I broke the input side stud somewhere around 450lbft, so I modified the input flange on the bridgeport so I could get a socket directly on it.



Currently, I'm at 5.5V, using a 6ft breaker bar and the pinion teeth have started to cut into my vice jaws. I'm planning to mill the pinion teeth square so I can seat it in the vice better and continue testing, but I'll be out of load cell range shortly...





And here's what the load curve looks like...this thing is STOUT.




Yeah, it's holding >800lbft

 

the torque it sees is multiplied by the gears in the transmission though right? so from the factory it shoud see in the neighborhood of

300 lbft from motor * 3.5 1st gear ratio * 1/3 pto gear reduction = 350 lb/ft

now that i type that out it seems the gear reduction in the PTO is what allows the coupler to work at all otherwise the torque would be extreme.

 

Yep, that's right. But honestly, if it's holding 800lbft at half its normal operating voltage, you could probably rely on it to hold something a little more than that.



There's a few things going on here and why I'm going down this path (again). We have blown up a few transfer cases at launch, and so have numerous other people.

1) The operation of the solenoid at launch is suspect. I've talked @@xfeejayx into doing some o-scope testing so we can finally confirm what the signals look like during the process of launching the car. I have some theories about this from how we've experienced it, and I think it's exacerbated by having stiff wall slicks on an unprepped surface. It feels like it doesn't engage until after the front wheels start to spin.

2) Hitting the drivetrain with a shock load can be made 'better' by lightening the rotating mass. Driveshaft and wheels are two super easy spots to lose rotational mass. Realizing the driveshaft has to be spun up at 3x the rate of the front wheels makes me think there's a huge load reduction if you go with aluminum or carbon fiber.

3) Transfer case flex - I want to measure this because we've talked about it for years. If it can hold a static 400+++lbft of torque, it's all about the impact load.

 

have you tried wiring a switch to manually lock the coupler for launches? or even for a whole run?

 

That's where I'm going with this We've been out of the MS6 for two years, so in a month or two we will try it out.

 

For your use case if it works out being locked then you could just get a solid driveshaft made