5EAT in 2wd problem...........

[AWDPete]

Ah OK, the first two tests I had my wife on the passenger side watching and I watched the drivers side. THe second two my spotter wasn't available.

No, after that run the abs was having trouble bringing all four wheels to a stop - i presume because the left front wasn't spinning?

Best I could tell without a spotter the rears felt like they where keeping up with the fronts, the data indicates this.


So why is it when the car is on the ground I get no rear drive but in the air I do, weak clutches?

[cardoc]

Originally Posted by AWDPete

no noises but the ABS was working pretty hard to stop the wheel rotation!

So you do have traction control!! This gets more bizarre by the day.

The brakes were active and you did not have your foot on the brake? That's traction control. VDC or otherwise. Where is your build spec sheet? Was it with the car when you purchased it?

Looking at the data log:

As the throttle increases, the pressure and speed sensor readings increased also. With the park brake set holding the rear wheels, the rear speed reading your getting on the log is the rear output speed sensor because the rear wheels are stationary which means the clutch is working, the TC program is trying to stop the wheel spin at the front to get traction on the rear, and your slipping seems to be in the differential.

I have seen some really strange things on a lot of cars, mostly Subaru. What I don't understand is why you aren't feeling the TC program when the front wheels are spinning. And even more so, why does your XT have a TC program?

Run the VIN number if you don't have the build sheet. You can call Subaru parts or service and they can decode the build. You may even call SOA and see if they can e-mail one. What you want is TC or not, Diff type and Center Diff type to be sure.

Fuse 33 in the fuse box under the dash is ABS/VDC. Fuse A in the under hood box is the FWD fuse but there is a note "except turbo models and 3.0 models". I would look to see if there is a fuse in Fuse A just for the h3ll of it and if not, put one in and see if anything stays lit up on the dash after you start it.

[cardoc]

Okay, bizarre maybe. I was typing and looking at the log he sent while Pete was apparently posting.

Are you sure the rear wheels were turning during the test you hit 20 mph?

The data is indicating rear output shaft speed, not wheel speed.

[AWDPete]

Alright, let me clarify my previous statements. When I said the ABS had a hard time slowing down the wheels I meant; AFTER the acceleration test while applying the foot brake the ABS was working hard to stop wheel rotation.

There is no VDC on this car. When I open RomRaider there is a ABS/VDC listing but when I plug in the cord that option disappears.

The first two test where done with a spotter watching the passanger side wheels, the second two without a spotter.

The second two tests where done accordingly: foot brake applied, throttle applied, footbrake released - no parking brake, the next was done using the parking braking. rear wheels where rotating - both of them.

I get the feeling I'm not testing exactly the way you would like me too.

In the first tests with the car idleing I could stop the rotation of all wheels with application of the parking brake.

[plain OM]

I would not take the data we have as conclusive. With the car raised, and little or no "resistance" (recall the requirement in the video?), the AWD control doesn't react the same as when there is some load. This was observed when we were examining the AWD solenoid duty cycle in FreeSSM (see this and related posts.)

The rear wheel speed in the data still is doubtful. I'm not sure it is telling us what the rear drive is doing. This was the uncertainty early on when it was first found that the wheel speed data was the same throughout while AWDPete indicated that the rear wheels were not turning.

We need to know what is happening underneath (drive shaft and rear wheels) when the solenoid pressure goes up and the parking brake is applied.

Here's a graph of the first log AWDPete attached above:



Again, we don't know where on this (time-wise) the parking brake was being applied. We are told that at some point both rear wheels were stopped. (Note: I have included only the front wheel speed as the rear data is the same.) We can see two instances where the solenoid target pressure rose to about 35 psi. Interestingly this is with the front wheel speed dropping to or at zero! (It's much the same in the other logs.) So what is the solenoid control reacting to? To find out I added a column, "delta" that is engine speed minus turbine revolution speed (rpm). This is indicating the slippage in the torque converter, and this is a proxy for torque. Here's what the new graph looks like:



So here we see a rather clear relationship between solenoid psi and delta. In order for the solenoid psi to increase significantly, there has to be a significant delta.

In this second graph, as before, we don't know what the rear drive shaft and rear wheels were doing when the solenoid psi rose.

I'm might be a bit behind, but I think we're still trying to determine whether there's a break somewhere in the rear drive train, one that provides sufficient friction to rotate the rear wheels when they're not being braked, but not when the parking brake is applied. Is this correct, or is it the consensus that the rear drive, including the drive shaft, rear differential and rear axles, is intact?

If it's felt the rear drive is intact, then the problem is inside the transmission, and we have to be looking at the solenoid data and actual pressure being delivered to the clutch. If it's felt that the rear drive is still not clearly intact, I would suggest going to the manual check involving stopping the drive shaft from turning and seeing what happens when one, and then both rear wheels are manually turned (in the same, forward, direction). (If the drive system is intact, when one wheel is turned the other should turn the same amount in the opposite direction. The wheels should not be free to turn when both are being turned in the same direction at the same time.)

[plain OM]

In the first tests with the car idleing I could stop the rotation of all wheels with application of the parking brake.

This is probably not conclusive.

With the engine idling, when the parking brake was applied, the rear wheels stopped. That's fine. At that point, the stopped rear would put a greater resistance on the planetary differential's output shaft. Recalling again the demonstration in cardoc's video, when the rear sun gear is stopped, the clutch hub, planetary carrier, and reduction drive gear (all mechanically attached to each other) should turn or turn faster if the input, the first sun gear, is turning. (We can see this in the video.)

However, let's assume that the AWD system is sensing that the rear drive has stopped, and increases the solenoid psi. This increases the AWD clutch engagement (I'm again making some stretching assumptions) which, therefore, locks the planetary carrier to the rear sun gear, and stops the front wheels! (Keep in mind that with the engine idling there isn't a lot of torque; it probably doesn't take much grip of the clutch to give this result.)

The fact that the engine was idling explains how this can happen. It's the same as when the car is stopped at a red light. The wheels are not turning; instead the torque converter allows for the engine to continue running while the transmission and wheels are not.

If the engine is revved up (as has been suggested), the engine torque available and transferred through the torque converter will be higher. At this point, the rear wheels might still not turn assuming that the parking brake is fully engaged and able to overcome the torque being applied. Also, if the AWD clutch is stronger than the torque, then the front wheels will also remain stationary. The torque converter will continue to slip up to its "stall point" (around 2400 rpm). If the AWD clutch isn't holding, the fronts will turn (again back to the video demonstration), but this also means the AWD clutch is slipping and as Seabass has warned, this is not something that should be sustained.

[Scout.]

1) I can't believe this post has gone on for 15 pages

2) I can't believe anyone thinks its a good idea to put a car on jack stands in a
Parking lot somewhere with a running car and in gear, can you say dangerous
to say the least ?


If there is truly a problem take it to a qualified shop that has the gear to examine, diagnose, and repair it properly. The post seems to have several strange twist to it already that lead me to believe this is probably going no where good.

Does anyone agree or should I just keep my mouth shut and ignore the post om here on out ?

[cardoc]

Originally Posted by Scout.

1) I can't believe this post has gone on for 15 pages

2) I can't believe anyone thinks its a good idea to put a car on jack stands in a
Parking lot somewhere with a running car and in gear, can you say dangerous
to say the least ?


If there is truly a problem take it to a qualified shop that has the gear to examine, diagnose, and repair it properly. The post seems to have several strange twist to it already that lead me to believe this is probably going no where good.

Does anyone agree or should I just keep my mouth shut and ignore the post om here on out ?

The car isn't going anywhere on the stands.

It's been to different shops with the same results. They don't know.

Plain, reread what you posted last. The problem is in the rear diff. If when Pete applied the park brake and the front wheels stopped, that means the clutch stopped the front drive since the rear shaft would be locked to he VTD.

I know we don't see rear diff issues and they seem to last forever, but something has happened to this one that is allowing slip on both rear wheels.

[Seabass]

A worn out rear diff could act as an open diff. But that doesn't explain the front slippage. Although, I still haven't read anything conclusive about the rear wheels not moving. It's very possible that one front and one rear is spinning though. The center diff would keep the front and rear speeds the same in those conditions.

[plain OM]

Plain, reread what you posted last. The problem is in the rear diff. If when Pete applied the park brake and the front wheels stopped, that means the clutch stopped the front drive since the rear shaft would be locked to he VTD.

cardoc: No problem there. My point is that if the parking brake shoes acting on the rear rotors stopped the rear wheels from turning, and that, through the rear differential, stopped the drive shaft from turning, and that, through the splined transmission output shaft stopped the second sun gear from turning, which through the applied VTD clutch stopped the front wheels, then the rear drive train would seem to be intact.

If the rear drive train isn't intact, then what is it that suggests the rear differential is the problem?

If there's a disconnect in the differential, that could prevent the drive shaft from turning the wheels. For example, if the pinion shaft gear breaks off the differential input shaft. In this case, the drive shaft would spin, but the wheels would not, or if they did turn for some reason (meaning the crown gear was being turned), they could be stopped easily without affecting the drive shaft. But the wheels do turn, and they appear to resist being stopped when the parking brake is partially applied.

A worn out rear diff could act as an open diff.

Seabass: I think you're referring to a rear differential with a limited slip function, and that the differential can act as a plain open differential when the limited slip mechanism (whatever type it is) is worn out. No problem with that as well. But the issue here isn't the LSD function. It's the transfer of power to the rear wheels, which even an open differential is designed to do.

Does anyone agree or should I just keep my mouth shut and ignore the post om here on out ?

I'm with you on the safety issue of running the raised car, and would much prefer the car be on a proper hoist, or, to test the integrity of the rear drive train, do so by blocking the differential and seeing if the rear wheels can be turned manually.

The objective at this point is to confirm if the rear drive train, from the drive shaft front yoke to the wheels is intact -- no broken CV joints, gears in the differential, etc., and that when the drive shaft is being turned from the transmission output, the drive train will transfer that action to the rear wheels.