Building a better Vanagon automatic transaxle

[h00drat]

I took this picture of it prior to pulling it apart:

I do remember checking the gap when I rebuilt it per the Bentley manual. According to my notes it was in spec.
_________________
<<Tyler>>
1987.5 Wolfsburg Weekender / CHC / 2.0 ABA
CHC Top Thread
Engine Conversion Thread (2.0 ABA Swap)
ABA Swap - Image Gallery
Auto Trans Rebuild - Image Gallery

[vwhammer]

Hmm there is some interesting wear on the inner teeth on the ring that broke.

Perhaps it happened after it broke.

most of this material is pretty hard but is there any damage to the gear on the forward drum that engages the inner teeth on the DD frictions?

Not sure how but maybe the forward drum was not fully engaged with the direct drive clutch pack?

I'm just wingin' it here.
With that kind of damage I feel like the reason will be pretty obvious once you find it, if that makes any sense.

[MarkWard]

If you didn’t fully engage all the plates when you installed the forward clutch into the drum, that could account for how it got damaged.
_________________
☮️

[kourt]

[MarkWard]

Yes it should be impossible, but mixing parts from vw and Audi automatics might obscure a miss fit. This is the better automatic thread and in the last 2 pages, 2 different members had failures within 3000 miles of being rebuilt and shifting fine till they don’t. These were catastrophic failures. I recall both failures were in the direct clutch. Bummer.
_________________
☮️

[h00drat]

I wonder if myself and vwhammer were the first two to use (or plan to use) the Audi DD? IIRC Kourt used the VW DD. Not sure about others.
_________________
<<Tyler>>
1987.5 Wolfsburg Weekender / CHC / 2.0 ABA
CHC Top Thread
Engine Conversion Thread (2.0 ABA Swap)
ABA Swap - Image Gallery
Auto Trans Rebuild - Image Gallery

[h00drat]

Watch this video. This is from my original assembly. It was never clearly determined whether this was OK or not, but it's worth noting that it concerned me, thus the video.

It seams entirely possible that with this amount of play in the reverse planet, that the forward clutch could have potentially shifted forward, allowing that bottom friction to spin freely, then get crushed by the forward clutch when it shifted back into place.

If this amount of play SHOULD NOT exist here, then I've found my problem. Though I don't know how to fix it.

Perhaps it was assumed that the reverse gear would hold it in place so that this forward movement wouldn't happen. I'll have to look at it all closer tonight.

https://photos.google.com/u/1/share/AF1QipMLYCsela...VleFVJb2FB
_________________
<<Tyler>>
1987.5 Wolfsburg Weekender / CHC / 2.0 ABA
CHC Top Thread
Engine Conversion Thread (2.0 ABA Swap)
ABA Swap - Image Gallery
Auto Trans Rebuild - Image Gallery

[Butcher]

That clutch was ruined because the clutch was partially engaged or the clutch pack was adjusted too tight. I would lean on the clutch pack being too tight.

When the clutch is not engaged, they should spin freely. It the clutch is tight, they will build up heat and eventually be ruined. It does not surprise me a bit that it would take 3k for that to happen.

To check the clutch, usually there is a preload on the clutch pack so the measurement can be made. If you do not rebuilt transmissions regularly it's hard to measure it properly without the load.

I do not think the spring count would cause this. Yeah, it may allow the clutch to disengage too slow/fast but I do not believe that would cause the clutch to overheat like that. It takes a lot of friction for a long time for that to happen.

Either way, those clutches were overheated because of a tight clearence. Caused by a hydraulic leak in that circuit [valve body?] or adjusted incorrectly.

[h00drat]

[MarkWard]

The forward clutch housing does not look so great to me. Also, it looks like your forward clutch has the shell of the pump needle bearing. The needle bearings and lower shell should be on the end of the oil pump. You need to pay attention to the vanagon manual. The early pumps had a machined surface and small thrust washer with 2 tangs that engaged the forward clutch hub center. You can mix and match parts, but you need to pay attention to what parts work with what. The Vanagon Bentley manual spells out the differences.

Bottom line, no one is born an expert. Experience is learned and can be painful. Working on a stock automatic has enough challenges without trying to interchange parts as you are learning. It takes a keen eye and some careful comparisons to know what will interchange.

ETKA shows both the forward drum and the DR drum to be different part numbers between the 86 vanagon and the 87 Audi 5000.

In my reading over the years, the only advantage I can see using Audi parts would be if you were able to utilize the 4 gear planetary gearset from the Audi forward clutch. Of course you'd need to use all the parts for that to be in harmony.

In my years I have seen techs skip the step of determining the shim package between the automatic and differential. If done properly, you should be able to determine if it stacked correctly. Interesting thread though. Mark
_________________
☮️

[Butcher]

After a closer inspection of those newest pictures, I cannot see how those teeth [on the friction plates] would be bent except for them not being aligned properly during installation. It's probably too late but if the teeth were pushed down, it would leave me to believe that they were forced down by improper installation.

They do have some twist to them, so if you are sure the clutches were engaged properly, that was a lot of friction for them to twist and push down/up like that.

If the clutch is locked, there is no fiction therefore they cannot over heat. There should only be friction during engagement/disengagement. Could the teeth weaken because of the amount of heat that was present? It could I guess but I have never seen that before. Typically the repairing that I have done is not taking several different transmissions to make one.

So again, if there is that much heat, it still leads me to believe the clutches were set too tight, there is always residual pressure in the clutch to keep them partially engaged, or lack of overall pressure to keep the engaged [piston seal torn?].

I'm just sharing how I would approach this issue. We're all different and therefore there are many more opinions out there.

[h00drat]

The teeth were definitely bent DOWN, not up.

I'll order new frictions and steels for the Audi DD so that I at least have a complete set of parts to work with.
_________________
<<Tyler>>
1987.5 Wolfsburg Weekender / CHC / 2.0 ABA
CHC Top Thread
Engine Conversion Thread (2.0 ABA Swap)
ABA Swap - Image Gallery
Auto Trans Rebuild - Image Gallery

[mikemtnbike]

good luck Tyler! I really appreciate it when people document their troubles and subsequent problem solving, we've all been there.
_________________
1991 Vanagon GL 2.1 AT Westfauxlia. "Frankie" Totaled https://www.thesamba.com/vw/forum/viewtopic.php?t=764510&highlight=carnage
1995 Eurovan Camper "Marzivan"
2020 GTI SE manual

[vwhammer]

Well I suppose I am with everyone else on this.
My guess is that the last friction was not lined up and the teeth got pushed down upon install of the forward drum.

They may have eventually lined up but the damage was already done.

The bent teeth and possibly bent friction area itself might have tightened up the pack to the point that it was constantly slightly engaged.
It's also possible that the teeth were rubbing on the steel below it which would also make it slightly engaged all the time.

This could have very likely worked for a while but obviously eventually failed.

I think with all this knowledge I am definitely going to run the 24 springs from the van just to be real sure that my pack is not slightly engaged.

Naturally you have to be careful when reassembling these but i don't think the Audi bit are the problem.

There are several others that have done the upgrade and are working fine.

On another note Matt Steedle told me when I first called him that he does not see the point in the Audi upgrade bits.
He runs good clutches and a kevlar band and said he puts a boat load of power through the van parts with no issues.
Of course I am not sure if that is in a van or bus or beetle or what.

Lastly I am pretty sure my brake piston has a little nub with a hole in it that sticks up into that other hole.
I will check and get back with you on that.
It seems like it would be a necessary fluid transfer orifice

[Butcher]

I'm not certain what springs I would recommend. Remember, you are changing the timing of the engagement/disengagement. This could effect the timing of the other clutch/band that is also changing during a shift change. Both must be in tune with each other so there is no harsh or slipping during engagement.

Again, just something to think about when you are swapping parts.

As for no benefit for upgrading to stronger internals, Audi/Porsche did when they added more power to the front of the transmission. Maybe it's overkill and sometimes the Germans overthink things, but I see no harm in 4 planets and more clutches.

[vwhammer]

[Butcher]

There was a time that Mercedes allowed technicians to modify holes in the intermediate plate [smaller] and drill ports larger in the front cover. Lots of little changes that when done properly made the transmission shift much better. A few mods were to replace springs in the valve body. Usually the spring came from some engineers brief case without a part number.

I was always amazed how such a small difference in hole sizes would change the entire engagement process. I also wondered how many attempts it took before they found the right combination.

So although I do not know if the amount of spring change would help or hinder the shift, it's always better to understand what is happening when you are making that decision. Having a dyno would certainly make it easier to make a change and checking out what happens. Much better than taking the transmission out of the car.

[h00drat]

Somebody mentioned that they thought I may have done the bearing between the pump and forward clutch wrong. Here are some pictures showing that, I think, I did it right.

Bentley:

My bearings:

My Pump:

Yes, you can see that part of the sandwich is still sitting in the forward clutch.

Now, I'm wondering if there is value in fully disassembling the valve body. Last time, I removed the separator plate and cleaned it well, but I DID NOT remove the cover plates on either side which release all of the springs and pressure valves. On the opposite Bentley page it shows all of the measurements.

This looks like the kind of thing that if not done correctly could cause some REAL issues.
_________________
<<Tyler>>
1987.5 Wolfsburg Weekender / CHC / 2.0 ABA
CHC Top Thread
Engine Conversion Thread (2.0 ABA Swap)
ABA Swap - Image Gallery
Auto Trans Rebuild - Image Gallery

[MarkWard]

I doubt the valve body is related to the failure. You have an early apply piston with out the bleed and a later taller shell that according to the post above deletes a wave washer. Your pump and needle bearing match looks correct. The manual shows how to assemble both drums on the bare pump housing and how to pressure test. Do this and get an idea of the assembled overall height. If the forward clutch assembly did not seat completely, the bearing on top of the pump would have been loose. Pay attention to that.
_________________
☮️

[h00drat]