Flange Wear from CV Joint Options

[frank_c]

Today I started the process of dropping the engine & transmission so that I can send the transmission off to Dr. Evil and take care of some misc engine work (rear main seal, oil return tubes, etc. ). As part of this work I was also hunting for a clunking noise I hear from the back when the car is stopped and as I let the clutch out and the car just starts rolling forward. I already checked and/or replaced all suspension and engine mounts, as well as the outer CV joint bolts (inners are safety wired) and found nothing loose. So next on my list was either the CV joints themselves, or possibly loose flywheel bolts.

When I removed the inner CV joints from the transaxle flanges, I was surprised to see some pretty deep gouges in the flanges from the CV joint inner race/star (see photos below). This seems like a good candidate for the noise I am hearing.

I have never seen this type of wear before with any of the other cars I've worked on - what would cause this? I noticed that there is not much grease in the joints, which I suppose is one possibility. Was the inner race put on backwards? Wrong parts? I did confirm that the axle shaft does move in and out freely at the outer CV joint.

Thanks for any insight.

Left flange:


Right flange:


Circled area shows where inner race is cutting into the flange.

[frank_c]

Thanks for all the responses. Please find my answers below to some of the comments/questions:

1. CV joint was previously disassembled and not reassembled properly as you speculate.

A: I was looking into to this some more this morning, and from the images of the inner race/spider I have a hunch on what might be going on. See below.

2. Flanges not correct / not seating deep enough into differential?

A: Other than the older 911 flanges (which should be ok), what else would fit?

3. Axle shafts too long?

A: Based on what I found this morning, this is seeming like a more realistic possibility. See below.

4. Wrong CV’s - maybe someone using the wider 944 CV’s without proper machining mods to 914 axles? (Not sure this is possible without doing the axle machining to get the CV onto the axle)

A: The CVs don't appear to be the wider 944 part - and as you mentioned, if they were the circlip would not fit without machined axles (I checked and the axles don't appear to be machined).

5. Don’t know if you’re running non-stock trans position / angles / ride height that is resulting in too much axle plunge?

A: I did have the car corner balanced and they set the ride perhaps a little lower, but nothing extreme.

6. Is it both sides? Have you pulled the axles and checked to see if it happened on the outboard CBs?

A: Yes, both sides (pictures in original post show L & R). I haven't pulled the axles yet, but based on how little grease there is on the inboard CVs, I will likely be doing that at least to get more grease in there.

Update:
I did some further inspecting of the CV joint this morning, and found something else alarming that I've never seen before that points me to believe that one or more of the parts are not correct here. Note that I purchased the car back in February, and I'm still in the 'discovery phase' in finding and fixing things.

As a side note, the rear wheels were very difficult to turn by hand with gearbox in N. Now with the inboard CVs disconnected they turn very easily. I suspect the CV spider interfering with the flange when the wheels were at full drop contributed to this.

Observation 1:
The inner conical washer has grooves worn in it from the CV balls (see photo). Maybe this is considered normal wear, but I have never seen this before on any of the Porsches and BMWs I've worked on. This seems to fall into the "axles shafts too long" bucket.




Observation 2:
From looking at 914 CV joint images this morning, I noticed that there is a difference in the spider shoulder height for the Porsche part vs. the GKN/Loebro part, the latter being much taller and appears to match what is installed in my car.

Notice the height of the yellow arrow in the two images below that I pulled from online vendor sites. The height of the spider on the Porsche part is much shorter and has a bevel cut in the edge (red circle).

Porsche part:


GKN/Loebro part:


And here again is the CV from my car with the taller shoulder height shown (& no bevel):


So is it the GKN/Loebro part alone that is causing the issue? If so, I would think others would have discovered this by now.

I would like to push the 'easy button' and simply order the completely assembled 914 axles and CVs from PMB, that way I know I would have the correct axles, but I see they use the GKN/LoebroGKN/Loebro CVs - so there is a concern there.

If my axles are correct (does anyone have the length of the 914 axle shafts handy?), then I would feel safer purchasing the Porsche CVs.

Thanks.