Ok. Let me take a step back here.

I'm using the same ring and pinion that came with the transmission. And I'm using all the same shim packs it came with. The P/S bearing was re-used as it was determined by the machine shop to still be in good condition. The only thing that was changed were the carrier bearing and races.

Based on the above my understanding is that I should not have changed any differential settings.

It turns out I do have a torque wrench that goes as low as 10 ft-lbs. So I torqued the nuts to 18 ft-lbs (which is the higher end of the spec and it was too tight for the ring gear to move).

Is it possible that the bearing races are not fully seated?

"Is it possible that the bearing races are not fully seated?"

This is possible if you think about it logically but if you used a press they should be full home. Even if you didn't use a press you would be able to tell by looking at them. Are the races full seated in the case and side cover? Are the bearings tight against the spacers on the diff? I'm guessing yes to both of these questions.

Hopefully Peter Zimmerman will chime in shortly. (he almost always does on the gearbox posts)

I hope I haven't ruined your day but the reason I made that post is because of other threads I have read on the subject of pre load and backlash. I'm re assembling my troubled gearbox at the moment and here is a link to my thread...

http://forums.pelicanparts.com/porsche-911-technical-forum/462734-915-shift-fork-check-unexpected-outcome.html

I have created a word document with lots of detailed information about this stuff.

I'll send you a message with my Hotmail email address shortly so I can send you this document. It's about 22 pages long with lots of diagrams and pictures.

This is a picture from the Porsche work shop manual. (stolen from another thread)

Please read the section under "Ring gear has to be adjusted, if"

http://forums.pelicanparts.com/uploa...1240402142.jpg

Peter, may I please get a copy of your work also?

Cheers
John

Just sent you a PM with my email address. I'm a bit reluctant to post it here.

Whew, this isn't easy to do via "long distance." FYI: Side cover nut torque will not affect bearing pre-load after the nuts are tight. Using only four (for example) equally spaced nuts to make the measurement might not give you a perfect measurement. Here are my questions:

1. Are you sure that both carrier bearing races are fully seated (one in the housing & one in the side cover)? You can see if they are by simply looking at the back side of each race through the axle flange hole.

2. At this point forget all about pinion depth and backlash, it just confuses this issue.

3. When you removed your old carrier bearings I will assume that you did them one at a time, pulled off any shims under them, and cleaned the shims thoroughly, correct?

4. Did you damage a shim while having trouble removing the one bearing?

5. Did you notice that the shims (not the .25mm super-thin ones) have a square cut on one side, and a bevel on the other. Are you absolutely sure that the bevel is facing the diff on each of the thick shims?

Answer these items and then we'll move on to how you are trying to get the ring gear to turn with the side cover torqued.

In order to correctly measure pre-load you need the following, the washer is homemade and is a snug fit at the ID (35mm), and I had to bevel the outer edge to face the diff because it was a little too wide (7.4mm) and I didn't have access to a lathe (my washer is 2.6mm thick):


http://forums.pelicanparts.com/uploa...1240412895.jpg
http://forums.pelicanparts.com/uploa...1240412935.jpg
http://forums.pelicanparts.com/uploa...1240412958.jpg
http://forums.pelicanparts.com/uploa...1240412986.jpg
http://forums.pelicanparts.com/uploa...1240413016.jpg

Hopefully this will shed some light on things.

The light is brighter!

Sorry, but you might have to pull those bearings back off. This will be very difficult, to not damage them, with the tools you have, so we'll try something else first.

Use a mag light, or similar, to light the area under each bearing. Look at the side/edge of the shims and check to see if they are standing proud from the carrier surface, or if they appear to be firmly seated against the carrier. I have an old diff in the garage (see next post) that I'll look at later to see if this approach will help us...

Matt: Please don't miss post # 48!

Here are a couple of pics for you; I've never experimented with putting the shims on square side down, but I imagine that will only be a few thousandths difference. That said, in this biz even a spec of hard dirt makes a difference.

If you have one or both shims upside down hopefully you'll be able to detect a space using feeler gauges.
http://forums.pelicanparts.com/uploa...1240426639.jpg
http://forums.pelicanparts.com/uploa...1240426667.jpg

I'll take pictures when I get home.

Peter Zimmerman

You post the best pictures... You should write a book about the 915 gearbox.

PM from Porsche Monkey replied to.

Thank you, sir!

Don't print more than 100 copies. We are the proud, the few, the 915 rebuilders. :)

Ok. I pull the differential back out of the case and looked at the bearings. Here are some pictures...

http://forums.pelicanparts.com/uploa...1240450902.jpg
http://forums.pelicanparts.com/uploa...1240450914.jpg
http://forums.pelicanparts.com/uploa...1240450925.jpg
http://forums.pelicanparts.com/uploa...1240450938.jpg
http://forums.pelicanparts.com/uploa...1240450953.jpg
http://forums.pelicanparts.com/uploa...1240450971.jpg
http://forums.pelicanparts.com/uploa...1240450985.jpg

But I don't think this is where my problem lies...see next post there will be more pictures.

I decided to put the differential back in the case and spin it without the side cover on.

It spun freely, but I would hear it get caught up on something at every cycle. It seemed like the speed sensor was rubbing against something...

http://forums.pelicanparts.com/uploa...1240451163.jpg

What the hell is this?

http://forums.pelicanparts.com/uploa...1240451185.jpg

And this?

http://forums.pelicanparts.com/uploa...1240451206.jpg

And this.

http://forums.pelicanparts.com/uploa...1240451223.jpg

Does that look right?

I'm slowly getting that sinking feeling.

Looks like someone has welded up the crack from the back (inside). It is a really thin bit of alloy casting between the diff magnets and the external sensor... But no reason it can't be cleaned up.

x2... The 915 castings are not perfect. Don't stress about it. If you can clean up that weld without compromising it then it should be fine.

The crack from the outside is a worry but the weld from the inside can be re done and ground flat. This won't cost mush to fix.

Think of it this way. It's so much better that you found this now rather than after it was back together and in the car. So in a way this is a good thing... Well done.

This reminds me of something a Doctor told me when I was in training to become a nurse...

More is missed from not looking that not knowing.

Just a thought... This is an SKF bearing.

Was the old one you removed an SKF? or the original OEM one from the factory?
I just looked at mine and I can't see any marking on it like yours in the photo so I'm assuming it's OEM
and therefore not manufactured by SKF. (or it may be unbranded but made by SKF)

You would assume that the bearings would be the same size externally regardless of the manufacturer but we need to confirm this measurement.

Do you still have the old ones? Are you able to accurately measure their stacked height?

The Genuine Porsche ones are: OEM-999-059-027-00 - $80.25
SKF Part number from our host: 999-059-027-00-M92 - $25.00
(Both available through our host)

Again... Peter Zimmerman should be able to confirm this measurement as he probably has a box full of old OEM and SKF bearings.

What a great thread!