Another 915 saga

[sjf911]

Since my son is home for spring break and the parts for my engine rebuild are no where in sight, we elected to proceed with the transmission. All of the pieces are in. I had known from a previous build to replace 1st and 2nd gear synchro's that my pinion bearing race was loose in the final drive housing although I had used sleeve retainer compound to put it back. I figured, given my current torque levels, that it had not held up and the disassembly proved it. The plan was to add a torque biased LSD, repair the pinion bearing race by sleeving it, add a one piece final drive bearing retainer plate, and a reinforced differential side cover with a 930 carrier bearing sourced from CMS a while back in anticipation of this moment.



As you can see, the pinion race was very loose and easily removed from the housing. I then sent the case off to CMS for a steel insert. Very impressed with their customer service and unbelievably fast turn around. It took only 8 days with shipping both directions.

Unfortunately, I found my main case input shaft bearing race to be in worse shape than the final drive housing. The case was severely deformed at the race and the race could be easily rocked back and forth with it almost 0.1" out of round. Since I had already had the final drive case done, I went ahead and sent the main case to Wevo. We got the case back earlier this week which allowed us to start the project.





The Wevo main case repair is certainly a work of art. The bad news is that one of the main selling points of this repair was supposed to be the availability of off the shelf "cheap" bearings but it turns out that these are NLA so you are stuck with the OEM's. So, the first job was to heat up the main case in the oven while the missus was out of the house and drop the frozen bearing races in. The locking plates were then bolted down.





The differential was then disassembled keeping track of the shim thicknesses on both sides. The ring side had a 3.0mm and a 0.25mm spacer and the opposite side had a 2.6mm and a 0.25mm spacer. I didn't have a fancy puck to use with the gear puller so we made due with the combination of part of the ERP rubber bushing tool and a small bearing race setter.



We then removed the ring gear.




The final drive housing carrier bearing race was replaced by heating the case and freezing the race. Unfortunately (or fortunately?), it would not fit in the oven so it went on the barbeque for a gentle heating. We then installed the ring gear discovering that the new ring gear bolts have a 20mm head rather than the original 19mm. This created an unanticipated delay (after store hours) and necessitated a trip to Sears to locate a 20mm socket. With the ring gear on, the original shims with new carrier bearings were then pressed on the LSD and installed into the case. Carrier bearing preload was then measured and came in at 30.5 in/lb which is spot on for SKF bearings.
Next will be removing the gear stacks to place the bearing plate and then measuring pinion backlash.

[prschmn]

You might want to consider taking both carriers to a shop with a height mic-saves time adjusting the shims.

[sjf911]

Originally Posted by prschmn

You might want to consider taking both carriers to a shop with a height mic-saves time adjusting the shims.

I thought of that but this really is an educational project for my son and I so we are going to do it the hard way.

[sjf911]

Alright, more data. I remeasured the original shim stack which was 2.85+3.25 for a total of 6.1mm. This resulted in a preload of >50in/lbs. Dropped to a 6.0mm stack and got 26in/lbs and 6.05mm at 37mm. The pinion has 0.16mm inscribed on it.
Using the workshop formula of the average -.1mm and average+0.1mm I tried 2.9 and 3.1 but got a backlash of 0.35mm. Dropped to 2.8 and 3.2 and the backlash dropped to 0.14mm. Unfortunately I don't have a 2.9mm shim to attempt to split the difference.
The best backlash I could get without a 2.9mm shim at 6.05mm stack height was 0.1mm with 2.85mm+3.2mm.

So is it better to have the bearing preload on the lower end of the range or the upper end?
Is 0.14mm backlash adequate with the bearing preload at the lower limit or should I source a 2.9mm shim and try more combinations?

[Trackrash]

I would go with the upper end of the preload with new bearings. According to Peter Zimmermann the preload spec is different according to the brand of bearings you are using. How-To: Porsche 915 Transmission Repair Tutorial Part 5 - Porsche Wiki . What bearing are you using? Is the spec different for the turbo bearing?
I find it interesting that the manual for VW Beetle transmissions lists different differential preload specs for new and for used bearings. This is not the case for Porsche transmissions.
Last year when I did my trans I could not find a source outside of Germany that had the shims in stock.
Hopefully some of the pros will chime in on this.

[sjf911]

Thanks Gordon, these are SKF so the range is 26 to 37 in-lbs. Technically I am in range with it at 26in-lbs and 0.14mm backlash but I am a little uncomfortable with it (insufficient personal experience). It seems I am stuck taking one end of the range limit for preload. I went ahead and ordered a 29mm spacer, unfortunately it will probably take 6 weeks or more to come from Germany unless someone has a 29mm spacer they can loan/sell to me.

[RWebb]

which 915 is this (exactly; or what year)?

also, doesn't setting the shims require that expensive and rare device or machine as a jig?

[sjf911]

Originally Posted by RWebb

which 915 is this (exactly; or what year)?

also, doesn't setting the shims require that expensive and rare device or machine as a jig?

If I was setting pinion depth you would need that uber expensive tool, but carrier bearing preload and the ring/pinion backlash can be done with minimal tools.