Not junk at all. There a couple of places (WEVO comes to mind, and maybe Aasco) that can repair. MikeZ just had his done......

Cheers

Checking pre-load
 

Grady,

You're right, I haven't checked the pre-load. I am only inferring at this point that the carrier has *some* pre-load, as there is no measurable movement side-to-side. The way I'm looking at pre-load now is that there are three possibilties for pre-load: none, just right, or excessive. We only know that on Brian's diff, it's either B or C. Knowing if it's B or C will require measurement, involving the removal of the pinion gear.

This is a good lead-in to one of the questions I had written down to ask later.

I mentioned I had compared the two sets of Porsche manuals which cover the 915. The earlier one '72 - '83 details the method of preload check using the torque gauge. In short, the taper bearing shims are calculated in part by using excess thickness shims and evenly tightening the cover plate until the proper drag as measured by the torque gauge is reached. Then the gap remaining between the differential case and the cover plate is measured and this gap is used to determine the sum of S1 and S2. And I understand that the pinion gear can't be in the way and must be removed for this process. Then the torque meter is used to further refine the spacer thicknesses.

The Carrera manuals, still covering in part the same 915 transmission, don't use this method for the pre-load. No torque gauge is involved, and once the differential side to side movement is determined and the sum of S1 and S2 is determined, a constant number of 0.40 mm is added. Then math is used to determine the individual spacer thicnesses. With this method, the only turning of the differential, according to the manual, is done initially before the measurements are made, to settle in the taper bearings. And here again the pinion gear must be removed to accomplish this.

Leading to the question: Do you consider either of these two methods acceptable for any 915? If the earlier method was accepted practice for the earlier 915, shouldn't it work for the later ones, and the later method for the earlier ones?

What I was trying to ask but perhaps not very clearly is that I wasn't saying that I had checked the pre-load, only that there was pre-load present, as there was no side-to-side differential carrier play, and that there was backlash present.

Here is an alternate low budget setup for backlash measurements utilizing the spacer under the transaxle flange concept:

http://forums.pelicanparts.com/uploa...1145284728.jpg

The angle stock and the hole spacing used here places the vertical portion of the angle almost exactly above the centerline of the axle flange.

HI the first diff setting system with the drag on the bearings is a time consuming job, also the manual only lists drag for SKF and FAG bearings (made in Germany) non for the after market bearings now avail.
The second system is easer to use for all bearings of which should be of better quality than the bearings avail back in the 70is, i have used both systems and use the second with no probes on road and comp box's.
Which way you go you will require shims to adjust, 2 of each and some 0.25 for fine ADD.
try your local OPC to borrow, if you are on good terms with them there shim book for the 915 box, for a deposit, as i would think that thay would poss not do 915 box's any more, my local OPC just gives the guy with the 915 box my No' as thay say no thanks, and at 80 to 90 UK pounds ($159.43) per hour + vat the job gets to pricey. good luck

regards mike

I'm beginning to understand why people throw these boxes away with this type of problem.

I'd like to focus specifically on the reason for this thread - the most economic way to stop the R&P whine. What EXACTLY has changed to cause this noise and how we get it back. Tearing the tranny down 2 or 3 times at $500 a pop to measure things IS NOT AN OPTION. We need to know the down-and-dirty way to re-set whatever has changed OR to accommodate and adjust for that change.

My feeling is that the pinion is too far up into the ring gear on this tranny. It would seem that all we need to do is re-shim the pinion gear so it moves away from the ring gear. The pattern of mesh can be determined with the color "painting" used by all hotrodders. Then the seasoned mechanics can tell us how much to move it back by changing shims. The goal is to tear this tranny down ONE time only.

The information gained so far is great, we're getting closer.

R&P photos
 

Although these photos are from the 915 I am "practicing on" getting familiar with the pinion measuring tool, and not from the one actually giving Brian the current noise issue, I thought I'd post them. Maybe someone can offer thoughts on the cause of the wear. I did check the pinion gear setting and found it to be 0.06 mm closer to the diff centerline than the # marked on the pinion. The factory spec says + or - 0.03 mm is acceptable. The 0.06 mm seems like a pretty small number to be the only cause of this.

Ring gear
http://forums.pelicanparts.com/uploa...1145413373.jpg

Pinion
http://forums.pelicanparts.com/uploa...1145413394.jpg

I've done about all I can do with regard to the measurements on the R&P, without a disassembly of the transmission. I reviewed the factory manuals, and Grady Clay has posted some quite helpful info on this thread, to see if measuring anything would help diagnose a noisy R&P. The limits on tearing the spare trans down hinged on the fact that any further teardown didn't actually relate to the transmission Brian is trying to repair. The pinion gear measurement on the spare was 66.55 mm compared to the factory setting of 66.60 mm. My guess is that 0.05 mm didn't cause the wear I saw on the R&P. Checking the back lash isn't possible without checking the pre-load, which isn't recommended without removing the pinion. If you have your gear box apart, and think any of the differential parts might need checking, you can check the pre-load for the differential carrier while the pinion gear is out of the way. The pre-load setup seems to be independent of what is done inside the gearbox proper, and needs to be done before the backlash can be measured.

So as a result of offering to help Brian figure out what steps to take to solve his R&P whine, I have learned much about R&P measurements -- what needs measuring and how to do it. However, this hasn't helped solve the issue. At least I have learned what has to be done to do to determine these measurements, rigged up some fixtures to hold the dial indicator, and used the Gary Fairbanks tool to measure pinion gear setting.

http://forums.pelicanparts.com/uploa...1145725061.jpg

The factory manuals say that unless one or more of the following have been replaced, there is no need to go through and measure what the factory did in setting up the R&P:

pinion and/or ring gear
differential case
bearings
cover plate
differential carrier

Of course, if any of these have been damaged, then some of the above will have to be replaced, necessitating the various shim sizes to be calculated.

Tyson Schmidt posted:
--------------------
"Sometimes the R&P have a wear pattern on them that is invisible. This is usually cause from being run with a flexing side cover, or worn pinion bearings and carrier bearings, which cause the ring and pinion to be outside the proper mesh when under load.

This wear pattern makes the factory setting numbers incorrect. In this instance, Prussian Blue is the best way to find the correct pinion depth setting and back-lash."

and in a later post " I'd pull the sidecover, and check the mesh pattern ala Prussian Blue."
--------------------

Others have mentioned in the thread about checking the wear pattern. The factory manuals don't mention this.
This is outside my knowledge base. I'm sure that those following this thread would like to know how this is done. I know very little about the practical side of analyzing these R&Ps by just looking at a wear pattern made visible by the blue dye. It's kinda hard to understand how this could substitute for actual measurements. So the question is, how does one analyze the wear pattern and determine ther proper pinion depth and back-lash? Inquiring minds want to know..........

And Grady posted:
--------------------
" Of course there is the non-Factory procedure of inspecting the contact pattern."
----------------

Does anyone have this procedure written down, or any photos of a correct pattern?

Jim et al,
I took my car out this morning and fully warmed it up to get some really detailed data on when and where the noise is present. Noise level is scaled from 1-10 with 1 being just audible enough to hear with windows up and interior as quiet as can be and 10 being the level you would play the radio to hear it over wind and engine noise with your windows down.
Here is what I observed:

1) The "whine" is a "1" accellerating and a "5" decellerating.

2) Whine does not begin until ~50mph.

3) Whine is completely gone when holding a constant speed.

4) Whine is completely gone when clutch is pushed in.

5) Whine noise level is gear dependant. Decelleration at 50mph produces a "2" in 3rd, "4" in 4th, "5" in 5th. (Engine noise in each gear may bias results slightly). There is no noise at all in 1st or 2nd gear up to 50mph.

6) Whine noise level on decelleration is speed dependant. 50mph produces a "5" in 5th while 100mph produces a "7" in 5th.

7) Whine on accelleration is not speed dependant and never more than a "2" under any conditions.

8) Tranny seems to be getting more quiet every time I drive the car. The 911 has only been driven 3 times since re-installing the tranny with WEVO parts. Each time it was driven for 20-30 minutes at all speeds and in all gears.

RarlyL8,

Did you ever get this problem licked?

Not yet.
I put the tranny back in as stated above. It has quieted down some on decelleration but there is still an obvious problem. I think it is simply an allignment issue with the R&P.

What I am doing now is repairing another transmission with a similar problem. The condition was ignored resulting in a heavily worn R&P. I am swapping out the R&P and re-shimming the allignment. When that is done I plan to swap out the tranny and re-shim the injured unit.

This is purely a learning excersize. It is not economically feasible to repair a 915. If I damage another tranny in the future I will just replace it with a used one for less than $1000 and be done with it. You can buy about a dozen used 915s for the price of a G50 conversion.

Does anyone have images showing good contact/wear patterns vrs bad using the blue dye? I'm considering swapping a 7:31 in to a late 915 for track use and images of a proper wear pattern would be helpful.

Keith

I found the following picture. Is the wear pattern highlited by the blue line good when centered on the gear tooth? Does this picture indicate a good wear pattern? Thanks.

http://www.dtmotorsports.com/~krl/yellowpaste.jpg

I have discovered the same problem in my 915 that RarlyL8 (Brian) had re: spun race for pinion shaft roller bearing. I did not have any differential symptoms, however. The only reason I discovered it was by disassembling the tranny to repair worn out 1st and 2nd gear synchros. The outer race still seems to fit into it's bore nicely, but you can spin it with your hand and it is easy to remove and I know it is supposed to be a press fit. It is tempting to simply 'bond' it back in, but I worry about what can happen when it starts to spin again. I understand Wevo has some process wherein they repair the bore to accept a fresh race? Can anyone clarify (Wevoid?)
My retired mechanic neighbor is looking to see if FAG makes an oversize outer race for this bearing. It made sense to me to give the bore a slight overbore (there seems to be sufficient surrounding aluminum to remove another .001-.002") and press in a new race. Unfortunately I haven't heard anyone on this board mention the availability of such a replacement bearing. Of course this will require going thru all the above steps to reset the R&P, which I would like to avoid as the diff wasn't a problem initially.
Anyone else bonded (Loctited) their races back in with good results? Anyone heard of oversize replacent races for pinion roller bearing?

I've had two 915s rebuilt by two different mechanics. Both say they simply glue the races in place with future failure dependant upon how well the glued area is prepared.

HI I have repaired the pinion bearing with a over size outer race from www.francistuthill.co.uk as he has them made, I have also made up a tool to grind the outer of the race down and have the outer hard chromed and then re-ground to the case size , as the case is not round now we check the size of the STD outer bearing 'new' and adjust to the case size , then grind , I have a few cases with both types of bearings in all are ok the chromed bearings are mostly in the rally car boxes and have checks after every 2nd event as 250-290 bhp through the box tends to brake things if not checked on a regular bases.

regards mike

HI It looks good if the back lash is set OK .:) IT is working ok if you have some track days put through it , may be a few more and drop the diff out to check, on a oil change.

regards mike

I think that's a great contact pattern. What did it look like on the decel side of the ring gear? How about the pinion?

I got a solution for this problem which I think is worth sharing. You can have your bearing race plated which will slightly increase the outside diameter and then have it press fit into the case again. You should put some paint on the inner side of the race in order to avoid the plating to catch in there...any way I was told that even if plating goes on the inside that would not be a problem since it will still be under spec for the bearing.

just my 2 cents

Thanks for the suggestion, Mario. I had heard of this process also, and it makes sense. I found a newer diff case which also has the upgraded (smaller) input bearing bore. Probably cost a lot more than plating the bearing race but I had also discovered a crack in the bulkhead between the bores, which made it pretty well hopeless to get any bearing to last in there. I have just swapped the diff assembly into the new case and spent countless hours confirming the ring and pinion setup. I won't know till spring whether for sure if it's gonna fly, but I think it's looking promising. Thanks for your 2 cents', surely worth at least a nickel!!

HI mmasse if you and dentist90 do go down the plating the bearing route, it works fine, as I have repaired box's with this type of bearing, you will need to hard chrome the outside, then grind it back to 80.02-80.04mm depending on how far the box is out off shape, to get it to work OK and heat the case to fit.

regards mike