Gear box broken . . . . Again

[kaneda]

Bit of an update . . . . .
New clutch arrived so gotta get of my lazy bum n take the old one off so i can machine the fly wheel at work and put it all back on.
Found a replacement gear box for cheep as I only want the case and a few odds n ends to repair mine .
Herd back from Peloquin about the diff . . . apparently it was the plates/washers under the bolts that seem to have been the issue , it almost looks like they were done up then backed off then done up again?? but anyways diff is getting stripped and new bearings ,new speedo gear ,new ARP bolts and the final drive wheel is getting ground so that its 100%.

[Jmac]

Hey mate, great to see you eyes on the prize, the result from Peloquin is that his reasoning?? or have the parts been checked and result from there?? i ask as this has scared me off doing anymore after seeing what happened to another fellow member here who is running a workshop, has sickened me off a bit, i fitted quaife back in the day and never had a drama, but now hearing too much of this is off putting. Not sure where to go with it, been watching this thread religiously to see the outcome and the fix. Let us know if that was an assessment looking at the parts if you can. Anyway back to you mate hope all goes well.
Cheers
Jmac

[kaanage]

apparently it was the plates/washers under the bolts that seem to have been the issue , it almost looks like they were done up then backed off then done up again??

Did the washers get reused on after the 1st repair?
They could have yielded and compressed, reducing the installed preload, after the bolts were torqued up.

I'm looking at the pics on the ARP site on failure modes vs your sheared bolt pic and the Impact Shear (#3) seems to be the best fit. The following point in the article seems particularly relevant

Failures due to impact shear occur in bolts loaded in single shear, like flywheel and ring gear bolts. Usually the failed bolts were called upon to locate the device as well as to clamp it and, almost always, the bolts were insufficiently preloaded on installation. Fasteners are designed to clamp parts together, not to locate them. Location is the function of dowels.

The comment about the function of bolts echoes Carroll Smith's words on the subject.

The VW design for the ring gear / differential assembly seems to rely on precise assembly which is probably why they use rivets as they can accurately predict the installed preload on the rivets (which are also designed to be used primarily in shear rather than tension, unlike bolts). I'd be checking if the stretch of the bolts could be specified rather than the torque for the diff reassembly - at least it be easy to measure.

[Jmac]

Did the washers get reused on after the 1st repair?
They could have yielded and compressed, reducing the installed preload, after the bolts were torqued up.

I'm looking at the pics on the ARP site on failure modes vs your sheared bolt pic and the Impact Shear (#3) seems to be the best fit. The following point in the article seems particularly relevant

The comment about the function of bolts echoes Carroll Smith's words on the subject.

The VW design for the ring gear / differential assembly seems to rely on precise assembly which is probably why they use rivets as they can accurately predict the installed preload on the rivets (which are also designed to be used primarily in shear rather than tension, unlike bolts). I'd be checking if the stretch of the bolts could be specified rather than the torque for the diff reassembly - at least it be easy to measure.

Mate well said, i agree with you here, Vw have riveted diffs from , well since i stripped down my 1st 020 which was in 1988. Now as you have to remove the rivets to remove ring gear then yes you need to refit precise.The riveting process does ensure a correct torque or preload, i worked for Boeing when i escaped the trade for a bit, and we fitted huck bolts, eddy bolts and other fasteners which was machine fitted which were preload determined so Cant agree more. Think what its doing. Im sure there will be a install instruction from the supplier to ease fitment and keep the product name good. The 12.9 bolts will have a torque and then an angle torque. We fitted quaife back home , back in the day and the fitment guide was pretty good, i remember fitting one to a Vw Scirocco and the guy fitting it was the guru and i stuck with him for years. The washers are a one use only, totally. What comes with the kit all gets fitted, no mismatch otherwise itll go bad, there is some serious load on that item. i hope it all works out here,
Cheers for now
Jmac

[Preen59]

The bolts sheer when they become loose. When the ring gear is under tension, the bolts have far less load on them.

I have heard of people grinding the ring gear mounting face to ensure it is flat, as in some cases they are not. This is less of a problem with a standard VW diff, as the material is softer and forms into the ring gear face when riveted together.

The Peloquin diffs are made from heat treated billet and are much harder. This means that if the ring gear mounting surface isn't flat, even when it is torqued up, it can start to move a little and when that happens, everything loosens up.. Then it's the old snowball effect.

[kaanage]

It's funny how everyone is dead scared of overtorquing (stretching) yet undertorquing (inadequate preload) is worse, especially if the fastener is locating as well as clamping as they do with the ring gear. Too bad VW didn't use dowels to locate the ring gear on the diff as the bolts would be a lot less stressed (pure tension) and this sort of mess would only happen with a really incompetent assembly rather than just a slightly imprecise one. But that's the price of having affordable mass production parts....

Interesting point about the mounting surface, Chris - I would have thought that area would have been machined flat by Peloquin/WaveTrac/Quaife etc.

The technical and manufacturing pages in the ARP Bolts website are really quite informative and interesting (if you have an interest in stress/fatigue analysis and metallurgy )

Please keep us informed, kaneda and all the best with it (and Christmas and New Year )

[kaneda]

Got this back in the mail today

Like a bought one . . . . . .
Just waiting on the new case and im ready to start putting things together .

[Preen59]

Interesting point about the mounting surface, Chris - I would have thought that area would have been machined flat by Peloquin/WaveTrac/Quaife etc.

Sorry, the ring gear itself isn't 100% flat (or can be the case). The Peloquins themselves are definitely machined surfaces.

I have heard of Peloquin grinding the ring gear's mounting face flat to ensure better purchase, and that has cured the problem in the past.

[kaneda]

Well thanks to Holen1 a replacement gear box arrived the other day so I could start stripping it for the case and speedo tube.
Some before and after shots of the case


After having trouble with cases I tried to design and new system . . . . . I call it the gear bucket

Current testing has brought up a few issues but im sure with a government grant I could work something out

But anyways with school holidays, finding the time to put it all back together seems to be the only issue so far .

[kaneda]

Found some more time this arvo to put a bit more effort into getting the car back together. . . . . . .

Did a bit of detailing of the case , painted it black and polished all the bolts n caps n the like the clear coated over the lot.



So while I was waiting for that to dry I finnished sorting out the clutch assembly and got all that back in


A few too many brews by this stage so its all ready to go as soon as I recover from Australia day