CV joint came off

[T77911S]

has anyone else had problems with the inboard CV joint coming off? i just had the motor out and i KNOW i tightened all the screws because the same thing happened to before with my 914-6. i put new gaskets but not new washers and screws. ( i put new washers on the 914-6 and one still came off). when i checked the other side it was very loose. after tightening, i went back through them again and i could tighten them some more, i did this 4 times, they turned less each time, but they were still a little loose each time through. could it be the new gasket? i had to put the side that came off back on without the gasket and it tightened up fine. i put locktite on them this time, im even thinking of safty wire.

[Porsche84]

I had this problem using the gasket type CV joint on a 911. I cleaned the gasket surface very good, difficult to do with all of the grease, and used an RTV compound instead of the gaskets. They did not leak grease and never came loose again. This is a fairly common problem with the gakset type CVs. That's probably why Porsche did away with the gasket in 85-86.

Jerry Austin

[Grady Clay]

This 14 page thread is worth studying.

”Reconstructing Constant Velocity (CV) Joints“

While long, I think we covered most issues and I learned a lot.

Best,
Grady

[Chuck Moreland]

I drill and safety wire the cv bolts.

[EFullerton]

I had a similar event happen in my '73. I had recently put the car together and had approximately 50 miles on the car. My Porsche mechanic buddy reminded me to check the CV Joint bolts in 500 miles. I was headed home after a little road test and tune. As I was shifting into 3rd, the joint let go and I coasted on to the island in the center of the roadway. To add insult to injury (my mechanical pride was hurting) some kid in a Honda with a big and I am sure purposeful wing says to the effect "Who is your Mechanic....I would recommend you go see XYZ Local Repair Shop". Shortly after a member of the PCA stopped and helped me push the car out from the center of the road and into a nearby parking lot. I called my buddy up, he brought his trailer to the scene and we loaded it up and took it home.

I was lucky and found all of the balls that came out in the middle of the roadway and no tread damage to the drive flange on the transmission. I checked the other joints and found some of the bolts were loose and definitely on there way out. After I reassembled it, I checked the CV's several times to make sure they did not come loose again. Other than a small dent in the header and some scuffs on the CV, no real damage to report.

I like Chuck's suggestion of drilling and safety wiring the CV bolts. Sounds like a good winter project.

Happy New Year!!!

[T77911S]

whats causeing them to come loose? the gaskets? grease on the threads? lock washers?

[T77911S]

that was a VERY long thread. to bad there isnt a condensed version posted somewhere else of everyones problems and what actually fixed it without haveing to read everyones speculations and comments.

i dont have the half moon plates. there is a think flange, about 1/4", that has the boot crimped into it that sits on the CV joint. i have seen the thin metal ones but that is not what i have.

[EFullerton]

I would think that it is a matter of getting the gaskest fully compressed and heat cycled. Once they are tightened for the second time they should be good until the joint is removed for whatever reason and a new gasket is installed.

One other point about my setup is that my joints are the 100mm 6 bolt from an '82 SC with Course splined drive flanges from a '76 or 77 915.

[mm86911]

Originally posted by Chuck Moreland
I drill and safety wire the cv bolts.

Excellent idea! I use a paint marker on all critical fasteners and perform a visual inspection before, during and after each track event.

[Grady Clay]

There isn’t a condensed version of this complex problem. That thread probably has the best info you will find.

As you can see, there are probably multiple causes to this problem. A prudent person uses every path to a solution. We should add any new ones to that thread.

I disagree with Chuck as safety wire solving the problem. Safety wire might be a good protective measure in a racing situation. The problem is the safety wire prevents re-torquing the CV bolts. The bolts need to be torqued regularly (more so in a race situation). With safety wire, it must be un-done to get the Allen wrench in the bolt. That is OK if you are willing to put up with the effort.

Did you study that 14-page thread? There is a lot of good info there.

Best,
Grady

[Walt Fricke]

My attempts to drill small holes through grade 12.9 Allen bolts have not been successful. But Chuck is a skilled machinist amongst his other talents, and not just an ordinary schmoe with a wobbly drill press and no jig or bit sharpening ability. So it must be easy enough for him (or else he purchases ones with holes already in them).

Caroll Smith claimed that all safety wire does is to keep fasteners from falling out. Of course that is better than having them fall out. I have no idea how long you could drive with loose bolts retained by safety wire before things started shearing. I suppose you'd hear the clunking and investigate first. Certainly better than being stranded and running the risks that mace-like object could cause.

Grady is certainly right about the effect of safety wire on the racing fetish for tightening every nut and bolt between each race weekend. If you use the gasket, you absolutely have to retighten at least once.

I see my observations are pretty much in line with those of others:

1) Once the gasket has compressed to the point that it is not contributing at all to the bolt stretch you are sort of measuring with your torque wrench, if torqued to spec the bolts won't come loose again.

2) But once I quit using the gasket I never had a problem again either - the first torquing did the trick.

I favor the 12 point bolts (and reuse them), since so far I have never stripped one (or had a used one break), but torque is torque. And I use no washers and only the thin, deformable cover plate. Despite the wisdom, knowledge, and experience so ably set forth on that long and valuable post (I have stored a copy of it for my reference) to which Grady contributed so much, I am not about to change, either.

Walt Fricke

[T77911S]

when i put the side back on (without gasket)that came off, it tightened up fine. when i went back thru, they were still tight.
the other side ( with gasket ) i had to go thru 4 times to keep them tight.
safety wire, if done properly will keep a bolt from loosening. look at the high use of it on aircraft. but what ever the case, i wouldnt want them coming off at 70mph much less 100+! both times i was going very slow.
the article was helpfull. i am going to clean the threads and use locktite and see how they hold. thanks guys

[EFullerton]

With the bolts properly safety wired, they would not have much of a chance to move more than a 1/8th of a turn (I would guess). This would be a far better situation than them backing out totally and that sinking feeling.

[randywebb]

Here is the problem: "but not new washers..."

I would always use new washers; clean off all grease; and check the torque after 100 miles.

I don't like drilling - it weakens the head. Think about all the problems in dealing with Allen cap screws (aka 'bolts') and you too may develop some reticence re drilling...

But one _could_ look around for specialty cap screws of the requisite strength with holes made in them during the forming process (before heat-treating). They should be plenty strong. Find a place that offers AN bolts (AN is an old category for high-strength bolts).

[Grady Clay]

Everyone should study the thread listed above. The basics are the driving torque is transmitter by the metal-to-metal contact of the CV joint to the flanges. It is the clamping force provided by the CV bolts that provides that.

The CV bolts will shortly fail in shear if the contact interface slips. They are not designed to operate in that mode.

The failure modes seem to be: The CV bolt loosens (no or worn Schnorr washer or grease on the threads or not sufficient torque), slippage at the contact surface (contaminated with grease), gasket improperly installed (clamped between the contact surfaces) or all sorts of other possibilities.

I would really like to see this discussion copied to and continued on the thread cited. This is a serious issue and can possible be a matter of life-and-death.

Can the moderators append this thread to ”Reconstructing Constant Velocity (CV) Joints“.

Eventually Randy (rcecale) and I will edit the first two posts to summarize the thread.

Best,
Grady

[T77911S]

i bought a bag of 50 bolts for $10 and washers for $5 from fastenal the 50mm bolts are whats on my car now, the threads are about 1-2 mm from going all the way thru. im gonna take the bolts back and get the 55mm. that should give me about 1-2 mm of bolt thread, i have room to do 60mm. anybody need some bolts and washers? much better price than here.
the side without the gasket but with locktitie on the threads has not loosened at all, the side loosened just a little. i have only driven it very little til i can put new bolts and washers on. wifie-poo wrecked the volvo so i have to drive it now.

[Grady Clay]

T77911S,

Please forgo internet lingo and post in plain US English. If that isn’t your first language, I understand. Clear communication is critical.

The critical issue is exactly what bolts and washers you have. I gather that you read the thread. As you can see, while the hardware is very important, so is the assembly procedure.

This is an issue that Porsche AG didn’t do a very good job with IMHO. It seems left to us to diagnose the problems and devise solution. PAG/PCNA is too silent on this issue.

Best,
Grady

[Zef]

I don't like drilling - it weakens the head. Think about all the problems in dealing with Allen cap screws (aka 'bolts') and you too may develop some reticence re drilling...


These bolt are working in shear load...so the head can be weakened a bit for more safety.

[randywebb]

you misunderstand my comment - I'm referring to someone who is trying to remove the cap screw

[MattAlpha]

T77911S,

Some pictures of the equipment you are using (washers especially) and the how they are installed on each side would help clarify what you are trying to explain.

-Matt