Bill Farrell

Since the Axle is the weakest part of the 996 Cup - I'd like to investigate the failure a little deeper - and share some additional photos of the failure point. As a Mechanical Engineer - my goal is to prevent future crashes and find a better solution than throwing out $1700 of Axles every year...

As Professor Austin stated - most failures occur near the transition from the stub to the CV housing

Here's the two key Root Cause Hypothesis / Action:

1) This is a point of stress concentration due to the design/manufacture. If there are microscopic machining grooves, the stress can rise significantly and over time fail.

ACTION: Inspect with liquid penetrant test the stub axle and to look for cracks that are not visible to the naked eye. Once a minute crack has started, they will fail. If you can't do that, toss them after 40 hours.

2) The Stub Axle thread stretches - and therefore, the Central bolt is no longer square against the Wheel Hub and therefore, you get a bending stress as the wheel rotates.

ACTION: Retorque the Central Bolt after the first outing - and recheck the Torque on a regular basis (under cold conditions).

Here's the photos of my failure. This appears very similar to Juha G and chrcooks's pictures. I'd to ask others to add their failure photos as well - so that the community can get a better solution.

Van

In talking with Bill, and seeing the part first-hand, I think there must have been some sort of flexing movement the hub part was doing, under the face of that nut, that fatigued the shaft and snapped it at that location (so close to flush with the threads of the nut).

He said the 997 Cups have a redesigned carrier, bearing, hub, and/or axle... Is there a way to retrofit those parts into the '6 Cups? Or does anyone make an aftermarket kit that utilizes the '7 design improvements? If either of those are a possibility, the next question is: what needs to be done to make such an upgrade legal for Club Racing? This seems like a pure safety concern and not a performance enhancement. (Unless you're *counting* on your competitors spinning off the track at inopportune times!)

smsarchitects

This is exactly where mine broke a few years ago at WGI. Let me see if I can find photos.

McRae

This is what it looks like when it fails at the wrong place...

Bill Farrell

McRae - Thanks for Posting. That's why we've got to solve the root cause. Mine failed at 38 hours - so the 40 hour rule isn't sufficient...

Rickard 993 Turbo

Dosnt the manual says 30 hours ?

bgiere

Has anyone ever used this crack detecting dye ?
https://www.pegasusautoracing.com/pr...asp?RecID=1207

fstockcarrera

Bill,no photo's here but mine broke exactly the same at the center bolt flange, LR, In the downhill @ 98 mph.
Is re torquing the center bolt at some interval a viable PM? Still won't trust them pass 35-40 hours. I'm running blacks this year with no grease changes. I should finish the season with 20+ days, more than 40 hrs

Also heard anecdotal recommendation to reduce the center bolt torque an wheel nut to 270 ftlb's with no ill effects????

analogmike

Great thread, thanks Bill, hope to get more info.

Are the 996 axles failing more than the 997???

Bill Farrell

Bob: Per your question on the Lower Torque

Here was the Post from "Professor Austin" on Sept 3rd
You should not reduce the torque value specified for your car. Jerry discussed this with the PMNA engineers. The 997s have lower torque values, but the parts are designed differently.

Cory M

Florescent penetrant under a black light is better, but I've used a similar Grainger kit in the field:

http://www.grainger.com/product/SPOTCHECK-Penetrant-3WU63?s_pp=false&picUrl=//static.grainger.com/rp/s/is/image/Grainger/3NE71_AS01?$smthumb$

Do some studying on how to read the results. People who inspect for a living are trained and certified.

amso3

I doubt that Liquid Penetrant testing is an option in the threaded area. The LP die wicks into the crack (if there is one) then the surface needs to be cleaned, then a dry powder type material is applied which sucks the die out of the crack. My guess is that you can't clean the threads adequately to get a good test. Mag testing with flourescent might work but the threads might give false indication.

This failure is preceded by the threads stretching. If I had this car I think I would check the torque of the nut on the axle every ten hours or so. If you mark the nut, back it off one turn or so, then retorque to see if the mark lands in about the same place I would feel that the axle threads have not started to stretch. If the nut turns a fair amount further, I'd remove the axle for further examination.

claykos

Great advice on the thread stretch....

I've had 2 fail before and both were at the inner spline, inside the CV boot. Luckily with this failure the car just stops moving and there was no damage.

Steve113

Russian Roulette if you ask me ............ damn

amso3

This is an interesting subject. I just ordered a Dye Penetrant kit because I have a leak in the threads of a coil over shock tube, so I've decided to try LP testing in the threaded area before I repair it. Just not sure you can clean the dye out of the threads adequately to get a good test. I'll let you know the results.