Howdy,

What sorta time was on the hubs from Mike that failed? Had you ever re-greased them or otherwise taken them apart for inspection? Had you been checking for play?

I'm wondering if I need to contact the folks that bought my car and tell them to stop running the hubs right now, or if its possible that they just need to grease them more frequently or something.

Mark

I had them on for about 1 year and a couple months. I think I calc'd 6000 miles +/-, and about 12 AX events, and the track day was the first track day on them. But of course I drive the car sort of aggressively on the street, but on street tires, not R comps like I have for track (Hoosier cups) or AX (Kumho 710s, 315 all corners). I had not regreased them nor taken apart. There seemed to be no play from worn bearings and I did check them. When I took the first one apart it seemed like it had enough grease in it.
Maybe I will post some pix later tonight when I get home.

Here's a couple pix from the first failed one at Summit:

(IMG:http://paveglio.com/postimages/DSC_3931-sm.jpg)
(IMG:http://paveglio.com/postimages/DSC_3936-sm.jpg)
(IMG:http://paveglio.com/postimages/DSC_3939-sm.jpg)

Wow no beech marks or necking, looks like they went without any warning. Maybe a problem in heat treatment and/or metallurgy may have made the spindle brittle.

I know it's fun to speculate, but as a materials engineer who reviews failure analysis reports on an almost daily basis, I would never attempt to read a fracture surface based on a 1X magnification snapshot. There's a lot of relevant information on that fracture surface you simply can't see in those photos. There's also some things clearly visible in those pictures that are red herrings. Like defusing bombs, failure analysis is is best left to the pros.

has there been any updates on this?

+1. I had these on the 'to buy' list over the winter.

It would be interesting if some other owners of these hubs would tear them down and do some mag particle or dye penetrant to check for cracks. It's really not a good idea as stated above to draw much if any conclusions from the pictures, but it looks like the spindle was cracked for a while. Anyone still running these should use some dye pen to check them.

The thing is, there is a "shoulder" ring that is slipped on that goes right were the fillet junction is from spindle to stud plate area. (You can see it in the pics.) Both mine failed the same way, whatever crack or defect is there could not have been inspected unless that ring was removed, which is pretty much impossible. Sure there could be other cracks, but that is the important area and it's covered up.

I have some dye penatrant on order. I'm going to do mine soon provided I can get the ring mentioned above off. One of the broken ones is at a metalugical lab now for analysis. The analysis is being done as a favor, so I don't know how long it will take.

So far we know a couple basic things:

1. This particular unit had cracks start in at least two places.
2. The hardness was slighlty below the design target.

Hopefully, further testing will reveal more. Once I know all the test results, the design will be modified if the testing shows a clear path to improvement. That could be one or several changes including induction hardening intead of through hargening and/or minor dimensional changes.

The location of the break is the same size as a Dana 70 axle shaft and made of stronger material. These are used in heavy duty trucks that see loads far in excess of what an F-body can create. I am confident that the cause and solution can be found.

This post has been edited by 00 SS: Nov 13 2010, 08:33 PM

Any new info?

Yes, but a Dana 70 axle shaft does not carry any weight, it only has torsional loads on it to drive the wheel. The weight is supported by a beefy spindle with big tapered roller bearings that are about 4 inches apart. Your spindle will have alot of bending forces on it and that ring groove is a huge stress riser right where the bending force is going to occur...Just my 2 cents.

Wow! Somebody from the 'loops! Not many BCers here!

I talked to Mike several days ago. Here's some of what came out of the conversation. Some of this is paraphrasing and some is sort of restating what Mike posted earlier.
Pete's analysis showed that the parts were not forged well enough (if forged at all), that they were almost as porous as a cast piece. Pete said they looked like (more or less) just a fatigue break, and did not have any particular defect that instigated the breakage; simply the rotating stresses wore it down "like bending a paper clip till it breaks". Mike designs them, and another company buys axle blanks and essentially cuts off the long part of the shaft, and then does machining to the specifications. The machining company was contacted (if I remember he said that) and is supposedly changing suppliers or making some other adjustments to get better quality pieces to start.
Mike said he has a couple options he is looking into to improve the part.

1. Redesign the hub to have a thicker/wider radius at the transition area and/or thicker at other areas. His primary problem is finding a suitable grease seal from an off the shelf application. Says having a custom seal produced would cost a fortune.
2. Better quality part to start, and proper heat treating and hardening. (In regards to this I suggested examining the "left over" part of the axle to verify it's been forged of proper strength. I think Mike said that's not entirely feasible due to what happens to metal under heat treating or cutting it apart. We had some technical discussion that I do not remember some of the exact details of; I may be mis-remembering this particular part of it.)
3. Induction hardening. Mike says this could costs thousands to set up and require a lot of blanks to do trial/error setup of the induction unit and cutting/testing of several to optimize the hardening. But would be a really good way to harden the part, too bad it's so expensive, it probably won't happen due to cost and what is probably low demand of this part in general. I suggested a group buy on hubs, Mike thinks the demand is probably still too low to justify or cover the high cost of it.

His dye penetrant test on his hubs showed no visible cracks or stress on his unit. Though it was probably made in a different batch of hubs than mine was.
Mike has not been able to take off work recently to visit suppliers or shops during daytime business hours, so this has been sort of a slow process.
Mike said it was OK to post this and I hope I am posting everything correctly. If not I'll be happy to edit or post corrections after the fact.

Not sure what he has accessible, but working in parts I know he can get a seal catalog with oil seal specs. It also has all the applications but can be reverse checked to see how common of a seal it is after finding 1 by dimensions. If not I can/will mail a catalog with CR numbers.

Thanks for the offer on the seal catalog. But that won 't be necessary. My supplier has all the CR, SKF and a few other catalogues and pretty much all of them in stock. The issue is in large part, finding time to get over there and start digging through everything.

Since I can change dimensions a bit, it's not quite as simple as looking up a specific ID, OD and depth.

Case Carburize is the answer to this. It will be much cheaper due to low set-up costs and no scrapping of parts to develop the 'recipe', and will result in a stronger part. I'd use 8620 steel since it is a very popular carburizing grade steel commonly used for gears. Case Carburizing doesn't require near the effort to get right for low volume since it is basically just puting the part in an oven surounded by an atmoshpehere, then quenching and tempering (I'm leaving out details, but that is the basics). All exposed surfaces get a hard case on the outside, HRC 60, and the core will be hard yet ductile HRC 25-40. Low volume is ok for this process because it is predictable, as I understand it how long it stays in the atmoshphere is the main varable and that determines how deep the case goes. I can't remember what the cross section of your part looks like but if you have 'button' area like you'd see in the c-clip areas of an axle shaft you'll want to machine the groove after carburizing or mask the area to prevent brittleness. If you need help finding a source I can offer a name of a shop here in the D.

Also, having a case and core in a part should perform better than just through hardening because of favorable residual stresses - compressive residual stress on the surface balanced by tensile residual stress in the core. Since materials almost always fail in tension having compressive residual stress in the high stress areas (the surface) will reduce the effective stress which will increase the fatigue life.

I can't remember what the hardess was of your part but the case carburized part will most cerntainly be harder = another advantage, Tensile stregth for HRC60 = 300 ksi.

Some more data supporting a case hardened part: The only axle parts that I know of that are through hardened are parts that primarly only see compressive stress = bearings. Axle shafts are induction hardened (case and core).

I evaluated the broken parts for Mike. The main issue I had with the material of the failed parts is that they were a relatively porous casting. Unfortunately the corrosion on the fracture surface prevented me from determining if there was porosity at the origin of the fatigue crack, but the porosity I saw was significant and could have played a role.

Making the hub from bar stock or a forging instead of a casting would eliminate any porosity issues.

You can make castings work, in fact the Duralast hub that I also examined was a casting but a much better one. However, it can take several iterations of trial castings and destructive evaluations before you get a casting right.

The same is true for carburizing, nitriding and induction hardening. I wouldn't feel comfortable with any of these processes without some level of destructive evaluation to make sure they were done correctly.

One of the factors that Mike will have to weigh is how well he believes he can monitor the quality and consistency of any given process he uses Large automotive companies have specialists to do just that. Some processes are easier to do that with than others.

I thought they were made from an axle shaft forging?


I'm really surprised to hear you say that a casting is acceptable for this part. If for no other reason one surface on the factory spindle has a bearing surface, and bearing life is very dependent cleanliness of the steel, and castings are very unclean. Brake rotor type 'old school' hubs are castings but are much more beefy i.e. not a ~1.25” dia shaft. Nothing about the Duralast part would surprise me however.

Most bearing components are heat treated to HRC 60. Pete, could you do a hardness check on the factory hub? Sean sent me one a couple years ago; the part I looked at was a forging and appeared to be induction heat treated. I’m betting it is HRC 60.


Not saying that it’s not a good idea but I know it is common not to cut carburized parts on low volume stuff. I’m no expert so don’t quote me on this, but my experience has been that it really depends on the source and how much experience they have with the material/ size/type/spec of the part and how comfortable they feel about it. On the other hand induction hardening is a total PIA for low volume and needs many cuts to get right. This hub is a safety critical part so care needs to be taken.


True statement.