FrenchToast

The spindles delivered on 944s thru 1985 and all 924s have one side hollow to allow for a speedometer cable. This hollow spindle can fail with repeated performance driving. Later spindles are solid; but only certain ones match the offset of the aforementioned vehicles. As with any component of this nature, used items should be crack-tested prior to use.

jhowell371

Old off road VW dune buggies had the same problem. A snug fitting bolt was inserted and welded into place to reinforce the weak spindle. YMMV

V2Rocket

I’ve spent some time looking into this issue. After seeing a number of reported failures it appears to me that there must be a stress riser at the small ridge where the outer bearing stops against the tapered part of the spindle.

All photos I’ve seen of broken parts have been broken at that location. The spindle is only 19mm diameter at that section, but I think the ridge must have a sharper than ideal corner to it.

The hollow spindles have very slightly less stiffness to resist the loads on the spindle which leads them to break, but even 85.5-86 non-hollow but still early size spindles break at the same place. I think the hollow detail is secondary to the stress riser.

I suspect sustained high loads start a crack at that ridge, and the hollow spindles just have less meat to resist the crack growing.

The late cars have 22mm outer bearings with 35mm inners— the same sizes and actual bearings as used on tons of other vehicles including GMT400 trucks on their spindles. But then the failure becomes the old sand cast aluminum hubs…

CyCloNe!

I remember Xschop did a study on the 944 spindles. I believe he said that later NA spindles were the strongest. Even over the 951 spindles. Not sure if that's true but interesting thought.

V2Rocket

on the 87+ cars, the NA spindle and hub are the same dimensionally as the S2/turbo. the only difference in 87+ turbo/S2 vs NA spindles is the spacing of the brake caliper mounts. so they should be just as strong.
interestingly the 951 M030 hubs seem to have a higher rate of reported failure than the non-M030, i suspect that comes down to being used harder than non-M030 (old wisdom for track used to be, get the M030 brake setup)