What is "indexed steering"?

Some people seem to think that bearing surface "brinelling" or dimpling, is caused by extreme shock forces where the balls are hammered into the races. This may be true in the case of a poorly adjusted or loose headset, but not in a properly adjusted headset.

From: Jobst Brandt

Dimpling is caused by lubrication failure and occurs while riding straight ahead. If you believe it comes from hammering the balls into the races, I suggest you try to cause some dimples by hammering with a hammer onto the underside of the fork crown of a clunker bike of your choice. Those who pounded in cotters on cottered cranks will recall no such dimpling on the BB axle and even though this is a far smaller bearing race than a head bearing and the blows are more severe and direct, no dimples were made.

Ball bearings make metal-to-metal contact only when subjected to fretting loads (microscopic oscillations) while in the same position, as in riding straight ahead on a conventional road. If you watch your front axle while rolling down the road at 20+ mph you will notice that the fork ends vibrate fore and aft. This motion arises not at the blade tips but at the fork crown and articulates the head bearing in fretting motions that are not in the normal direction of bearing rotation. Any substantial steering motion replenishes lubrication from adjoining areas.

Lubrication failure from fretting causes welding between the balls and races and these tiny weld spots tear out repeatedly. The result is that at the front and rear of the races elliptical milky dimples occur. Were these brinelling (embossed through force) they would be shiny and round. Various testimonials for the durability of one bearing over another may be based on good experience, however, the differences in most of these was not in the design of the bearing but rather the type of lubricant used. A ball bearing is not suitable for this use. This is in spite of their use in almost all bicycles.

To reduce point loads and to protect the rolling elements from fretting motion, roller bearing head bearings have been built. In these the rotary motion is taken up in needle bearings on conical races and the fork articulation is absorbed by an approximation of a spherical cup (the steel race) against the aluminum housing. Both of these bearings are ideally loaded. The rollers all remain in contact and carry rotary motion while the plain spherical bearing remains in full contact carrying low pressure fore and aft motion.

I am disappointed that roller bearings until now have not been suitably perfected to rid us of the age old bearing failure. Maybe some day soon Sun Tour, Campagnolo, Shimano or Stronglight will emerge with an easily adjustable and fully compatible bearing. The one I am using is durable but not easily adjusted and it has too great a stack height to qualify for a recommended replacement.