garage sale GT said:
Jobst Brandt seemed to think that play in a cup and cone bearing was a good way to destroy it because all the weight is borne by one or at most two balls at a time, causing much higher pressure and stress than normal. He said to have slight preload. BTW, why can't a used cone be set to correct tolerance?
Correct:
"Bicycle wheel bearings, as most, require a slight preload so that more than one ball under the cone (inner race) will support its load. With proper preload, slight drag should be perceptible. Preload drag is small compared to drag caused by wheel loads, neither of which are significant regardless of adjustment. In contrast bearing life is
affected by proper adjustment. Adjusting ball bearings to spin freely unloaded does not reduce operating friction because a bearing with proper preload has lower drag when loaded than one with clearance.
For high quality bearings, preload should be just enough to cause light drag when rotating the axle between thumb and forefinger. Low grade bearings will feel slightly lumpy with proper preload.
Wheels with quick release (QR) axles present an additional problem in that closing the QR alters bearing clearance. Closing the lever requires increasing manual force with a slight over-center feel near the end of the stroke. This lever force arises from compressing the hollow axle and stretching the skewer. The ratio of elastic length change between axle and skewer is that of their cross sectional area and active lengths.
Although small, axle compression on QR hubs is large enough to alter bearing clearance and should be onsidered when adjusting bearings.
Bearings should be adjusted just loose enough so that closing the QR leaves the bearing with a slight preload. Excessive preload from QR closure is the cause of most wheel bearing failures not caused by water intrusion. Clearance, in contrast can be felt as disconcerting
rattle when encountering road roughness.
To test for proper adjustment, install the wheel and wiggle the rim side-to-side to determine that there is no clearance (rattle), then let the wheel rotate freely to a stop. If the wheel halts with a short (indexed) oscillation, bearing preload is too high.
Although adjusting QR force is a safety consideration, it is also one of bearing life. It should be kept at a constant level once the desired closure force has been determined. Rear vertical dropouts require a lower and more predictable closure force than was formerly required with axles that could move forward from chain tension.
Because vertical dropouts do not rely on friction to resist chain load, many hubs now have smooth faced jam nuts that do not damage dropout faces as older knurl faced ones did."
from
http://draco.acs.uci.edu/rbfaq/FAQ/8c.7.html