Wheel-X was manufactured in a facility in Mexico, off-site from the company’s USA headquarters.  One of the steps in the production process was bonding a single-shear lap-joint between an overlapping portion of the carbon fiber rim sidewall and the aluminum rim channel. ASTM International, formerly known as American Society for Testing and Materials, is an international standards organization that develops and publishes voluntary consensus technical standards for a wide range of materials, products, systems, and services has a reference for adhesively bonded lap joints.  

This single-shear lap-joint between the carbon rim sidewall and aluminum rim channel had multiple potential failure modes (all of which could be minimized or eliminated through rigorous manufacturing process control- but they weren’t), and the result of the joint failure meant that the aluminum channel separated from the carbon-fiber sidewalls.  There were several instigating events which caused the onset of the failure.  None of these instigating events would have caused a failure onset if one step of the manufacturing process would have been rigorously monitored to ensure that the process was in accordance with industry standards, and accurately repeatable. 

Here’s a description of how the process was supposed to go:

  • In the originally designed production process a synthetic layer of material known as peel-ply was applied to the uncured inward facing portion of the rim surface in the lay-up process and then co-cured, to be later removed just prior to bonding, exposing a clean, rough bonding surface on the inside of the carbon rim.  
  • In the originally designed production process the surface of the aluminum rim channel was chemically etched.  This etching process creates a rough and clean bonding surface.  Once etched, the aluminum rim channel must be sealed and stored in an inert environment, only exposing the etched surface to the ambient air immediately prior to bonding.

Here’s a description of how the process failed:

Eventually, in an effort to increase the manufacturing pace, reduce ancillary costs of materials, and decrease the aptitude required in the manufacturing process both steps 1 and 2 were omitted, and a substitution was created.

The substitution process was a follows:

Both the carbon-fiber rim and the aluminum channel were treated for bonding using a glass-bead blasting process which eventually gave-way to a sand-blasting process. glass bead vs. sand blasting

This process lacked the preparation for bonding in multiple ways:

  • The substrate surface texture was inappropriate
  • The sandblasting actually damaged and weakened the aluminum
  • The blasting material, initially glass, and later sand did not meet cleanliness standards, and the rims required subsequent cleaning (a step that was not conducted properly or was omitted) prior to bonding

There were several options encompassing the design and the manufacturing, and this includes process control.

  • The carbon sidewall-to-aluminum rim bond could have been eliminated and the rim could have been molded in one-piece
  • The bonding process could have been an automated or semi-automated process so that a human being could not have made changes and substitutions.
  • The process control and manufacturing oversight could have been such that when chemical etching changed to glass bead or sand blasting, an alarm bell would have sounded to ask the question, why are we not purchasing etching chemicals, and why are we purchasing glass beads or sand?
  • There could be non-destructive testing (NDT) to ensure the bond integrity if the bond was identified as a failure point that could induce a catastrophic event.

The list could go on-and-on; In hindsight the options to manufacture a reliable and safe wheel were many.

Paul Lew, "Bicycles make friends"