Rims 101

The tire well or drop center (Marked in RED) is the feature that allows fitting of a tire onto the rim. The deeper the well the easier it is to mount the tire. As we make tubeless rims we start to have more strict limitations on the form of this tire well. The well needs to crowd the beads toward a narrow area to allow for hand pump inflation by offering a passive seal. For this reason, the well should have as small a flat bottom as practical.

Bead well (Marked in RED) is the most important part of the rim with regards to tire fit. This should be at the target diameter of the rim. So this should be Ø559 for MTB, Ø584 for 27.5 and so on. The shape can vary depending on clients needs, important features are width >3.2mm, hump size (0.5mm), and outer radii no larger than 1.7mm.

Crochet hooks are a relatively new development in rims. Originally developed for early foldable or high-pressure clinchers in the late 1970’s, these were intended to act as additional security in holding the bead in place. Hookless rims are neither a new idea nor absent from high-pressure tires. Both hookless and crochet hook rims are included in international standards for bicycle rims. We can build rims with or without hooks as per your needs. While we are not confident hooks perform any meaningful function, on rims using a proper bead shelf. Before the bead shelf design the hook served to capture the bead while inflation to be sure it did not escape the bead wall area of the rim. we build rims for clients and follow clients requests. For more information on hook function etc, click here.

We are confident that the key diameter that needs explicit definition and therefore must be better controlled is bead well, also called bead seat diameter. Insufficient coordination between rim and tire makers combined with no explicit tire bead diameter specification makes for less confidence among rim makers in the determination of proper diameters.  Ultimately outside diameters (in form of hook height) can vary considerably and maintain good tire fit provided that the tire well is deep enough to allow an easy fit. Often it’s been claimed that rim outside diameters are not consistent, but this is simply not true. Rim designs vary considerably, and with these designs, differences come varied diameters, even from a single brand. But in a single model these differences are quite small, usually within 0.5mm effective diameter or 1.57mm in circumference, well within the international standards.

We refer to the area where the spoke nipple passes through the rim as the spoke bed (shown in RED). Over the last decades, we have found a few nice methods of lowering stress concentrations in the spoke bed while maintaining a light weight profile. Even with these improvements, we would normally prefer not to make spoke beds <1.85mm thick, and even at this thickness we must always consider if we have single pitch spoke placement in the rim or grouped spokes, paired spokes etc. These things all have a large effect on localized stress at spoke holes.

Last but not least the geometry of the spoke bed had a significant impact on wheel stiffness. When a nipple can rock back and forth due to two point contact, the support offered by the spoke is limited. Each spoke nipple can act like a hinge. This is because a simple concave form to the spoke bed when rolled into a hoop form to make a rim becomes a messy “hyperbolic paraboloid”. This in a form most commonly seen as a pringle’s potato crisp. This form can not offer a firm foundation for a tapered cylinder that is a spoke nipple head, even the Polyax form does little to fix a hyperbolic paraboloid spoke bed. As can be seen in image below, left and right of nipple is in contact while fore and aft are gaps, this is the two point contact we speak of. We’ve created an elegant solution for this that eliminates the need for spoke washers, and eyelets as structural reinforcements.

Most extruders have a lower limit of 0.8mm for extrusion thickness.  Felloe’s founders have worked to lower this limit through refined tooling and tuning chemistry of the alloys. We often extrude thinner than 0.8mm and commonly at 0.7mm, in special circumstances, we process aluminum as thin as 0.45mm. The key driver to such thin profiles is the total distribution of mass so these thin sections require balancing all of the profile thicknesses.

Rim profiles must strike a balance between function and processability. The perfect rim might not be an extrudable profile, while the best profile to extrude is hardly well suited to high-performance rims. The image on here shows a CAD thickness analysis, this can also serve as an effective extrudability analysis.  During the design phase of any project we will work closely with you to find the best performance and maintain stable production.

rim extrusion tolerance standards

For about 35 years we’ve questioned the application of a roundness or circularity tolerance to a hoop such as a bicycle rim. While this can make sense in something without a center as a datum such as tubes, when you look at a rim, finding the center is both not easy, and hardly repeatable. The intent of standards in drawings, and tolerances is to allow engineers, and manufacturers to have an explicit set of definitions. So if a roundness measurement is varied each time it is performed on a single sample, we have a problem.
Our opinion is that the proper tolerance form for a rim is the profile of a line or profile of a surface. This allows an inspection to determine the fitness of a product to a specified form with very little error, while roundness or circularity as a standard requires equipment far larger than any wheel or rim maker I have ever worked with has.  This reduces an explicit definition for rim “roundness” to these incredibly few but powerful geometric tolerances.

Simply stated we must only measure the deviation from a predicted curve (in this case r296.5). This is expressed as a total range so a +/-0.5 is the same as the symbol profile of a surface and 1.0. Here we constrain the joint area to a more strict tolerance as a local range.