INDEPENDENT BICYCLE DEVELOPMENT
The Wallasch Bike Project
Project direction: Jeffrey Wallasch, Wallasch Precision Manufacturing.
Our goal is to build a bike company. This is where we share the design thinking, mechanical studies, and practical design and manufacturing work behind that ambition.
The project is still being developed and reviewed. It is not a ride-ready bicycle or a released production design.
Built one considered decision at a time
We are developing an aluminum frame concept with attention to suspension geometry, service access, manufacturability, and how the individual parts work together. The work moves between CAD, mechanism analysis, and practical design review as we identify the next questions to resolve.
All profits from this business, after operating expenses, are reinvested in the Wallasch Bike Project.
HISTORICAL SUSPENSION STUDY
Suspension development in Linkage
We used Bikechecker Linkage X3 to establish the rear suspension mechanism and explore how a small junction change affected its calculated behavior through the travel. This historical comparison is separate from the later R11 adjustable-pivot, service and manufacturing studies shown elsewhere on this page.
Getting the mechanism right
The model represents a Horst-link four-bar. The main frame carries the fixed pivots and forward shock mount; the chainstay connects the frame to the Horst pivot near the dropout. The upper rear assembly carries the rear axle and moving shock eye, while a short rocker connects that assembly back to the frame.
An important setup correction attached the moving shock eye to the rear bar carrying the wheel, rather than the rocker arm. That changed which rigid body drives the shock. Early failed calculations were setup troubleshooting, not results for the corrected design. The corrected inputs used a 230 mm compression shock with 65 mm maximum stroke and its fixed eye on the main frame.
Travel and the leverage curve
In the archived original rear-only compression state, Linkage displayed 170.0 mm vertical rear-wheel travel, 171.4 mm along the curved axle path and 65.0 mm shock compression, with zero front-fork travel. Vertical wheel travel and distance along the axle path are different measurements.
ORIGINAL CONFIGURATION
22.1% progressivity
The displayed leverage curve starts at approximately 2.93, falls to about 2.24 near the end, then rises slightly to about 2.28. The small late upturn means the ratio was not decreasing throughout the entire displayed travel.
LOWERED JUNCTION
27.3% progressivity
Lowering the rear-bar/rocker junction by 10 mm retained the four-bar architecture and shock attachment. The revised displayed leverage curve runs from approximately 2.90 to 2.11, with the original late upturn absent.
Leverage ratio describes wheel movement per unit of shock movement. A falling ratio means progressively more shock movement for each additional millimeter of wheel travel. The comparison helped identify a development candidate; it did not determine a finished shock tune or prove ride performance.
Pedaling and braking tradeoffs
We examined chain growth, pedal kickback and anti-squat across several gears, including 34/24, 34/51 and 34/10. These effects depend on the selected gear and the model assumptions, so results from different gears cannot be treated as a like-for-like improvement.
In 34/51, the lowered configuration's anti-squat curve declines from approximately 131% to 90%, broadly similar to the original. Its preserved tooltip reads 95.1% at 152.0 mm wheel travel; that is not the 170 mm endpoint. The lowered 34/51 kickback curve reaches approximately 31°, while chain growth is about 28 mm for the upper run and 55–56 mm in total.
The original anti-rise graph ends around −25% at its displayed 171 mm endpoint. The lowered graph remains positive across its displayed range and ends around 16% at 170 mm. These are approximate graph readings at the stated endpoints, not measured braking results.
Chain-length trials also produced short-chain warnings in a large sprocket and a long-chain warning in a small sprocket. Suspension travel, chain length and derailleur take-up therefore remained an unresolved system-level check. No final chain length or real derailleur compatibility was established.
What these results can tell us
The comparison used an assumed combined rider-and-bike center of gravity and the app's Horizontal mode. Gear selection, front geometry, center of gravity and operating mode all matter when comparing these curves. The spring settings entered during the study were model inputs, not a released shock recommendation.
A later, separate rounded-coordinate C206/R6 calculation returned approximately 169.929 mm travel from 65 mm shock stroke and leverage of approximately 2.900 to 2.112. It supports the geometric trend but is not a Bikechecker export or a re-run of the saved Linkage projects. The later R11 plots below likewise belong to their own CAD study.
Development evidence only. The archived screens document a functioning mechanism model and the effect of the lowered junction. Strength, loads, fatigue life, bearing and weld details, drivetrain compatibility, shock tune and ride safety still require separate qualification and physical testing.
How the model works
These animations show existing CAD development models, not a tested bicycle or a workshop procedure. The R11 suspension baseline and R12 service study are shown separately from the later R14D yoke candidate.
How the linkage moves
The fixed front frame carries two pivot locations. The chainstay link rotates from the main pivot and connects to the upper/rear body. A rocker connects that body back to the fixed frame, forming a planar four-bar linkage. The rear axle moves with the upper/rear body; the shock changes length as the linkage moves.
The animation follows stored solved positions for three settings: neutral and provisional main-pivot offsets of minus or plus two millimeters. Each setting is shown separately; the cut between settings is not an adjustment under load. A removable keyed cartridge changes the pivot location, with matching caps identifying the setting. Radial location, axial cap retention and the axle's bearing-clamp stack have distinct roles. Actual bearings, fits, fasteners, preload and retention remain unresolved.
The model assumes rigid links and a shock-length range of 230 to 165 mm. Its calculated vertical axle travel is approximately 168.03 mm at minus two, 169.93 mm at neutral and 171.87 mm at plus two. Playback timing is illustrative: this is not a damping, speed, force, flex, pedaling-efficiency or ride-performance simulation. Reference parts are simplified; real-component compatibility and continuous clearances are not established.
Cartridge withdrawal and return path
Blue identifies the cartridge, pink the nominal bearing envelopes, grey the spacers, and orange the setting-matched R12 drift concept. The translucent housing and lower transition provide context. The moving group translates along the existing studied axial path. Reversing that path illustrates where the group would return; it does not validate installation, pressing, bearing contact or assembly force.
The view starts after caps and fasteners, axle hardware, the chainstay link and associated drivetrain items have been omitted. Those prerequisite removals are assumptions of the service study and are not demonstrated or validated here. Other frame and component geometry is hidden for clarity. The bounded sweep evidence concerns the housing and lower transition, not whole-bike service clearance.
The R12 tools match their pivot settings. A neutral-only tool must not be treated as suitable for both offsets. Existing axle/driver/nut-tool checks remain neutral-only; offset access is unverified. Real races, seals, fits, fixture alignment, tool stiffness and pressing forces remain open. This is a partial disassembly/return-path visualization, not complete bicycle disassembly or assembly instructions.
Where the R14D yoke fits
The separate R14D still shows an integrated recessed web joining nodes on the same upper/rear rigid body while retaining articulated joints. It is not silently installed in the R11 animation. R14D remains a review candidate with unresolved roots, machining and fatigue qualification; the R14E blend trial remains held. The motion clips do not demonstrate yoke removal or establish its service access.
No structural, fatigue, weld, heat-treatment, manufacturing, service or ride approval is claimed.
Development work so far
R11 — one cartridge, two retaining caps
The concept uses one through-cartridge (AP01), a right retaining cap (AP02R) and a left retaining cap (AP02L). Both sides show the same assembly, not two independently adjustable cartridges.
Three matched settings. Neutral, −2 mm and +2 mm each use a matching set of all three parts. Changing the setting replaces the cartridge and both marked caps together. The housing, exterior key profile and six cap-screw positions remain fixed; the bearing seats and cap openings shift the pivot axis. These are discrete replacement sets, not an on-bike adjustment knob.
Different retention roles. Key surfaces locate the cartridge radially. The two caps, with three screws per side, retain it axially. A separate axle, spacer and bearing-inner-ring stack provides the clamp arrangement. Actual bearing races, seals, fits, hardware and preload remain unresolved.
Orientation still matters. Side, signed setting and upward marks must agree. The asymmetric key limits half-turn rotation but does not prevent end-for-end reversal, so an installed-axis check remains necessary.
Service evidence is partial. R12 studies a bounded withdrawal and return path after assumed prerequisite removals. It does not validate whole-bike clearance, bearing extraction forces or a safe changeover procedure. Static separation views explain the parts; they are not service instructions.
What the motion model calculates
For an assumed shock length of 230 to 165 mm, the calculated vertical rear-axle travel is 168.03 mm at −2 mm, 169.93 mm at neutral and 171.87 mm at +2 mm. The leverage plot is the local change in vertical axle travel divided by shock compression; it uses adjacent differences from 651 poses per setting. The axle-path plot starts each setting at its own top-out position. These calculations describe geometry only. They do not predict damping, flex, loads, anti-squat, braking response, fatigue or ride performance.
R12 — service checks need their own scope
A neutral service drift cannot be assumed suitable for both offset settings. In the model, the legacy neutral drift overlaps the nominal bearing end-face envelope at either offset; the setting-matched R12 concepts remove that modeled contact. Separate straight-withdrawal envelopes were checked for each setting. This does not verify real bearings, seals, fits, pressing forces or the prerequisite disassembly. Offset axle, driver and nut access remains unverified. This is a design finding, not a servicing procedure.
R14D — integrating the upper yoke
The current yoke review candidate connects two nodes on the same upper/rear rigid body with a recessed web while retaining the articulated joints. Optional weight pockets were removed after review found that they cut into the web roots. The resulting candidate has no sealed cavities and an estimated mass of 421.78 g, based on modeled volume at an aluminum density of 2.70 g/cm³. Across 1,953 sampled poses, its added material produced no detected new contacts; 21 exact-solid pose spots also found no added interference. These bounded checks are not continuous clearance, tolerance, flex or whole-frame validation. The roots, actual hardware and full machining process remain unqualified.
R14E — recording a trial that was not adopted
A root-blend trial produced a valid solid but left four sharp junctions. Six of 99 sampled finishing-tool checks exceeded the source screening threshold of 0.01 mm3 summed tool-envelope overlap; smaller overlaps also remain. Larger full-root blends and subsequent local fillet attempts did not yield an acceptable result. R14E is held, and R14D remains the current review candidate. Neither visual smoothness nor a passed movement check establishes fatigue safety.
What remains to be established
Further work needs actual shock and pivot hardware drawings, bearing and retention details, final kinematics and clearances, material and weld/heat-treatment data, defined loads and fatigue spectra, and a verified machining and test plan. No structural, fatigue, weld, heat-treatment, service or rideability approval is claimed.
Adjustable pivot, up close
One through-cartridge. Two retaining caps. These R11 CAD views show the matched parts, the three provisional settings, and how the same assembly looks from both sides.
Concept illustrations: interfaces, bearings, retention, manufacturing and service remain unqualified. Open the full-detail plates to inspect small setting marks.
One cartridge and two caps
Three matched settings
The same assembly, from both sides
Seating and retention
Exploded cap detail
CAD and mechanism studies
R11 frame concept
R11 adjustable pivot
Modeled shock and wheel travel
Modeled leverage ratio
Modeled rear-axle path
R12 service finding
R14D yoke review candidate
R14E held root-blend trial
Manufacturing development
Preliminary design and manufacturing review. These concepts are not released for manufacture or riding.
Documented: the R11 manufacturing scope combines machined sleeves, shells, pivot housings, tabs, links and compact junction pieces with hollow tube or formed members. Its planning register separates 18 parts in a preliminary aluminum machining scope, nine tube/formed pieces, and eight material-unresolved pivot/cartridge pieces. These are R11 planning categories, not an approved production bill of materials; later yoke studies are separate candidates. Purchased suspension, wheels, dropper, controls and drivetrain remain separate from custom frame manufacture. Modeled reference-group counts are not purchasing quantities, and their interface compatibility is not established.
Proposed routes include turning sleeves and shells, provisional three-axis or multiple-setup machining for joints and links, and cutting/mitring tube stock. The lower down-tube transition and shaped main tubes still need a fabricator-confirmed hollow-stock or forming route. Raised Wallasch marking is design intent with its manufacturing method unresolved. The economic aim is to concentrate machining at interfaces while using hollow members between them. No verified cost or weight advantage is claimed; forming, fixtures, setup count and finishing still need evaluation.
Machining and assembly approach
Documented: R13 corrected the distinction between a visual/interface envelope, a pre-weld blank and a finished functional part. The existing bottom-bracket shell and housing hole envelopes are not machining-pilot definitions. That review preserved the baseline geometry rather than inventing stock allowances or final fits.
Proposed sequence: select the real components and fabrication route; define separate blanks with justified cleanup stock; fixture and weld using a qualified process; complete the approved thermal/distortion-control sequence; then finish functional bores, faces, threads and locating features as required. Final machining and inspection are separate work from raw-part machining. Removable pivot internals and purchased components are to be excluded from thermal operations. This is a planning sequence, not an assembly procedure: setups, allowance values, gauges, fastener selection, locking and torque remain unapproved.
Adjustable-pivot cartridge
Documented: one removable through-cartridge (AP01) locates the main pivot within a housing. Right and left retaining caps (AP02R/AP02L) match its selected setting. The key surfaces provide radial location; the caps provide axial retention. Cap retention and the axle's bearing-inner-ring clamp stack are separate design questions. The concept includes neutral and provisional offset settings.
Unresolved: actual bearing geometry, materials, removable fits, race contact, spacer lengths, cap seating, fastener engagement and worst-case tolerance stacks. Keying does not eliminate every assembly-orientation error. The R12 service study found that tooling must match the pivot setting; nominal modeled contact and withdrawal checks do not establish real bearing serviceability. Offset axle, driver and nut access remains unverified.
Integrated-yoke tradeoffs
Documented: the R11 upper-node geometry contained sealed cavities that block exact conventional billet machining. The separate R14D candidate instead uses an integrated recessed web joining nodes on the same moving body while retaining articulation. Its model has no sealed cavities. Optional pockets were removed after the study found loss of material at the web roots. Filling them removes machining work but increases modeled mass; it does not qualify the roots.
The R14E blend trial remains held. Residual sharp junctions and failed sampled tool access prevent treating its smoother appearance as an accepted improvement. Those checks used an assumed tool envelope, not an approved cutter, fixture or complete machining path. Smaller tools or alternative access would introduce further process tradeoffs and have not been adopted as a solution. R14D remains a review candidate, with root design, machining strategy and fatigue behavior unresolved.
Material, welding and heat treatment
Documented: 6061-T6 was used as a feedstock pricing assumption for a limited machining scope. It is not approval of the welded frame's alloy, final temper or properties. Pivot/cartridge materials remain unresolved. No filler, welding procedure, thermal schedule or final material condition has been released, and simple aging is not assumed to restore the required condition.
Required decisions include material traceability, weld preparation and access, procedure qualification, fixtures, distortion evidence, thermal route, final-machining allowances and inspection. The illustration colors are presentation choices rather than a released coating or finish specification.
Gates before manufacturing or ride release
The project still needs confirmed component drawings and interfaces; final frame size, kinematics and clearance budgets; functional fits and thread/gauge definitions; a complete tool/holder/fixture plan; and an approved material, welding and thermal route. Structural and fatigue work needs defined use, system mass, correlated load cases, material and weld data, and a qualified analysis/test program. Valid CAD solids, sampled movement checks and dimensional inspection cannot substitute for that work.
Development, with the limits visible
These CAD models and bounded calculations document design decisions. No structural, fatigue, weld, thermal-process, service or rideability approval is claimed.
Designs and results may change. Geometry files and manufacturing prints are not provided on this page.
Have a design challenge? Talk to WPM →