Machined Crank Arms and Pedal Components
A crank arm is a long lever with an envelope fixed at both ends, by the bottom bracket on one side and the rider's foot on the other. There is no room to solve a strength problem by adding aluminum, which is why cranks were forged long before anyone owned a CNC machine and why machined cranks are the part most likely to fail a fatigue test.
What this covers
- Crank arms, single and pairs
- Spiders and direct mount interfaces
- Pedal bodies and pedal axles
- Crank bolts, caps and preload hardware
- Crank boots and protectors
Interfaces you do not get to choose
- Pedal thread9/16 in x 20 tpi, right hand drive side, left hand non-drive
- Modern spindle typesHollowtech II 24 mm, DUB, GXP
- Legacy spindle typesISIS 22 mm 10-spline, Octalink, square taper 2 degrees
- BB shell widths68, 70 and 73 mm threaded
- Crank length range155 to 180 mm, 170 and 175 most common
Full tables in the standards reference.
Material choice
What each option actually buys you on this part, rather than in general.
| Material | Why it gets used here |
|---|---|
| Aluminum 7075-T6 | The normal choice for a machined crank. Roughly 570 MPa tensile against 310 for 6061, and the section cannot grow. |
| Forged 6066 or 7050 then machined | What most volume cranks actually are. Grain flow follows the shape, which is worth more in fatigue than alloy strength alone. |
| Titanium Grade 5 | Axles and hardware. Rarely worth it on the arm itself, where stiffness rather than strength sets the section. |
Where the tolerance belongs
These are typical working ranges, not promises. Tolerance these dimensions and leave the rest at a general tolerance, because every step tighter than you need buys machine time you will pay for.
| Feature | Typical | Why it matters |
|---|---|---|
| Spindle interface | ±0.02 mm | A loose interface creaks and then fails. Follow the spindle standard exactly, do not reinterpret it. |
| Pedal thread position | ±0.1 mm, perpendicular | An angled pedal thread loads the axle in bending for the life of the part. |
| Q factor and arm length | ±0.2 mm | Fit dimensions, measured center to center. |
| Chainring mount face | Flat, ±0.05 mm | Sets the chainline together with the ring offset. |
Finish
Type II only, unless you have tested otherwise. Published fatigue work on 7075-T6 cranks under European test conditions found cycles to failure falling as anodized depth rises. Test the part in the finish you will ship.
Realistic quantities
Machined cranks stay economical below roughly 2,000 pieces a year. Above that, a forged blank plus machining almost always wins on both cost and fatigue performance.
What goes wrong on this part
Failure modes specific to this component, collected so you can design them out rather than discover them.
- Anodize thickness is a test variableThis is the single most under-appreciated fact about machined cranks. Hard anodizing a crank because it looks tougher makes it worse where it matters.
- Aluminum has no fatigue limitThere is no stress level that runs forever. Design to a cycle count, not to a static safety factor.
- Section changesBlend them over a length. A step in the arm is where the crack starts.
- Test budgetISO 4210 fatigue testing consumes samples. Budget the pieces and the lab time before you freeze the design.
What to send
A drawing with these on it gets quoted in a day. One without them gets quoted with padding in it, or gets four rounds of questions first.
- STEP file plus a PDF drawing
- Spindle interface named exactly, with its standard dimensions
- Crank length and Q factor
- Pedal thread and its perpendicularity requirement
- Whether the part will be fatigue tested and to which standard
- Mill certificate requirement, stated up front
Printable version: the RFQ checklist.
Go deeper
- Cost
CNC or Forging for Bike Parts: Where Cost Crosses
Forging is cheaper per part and stronger. Machining has no tooling bill and no lead time on revisions. Here is where the two cross for a bike component.
- Standards
ISO 4210 and EN 15194 for Machined Bike Components
What ISO 4210 and EN 15194 actually test, which machined components they touch, who is responsible and how finish and material choices affect the result.
- Design
Weight vs Strength in Machined Bicycle Components
Where material can leave a machined bike part and where it cannot. Section shape, stress concentration, aluminum fatigue and why the last 10 grams cost the most.
Other parts
Stems and faceplates
Two clamp bores against fixed standards, a faceplate that has to close evenly and a rider on top of it.
Chainrings and spiders
A flat part with the tightest tolerance on the site. Tooth width decides whether a 1x chain stays on.
Hub shells and freehubs
Turning plus milling in one part. Concentricity between the bearing seats decides whether it drags.
Brackets, mounts and hardware
The long tail. Low part cost, fixed interfaces, and finishing batch minimums that dominate the quote.
E-bike drive hardware
Machined like a bike part, loaded like a light vehicle part, regulated like an electrical product.