From Design to Manufacturing: Essential CNC Machining Tolerance Knowledge You Must Know

In the field of precision manufacturing, if 2D drawings are considered the language of engineers, then tolerances are the “hidden language” within them. They determine whether a product can be assembled smoothly, remain stable and reliable, and whether production costs can be effectively controlled. Many designers and engineers focus primarily on dimensions at the drawing stage, often overlooking the specification of tolerances. In actual production, the reasonableness of tolerance selection can directly affect the performance and cost-effectiveness of the finished product.

For example, assembly holes used for fastening screws typically have their diameters controlled according to the ISO 2768-mK standard. If the tolerances are too tight or too loose, proper assembly may be prevented, thereby increasing production costs and delivery time.

CNC machining is known for its high precision and efficiency, but even so, the produced parts must still follow certain tolerance standards. Tolerance refers to the allowable deviation range of a part’s dimension, that is, the acceptable difference between the actual dimension and the design dimension.

The standard prototyping and production tolerance at Horizon factories is ±0.1 mm. For metal parts, we typically follow ISO 2768-1:1989 at fine (f) grade, while for plastic parts we apply the medium (m) grade.

If higher precision is required, we can provide a standard precision machining tolerance of ±0.05 mm. For specific features, hole tolerances can be held to ±0.02 mm, and positional tolerances can be maintained within ±0.05 mm when features are located on the same side of a part.

In addition, depending on the geometry and material of the part, we can often achieve even tighter tolerances. If your design requires special tolerances, please clearly specify them on your drawings or models when uploading files for quotation.

As shown in the figure above, we use three types of notation: Φ4±0.05 mm represents a bilateral symmetrical tolerance, Φ4+0.1/-0.05 mm represents a bilateral asymmetrical tolerance, and Φ4.1/Φ3.98 represents a limit tolerance.

All of the above notation methods are acceptable, but they must be clearly indicated on the design drawing. At the same time, attention should be paid to cumulative tolerance issues, and tolerances should be reasonably allocated across different stages to ensure that the final functional dimensions meet design requirements.

In part machining, tolerances involve not only dimensions (length, width, hole diameter, etc.) but also surface roughness, which has a significant impact on a part’s functionality, assembly, and appearance. Under standard machining conditions, the surface roughness of flat and perpendicular surfaces is typically 63 µin (approximately 1.6 µm), while curved surfaces are 125 µin (approximately 3.2 µm) or better. This level is sufficient for most functional parts. Surface roughness directly affects friction, wear, sealing performance, and fatigue life. Functional surfaces usually require lower roughness to ensure reliable fit.

For aesthetic surfaces, especially on metal parts or consumer products, visual appearance and tactile quality can be improved through light bead blasting, fine shot peening, or polishing. During the design phase, roughness levels should be selected appropriately according to part function and appearance requirements, and clearly indicated on drawings (e.g., Ra, Rz, or µm). When necessary, the machining method or surface treatment process should also be specified. Additionally, the effect of surface roughness on dimensional accuracy, fit, and assembly must be considered, with cumulative tolerances allocated reasonably to ensure that the final assembly and performance meet design requirements.