Roughing vs. Finishing Strategies in CNC Machining
Every CNC machining operation is a sequence of material removal stages: roughing removes bulk material as fast as possible, semi-finishing approaches final dimensions, and finishing achieves the specified tolerance and surface finish. UTEC Industrial provides precision CNC machining services for large and oversized industrial components in the Pacific Northwest, with in-house heat treatment and induction hardening integrated into the machining workflow. Each stage optimizes different objectives, uses different parameters, and may use different tooling. This article covers the objectives of each stage, the cutting parameters that differentiate them, stock allowances that connect them, and the decisions that determine how many passes are needed.
What is the fundamental purpose of the roughing stage and what parameters define it?
Roughing is the bulk material removal stage — its objective is to remove as much material as possible per unit time at the lowest cost per unit volume removed, while leaving a controlled stock allowance for subsequent passes. The parameters that define roughing: large depth of cut (0.100–0.500 inches in turning, 0.050–0.250 inches in milling), moderate to high feed (0.010–0.020 ipr in turning, 0.008–0.015 ipt in milling), cutting speed chosen for maximum tool life at the roughing depth and feed (typically 20–30% below the finish-turning speed for the same material and insert grade), and tooling optimized for toughness over wear resistance (ISO P35–P45 for steel turning, square shoulder mills with strong inserts for milling). Surface finish and dimensional accuracy are not objectives in roughing — the roughed surface may be Ra 125–500 µin with dimensional variation of ±0.005–0.020 inch, both of which are corrected by subsequent passes. The core efficiency metric for roughing is the material removal rate (MRR) in in³/min: MRR = depth of cut (inches) × width of cut (inches) × feed rate (in/min). Maximizing MRR within the machine's spindle power limits is the roughing optimization goal. For a 4140 turning operation on a CNC lathe with 15 HP available at the spindle: at 450 SFM, 0.015 ipr, 0.250-inch depth on a 6-inch diameter workpiece, MRR ≈ 6.7 in³/min, requiring approximately 10–12 HP — near the machine limit. This parameter set maximizes roughing productivity without exceeding available power (Machinery's Handbook, 31st ed., Industrial Press, 2020; ASM Handbook, Vol. 16, ASM International, 1989).
What determines the stock allowance left after roughing?
The stock allowance left after roughing — the material remaining on each surface before the semi-finishing or finishing pass — must be large enough to ensure the finish pass cuts below the roughed surface's damaged zone and work-hardened layer, but small enough to minimize finish-pass cycle time and insert wear. Minimum stock allowance for a finish turning pass: the finish pass must cut below the work-hardened layer produced by the roughing tool (typically 0.005–0.020 inch deep in alloy steel), the scale or chatter marks left by the roughing pass (0.002–0.010 inch), and any geometric error in the roughed surface (taper, runout, diameter variation — typically 0.003–0.010 inch in a well-controlled rough turning pass). The sum of these factors sets the minimum finish allowance at 0.010–0.030 inches per side for turning (0.020–0.060 inches on diameter). Standard practice: 0.020–0.030 inches per side for turning in steel (0.040–0.060 inches on diameter); 0.010–0.020 inches per side for milling flat surfaces. For parts that will be heat treated between roughing and finishing: larger allowances are required — 0.030–0.100 inches per side depending on part size and heat treatment distortion, as covered in the machining-hardened-steel and stock-allowance-post-hardening articles. If the allowance is too small (under 0.010 inches per side in steel): the finish pass intermittently rides over high spots left by the roughing pass, causing the tool to alternate between cutting and rubbing — degrading surface finish, accelerating wear, and producing inconsistent dimensions. The minimum 0.020-inch-per-side finish allowance is a practical floor for production alloy steel turning (Machinery's Handbook, 31st ed., Industrial Press, 2020).
What parameters differentiate finishing from roughing and why?
Finishing parameters are optimized for surface finish and dimensional accuracy, not MRR. The key parameter changes from roughing to finishing: feed rate is reduced to 30–60% of roughing feed (0.004–0.008 ipr in turning vs. 0.010–0.020 ipr in roughing), because surface finish Ra scales with feed² — halving the feed reduces theoretical Ra by 75%. Depth of cut is reduced to 0.010–0.030 inches, enough to cut cleanly below the roughed surface without inducing the tool deflection of a heavy cut. Cutting speed is increased by 10–20% relative to roughing speed to improve surface finish (higher speeds produce cleaner chip separation and reduce built-up edge tendency in the finish cut). Tooling is changed: a fresh insert or a new edge of a used insert is mandatory — a worn insert produces rougher Ra and causes dimensional drift in the finish pass. For turning: the insert nose radius may be increased (0.031 or 0.047 inches) to improve the theoretical surface finish at a given feed, or a wiper insert may be used for Ra under 32 µin without reducing feed below 0.005 ipr. For milling: a dedicated finish end mill or face mill with wiper geometry replaces the roughing tool — the wiper flat on the insert edge averages the feed marks from adjacent passes, producing Ra 16–32 µin at 0.006–0.008 ipt where standard geometry would produce Ra 32–63 µin. The depth of cut reduction in finishing also reduces cutting force — for precision bores and OD features held to IT7 (±0.001 inch), the finish pass must produce a consistent cutting force so that tool deflection is consistent and predictable. A light, consistent finish pass on a rigid setup produces dimensional consistency pass-to-pass within ±0.0005 inch; a heavy, variable-force finish pass produces ±0.002–0.005 inch variation (ASME B46.1-2019; Machinery's Handbook, 31st ed., Industrial Press, 2020).
When is a semi-finishing pass needed between roughing and finishing?
A semi-finishing pass is inserted between roughing and finishing when the roughing pass leaves too much stock for a single finishing pass to correct efficiently, or when the roughed surface has geometric errors (taper, runout, waviness) that would cause the finish pass to vary excessively in depth of cut. When semi-finishing is needed: workpieces with more than 0.500-inch diameter stock to remove — single-roughing passes rarely leave a sufficiently uniform surface for a direct finish pass at that stock level. Parts with high dimensional accuracy requirements (IT6 or tighter): the semi-finishing pass corrects the roughed geometry to within 0.005–0.010 inch of final size, so the finish pass operates at a consistent, predictable depth of cut rather than having to accommodate the geometric variation of the roughed surface. Parts with interrupted surfaces (keyways, cross-holes, flats) where the roughing pass produces localized geometric irregularities. Large-diameter turning (over 12 inches): centrifugal runout and the thermal growth from roughing can leave the roughed surface with 0.010–0.030-inch diameter variation — a semi-finishing pass corrects this before the finish pass. Semi-finishing parameters: intermediate depth of cut (0.030–0.060 inches per side), feed at 70–80% of roughing feed, fresh or less-worn insert. The semi-finishing pass is not attempting the final surface finish — it is creating a geometrically consistent surface for the finishing pass to work from. For a 4-inch diameter 4140 shaft being turned from 4.500-inch rough: rough to 4.060 (0.220-inch stock per side), semi-finish to 4.020 (0.020-inch stock per side), finish to 4.000 ±0.001 inch — three passes, each with a defined objective (Machinery's Handbook, 31st ed., Industrial Press, 2020; ASM Handbook, Vol. 16, ASM International, 1989).
How do roughing and finishing strategies differ for milling versus turning?
The roughing-finishing sequence applies to both milling and turning, but the specific strategies differ by operation. In turning (rotary workpiece, fixed tool), the roughing-finishing distinction is primarily in depth of cut and feed per revolution — the cutting geometry remains similar. The main strategic variation in turning is the use of multiple roughing passes at decreasing depths of cut as the part approaches finish size. In milling (fixed workpiece, rotating tool), roughing and finishing often involve completely different tool geometries and strategies. Roughing end mills: designed with serrated (corncob) flute geometry, large chip pockets, and high flute count for aggressive material removal — they are not suitable for finishing because the serrated flutes leave a rough, irregular surface. Finishing end mills: sharp, polished flutes with tight tolerances on diameter and runout, designed for light finishing passes at high speed. The roughing/finishing tool change is standard practice in milling, whereas in turning the same insert (indexing to a fresh edge) is more commonly used for both stages. In milling, adaptive roughing (also called high-efficiency milling or trochoidal milling) represents a specialized roughing strategy: a small radial engagement (15–25% of tool diameter) at a large axial depth (100–150% of tool diameter) and high feed rate. This strategy maintains constant chip load and constant tool engagement, extending roughing tool life by 50–200% compared to conventional full-width slotting at the same material removal rate — particularly effective for alloy steel and stainless steel where tool life in aggressive full-width cuts is short (ASM Handbook, Vol. 16, ASM International, 1989; Madison, CNC Machining Handbook, Industrial Press, 1996).
What is the role of thermal stabilization between roughing and finishing?
Heat generated during roughing causes the workpiece to expand — and if the finish pass is taken before the part has returned to ambient temperature, the finish dimensions will be measured on a hot, expanded part that will be undersized when it cools. The magnitude: 1045 and 4140 alloy steel expand at approximately 6.5 × 10⁻⁶ inch/inch/°F. A 6-inch diameter 4140 shaft that is 50°F above ambient from roughing heat will be approximately 0.002 inch larger in diameter than its room-temperature dimension — well within the ±0.001-inch finishing tolerance of a typical precision shaft. The appropriate approach: after roughing, allow the part to stabilize to within 5–10°F of ambient before taking finish measurements and finishing passes. For small, thin-section parts (under 2-inch diameter, under 12 inches long): stabilization takes 5–15 minutes with the part still in the chuck and spindle stopped (air cooling). For large-section parts (over 6-inch diameter, over 24 inches long): 20–60 minutes or longer depending on the roughing heat input and section mass. UTEC Industrial applies thermal stabilization as standard practice before finish turning on large-diameter 4340 and 4140 components — the practice eliminates the re-work that results from finishing a thermally expanded part to a dimension that drifts out of tolerance as it cools (Machinery's Handbook, 31st ed., Industrial Press, 2020).
How does the choice between one and two finishing passes affect quality and cost?
The decision between a single finish pass and a two-pass finish (semi-finish plus finish, or two finish passes) balances dimensional accuracy against cycle time and insert cost. A single finish pass is appropriate when: the roughed surface is geometrically uniform (consistent diameter or flatness within 0.005 inch); the required tolerance is IT8 or looser (±0.001 inch or looser); the finish stock allowance is 0.015–0.030 inch per side (enough to cut cleanly below the roughed surface without excessive depth-of-cut variation). A two-pass finish is appropriate when: the required tolerance is IT7 (±0.001 inch) or tighter; the part is large-diameter or has significant thermal growth concern; the material is stainless steel or other work-hardening grade where the first pass work-hardens the surface and a second pass at reduced depth cuts through the work-hardened layer to the final surface; or the roughed surface has geometric errors exceeding 0.010 inch that make a consistent single finish depth impossible. The cost of a second finish pass: for a typical turned shaft, a light spring pass (0.005–0.010-inch depth, 0.004–0.006 ipr, 30–60 seconds cycle time) adds $2–8 of machine time and insert wear — trivial relative to the risk of scrapping a finish-machined part that is out of tolerance. The rule of thumb: for any part held to IT7 or tighter, a two-pass finish strategy (semi-finish to within 0.005 inch, then spring pass to final size) is almost always justified by the reduction in scrap rate.
- Toolpath Strategy & Machining Efficiency — advanced roughing strategies for milling
- Cutting Tool Geometry: Rake, Relief, and Nose Radius — how geometry affects finish pass performance
- Machining Tolerances: What to Specify and What They Cost — the tolerance context that drives finishing pass selection
- Surface Finish Parameters: Ra, Rz, and What They Mean — the surface finish measurements the finishing pass must achieve
References
- Machinery's Handbook, 31st ed. Industrial Press, 2020.
- ASM International. (1989). ASM Handbook, Volume 16: Machining. ASM International.
- ASME B46.1-2019: Surface Texture (Surface Roughness, Waviness, and Lay). ASME.
- Madison, J. (1996). CNC Machining Handbook. Industrial Press.
- Altintas, Y. (2012). Manufacturing Automation, 2nd ed. Cambridge University Press.
Need Precision CNC Machining?
UTEC Industrial provides large-scale CNC machining services from our 25,000 sq ft facility in Spokane Valley, WA — equipped with Mazak, Monarch, and Mori Seiki machining centers, plus a gantry bandsaw cutting sections up to 50" × 84".