A large tool clears material fast but cannot reach into small features. A small tool reaches everywhere but removes material slowly and breaks when overloaded. Every CNC program in every material has to reconcile those two facts, and the way that reconciliation is handled determines both cycle time and how long the small tools survive. In stone, where the tools are diamond and expensive and the workpiece is a slab that cannot be replaced from a bar stock rack, getting it wrong is costly in two directions at once.
Rest machining is the strategy that resolves the conflict properly. Rather than sending a finishing tool across the whole part, it directs that tool only at the material a previous, larger tool was physically unable to reach. Every other region is skipped, because it is already at size. The result is less air cutting, dramatically lower and more consistent load on the small tool, and a cut quality that does not vary depending on how much stock happened to be sitting in a given corner.
What Rest Machining Means and Why It Exists
A roughing pass with a large tool leaves a part that is close to shape but not identical to it. Wherever the finished geometry has an internal radius smaller than the roughing tool's radius, the tool could not enter, and material remains. The same happens at the bottoms of features shallower than the tool could reach and in narrow slots the tool could not fit. That leftover material is the residual stock, and it is distributed unevenly across the part.
A naive finishing strategy ignores this and simply runs the small tool over the entire surface at a constant programmed feed. In the large open areas the tool is cutting nothing but air or a whisker of stock. In the corners it suddenly encounters everything the roughing tool left, which can be several times the depth of cut the finishing pass was programmed for. The load on that small tool spikes exactly where it is least able to tolerate it.
Rest machining calculates where material actually remains, using a model of what the previous tool removed, and generates a toolpath that visits only those regions. In stone terms, the profile bit or small-diameter tool spends its life cutting the corners it was brought in for, at a consistent engagement, rather than travelling the whole part hunting for work and then being ambushed in the corners.
Why This Matters More in Stone Than in Metal
The Tooling Is Brittle and Expensive
Diamond tooling does not behave like carbide under a sudden overload. A metal cutting tool that meets unexpected stock will usually deflect, chatter and perhaps chip. A diamond tool meeting an unexpected step in stone can lose segments, crack a core or shatter, and it takes the workpiece surface with it. The cost of a single such event, counting the tool and the slab, generally exceeds whatever cycle time was saved by not programming the strategy properly.
Load consistency is what protects the tool, and rest machining is fundamentally a load-consistency strategy. By ensuring that the small tool encounters a predictable amount of material everywhere it cuts, it lets you choose a feed and depth that are genuinely appropriate rather than a compromise between the light regions and the worst corner.
The Workpiece Is Irreplaceable
A machined metal part that fails can be remade from the next piece of bar. A slab is a unique piece of natural material, frequently selected by a client for its specific figure, and often the only one of its kind in the building. That asymmetry should push stone programming toward conservatism in exactly the places where a metal programmer would push for speed. Rest machining lets you be conservative where it counts without paying for that conservatism across the whole part.
The Material Is Not Uniform
Stone contains harder and softer minerals, healed fractures, and occasional voids. A tool encountering variable material at a consistent engagement can cope; a tool encountering variable material at a variable engagement is being asked to absorb two sources of uncertainty at once. Removing the engagement variable through rest machining leaves only the material variable, which is the one you cannot control anyway.
Coolant reach is a further reason rest machining suits stone. A small tool working deep in a corner is in the hardest place on the part for water to reach, and diamond tooling starved of coolant fails quickly and expensively. By limiting the time the small tool spends in those confined regions to only what is genuinely necessary, the strategy reduces the total exposure to marginal cooling rather than relying on the coolant system to perform perfectly for the whole of a long finishing pass.
Chip and slurry evacuation follows the same logic. Residual stock sitting in a corner becomes slurry that has nowhere obvious to go, and a tool recutting its own slurry wears faster and produces a poorer surface. Directing the tool through those regions in controlled passes, with adequate flushing between them, keeps the cutting zone cleaner than a single heavy pass that fills the corner and then works in the resulting paste.
| Strategy | Small Tool Behavior | Consequence |
|---|---|---|
| Full finishing pass everywhere | Mostly air cutting, then heavy load in corners | Long cycles and unpredictable tool failure |
| Rest machining | Consistent engagement only where stock remains | Predictable load, longer tool life, shorter cycle |
| Roughing tool sized close to final radius | Little residual stock but slow roughing | Poor overall cycle time |
| Multiple intermediate tools | Stock reduced in stages | Best load control; more tool changes |
| No roughing at all | Small tool removes everything | Very long cycles and heavy tool wear |
Pro Tip
Have your CAM system display the residual stock model before you post the program. Most packages will show the remaining material as a shaded solid after each operation. Looking at that picture for thirty seconds tells you where the finishing tool is going to be loaded, whether an intermediate tool is warranted, and whether any region has more stock than the tool you selected can safely take. That check catches more potential tool failures than any amount of parameter tuning after the fact.
Programming Rest Machining Well
Keep the Stock Model Honest
Rest machining depends entirely on the software knowing what material is actually left, which means the stock model must reflect reality. If the roughing operation was edited, if a pass was skipped, if the operator adjusted a depth at the machine, or if the raw stock definition does not match the slab in front of you, the calculated residual stock will be wrong and the finishing tool will meet something the program did not predict.
Verify stock definition against the actual material at setup, and treat any manual change at the machine as invalidating the model. Where a program has been modified on the shop floor, the safest response is to regenerate the finishing operations from the amended roughing rather than assuming the difference is small.
Consider an Intermediate Tool
Where the gap between the roughing tool and the finishing tool is large, the residual stock in the corners may still be more than the small tool should take in one pass, even with rest machining directing it correctly. Introducing a mid-size tool that rest-machines after roughing, followed by the small tool rest-machining after that, steps the stock down in manageable increments.
The trade-off is an extra tool change and an extra operation against reduced risk on the most vulnerable tool in the program. On expensive slabs and on tooling with long lead times, that trade is usually worth making. On simple geometry with modest corner radii it may not be, and the residual stock model is what tells you which situation you are in.
Entry, Exit and Direction
Corners are where rest machining sends the tool, and corners are also where engagement geometry changes most sharply. Program entries into those regions with an arc or a ramp rather than a straight plunge, since a plunging entry into a corner full of residual stock is close to a worst case for a diamond bit. Similarly, plan exits so the tool leaves through material rather than breaking out at an unsupported arris.
Climb and conventional direction both have advocates, and the right answer in stone depends on the material and the finish required. What matters more is consistency: mixing directions within a rest machining operation produces varying surface quality and varying load, which undermines the whole point of the strategy. Choose one, prove it on the material, and apply it uniformly.
Verification, Load Data and Long-Term Practice
Simulate before you cut. A full material-removal simulation will show both the residual stock the finishing tool is expected to encounter and any collision risk introduced by directing a short tool into deep corners. Reach and holder clearance become live concerns in rest machining precisely because the strategy sends the tool into the confined geometry the larger tool avoided.
Pair rest machining with load monitoring where the machine supports it. The two are complementary: rest machining makes load predictable, and monitoring verifies that the prediction is holding. A load trace that shows a spike where the model predicted uniform engagement is direct evidence that the stock model and the reality have diverged, which is exactly the failure this strategy is meant to eliminate.
Record what worked. Cycle time, tool life in linear feet or parts, and the residual stock depth you accepted for a given tool and material together form a set of parameters you can reuse. Stone shops frequently reprogram similar geometry from scratch each time, which discards the tuning that previous jobs paid for. A short library of proven operations for common profiles and features is one of the highest-return investments a programmer can make.
Review tool life against the strategy rather than against the calendar. If a small profile tool that used to complete a certain number of jobs starts failing early, look first at whether roughing changed, whether the stock definition drifted, or whether someone edited a depth. Tool life on a rest-machined operation is a fairly sensitive indicator that something upstream has moved.
There is a quality argument as well as a tooling argument. A finishing tool that meets varying stock produces varying deflection, and deflection translates directly into dimensional variation on the finished profile. On an edge that will be viewed along its length under directional light, that variation reads as a wavering line. Consistent engagement is what produces a consistent profile, and rest machining is how consistent engagement is achieved on geometry with mixed radii.
Fixturing interacts with the strategy too. Rest machining concentrates cutting force in specific regions rather than distributing it across the part, which can produce localized lifting or movement if vacuum pod placement was planned around a uniform finishing pass. Review pod layout against where the rest machining operations will actually apply force, particularly on narrow pieces and on parts with large cutouts.
Train the whole programming team on the same conventions. Rest machining depends on a chain of operations behaving as expected, and a program written by one person and edited by another with different assumptions about stock is where the model quietly breaks. Documented conventions for stock definition, tool sequencing and how modifications are handled are what keep the strategy reliable across a team.
The underlying principle generalizes well beyond CNC work. Direct the expensive, fragile resource only at the work it uniquely can do, and let the robust, cheap resource handle the bulk. That is as true of a profile bit in a corner as it is of a skilled fabricator's time on a shop floor, and programming that reflects it produces both better parts and lower costs.
The strategy protects tooling, but the tooling choice comes first. Browse our CNC tooling and router bits by material and operation, and reach out if you need help matching bond and geometry to the features you are machining.
Refining your CNC programming or replacing tooling that keeps failing early? We can help you match bits and bonds to your material and your machining strategy.
Shop Dynamic Stone Tools →Free Tool
Diamond Blade Selector — Narrow diamond tooling options by machine, material and operation so your roughing and finishing tools are matched before you write the toolpath.
Open Blade Selector →