Machining cost basics

How to Calculate CNC Machine Cycle Time

This guide is for engineers and buyers who need a cycle time they can defend in a quote review. We break cycle time into its four parts, give the formulas, and show which inputs usually get fudged. Read it and you can build an estimate within 10% of the real run.

Get Instant Quote Talk to an Engineer

4 time componentsFormula per operationTypical error sources

how to calculate cnc machine cycle time on a machining center

On this page

1
Key takeaways
2
What cycle time includes and what it does not
3
The four components you have to add up
4
What throws the estimate off
5
Software and shop-floor checks
6
When the calculation changes the decision
7
Step by step: calculate CNC machine cycle time
8
Typical cycle time components by operation

Quick answer

Key takeaways

Cycle time is four numbersCutting time + tool changes + rapid and approach moves + load/unload. Miss any one and the estimate drifts.

Cutting time comes from feed and lengthTool path length divided by feed rate gives minutes. Everything else is bookkeeping around it.

Non-cut time is often 30–50%On small parts with many tools, tool changes and rapid moves can match the cutting time.

Rough numbers carry ±20%Feature-based estimates without a tool path are fine for quoting, not for scheduling.

Verify with one real runCut a sample, log the machine timer, and compare. The gap tells you which input is wrong.

Definitions

What cycle time includes and what it does not

Cycle time is the machine time for one part: from the moment the spindle starts the first cutting move to the moment the finished part is off the table. If the part runs two operations on two machines, each operation has its own cycle time. Add them for the total machining time.

Do not mix cycle time with lead time. Lead time covers material purchase, programming, queue time, machining, finishing, inspection, and shipping. Cycle time is only the machining slice. A part with a 12-minute cycle time can still ship three weeks later if the material is not in stock.

Setup time is also separate. On a 50-piece run, a 45-minute setup spreads to roughly 0.9 minutes per part. On a one-off prototype, that same setup adds 45 minutes. This is why job quantity changes the effective cost per part even when the cycle time stays fixed.

Cycle time in this article means spindle-to-spindle time on a machine that is already set up and running. That is the number the machine timer shows.

1
IncludedCutting, tool changes, rapid and approach moves, load/unload, in-process pauses.
2
ExcludedSetup, programming, queue, finishing, inspection, shipping.

The parts

The four components you have to add up

Cutting time is the only component that removes metal. Calculate it as tool path length in mm divided by feed rate in mm/min. For turning, length is the axial distance the tool travels per pass, multiplied by the number of passes. For milling, use the actual path length including entry and exit arcs, not the straight-line distance across the pocket.

Tool change time is the number of tool changes multiplied by the machine's change time. A standard 40-taper mill with a 24-tool carousel runs about 3–6 seconds per change. A mill-turn or a machine with a large tool magazine can run 8–12 seconds. Count every change, including the one at the end of the cycle.

Rapid and approach moves are the non-cutting travel between features. A rapid rate of 20,000 mm/min sounds fast, but acceleration and deceleration dominate on short moves. On a part with many small features, allow 20–40% more time than the theoretical rapid calculation.

Load and unload time depends on the workholding. A vise with soft jaws on a 5 kg block is 30–60 seconds. A fixture that needs eight bolts and a dial indicator can be 5–10 minutes. On high-volume runs, a hydraulic fixture or a pallet changer drops this to a few seconds per part.

1
CuttingPath length ÷ feed rate. The only value-adding time.
2
Tool changeCount × 3–12 s depending on the machine.
3
Rapid travelAdd 20–40% over theory for accel and decel.
4
Load/unload30 s for a vise, up to 10 min for a complex fixture.

Accuracy

What throws the estimate off

Feed rate is the biggest lever. The chart value is a starting point, not a guarantee. A 12 mm end mill in 6061 aluminum may be rated at 3,000 mm/min, but a machine with a 12,000 rpm spindle and a light fixture will chatter before it reaches that. If the real feed is 2,000 mm/min, the cutting time is 50% longer than the estimate.

Tool path length is the second lever. CAM software shows the path, but the post-processor may add retracts, safe-Z moves, and lead-in arcs that the preview does not count. On a part with 30 small pockets, the difference between the theoretical path and the posted path can be 15–25%.

Material condition matters. A 6061-T6 block machines differently from a 6061-T6 casting with a hard skin. A 17-4PH stainless part in the H900 condition is harder than the annealed condition. If the drawing does not state the condition, the estimate can be off by 30% or more on stainless and titanium.

Machine dynamics are the third lever. A 3-axis machine with a 40-taper spindle and a 4,000 mm travel has different acceleration than a compact 5-axis machine. On a part with many short moves, the larger machine may be slower because it cannot accelerate as fast between features.

1
Feed rateUse the real value the machine can hold, not the chart maximum.
2
Path lengthRead the posted path, not the CAM preview.
3
Material conditionAnnealed vs. hardened can change time by 30%.
4
Machine dynamicsShort moves favor machines with high acceleration.

Tools

Software and shop-floor checks

CAM software gives the most accurate cutting time because it uses the actual posted tool path. Most packages report a time per operation and a total. Treat the total as a lower bound: it usually excludes load/unload and may exclude some rapid moves.

Machine simulation adds the non-cutting moves and checks for collisions. It is worth running on any part with more than 10 tools or a 5-axis tool path. The simulation time is close to the real cycle time if the machine model and the acceleration values are set correctly.

On the shop floor, the machine timer is the ground truth. Run one part, read the timer, and log the result against the estimate. After three or four jobs, you will see a consistent offset for a given machine and material. Use that offset to correct future estimates.

For quoting, a feature-based estimate is usually enough. Count the features, apply a time per feature from a table, and add 20% for tool changes and non-cutting moves. This gets you within 20% without a full CAM model.

1
CAM timeAccurate for cutting, usually excludes load/unload.
2
SimulationAdds non-cutting moves and collision checks.
3
Machine timerThe ground truth. Log it and build an offset.
4
Feature-basedGood enough for quoting, within about 20%.

Application

When the calculation changes the decision

Cycle time drives the choice between 3-axis and 5-axis machining. A part with features on five faces can run on a 3-axis machine with multiple setups, or on a 5-axis machine in one setup. The 5-axis cycle time may be longer per operation, but the total is often shorter because setup and re-fixturing time disappear.

Cycle time also drives the choice between machining and another process. A part with a 40-minute cycle time on a mill may be a better fit for die casting or vacuum casting at volume. The crossover point depends on the annual quantity, the tolerance, and the surface finish.

On high-volume runs, cycle time sets the machine count. If a part takes 8 minutes and you need 5,000 parts per month, one machine running 20 hours per day can make about 150 parts per day, or 3,000 per month. You need two machines. This is the calculation that decides capacity.

For prototypes, cycle time matters less than the number of setups. A one-off part with a 2-hour cycle time is normal. The goal is to get a functional part, not to optimize the machine.

1
3-axis vs. 5-axisFewer setups can beat a faster single operation.
2
Machining vs. castingCompare at the annual quantity, not per part.
3
Capacity planningCycle time × parts per month sets the machine count.
4
PrototypesSetups matter more than cycle time.

Method

Step by step: calculate CNC machine cycle time

1
Split the part into operationsList every operation in order: facing, roughing, finishing, drilling, tapping, boring, chamfering. Each one gets its own cutting time. A part with 14 features usually becomes 20–30 separate cutting segments.
2
Get the tool path length for each cutPull the length from CAM, or measure it on the drawing for simple shapes. For a pocket, add the perimeter passes plus the stepover moves. For a drilled hole, use the depth plus 0.3 × drill diameter for the point and clearance.
3
Divide length by feed rateFeed rate in mm/min comes from the tool supplier's chart for the material. For 6061 aluminum with a 12 mm carbide end mill, 1,500–3,000 mm/min is normal. For 316 stainless, expect 200–500 mm/min. Length ÷ feed = minutes.
4
Add the tool change timeCount each tool change and multiply by the machine's change time. Use 3–6 s for a standard carousel, 8–12 s for a large magazine. Add 2–4 s per change if the tool needs a spindle probe or a coolant-through check.
5
Add rapid and approach movesEstimate the non-cutting travel between features. For a part with 20 or more small features, take the theoretical rapid time and multiply by 1.2–1.4. This covers acceleration, deceleration, and tool retracts.
6
Add load and unloadMeasure the actual load/unload once on the machine. For a vise, 30–60 s. For a multi-bolt fixture, 5–10 min. Divide by the number of parts per load if the fixture holds several.
7
Add the in-process pausesInclude any dwell for a spindle probe, a tool breakage check, or a coolant wash. A probe cycle is typically 8–15 s. A tool breakage check adds 3–5 s per tool.
8
Compare against a real runCut one part, read the machine timer, and compare. If the real time is more than 15% above your estimate, the usual cause is a feed rate that the machine cannot hold, or a tool path that has more retracts than the CAM preview showed.

Reference

Typical cycle time components by operation

Values are starting points for estimation, not guarantees. Actual values depend on the machine, tool, and material.

Component	Typical range	How to estimate	Common error
Cutting time	30–70% of cycle	Path length ÷ feed rate	Using the drawing length, not the path
Tool change	5–20% of cycle	Count × 3–12 s	Forgetting the last tool change
Rapid travel	10–25% of cycle	Theory × 1.2–1.4	Ignoring accel and decel on short moves
Load / unload	5–15% of cycle	Measure once on the machine	Assuming a vise when it is a fixture
In-process pause	2–8% of cycle	Probe 8–15 s, check 3–5 s/tool	Leaving probe cycles out of the total
Chip-to-chip	Add 5–10%	Cover coolant and retract moves	Treating it as zero on short cycles

The estimate is only as good as the inputs

Use the four-component method, verify against one real run, and correct the offset. That is the difference between a quote that holds and one that drifts.

Get Instant Quote Talk to an Engineer

FAQs

Frequently asked questions

What is the difference between cycle time and lead time?

Cycle time is the machine time for one part, from the first cutting move to the finished part off the table. Lead time is the total time from order to delivery, including material purchase, programming, queue, machining, finishing, inspection, and shipping.

A part with a 10-minute cycle time can still have a three-week lead time if the material is not in stock or the finishing queue is long.

How accurate is a CAM cycle time estimate?

CAM software is accurate for cutting time because it uses the posted tool path. It is usually within 5–10% of the real cutting time.

The total is less accurate because it often excludes load/unload, in-process probing, and some rapid moves. Add those separately to get a realistic total.

How do I estimate cycle time without CAM software?

Use a feature-based method. Count the features on the part, apply a time per feature from a table built from past jobs, and add 20% for tool changes and non-cutting moves.

This gets you within about 20% for most parts. It is suitable for quoting, not for scheduling a machine.

Why is my real cycle time longer than the estimate?

The three common causes are a feed rate the machine cannot hold, a posted tool path with more retracts than the CAM preview showed, and a material condition that is harder than the one used in the estimate.

Compare the machine timer against the estimate on one part. The gap tells you which input to correct.

Does cycle time include setup time?

No. Setup time is separate from cycle time. On a 50-piece run, a 45-minute setup spreads to about 0.9 minutes per part. On a one-off prototype, the same setup adds 45 minutes.

For cost per part, add setup time divided by the quantity to the cycle time.

How does cycle time affect the price of a machined part?

Machine time is one of the largest cost drivers. Cutting time, tool changes, and load/unload all consume machine hours that have to be paid for.

Reducing cycle time by 20% on a high-volume part can reduce the price by a similar amount, if the material and finishing costs stay the same.

Need a cycle-time estimate you can defend?

Send us the drawing and we will return a quotation and a DFM analysis within 12 hours. No minimum order quantity, from one prototype to 10,000+ part runs.

12-hour quote100% inspectionNo minimum order quantity

Get Instant Quote Talk to an Engineer