A buyer-protective acceptance specification is a written set of measurable pass/fail tests agreed and signed before the machine order is placed.
A buyer-protective acceptance specification is a written set of measurable pass/fail tests agreed and signed before the machine order is placed. Without it, catalogue accuracy claims carry little legal weight, and you may end up accepting a machine that cannot hold your workpiece tolerances.
This memo gives you a practical, step-by-step method with worked numbers. It is written for machine shop owners and purchasing engineers in the DACH and UK markets, where formal acceptance protocols are common but often too vague to protect the buyer.
This is a technical due-diligence memo from the Gree CNC distributor team. It explains how to write an acceptance specification that protects your investment. It uses typical industry ranges for illustration. Those ranges are not guarantees. Every figure that matters must be confirmed against the official Gree catalogue or a written factory test report before you sign the purchase order.
The method works for any CNC machining centre, but the examples focus on a 5-axis vertical machining centre for aluminium housings, a common application in DACH and UK job shops.
A weak acceptance specification uses words like “high accuracy”, “stable”, “suitable for precision work”. These words cannot be measured and cannot be enforced.
A strong specification defines:
Every test must have one of two outcomes: pass or fail. There is no third category called “acceptable in practice”.
Start with your parts, not the machine. List:
A machine shop making hydraulic valve bodies with ±0.02 mm bore positions needs a different acceptance spec than one making large structural brackets with ±0.2 mm.
A common rule in DACH/UK procurement is that the machine’s total uncertainty should be no more than 1/4 to 1/3 of the tightest workpiece tolerance band. Uncertainty here means the sum of positioning error, repeatability, thermal drift and tool deflection, not just the positioning accuracy from a catalogue.
Worked example:
So the machine’s combined positioning and repeatability error on that axis should stay below about 0.020 mm. A positioning accuracy of 0.010 mm and repeatability of 0.005 mm would be a reasonable target for this tolerance. These are typical mid-range values, not a Gree guarantee. Confirm the official catalogue for the specific model.
Use international standards where possible. They define the measurement method and reduce arguments later.
Each standard has a defined test procedure. Refer to the standard by name in the acceptance spec. Do not accept a supplier statement like “accuracy is 5 microns” without a standard reference.
Specify the instrument for each test. Common choices:
State who supplies the instrument, who calibrates it, and who witnesses the test. If the buyer does not witness the factory test, the on-site acceptance test must repeat the same measurements.
Machine accuracy changes with temperature. A machine tested at 20 °C may behave differently at 28 °C in a non-air-conditioned shop.
The acceptance spec must state:
Without these controls, a supplier can claim the machine failed because “your shop is too hot”. With them, the test is repeatable and fair.
Accuracy is not enough. You buy a machine to make parts at a certain rate.
Define a test part program that includes the features you actually machine: face milling, hole drilling, tapping, pocketing, 5-axis contouring if used. State the tool list, spindle speeds, feeds and coolant condition.
Pass criteria for cycle time:
State exactly what happens if a test fails:
Do not leave retest rules to “good faith negotiation” after a failure. Write them down.
Assume the buyer machines an aluminium housing with a bore position tolerance of ±0.05 mm. The acceptance specification for the machine could look like the table below. All “typical industry range” values are illustrative. All “required value” figures must be confirmed against the official Gree catalogue for the specific model and configuration.
| Test item | Measurement method | Typical industry range (not a guarantee) | Required value in this example | Status |
|---|---|---|---|---|
| Linear positioning accuracy X/Y/Z | Laser interferometer, ISO 230-2, 5 positions per axis, 3 runs | 0.008–0.015 mm full stroke for mid-range VMC | ≤ 0.010 mm | Confirm official catalogue |
| Linear repeatability X/Y/Z | Same as above | 0.003–0.008 mm | ≤ 0.005 mm | Confirm official catalogue |
| Circular interpolation roundness, XY plane | Ballbar, ISO 230-4, 300 mm circle, 1000 mm/min | 0.005–0.015 mm roundness deviation | ≤ 0.010 mm | Confirm official catalogue |
| Spindle runout at taper | Test bar, dial indicator, 300 mm from gauge line | 0.002–0.005 mm | ≤ 0.003 mm | Confirm official catalogue |
| Thermal drift after 30-minute warm-up | Repeated probing of reference bore at 20 °C, X/Y axes | 0.005–0.015 mm drift | ≤ 0.010 mm | Factory data requested, not yet received |
| Cutting test on customer material | Mill 5 bores, 3 pockets, measure on CMM | Cpk ≥ 1.33 for ±0.05 mm tolerance is typical target | All features within ±0.04 mm, Cpk ≥ 1.33 | Not yet known; must be agreed and written |
| Cycle time test | 10 consecutive cycles of agreed test part | Varies widely by part and machine | Within ±5% of written quotation time | Confirm against official quotation |
Cpk is a statistical measure of how well a process stays within tolerance. A Cpk of 1.33 means the process spread is small enough that about 99.99% of parts should be in tolerance, assuming a normal distribution.
This section is deliberately explicit. As of the date of this memo, the following items are not confirmed for the specific Gree model and configuration we are discussing. We will not present them as facts.
Run the buyer self-check & book a machine selection call — contact the Gree CNC team.