Which reference parts to prepare for a laser-cutting demonstration before purchase ART-043

What the demonstration must prove

Before sending files, formulate the questions the demonstration must answer. For example: whether the geometry of a critical contour is maintained; whether the edge is suitable for painting or assembly; how thin tabs behave; whether holes of a particular type can be cut consistently; what the actual cycle time is for a serial part; whether the part requires an additional manual operation. Without a question, the result becomes a subjective “like it / do not like it.”

Do not try to obtain answers to every production and economic question from one test. A demonstration does not reproduce a year of workshop operation, confirm future cost for every batch, or guarantee the same result on sheet from another supplier. It provides a controlled point for checking agreed samples, materials, configuration, and criteria.

Selecting parts: four mandatory groups

The first group consists of typical serial parts. They are needed to assess full-cycle productivity, nesting, consistency, and post-cutting operations. The second group contains parts with critical features: small holes, narrow slots, frequent corners, fine contours, thin tabs, or visible edges. The third group covers thicknesses and materials that impose the greatest process requirements. The fourth consists of products that determine development: a new product range or transfer of an operation from another technology. It should not be disguised as current workload, but it can be checked separately.

| Reference part type | What it checks | Data to provide with the file | What cannot be concluded from this test alone | |---|---|---|---| | Typical serial part | Cycle time, nesting, overall quality for the main flow | Quantity per batch, material, thickness, subsequent operation | Annual workshop output without accounting for load and logistics | | Part with critical holes/slots | Geometry of sensitive features and the need for rework | Drawing with critical features marked, inspection method | Universal ability to make any small feature in all materials | | Part with a visible edge | Suitability of the edge for subsequent painting, assembly, or appearance | Description of the front side and allowable marks | An arbitrary “perfect edge” standard without an agreed criterion | | Thick/special material | Limits of the specific process and configuration | Grade, actual thickness, coating, sheet origin if available | Result on another grade, heat, or material condition | | Large nested part | Nesting, shape stability, and sheet handling | Sheet dimensions, orientation, tab requirements | Full automation efficiency without a loading/unloading test |

Provide more than DXF

DXF or another agreed file is important, but by itself it does not explain the designer’s intent. For every part, add a PDF drawing, current revision, material, nominal thickness, quantity, blank type, critical features, and a brief note on subsequent operations. If the contour must be changed for manufacturability, discuss that separately: a demonstration does not entitle a supplier to alter the design part unilaterally.

File names must be readable and unambiguous: part number, revision, material, and thickness. Do not send a folder of files such as `new_final_last.dxf`; a revision error can produce a perfectly cut but wrong part. Where confidential information is involved, agree on the exchange procedure, access rights, and whether a simplified test fragment can be used. An overly simplified fragment, however, may fail to represent the real nesting or critical geometry transitions.

Test material must be agreed

Material is not a backdrop for a demonstration, but part of the condition. A particular thickness, grade, coating, surface condition, and actual batch can affect the result. Testing a product on the supplier’s material is not necessarily wrong, but it must be recorded in the protocol. If the result critically depends on material, it is advisable to provide a sample or sheet from your usual supply under an agreed procedure.

The same applies to gas. Oxygen, nitrogen, and compressed air are used in different process scenarios, and the choice affects the edge, subsequent operation, cost, and process settings. Do not ask to “show the best edge” without specifying the conditions under which it was obtained. The protocol must state the material, thickness, assist gas, and key test limitations. Gas selection is discussed in more detail in a separate article [“Oxygen, nitrogen, or compressed air”](#internal-link-cand-041).

Agree quality criteria in advance

The result is assessed by more than appearance. Define what matters most for each reference part: contour size, holes, flatness, lower-edge condition, thermal mark, readiness for bending, welding, or painting. If a part is welded, surface cleanliness and the subsequent joining zone may matter; if it has a decorative function, the front side matters; if it enters an assembly, datum dimensions and repeatability matter.

Replace the words “accurate,” “clean,” and “burr-free” with verifiable criteria. They may be numerical if they already exist in the design documentation, or descriptive with an agreed inspection method. For visual assessment, it is useful to prepare an example of acceptable and unacceptable results on a real part. Do not create new arbitrary tolerances merely for the demonstration: they must belong to the product requirements, not to a marketing scenario.

In the general production context, requirements for geometry, tolerances, and surface quality are formalised by applicable standards and company documentation. For example, ISO 21920-1 covers general requirements for geometrical product specifications for surface texture, but it is not a ready-made acceptance template for a particular laser-cut part. ISO 21920-1. The applicability of any standard is determined by the responsible specialist.

How to see the economics without substituting promotional speed

Ask to see not only cutting speed, but the complete agreed cycle time for the selected part: program preparation, positioning, piercing, cutting, tabs, unloading, and, where necessary, sorting. Not every operation will be reproduced at a demonstration, but the supplier must clearly distinguish actually measured time from modelling or assumptions.

Compare identical conditions. It is incorrect to compare one result using nitrogen with another using oxygen, one sample on different material, or different contour revisions. Nor should conclusions about the whole economics be drawn from one complex part. A serial reference part shows the rhythm of the main flow; a complex one shows the process margin; a thick one shows the boundary of the required scenario. Together, they provide a sufficiently well-founded picture.

Production attendance and a protocol matter more than video

Where possible, the demonstration should be attended by a production representative, process engineer or designer who understands the critical requirements, and the person responsible for the investment decision. Video is useful for recording, but it does not replace inspecting the part, asking questions, and receiving a signed protocol.

The protocol need only state the date, machine and configuration, material, thickness, file and revision, gas, list of parts, measured or inspected criteria, deviations, conclusion, and open questions. It need not become a multipage expert report. The main thing is that a month later everyone understands what was tested, under which conditions, and what was actually confirmed.

Public descriptions of manufacturers’ 2D laser systems show that the actual configuration may include not only the source and machine frame, but also material-flow automation, software, and other options. TRUMPF and Bystronic describe systems precisely as integrated complexes. Therefore, if you are assessing not only cutting but production throughput, record which options are included in the tested configuration and which are not.

Typical mistakes

• Send only the most complex part. It may not represent the key serial flow.
• Select only simple parts. Critical contours and the real process margin then remain unseen.
• Do not specify the revision and material. The result cannot be compared correctly.
• Judge a part “by eye” without a criterion. Different participants will always see different “quality.”
• Treat maximum speed as an economic indicator. A real-part cycle and explanation of conditions are needed.
• Change files after the demonstration without a new protocol. This destroys comparability of results.
• Ignore the subsequent operation. An edge acceptable for one route may be unsuitable for another.

Separate the demonstration, FAT, and SAT

These three formats can use the same reference parts, but they answer different questions. A preliminary demonstration helps compare the process suitability of several solutions and formulate the technical specification. Its format may be flexible: some samples are cut in a demonstration centre, others on an available machine, and the result is discussed with a process engineer. It is important to understand honestly that this is not yet delivery acceptance.

FAT before shipment is already a contractual check of the agreed machine and agreed specification. Its list of parts is better narrowed to key items, because each must have a description of conditions and criteria. If a preliminary file was used during the demonstration and the project was then changed, FAT must use the updated revision and record it. Do not automatically transfer demonstration results to another configuration: different sources, heads, software, automation options, and material can change the result.

SAT after installation confirms that the system works in your environment. Here, reference parts should help check not only contour quality but process readiness: whether a correct program has been created, whether agreed actions are clear to the operator, whether material is prepared consistently, and whether service-support channels have been selected. SAT should not be overloaded with many exotic tests. A few products that will actually enter the first production batch are more valuable.

Each format should have a separate protocol. The minimum fields are the same: date, location, participants, file version, material, thickness, gas, machine/configuration, criterion, result, and open questions. The meaning of the conclusion differs. In a demonstration it is a basis for selection and clarification of the technical specification; in FAT it is confirmation before shipment in accordance with the contract; in SAT it is evidence of commissioning under site conditions. This separation does not create needless bureaucracy; it protects both parties from comparing incomparable results.

One further practical point: do not require the demonstration centre to reproduce the future workshop perfectly. Instead, record which factors were not reproduced: another sheet format, no automated store, another operator route, or another gas supply. This does not invalidate the test; it sets correct boundaries for the conclusion. If such a factor is critical to the investment, it becomes a separate item in the technical specification, FAT, or SAT plan.

After each stage, retain cut samples with a label of the test conditions. A physical part together with the protocol is often more useful than a set of photographs: it can be shown to a process engineer, welder, or coating specialist and compared with the production standard.

Photographs should supplement the protocol, not replace it.

Preparation algorithm

1. Extract typical and critical parts from the actual product mix for the recent period. 2. Select 8–15 items from four groups: serial, critical, difficult materials/thicknesses, and prospective. 3. Check every file with the designer or process engineer: revision, material, thickness, contour, and critical features. 4. Prepare a DXF/DWG package, PDF, parts table, and short requirements for each item. 5. Agree the material, gas, demonstration format, and criteria before testing begins. 6. Require the conditions of every sample to be recorded in the protocol. 7. Compare results from different suppliers only under identical parts and conditions. 8. Transfer the agreed list of reference parts and acceptance criteria into the technical specification and contract.

Limits of this article

This material does not establish tolerances, cutting modes, speeds, gas pressures, or acceptance criteria for a particular product. It does not replace design documentation, a process audit, or official testing under a contract. All design changes, material selection, and safety rules must be agreed by responsible specialists.

After selecting reference parts, check whether the technical specification contains the complete delivery scope and whether the supplier’s service model can support the selected configuration after start-up.

How to prepare a supplier package in one working day

Create one folder with four subfolders: “Cutting files,” “PDF drawings,” “Conditions table,” and “Protocols.” In the table, one row for each part number is enough, containing the revision, material, thickness, quantity, critical features, subsequent operation, and a contact person who can explain the requirement. This format is better than lengthy correspondence: the supplier sees the complete context, and your team can quickly find any item.

Before sending it, conduct a ten-minute internal review. The designer checks revisions, the process engineer checks the specified operations, and procurement makes sure every participant receives the same package. If a file cannot be transferred because of confidentiality, mark this immediately and agree how to work with it. Do not replace a critical contour with a random example without explanation: the demonstration may then produce a falsely optimistic result.

After the demonstration, add the actual conditions and status to the same table: “confirmed,” “requires clarification,” or “not tested.” This creates a simple bridge between preliminary assessment, FAT, and SAT. If a part does not meet an agreed criterion, do not conclude that the whole machine has failed. First record exactly what remains open: geometry, material, edge, subsequent operation, or configuration. The next check will then be technical rather than emotional.

Need help selecting equipment?