Why this question arises between two parties
The designer works from functional geometry, while the manufacturer works from an actual process. A value copied from another part or system may not give the same result. The agreed outcome is the finished geometry, not a formula alone.
The designer normally specifies finished flange dimensions, hole positions, datums, tolerances and the part’s function. The technologist or contractor determines the route from flat sheet: operation sequence, development and verification. Both roles are necessary, but their documents must not be confused. A customer supplying only a flat pattern is supplying a production assumption; a contractor preparing one from its own process data still has no authority to change functional geometry without approval.
Terms that must be kept separate
| Item to agree | Why it is needed | What must not be assumed silently | |---|---|---| | Parameter definition | To read the term the same way | Identical meaning in every design system | | Starting conditions | To bind it to material and process | Automatic suitability of another table | | File revision | To identify the current development | The latest chat attachment is approved | | Inspection | To verify real geometry | A correct formula guarantees series output | | Decision owner | To approve a change | Silence means agreement |
What to send the manufacturer
Send the finished-part drawing, material designation and condition, thickness, required radius, critical dimensions and tolerances, quantity, revision and any existing development. Explain which surfaces, holes, seams and appearance conditions are functional. A model without acceptance dimensions is not enough.
Add a 3D model when it clarifies the form and a flat file when one exists, but label every file as approved for production, reference-only, or a contractor proposal. Name which requirements are functionally critical: one flange may control assembly while another may allow more variation. Geometric requirements must be communicated unambiguously, but no standard can supply dimensions that are absent from the design.
How to record a process assumption
Record the material, thickness, radius, tool and machine conditions, the parameter or allowance used, the development revision and its owner. State whether it is a proposal, a trial result or an approved production condition. Do not present it as a universal material constant.
A useful record identifies the exact part and revision; verified material, thickness and coating; process and tooling set; the changed part of the development; controlled finished geometry; and the person who approves the result. Mark an unknown item as uncertainty rather than inserting a typical value. A short six-part record—part, inputs, process, development, control and changes—is more useful than a page of formula debate because it distinguishes a design error from an unfulfilled process assumption.
When a trial part is required
A trial is appropriate for a new material, thickness, radius, tolerance, tool or sequence, and for expensive series work. Define dimensions, datums, surface requirements and measurement method beforehand. A successful sample applies only to the recorded conditions.
The trial is not a universal calibration. A sample made with another thickness, tool or revision does not prove new conditions. Its record should separate “within requirements” from “a design change is required.” An out-of-tolerance measurement does not by itself prove the K-factor is wrong: revision identity, units, material, measurement method and the drawing itself may be the cause.
How to handle a revision change
| Revision | Change | Consequence for development | Decision | |---|---|---|---| | A | Initial drawing | Initial proposal created | Await trial or approval | | B | Functional dimension changed | Changed flanges need verification | Approve a new file | | C | Material or thickness changed | Earlier assumptions do not transfer | Separate technical assessment |
Revision letters in this example are only labels. The important point is that the file, correspondence and decision share one identifier. Changing a drawing, thickness, material grade, coating, radius or available tool creates a new starting condition; replacing a file in a shared folder is not enough. Record which old assumptions no longer transfer, whether finished geometry changes, and who may use an earlier trial only as reference. Designers, technologists and purchasing staff must all refer to the same revision.
Common mistakes
Do not copy a parameter from an unrelated job, confuse finished dimensions with blank dimensions, mix revisions, or approve a result only by appearance. A number without its process conditions is incomplete information.
Do not rely on a table found on a website as a guarantee either. Tool suppliers offer different profiles for different applications, including deep forming and reduced marking; that illustrates that tooling selection depends on the actual task. It does not establish that a particular tool is suitable for a particular part.
How to split customer and manufacturer roles
| Stage | Customer decision | Manufacturer action | Result | |---|---|---|---| | Requirement | Define mandatory geometry and function | Check data completeness | Agreed drawing | | Process review | Accept or reject assumptions | Prepare development in its process | Development version | | Trial | Check critical characteristics | Make sample under recorded conditions | Measurement record | | Series | Release one revision | Repeat agreed route | Controlled package |
This division does not mean that a contractor accepts every file without comment, or that the customer approves technical settings it has not seen. A contractor’s proven table or process database is useful, but its assumptions should not be hidden. The customer need not request confidential internal settings; a clear statement of the conditions, checked result and changes that stop automatic reuse is enough. The request can state that parameters apply to this part only and are not transferred to other materials without reassessment. A recorded decision also makes a repeat order safer, because it allows the parties to check that conditions are genuinely unchanged rather than copying an old development.
How to resolve a difference after a trial
Compare the measured part with the approved drawing and recorded conditions, then identify whether the difference is design, material, tooling, sequence or measurement. Decide whether to change the design, process assumption or trial plan; never conceal it in an overwritten file.
Start with package identity, not with a search for a “wrong number”: confirm the drawing revision, material and thickness, tooling, operation sequence, datums and measurement method. Record the observed fact rather than a guessed cause. “The distance between datums differs from the requirement” is a fact; “the K-factor must change” is only one hypothesis. Other causes include an incorrect revision, wrong units, springback, material variation, datum error or an unsuitable measurement method. If the difference repeats, decide separately whether a new trial, a new development revision or a design review is required. A new flat pattern must not change a finished functional dimension without designer approval, and a design change must not be presented as a minor process adjustment. Mark old files reference-only or cancelled after the decision.
Why the parameter is not a universal rule
Different material batches, thicknesses, radii, tooling and methods can change the effective result. A reference table is a starting point, not proof that a particular part will meet its requirements.
The same commercial material name does not define every batch property, and equal thickness does not guarantee an equal process. Geometry, flange width, sequence, tool wear and tooling configuration also matter. A contractor’s internal table can therefore be valuable without being transferable outside its stated conditions. When work moves to another manufacturer, send the finished drawing, material, thickness, control dimensions, revision and trial history—not merely a number. The new manufacturer may use that history as reference but must assess it under its own conditions.
What cannot be decided without data
Without a drawing, material, thickness, radius, process conditions and acceptance criteria, no reliable K-factor, allowance, blank size, price or delivery promise can be made. Safe operation settings remain the responsibility of competent specialists.
The article does not authorize independent press programming or a change to equipment settings. Parts carrying strength responsibility or safety implications require assessment against the applicable product documentation by competent specialists. A numerical value is useful only after the finished-part requirement and the conditions of its verification have been made explicit.
Approval checklist
- [ ] Finished-part requirements and critical dimensions are clear.
- [ ] Material, thickness, radius, quantity and revision are identified.
- [ ] The meaning and conditions of the agreed parameter are recorded.
- [ ] The current development file has one identified owner and revision.
- [ ] Trial-part criteria are defined where uncertainty remains.
- [ ] A changed input triggers a new assessment.
Need a manufacturing assessment for a part?
Send the drawing, material, thickness and quantity. A specialist will clarify the data needed for a technical manufacturing assessment.
Send a manufacturing request