How We Control Tolerances on Magnetic Boxes

How We Control Tolerances on Magnetic Boxes

A magnetic box is built from components that each carry variation, and those errors accumulate rather than average out. This guide explains why, names the six stages where tolerance is lost, gives working values (±0.5 mm components, ±1.0 mm large), the measurement tools and method, and how we hold tolerance in production.

Reference and Next Steps

Tolerance Stack-Up in Practice

On a double-door rigid box the closure gap is the sum of grooving drift, wrap thickness, and magnet placement error; plotting each on a simple stack-up chart shows which stage to control first. We keep that chart per project so tolerance decisions are visible, not argued.

Matching Tolerance to Product Risk

A protective transit box and a display luxury box need different tolerances for the same dimension. Set the number from the consequence, whether the product would be damaged or the box would simply look slightly off, and document it so bulk is judged fairly.

Where to Start on a New Project

Begin with the internal fit and the closure, because those are where most returns start, then tighten cosmetic registration. Send the drawing with the functionally critical dimensions marked and we will propose a tolerance plan rather than a generic figure.

Why Tolerance Is an Accumulation Problem

A wrapped rigid magnetic box is assembled from several components — greyboard panels, a wrap, magnets, hinge grooves, and an insert — and each carries its own variation. Those variations add, they do not average out, so two panels that pass individually can still build a box whose lid binds or whose walls sit proud. This is the core reason a tolerance conversation about magnetic boxes must be about the assembled unit, not the flat parts.

The practical consequence is that inspecting cut board alone is necessary but not sufficient. The closure, the lid registration, and the internal clearance are all functions of stacked errors from cutting, grooving, wrapping, and assembly. Hold the right stages and you hold the box; tighten the wrong ones and you spend money for no visible gain. For the closure side of the same box, see our strength test guide — and note that a misaligned lid is a tolerance story, explained in why lids become misaligned.

Flat Component vs Assembled Box

Specifications often say "dimensions ±1 mm" without stating which dimension — and that ambiguity is where disputes begin. A component dimension is a cut panel before wrapping; an internal dimension is the usable space the product sees; an external dimension is the overall footprint; and lid-to-base registration is a visual alignment, not a single linear number. They are different measurements with different achievable tolerances.

Table 1: Four types of dimensional basis. The specification must state which one, otherwise supplier and buyer each calculate their own.

Six Stages Where Tolerance Is Lost

Knowing where variation enters lets you spend control where it matters. Board cutting sets panel size; V-grooving sets hinge fold position and depth (groove depth is about half the board thickness); wrapping adds cumulative error from paper caliper, adhesive, and pressure; magnet placement decides whether pairs meet; assembly fixturing decides repeatability; and the insert can shift internal clearance. Tool wear at cutting and grooving drifts over long runs, so inspection intervals matter.

Table 2: Six stages that introduce error. Prioritise controlling the one or two that dominate.

Working Tolerance Values You Can Use

The figures below are commonly achievable working values for wrapped rigid construction; they are indicative, not guarantees, and should be agreed per project based on size, structure, and volume. Die-cut rigid components commonly hold ±0.5 mm. Finished external dimensions run wider because wrap accumulates. Lid registration is defined visually against an agreed sample rather than a default industry number, and internal clearance is derived from the product's own tolerance, not copied from a table.

Table 3: Common working tolerances for rigid boxes. Confirm item by item against the design.

Measurement Tools and Method

Holding tolerance starts with measuring it the same way every time. A vernier or digital caliper reads component and external sizes to 0.01 mm resolution; feeler gauges read the rest gap at the lid seam; a height gauge or comparator checks flatness and squareness on the assembled unit; and a CMM or optical comparator is used where tight geometric limits are specified. The method — where the measurement is taken, how the box is held, and whether on flat or assembled units — must be written, because the same box reads differently under different methods.

How Tolerance Is Held in Production

Control is a set of habits, not a single machine. Fixtures register parts instead of relying on operator judgement at every repeatable position. A retained golden unit sits on the line and on record so drift is visible against something real. Tool-wear monitoring catches grooving and cutting drift before product drifts out. In-process sampling measures through the run, not only at the end. Lot control holds board source and grade, the largest single source of dimensional change. And final inspection is on the assembled box, because components passing individually do not guarantee a good build.

1. Fixturing over skill — register the part, remove operator variance
2. Retained golden unit — drift is visible against a real reference
3. Tool-wear monitoring — planned checks catch groove/cut drift
4. In-process sampling — measure through the run, not just at the end
5. Lot control — hold board source and grade to stop the biggest shift
6. Assembled-unit inspection — final check on the built box, not flat parts

Writing Tolerance Into Your Spec

Six additions turn a drawing into something a factory can actually hold. State which dimension (internal, external, component). State the tolerance as a number with units. State the measurement method (where, how held, flat or assembled). State the fit requirement as clearance derived from the product. State the visual criterion for registration in measurable terms. Agree the sampling plan — how many units, when, and what happens on failure. None of these cost anything to write and each prevents a recoverable cost later.

Table 4: Six supplementary items to turn a drawing into an executable specification.

Most tolerance disputes are method disputes; once both parties measure the same way, they usually resolve. For structural choice that affects how hard tolerance is to hold, compare book-style vs double-door magnetic boxes.