NdFeB allows compact high-energy designs, but the material is brittle and corrosion-sensitive. Design the magnetic circuit and the mechanical protection together, and decide early which parts of the specification the assembly depends on.
Design the magnetic circuit first
Start from the flux path rather than from the magnet. Where does the field have to go, what does it have to cross, and what metal or air sits in its way? The magnet occupies only part of that circuit, and changing the air gap, the pole piece or the return path usually changes the result more than changing the grade does.
A common pattern in prototypes is a large magnet used to compensate for a weak circuit. That extra volume tends to survive into production, where it reappears as cost, weight and assembly difficulty.
Decide the magnetisation direction early
The magnetisation direction is a manufacturing input, not an annotation added before release. Axial, diametral, multipole and custom pole patterns each constrain how the magnet is positioned, how it is retained and how much the assembly suffers when a part is installed the wrong way round. Deciding it late is one of the most common reasons a working prototype cannot be reproduced in volume.
Geometry, air gap and the flux that actually arrives
Published grade values are measured in reference conditions. In service, the usable flux depends on the length-to-diameter ratio of the magnet, the size of the air gap, the return path and the temperature of the magnet itself. Two assemblies holding identical magnets can differ widely. When a design is tight, that is the first place to look.
Treat brittleness as a design constraint
Sintered NdFeB is hard, brittle and electrically conductive. Edges chip, thin sections crack under bending, and fast-moving fields can heat the magnet through eddy currents. Design for handling: chamfers or rounded edges, retention by adhesive, housing or mechanical capture rather than relying on the magnet to hold itself, and no sharp press fit that puts tension into the part.
Corrosion is a materials choice, not an afterthought
Bare NdFeB corrodes quickly in humidity. Nickel-copper-nickel is the common default, zinc is used where cost matters more than durability, and epoxy or another organic coating suits salt, chemical or outdoor exposure. Choose against the environment the part will meet in service, including storage and shipping, and check what the assembly process does to the coating, because solvents, heat and abrasion all act on it.
Temperature in the assembled part
The temperature a magnet reaches is usually higher than the ambient figure on the enclosure. Motors, coils, curing ovens and soldering steps all add heat, and the magnet responds to the temperature of the magnet itself. Design the circuit so the working point stays clear of the knee of the demagnetisation curve at the hottest condition, and revisit the design when the current, the duty cycle or the enclosure changes.
Tolerances that decide assembly success
Sintered magnets are typically machined to tolerances looser than the rest of the assembly expects. Where a magnet sits in a pocket, a raceway or a sensor position, a stack of tolerances decides the result: magnet size, coating thickness, pocket size, adhesive thickness and the thermal expansion of both materials. Specify the tolerance the assembly needs rather than the tolerance that looks tidy on a drawing.
What to validate before tooling
Before tooling is committed, test what production depends on: force or flux in the assembled condition, fit and position repeatability, coating adhesion, drop and handling damage, thermal cycling at the hottest condition and, where the environment demands it, corrosion testing. Record the measurements, because a documented prototype result is what makes a later production change reviewable.
Where volume changes the design
Volume changes the economics of geometry. Machined prototypes allow shapes and tolerances that a press tool cannot deliver at the same cost, so designs usually simplify as volumes rise: standard sizes, a coating that can be applied consistently, and tolerances aligned with the process. Start that conversation before the drawing is frozen rather than after.
Documentation that survives a supply change
A specification that lives in an email thread is fragile. Put the drawing, the grade or performance target, the magnetisation direction, the coating, the tolerance and the inspection expectations into one document. When a part is re-quoted, replaced or produced on another line, that document is the only thing keeping the assembly behaving the same way.
Sample quantity and the first production lot
Prototypes are built in small numbers, which hides variation. Order enough samples to see the spread: several parts from the same lot for assembly trials and, where possible, parts from a second lot to check that coating thickness, dimensions and magnetic output stay inside the window you need. Keep one part as a reference and measure it again when the first production lot arrives. That single comparison catches most of the drift that would otherwise only appear in the field.
The parts around the magnet
A magnet never works alone. Washers, keeper plates, pole pieces, screws, springs and adhesives all take part in the circuit, and the material of each one matters: ferritic steel carries flux, most austenitic stainless steel does not, and a plated screw behaves differently from a bare one in a humid enclosure. Where clearance is tight, allow for coating thickness and for the fact that a sintered magnet is not perfectly flat or square. The tolerances of the parts around the magnet often decide the result more than the tolerance of the magnet itself.
Where designs usually break
The recurring causes are familiar: a grade chosen for strength without checking the hottest condition, a magnetisation direction decided after the housing was drawn, a coating chosen for the price of the first sample order, a tolerance that cannot be held in production, and no measurement to show what the prototype actually did. Each one is cheap to solve early and expensive to solve late.
This guide summarises general engineering practice for sintered NdFeB assemblies. Confirm the numbers that matter for your part against the manufacturer datasheet, the prototype measurements and the first-article inspection.