Magnetic mounts: how to choose, install and maintain them

A magnetic mount is a neodymium magnet housed in a steel cup. The cup is not decoration - it closes the magnetic circuit on one side and channels the flux towards the contact face. The same magnet without a housing holds noticeably less, and the same magnet in a double-sided assembly behaves differently again. That is why the first question when specifying a mount is not "how strong is the magnet" but "how is this magnet housed, and what am I bolting it to".

This guide explains what to look for on a datasheet, how we measure pull force ourselves, why shear (side) load capacity is a fraction of the vertical figure, and how to calculate the force before you buy. You will not find comparisons with named suppliers here - you will find a procedure you can repeat in your own workshop.

Safety warnings

A magnetic mount is not a toy, and it is not a structural fixing in the sense used by building regulations. The rules below follow from the physics of a neodymium magnet and from how a steel housing behaves under load.

• Never use magnetic mounts to lift people, or to suspend loads above people. A magnet holds by adhesion and friction - there is no mechanical backup that engages once contact is lost.
• Always apply a safety factor. For static loads on clean, dry steel, allow at least a factor of three against the rated pull force. With vibration, variable loading or any risk of impact, allow more.
• Crushed fingers. A mount rated at tens of kilograms accelerates towards steel faster than you can stop it by hand. Approach the surface edge-first, not flat, and keep your fingers to the side of the housing.
• Pacemakers, defibrillators, insulin pumps. Keep at least 30 cm (12 in) away from active implants. This applies to bystanders in the work area as well as to the operator.
• Magnetic media, payment cards, instruments. Keep away from magnetic stripe cards, Hall effect sensors and moving-coil instruments.
• Temperature. Standard NdFeB neodymium magnets used in mounts have a working limit of 80 °C (176 °F). Exceeding it causes a permanent loss of force - the magnet does not recover on cooling.
• Brittleness. The magnet material is hard and brittle, much like ceramic. A knock against a steel edge can crack it inside the housing - the mount looks intact from the outside and holds far less.
• Loss of grip through debris. Swarf, rust and mill scale collect on the contact face and lift the mount by fractions of a millimetre. That is enough to lose a large share of the force. Clean the contact face before every loaded installation.

How to specify a mount

Catalogues usually describe a mount with two numbers: diameter and pull force. That is not enough to compare two products. Below are four things that decide whether a mount will work in your application.

Magnet height versus housing height

Two mounts with the same outside diameter can contain magnets of completely different heights. A housing 8 mm (0.31 in) tall may hold a 5 mm magnet, or a 3 mm magnet with a thicker steel base - and that is a difference in force, not in appearance.

Rule of thumb: at a given diameter, breakaway force increases with magnet height, but not linearly. Doubling the height does not double the pull force. On the other hand, reducing the height below a certain point collapses the pull force sharply, because the magnet can no longer sustain flux through the circuit and starts closing it on itself. This behaviour is described by the permeance coefficient - more on that further down, under calculators.

What this means when buying: if a supplier quotes only the outside dimensions of the housing, ask for the height of the magnet itself. Without it you cannot compare two mounts of the same overall size.

Flux matters as much as material grade

The magnet grade - N38, N42, N45, N50 - describes the material, not the finished mount. The grade tells you the band within which remanence Br falls. Our calculators cover grades from N30 to N50, and for each one we give the Br range in kilogauss, because it is a range and not a single value.

The conclusion is inconvenient for marketing but true: a higher magnet grade does not mean a stronger mount. An N38 mount with well-chosen geometry and a properly closed circuit can hold better than an N50 mount whose magnet is too short for its diameter. Grade is one factor among several, not a ranking.

Eyebolt, thread and the joint with the housing

Mounts usually fail on the mechanics, not on the magnet. Three places worth inspecting:

• The eyebolt seated in the housing. Check whether the eye is set into the steel and swaged, or merely bonded. An adhesive joint will carry a static load but lets go under a snatch.
• The internal thread. In mounts with M4 to M10 threads, the effective thread engagement matters more than the diameter alone. Three turns of thread in a thin base plate will strip during tightening, not in service.
• Bolt and nut. Fasteners in A2 stainless steel (AISI 304) resist corrosion but have lower tensile strength than grade 8.8 carbon steel bolts. Loaded near the rated pull force, the bolt is the link that gives way first.

Coating and environmental resistance

Bare NdFeB corrodes in humid air within weeks, so it always carries a coating. The standard for mounts is NiCuNi - a three-layer nickel-copper-nickel electroplate. It is hard, it carries flux well and it is sufficient for indoor use.

The housing itself is usually structural steel, most often zinc-plated. That is adequate in a workshop or a production hall. For outdoor work, exposure to water or fluctuating humidity you need either a stainless housing or a barrier coating - epoxy or rubber. Note that every layer between the magnet and the steel it grips pushes the magnet away from the surface and reduces the pull force. A 1 mm (0.04 in) rubber coating takes away a substantial share of the force - in exchange it improves grip in shear and does not mark paintwork.

How we measure pull force

Rated pull force is a measured figure, not a calculated one - which makes the test conditions as important as the result. Below is the procedure we use. You can repeat it and compare your figure with ours.

Test conditions

• Substrate: a plate of S235JR structural steel at least 10 mm (0.39 in) thick. Thickness matters: with thin sheet the steel saturates magnetically and does not carry the whole flux, so the reading is lower. A 10 mm plate is thick enough to stop being the limiting factor.
• Surface: clean, dry, free of rust, mill scale, oil and paint. Not ground - as-supplied surface, degreased.
• Orientation: axial. The force acts perpendicular to the contact face, along the magnet axis. The mount sits flush over its whole face.
• Measurement: with a force gauge, increasing the load until breakaway. We record the peak value read immediately before separation.
• Temperature: ambient, around 20 °C (68 °F).

Every one of these conditions raises the figure relative to real-world use. That is why the rated pull force is a maximum, laboratory value - not a value to design against.

What lowers the figure in service

The cells marked [MEASUREMENT PENDING] are deliberately left empty. We would rather publish no number than an estimate we have not measured - an empty cell tells the truth, while a figure written in "for completeness" ends up in somebody else's calculation as fact.

Vertical pull versus side (shear) load

This is where most failures happen - and the most common misunderstanding at the specification stage.

Rated pull force is quoted for an axial load, perpendicular to the contact face. That is how a mount behaves when bolted to a ceiling with the weight hanging directly below it. The magnet then holds with its full adhesion.

If the mount is instead placed on a vertical wall, with the weight hanging off to the side, the situation is entirely different. The magnet still presses against the steel with the same force, but the load acts parallel to the surface and tries to slide the mount down. What holds it is now friction - and the coefficient of friction between clean, dry steel faces is a good deal less than one.

In practice: with a side eyebolt under a shear load, real holding force drops to 15-25% of the catalogue figure.

The practical conclusion: a mount rated at 50 kg (110 lb), placed on a vertical wall, will safely hold a few kilograms, not tens of them. If you are designing a side-loaded fixing, start from 15% of the catalogue value and apply your safety factor on top of that.

A separate word on moments: a weight on a cantilever arm does not only shear the joint, it tries to tip the mount over. The magnet then lifts away from the surface at one edge - and a gap of a few tenths of a millimetre at one edge is enough to lose the rest of the force in a cascade. That is why cantilevered fixings are designed with far more margin than axial ones.

Calculating pull force before you buy

You do not have to guess. There are five calculators on dhit.pl, two of which apply directly to specifying a mount.

Pull force calculator, and the "mounting type" field

The calculator at /kalkulator-magnesow/ works out breakaway force. It covers cylinders, blocks and rings, and material grades from N30 to N50 - for each grade we give the remanence Br range in kilogauss and a working limit of 80 °C (176 °F).

The most important part for mount selection, though, is a field you will not find in competing calculators: mounting type. You choose one of three:

Bare magnet (no housing)

The neodymium magnet on its own, placed against steel. The flux closes through the air around the magnet and part of the force escapes sideways. This is the reference case - and the weakest of the three.

Single-sided mount (magnet in a steel cup)

The magnet is set into a structural steel cup. The housing closes the flux on one side and directs it to the contact face. The same magnet delivers a noticeably higher breakaway force than in the bare case - and that is the whole point of a mount.

Double-sided mount (magnets on both faces)

An assembly with magnets on both sides of the housing, drawing two parts together or holding onto both surfaces. It behaves differently from a single-sided mount - the magnetic circuit is closed through both workpieces.

Changing the mounting type changes the result even when every other input stays the same. This is exactly the difference described earlier under magnet height: the same puck gives three different pull forces depending on how it is housed. Without this field, comparing calculators leads to conclusions that do not apply to a mount at all.

Field strength at a distance

The calculator at /kalkulator/sily/ gives field strength and flux density at a chosen distance. It covers blocks, discs, rings and spheres, grades N30 to N50, and you pick the gap from a list: 0, 1, 2, 5, 10, 20 or 50 mm.

Why it matters for mounts: it shows how quickly the field falls off with distance. If there is any layer between magnet and steel - paint, rubber, a sticker, a film of dust - that is not contact, that is a gap. The calculator shows how much field remains at 1 mm and how much at 2 mm. The answer tends to surprise people running the numbers for the first time.

Unit converters

Datasheets from different sources quote parameters in different systems: gauss and oersted in CGS, tesla and kiloamperes per metre in SI, energy product as MGOe or kJ/m³. The calculators at /kalkulator/jednostek-si/ and /kalkulator/jednostek-si-na-cgs/ convert between them in both directions.

Useful when comparing documentation from two suppliers who quote the same thing in two systems - and when checking whether "12 kGs" and "1.2 T" are the same value. (They are.)

Magnetic moment and permeance coefficient

Two more technical calculators, useful in design work rather than routine selection.

Magnetic moment - /kalkulator/moment-magnetyczny/ - computes m from m = Br · V / μ₀, in ampere square metres. It covers cylinders, blocks, rings and spheres. Remanence Br is taken as the midpoint of the grade band, and that value is shown explicitly alongside the result so you can see where it came from. An important caveat: magnetic moment does not predict pull force. It describes the magnet as a source of field, not as a holding device - there is a separate calculator for pull.

Permeance coefficient - /kalkulator/wspolczynnik-permeancji/, currently under construction - quantifies something mentioned earlier under magnet height. Pc depends on geometry, not on material grade. The clearest illustration: two blocks of exactly the same volume, 200 mm³, but with different proportions, differ in permeance coefficient by a factor of eleven - a 20 × 10 × 1 mm block has a Pc of about 0.13, while a 10 × 5 × 4 mm block has a Pc of about 1.49. The same amount of magnet material, the same grade, and completely different behaviour in a magnetic circuit.

An honest caveat: the permeance calculator does not locate the knee of the demagnetisation curve, because we do not have B-H curves for our materials. It gives the geometric permeance coefficient and stops there. The point at which a magnet begins to demagnetise permanently would require measuring that curve - and we have not done so.

Installation, use and maintenance

Fitting

• Degrease the contact face. Oil, grease and cutting fluid form a layer that behaves like a gap. A rag and a degreaser will do.
• Remove swarf and rust. A magnet draws swarf out of its surroundings and collects it around the rim of the contact face - a natural process, but one you need to keep in check.
• Approach edge-first. With higher-rated mounts, presenting the face flat ends in a snatch and crushed fingers. Set the mount down on its edge and roll it flat.
• Tighten by feel. A thread in a thin steel base strips more easily than you would expect. A nut pulled up with a ring spanner at full arm's length is usually too much.
• Check that it sits square. A mount bolted at an angle does not seat over its whole face - and partial contact means partial pull force.

Removal

You will not pull a mount rated at tens of kilograms straight off. Slide it sideways across the surface - in shear the force is many times lower, for exactly the same reason that side load capacity is 15-25% of the rated figure. With larger mounts use leverage: a wedge, or a flat screwdriver under the edge of the housing.

Lever against the housing, never against the magnet. The magnet material is brittle and will chip at the edge.

Storage

• With a keeper, or in pairs. A mount stored on a steel keeper plate has a closed magnetic circuit and does not attract dust or swarf. With two mounts, store them stuck together with opposing poles.
• Away from electronics and media. Especially magnetic stripe cards, measuring instruments and hard drives.
• Dry. The NiCuNi coating protects the magnet but is not hermetic - and a zinc-plated housing corrodes in contact with water like any other galvanised steel.
• Not above 80 °C (176 °F). This includes storage in a car in summer and anywhere near heating equipment.

Periodic inspection

A mount in a permanent installation is worth inspecting from time to time. What to look at:

• Whether the contact face is clean, or a layer of dust and swarf has built up.
• Whether the eyebolt is free of play in the housing and shows no sign of pulling out.
• Whether the thread has been stretched, and whether vibration has backed the nut off.
• Whether the housing coating has corroded at the contact face - corrosion lifts the mount and reduces pull force.
• Whether the mount has taken an impact - a cracked magnet inside the housing is invisible from outside but shows up as a sudden loss of force.

How do I remove a mount that will not come off?

Do not pull straight out - slide it sideways along the surface. In shear the holding force is many times lower than in tension, precisely because friction is doing the work rather than adhesion. With larger mounts, lever the housing up with a wedge or a flat screwdriver worked in under the steel cup. Do not lever against the magnet itself - it is brittle and will chip at the edge.

Materials referenced in this guide

NdFeB (neodymium-iron-boron)

A magnetically hard material based on the intermetallic compound Nd₂Fe₁₄B. The strongest permanent magnets available commercially. Produced by powder metallurgy: pressed powder is sintered in a magnetic field, then machined to size and magnetised. Hard and brittle like ceramic - it cannot be machined with cutting tools, only ground. It corrodes strongly in humid air, hence the protective coating in every case. Standard working limit for grades without high-temperature additives: 80 °C (176 °F). Grades are designated with the letter N and a number (N30 to N50 in our calculators), where the number corresponds approximately to maximum energy product in MGOe.

NiCuNi (nickel-copper-nickel)

The standard protective coating for neodymium magnets, electroplated in three layers: nickel, copper, nickel. Total thickness of the order of tens of micrometres. Hard, silver in appearance, with good abrasion resistance and sufficient corrosion resistance indoors. It is not hermetic - under prolonged immersion, or where the coating is damaged, the magnet can begin to corrode beneath it. Outdoor and wet applications call for additional barrier layers: epoxy, rubber, or a stainless steel housing.

A2 stainless steel (AISI 304, X5CrNi18-10)

An austenitic chromium-nickel steel containing around 18% chromium and 8-10% nickel. The A2 designation applies to fasteners - bolts, nuts and washers. Good resistance to atmospheric corrosion and to moderately aggressive environments. Effectively non-magnetic as supplied, which has two consequences: a magnetic mount will not hold on a surface made of it, and A2 fasteners do not disturb the mount's magnetic circuit. Its tensile strength is lower than that of grade 8.8 carbon steel bolts - loaded near the rated pull force, an A2 bolt may be the link that gives way first.

S235JR

A low-carbon non-alloy structural steel, among the most widely used in construction and steelwork. Minimum yield strength 235 MPa; the JR designation refers to impact toughness at room temperature. Ferromagnetic, with high magnetic permeability and high saturation induction - which makes it a good reference material for measuring magnet pull force. We use S235JR plate at least 10 mm (0.39 in) thick as the standard substrate in our pull force measurements, because at that thickness the steel carries the entire magnetic flux and does not limit the result. Figures obtained on this substrate are maxima - every departure from it (thinner sheet, a different grade, any layer on the surface) reduces the pull force.

Zinc-plated steel

Structural steel with a zinc coating, applied by hot dip or electroplating. The standard construction for magnetic mount housings. Zinc protects the steel cathodically - it corrodes in place of the steel, including where the coating is locally damaged. Adequate for indoor use, in workshops and production halls. Under constant contact with water or in an aggressive atmosphere the coating is consumed and the housing begins to rust - corrosion at the contact face lifts the mount off the surface and reduces pull force.

Calculators referenced in this guide

• Pull force calculator - breakaway force for cylinders, blocks and rings, grades N30 to N50, with a mounting type selector: bare magnet, single-sided mount, double-sided mount.
• Field at a distance calculator - field strength and flux density for blocks, discs, rings and spheres, at gaps from 0 to 50 mm.
• SI unit converter and SI to CGS converter - gauss, tesla, oersted, kA/m, MGOe and kJ/m³.
• Magnetic moment calculator - m in A·m² for cylinders, blocks, rings and spheres. Does not predict pull force.
• Permeance coefficient calculator - under construction. Pc depends on geometry, not on material grade. Does not locate the knee of the demagnetisation curve.