{"id":8285,"date":"2026-05-06T16:19:42","date_gmt":"2026-05-06T08:19:42","guid":{"rendered":"https:\/\/www.endlesswiresaw.com\/?p=8285"},"modified":"2026-06-06T11:38:30","modified_gmt":"2026-06-06T03:38:30","slug":"ndfeb-cutting-guide","status":"publish","type":"post","link":"https:\/\/www.endlesswiresaw.com\/tr\/ndfeb-cutting-guide\/","title":{"rendered":"NdFeB Magnet Cutting Process Using Endless Diamond Wire"},"content":{"rendered":"

Sintered NdFeB is one of the hardest permanent magnet materials to cut cleanly. The grain boundaries are brittle, the Nd-rich phase oxidizes fast when exposed to moisture, and any thermal input during cutting can permanently damage the microstructure \u2014 reducing the magnet’s achievable performance when it’s finally magnetized downstream. We’ve processed hundreds of NdFeB blocks on our endless diamond wire saws<\/a> \u2014 from small 10 \u00d7 10 mm lab samples to 60 \u00d7 40 mm production blocks \u2014 and the lesson is always the same: the cutting method you choose determines whether you end up with usable parts or expensive scrap.<\/p>\n\n\n\n

This guide covers what actually matters in NdFeB cutting: which methods work, which don’t, what process parameters to use, and how to avoid the chipping, cracking, and oxidation problems that plague most magnet machining operations.<\/p>\n\n\n\n

\"Vimfun<\/figure>\n\n\n\n

Why Is NdFeB So Difficult to Cut?<\/h2>\n\n\n\n

NdFeB isn’t metal in any practical machining sense. It’s a sintered powder metallurgy product \u2014 as classified under IEC 60404-8-1<\/a> for hard magnetic materials \u2014 essentially ceramic particles bonded at grain boundaries by a thin Nd-rich phase. That microstructure creates three problems that interact during cutting:<\/p>\n\n\n\n

Brittleness without warning.<\/strong> NdFeB has a Vickers hardness around HV 570\u2013650, comparable to hardened tool steel. But unlike steel, it has near-zero ductility. There’s no plastic deformation before fracture \u2014 cracks initiate at grain boundaries and propagate instantly. A 5\u201310 cm drop onto a hard surface can shatter a finished magnet. During cutting, any lateral force or vibration at the wrong moment produces edge chipping that no amount of post-processing can fix.<\/p>\n\n\n\n

Oxidation sensitivity.<\/strong> The Nd-rich grain boundary phase is chemically reactive. Expose a freshly cut NdFeB surface to humid air, and you’ll see discoloration within hours. Use water-based coolant without corrosion inhibitors, and the cut surface develops a grey oxide layer that weakens the grain boundaries further. This is why coolant selection during NdFeB cutting isn’t optional \u2014 it’s a process-critical decision. We’ll cover this in detail in our cooling and lubrication guide<\/a>.<\/p>\n\n\n\n

Thermal sensitivity of the microstructure.<\/strong> NdFeB cutting happens before magnetization \u2014 the blanks coming off the sintering furnace are unmagnetized, so demagnetization during cutting isn’t the concern. What IS a concern is thermal damage to the grain boundary microstructure. The Nd-rich phase that bonds grains together begins to degrade above 200 \u00b0C, and excessive localized heating from EDM or aggressive blade cutting can create micro-voids and phase changes at grain boundaries. These defects don’t show up visually, but they reduce the magnet’s achievable energy product (BHmax) when it’s finally magnetized downstream. We’ve seen batches lose 3\u20138% of their rated grade performance because of cutting-stage thermal damage that wasn’t caught until final QC.<\/p>\n\n\n\n

One practical issue worth mentioning: NdFeB cutting swarf contains fine iron-rich particles. While the workpiece itself isn’t magnetized during cutting, the swarf can still cause problems \u2014 it clogs coolant filters quickly, embeds in soft guide wheel grooves, and creates abrasive contamination if not managed. A good filtration system in the coolant loop is essential, not optional.<\/p>\n\n\n\n

How NdFeB Cutting Methods Compare<\/h2>\n\n\n\n

Not every cutting technology handles NdFeB equally. Here’s what we’ve observed across different methods, based on both our own testing and feedback from customers who switched to diamond wire after struggling with other approaches.<\/p>\n\n\n\n

Inner-Diameter (ID) Blade Cutting<\/h3>\n\n\n\n

This is the traditional go-to for small NdFeB parts. An annular blade with diamond abrasive on the inner edge rotates at high speed while the workpiece feeds through. It works \u2014 but with limitations.<\/p>\n\n\n\n

Blade thickness is typically 0.3\u20130.5 mm, which means kerf loss is significant when you’re cutting expensive rare-earth material. For a 50 mm block sliced into 2 mm wafers, you lose roughly 15\u201320% of the raw material just to sawdust. The lateral cutting force from the rigid blade also creates a real chipping problem on entry and exit edges, especially on blocks thinner than 5 mm.<\/p>\n\n\n\n

Feed rates are typically limited to 1\u20132 mm\/min for clean cuts. Push harder, and you get subsurface micro-cracks that show up as coating adhesion failures downstream.<\/p>\n\n\n\n

\"magnet<\/figure>\n\n\n\n

EDM Wire Cutting<\/h3>\n\n\n\n

EDM works well for complex shapes \u2014 arcs, slots, custom profiles. The problem is physics: EDM removes material through electrical discharge, which means localized melting. The heat-affected zone on NdFeB typically extends 20\u201350 \u03bcm from the cut surface, creating recast layers with altered microstructure. Even though the blanks aren’t magnetized yet at this stage, the thermal damage to the Nd-rich grain boundary phase permanently degrades the material’s magnetic potential \u2014 you won’t get the full energy product when these parts are finally magnetized.<\/p>\n\n\n\n

There’s also the carbonization issue. The recast layer on the EDM-cut surface contains carbon deposits that interfere with NiCuNi plating adhesion. Customers doing precision motor assemblies have told us this is a persistent reject cause.<\/p>\n\n\n\n

Reciprocating Diamond Wire Saw<\/h3>\n\n\n\n

EDM also can’t cut ferrite<\/a> or any non-conductive magnetic material. If your production mix includes both NdFeB and ferrite, you need two different cutting setups.<\/p>\n\n\n\n

Long-wire saws using 1000+ meter wire spools with reciprocating motion were an improvement over blades for larger blocks. But the direction reversal \u2014 the wire stops, reverses, accelerates again \u2014 creates two problems. First, the reversal zones leave visible marks on the cut surface, creating periodic waviness that researchers have measured at PV values of 5\u201315 \u03bcm depending on wire speed and tension settings. Second, the maximum effective wire speed is limited to about 20 m\/s because the wire can’t accelerate fast enough between reversals.<\/p>\n\n\n\n

The equipment is also substantially more complex and expensive than what most small-to-medium magnet manufacturers can justify.<\/p>\n\n\n\n

Endless Diamond Wire Cutting \u2014 How It Solves These Problems<\/h3>\n\n\n\n

The core difference is simple: a short, closed-loop diamond wire<\/a> (typically 1\u20135 meters) runs continuously in one direction. No reversal. No acceleration cycles. No periodic marks.<\/p>\n\n\n\n

On our SG20-R<\/a> machines, we run NdFeB cutting with these typical parameters:<\/p>\n\n\n\n

Parameter<\/th>Typical Range<\/th>Notes<\/th><\/tr><\/thead>
Wire diameter<\/td>0.35\u20130.50 mm<\/td>Thinner wire = less kerf loss, but shorter life<\/td><\/tr>
Wire speed<\/td>30\u201360 m\/s<\/td>Higher speeds improve surface finish<\/td><\/tr>
Wire tension<\/td>100\u2013150 N<\/td>Lower than glass or quartz \u2014 NdFeB is relatively soft<\/td><\/tr>
Feed rate<\/td>1.5\u20133.0 mm\/min<\/td>Can push to 5 mm\/min on smaller cross-sections<\/td><\/tr>
Coolant<\/td>Oil-based (white mineral oil)<\/td>Prevents oxidation of freshly cut surface<\/td><\/tr>
Surface roughness (Ra)<\/td>0.3\u20130.5 \u03bcm<\/td>Often eliminates the need for grinding<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

A few things stand out from these numbers. The wire diameter at 0.35 mm gives a kerf of roughly 0.40\u20130.45 mm \u2014 about half the material loss compared to ID blade cutting. For NdFeB priced at $50\u201380\/kg for higher grades, that kerf reduction pays for itself quickly.<\/p>\n\n\n\n

The feed rate of 1.5\u20133.0 mm\/min might seem conservative compared to some competitor claims. In our experience, pushing beyond 3 mm\/min on blocks larger than 30 mm cross-section leads to increased wire deflection and measurable waviness on the cut surface. Research published in Materials<\/em> (MDPI)<\/a> confirms that feed rate is the dominant factor affecting both surface roughness and waviness in diamond wire sawing of NdFeB \u2014 more so than wire speed or workpiece size.<\/p>\n\n\n\n

\"magnet<\/figure>\n\n\n\n

What Happens at the Cut Surface<\/h2>\n\n\n\n

Understanding what’s actually going on at the microscopic level helps explain why process parameters matter.<\/p>\n\n\n\n

When diamond grits on the wire contact the NdFeB surface, material removal happens through a combination of micro-cutting and brittle fracture. At the grain level, the Nd\u2082Fe\u2081\u2084B phase is relatively hard but cleavable, while the Nd-rich grain boundary phase is softer and more ductile. The diamond grits cut through the grain boundary phase easily, but when they hit the main phase grains, the removal mechanism shifts to fracture along crystallographic planes.<\/p>\n\n\n\n

This dual-mechanism removal creates a characteristic surface morphology: relatively smooth plateaus where micro-cutting dominated, interspersed with small fracture pits where grains pulled out at the boundary. The density and depth of these fracture pits is what determines your final Ra value \u2014 and it’s directly controlled by feed rate and wire tension.<\/p>\n\n\n\n

Lower feed rates mean less force per grit, which keeps more of the removal in the micro-cutting regime rather than fracture. That’s the trade-off: productivity versus surface quality. For motor magnet segments where surface flatness directly affects air-gap consistency, we recommend staying at or below 2 mm\/min.<\/p>\n\n\n\n

Fair warning: if you see periodic waviness on your cut surfaces (regular ridges at 0.5\u20132 mm intervals), the cause is almost always wire lateral vibration, not feed rate. Check your guide wheel grooves<\/a> \u2014 worn grooves let the wire oscillate sideways, and the oscillation pattern transfers directly to the workpiece surface.<\/p>\n\n\n\n

Coolant Selection: Oil vs. Water for NdFeB<\/h2>\n\n\n\n

This is where NdFeB cutting diverges sharply from cutting other hard materials like sapphire<\/a> or quartz<\/a>.<\/p>\n\n\n\n

For most ceramics and crystals, water-based coolant works fine \u2014 it dissipates heat efficiently and is easy to manage. NdFeB is different because of the Nd-rich grain boundary phase. Water molecules react with exposed neodymium at the cut surface, forming Nd(OH)\u2083 and releasing hydrogen. This isn’t just surface discoloration \u2014 it’s active corrosion that weakens the grain boundaries and can cause delayed cracking hours or days after cutting.<\/p>\n\n\n\n

Our recommendation: use oil-based coolant (white mineral oil or cutting oil) for all NdFeB cutting operations.<\/strong> The oil film protects the freshly exposed surface from moisture contact during cutting and provides adequate lubrication for the diamond wire. Heat dissipation is slightly lower than water, but since the endless wire cutting process generates minimal heat to begin with, this isn’t a practical issue.<\/p>\n\n\n\n

If you must use water-based coolant (some production environments require it for fire safety or environmental reasons), add a corrosion inhibitor at minimum 3% concentration and ensure cut parts are dried and oil-coated within 30 minutes of cutting. Any longer, and you risk surface degradation that affects downstream coating adhesion.<\/p>\n\n\n\n

For a detailed comparison of coolant options across different magnetic materials, see our cooling and lubrication technical guide<\/a>.<\/p>\n\n\n\n

Fixturing: Clamping Fragile Sintered Blanks Without Damage<\/h2>\n\n\n\n

Since NdFeB blanks are unmagnetized during cutting (magnetization is a downstream step), you don’t have to worry about magnetic attraction between workpiece and fixture. The real challenge is mechanical: sintered NdFeB is brittle, and excessive clamping pressure can initiate cracks before the wire even touches the material.<\/p>\n\n\n\n

We use aluminum worktables as standard on all our magnet cutting setups \u2014 not because of magnetic concerns, but because aluminum is softer than NdFeB and won’t chip the workpiece edges during loading. For clamping force, less is more. The goal is to prevent workpiece movement during cutting, not to crush it into position.<\/p>\n\n\n\n

For thin blanks (under 3 mm final thickness), mechanical clamping often applies too much localized stress. Adhesive mounting on a sacrificial substrate works better \u2014 wax or UV-release adhesive holds the workpiece firmly during cutting and releases cleanly afterwards without damaging the cut surfaces. This approach also eliminates the vibration that can occur when clamp contact points shift slightly during the cut.<\/p>\n\n\n\n

For larger blocks and batch production, a simple vise with rubber-padded jaws provides adequate holding force without the risk of edge damage.<\/p>\n\n\n\n

More detail on fixture design options in our workpiece fixture guide<\/a>.<\/p>\n\n\n\n

\n
\"YouTube