{"id":8400,"date":"2026-05-23T17:03:54","date_gmt":"2026-05-23T09:03:54","guid":{"rendered":"https:\/\/www.endlesswiresaw.com\/?p=8400"},"modified":"2026-07-13T14:24:40","modified_gmt":"2026-07-13T06:24:40","slug":"smco-ferrite-cutting","status":"publish","type":"post","link":"https:\/\/www.endlesswiresaw.com\/tr\/smco-ferrite-cutting\/","title":{"rendered":"Ferrite Magnet Cutting for SmCo and Ferrite Magnets"},"content":{"rendered":"

SmCo magnets crack without warning. We’ve had a customer send us a batch of Sm2Co17 arc segments \u2014 40 pieces total \u2014 and ask why their blade saw was destroying 1 in 4 during slicing. The answer was thermal shock. SmCo has a fracture toughness around 1.0 MPa\u00b7m^(1\/2), roughly the same as window glass, and any heat-generating process concentrates stress right where you don’t want it. Ferrite magnet cutting presents a similar challenge: strontium ferrite (SrFe12O19) is cheap and everywhere, but it chips like a ceramic tile if the feed force is even slightly too high.<\/p>\n\n\n\n

Both materials demand a cold, low-force cutting method. That’s where diamond wire comes in \u2014 and why we’ve developed specific parameter sets for SmCo and ferrite that keep edge chipping below 50 \u03bcm on production parts.<\/p>\n\n\n\n

This article covers the cutting characteristics of SmCo and ferrite magnets, the specific wire saw parameters that work, and the mistakes that lead to cracked parts. If you’ve already read our overview of diamond wire cutting for magnetic materials<\/a>, this goes deeper into the two material families that cause the most trouble on the shop floor.<\/p>\n\n\n\n

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

Why Are SmCo and Ferrite Magnets So Difficult to Cut?<\/h2>\n\n\n\n

The short answer: they’re both ceramics, mechanically speaking.<\/p>\n\n\n\n

SmCo (both Sm1Co5 and Sm2Co17 grades) is a sintered intermetallic compound. It’s hard \u2014 typically HRC 55\u201365 depending on the grade \u2014 and has almost no plastic deformation capacity. When stress exceeds the fracture threshold, the material doesn’t bend. It snaps. The fracture toughness of Sm2Co17 sits around 1.0\u20131.2 MPa\u00b7m^(1\/2), which is comparable to NdFeB magnets<\/a> but with one critical difference: SmCo’s Curie temperature is 700\u2013800\u00b0C versus NdFeB’s 310\u2013340\u00b0C. That means SmCo gets used in aerospace actuators, military-grade sensors, and downhole drilling tools where temperatures routinely exceed 150\u00b0C. You can’t afford edge damage on these parts \u2014 they go into assemblies where dimensional tolerance is tight and surface cracks propagate under vibration loading.<\/p>\n\n\n\n

Ferrite is a different animal. Barium ferrite (BaFe12O19) and strontium ferrite (SrFe12O19) are true ceramics \u2014 oxide-based, sintered at 1200\u20131300\u00b0C, with Vickers hardness around HV 500\u2013600. They’re the most widely produced permanent magnets in the world by volume, largely because the raw materials (iron oxide + strontium\/barium carbonate) are abundant and inexpensive compared to rare earth elements. But that low cost comes with a catch: ferrite is extremely brittle. We’ve seen ferrite arc segments crack from nothing more than clamping pressure during fixturing.<\/p>\n\n\n\n

The table below summarizes the key mechanical differences:<\/p>\n\n\n\n

Property<\/th>SmCo (Sm2Co17)<\/th>Ferrite (SrFe12O19)<\/th>NdFeB (N35\u2013N52)<\/th><\/tr><\/thead>
Hardness<\/td>HRC 55\u201365<\/td>HV 500\u2013600<\/td>HRC 57\u201361<\/td><\/tr>
Fracture toughness (K_IC)<\/td>1.0\u20131.2 MPa\u00b7m^(1\/2)<\/td>0.9\u20131.1 MPa\u00b7m^(1\/2)<\/td>1.0\u20131.5 MPa\u00b7m^(1\/2)<\/td><\/tr>
Curie temperature<\/td>700\u2013800\u00b0C<\/td>450\u2013460\u00b0C<\/td>310\u2013340\u00b0C<\/td><\/tr>
Density<\/td>8.2\u20138.4 g\/cm\u00b3<\/td>4.8\u20135.0 g\/cm\u00b3<\/td>7.4\u20137.6 g\/cm\u00b3<\/td><\/tr>
Primary failure mode in cutting<\/td>Thermal microcracking<\/td>Edge chipping \/ spalling<\/td>Edge chipping + oxidation<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

Both materials fail by brittle fracture. But they fail differently \u2014 and that difference matters for how you set up the cut.<\/p>\n\n\n\n

How Does Diamond Wire Cutting Solve the Cracking Problem?<\/h2>\n\n\n\n

Traditional abrasive blade saws generate localized heat at the cut zone \u2014 typically 200\u2013400\u00b0C at the contact point, even with coolant. For SmCo, that thermal gradient creates residual stress that initiates microcracks below the cut surface. For ferrite, the blade’s lateral force causes edge spalling because the material has essentially zero ductility.<\/p>\n\n\n\n

Diamond wire cutting eliminates both problems. The endless diamond wire loop<\/a> runs unidirectionally at controlled speed, distributing the cutting load across thousands of diamond grit points simultaneously. The contact area is tiny \u2014 a 0.35 mm diameter wire touches the workpiece across less than 0.5 mm\u00b2 \u2014 so cutting forces stay below the fracture threshold.<\/p>\n\n\n\n

We typically run ferrite magnet cutting and SmCo cutting on our SG20<\/a> or SG20-R<\/a> desktop precision wire saws. These machines give you the control you need: adjustable wire speed from 0 to 60 m\/s, tension control from 0 to 150 N, and programmable feed rates down to 0.1 mm\/min. That level of granularity matters when the difference between a clean cut and a cracked part is 0.5 mm\/min of feed speed.<\/p>\n\n\n\n

\"NdFeB<\/figure>\n\n\n\n

What Are the Recommended Cutting Parameters?<\/h2>\n\n\n\n

Here’s where things get specific. We’ve tested hundreds of SmCo and ferrite samples across different grades and geometries. The parameters below are our production-validated starting points \u2014 not textbook numbers.<\/p>\n\n\n\n

SmCo Cutting Parameters<\/h3>\n\n\n\n
Parameter<\/th>Recommended Range<\/th>Notes<\/th><\/tr><\/thead>
Wire diameter<\/td>0.35\u20130.50 mm<\/td>0.42 mm is our default for arc segments<\/td><\/tr>
Wire tension<\/td>100\u2013130 N<\/td>Lower than NdFeB \u2014 SmCo is more crack-sensitive<\/td><\/tr>
Wire speed<\/td>30\u201350 m\/s<\/td>Don’t exceed 55 m\/s; diminishing returns on surface finish<\/td><\/tr>
Feed speed<\/td>1.0\u20132.5 mm\/min<\/td>Start at 1.5 mm\/min and adjust based on edge quality<\/td><\/tr>
Coolant<\/td>Water-based coolant<\/td>White mineral oil also works; avoid dry cutting<\/td><\/tr>
Kerf width<\/td>0.40\u20130.55 mm<\/td>Depends on wire diameter + grit size<\/td><\/tr>
Surface roughness<\/td>Ra 0.4\u20130.8 \u03bcm<\/td>Achievable without secondary grinding<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

One thing that tripped us up early: SmCo generates fine metallic-magnetic debris during cutting. These particles are attracted to the wire, the guide wheels, and basically every metal surface on the machine. If you don’t flush the cut zone adequately, the debris acts like a secondary abrasive and accelerates wire wear. We now recommend a minimum coolant flow rate of 2 L\/min directed at the cut entry point. For more on coolant setup, see our cooling and lubrication guide<\/a>.<\/p>\n\n\n\n

Ferrite Cutting Parameters<\/h3>\n\n\n\n
Parameter<\/th>Recommended Range<\/th>Notes<\/th><\/tr><\/thead>
Wire diameter<\/td>0.35\u20130.50 mm<\/td>0.35 mm for thin tiles (<3 mm thickness)<\/td><\/tr>
Wire tension<\/td>100\u2013140 N<\/td>Slightly higher than SmCo \u2014 ferrite tolerates more tension<\/td><\/tr>
Wire speed<\/td>35\u201360 m\/s<\/td>Higher speed improves edge quality on ferrite<\/td><\/tr>
Feed speed<\/td>1.5\u20133.0 mm\/min<\/td>Can push to 3.5 mm\/min on thick blocks (>15 mm)<\/td><\/tr>
Coolant<\/td>Water-based coolant or white mineral oil<\/td>Water-based preferred for ferrite dust management<\/td><\/tr>
Kerf width<\/td>0.40\u20130.55 mm<\/td>Same as SmCo<\/td><\/tr>
Surface roughness<\/td>Ra 0.5\u20131.0 \u03bcm<\/td>Ferrite surface is inherently grainier than SmCo<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

Ferrite is more forgiving than SmCo on feed speed \u2014 you can typically run 30\u201350% faster. But the edge chipping risk goes up sharply above 3 mm\/min on parts thinner than 5 mm. We’ve seen otherwise good cuts produce 100\u2013200 \u03bcm chip-out at the exit edge simply because the feed rate was 0.5 mm\/min too high.<\/p>\n\n\n\n

Fair warning: ferrite dust is abrasive and conductive. It gets everywhere. Clean the machine after every batch, and make sure your coolant filtration system can handle the particle load. We’ve had a customer skip cleaning for a week and end up with clogged coolant lines and a seized recirculation pump.<\/p>\n\n\n\n

For a detailed discussion of how wire speed, tension, and feed rate<\/a> interact, we’ve published a separate technical guide.<\/p>\n\n\n\n

How Do SmCo and Ferrite Cutting Compare to NdFeB?<\/h2>\n\n\n\n

If you’ve already cut NdFeB magnets, you’re halfway there. The cutting mechanics are similar \u2014 all three are brittle sintered materials removed by diamond abrasion. But the devil is in the parameter details.<\/p>\n\n\n\n

Factor<\/th>NdFeB<\/th>SmCo<\/th>Ferrite<\/th><\/tr><\/thead>
Feed speed tolerance<\/td>1.5\u20133 mm\/min<\/td>1.0\u20132.5 mm\/min<\/td>1.5\u20133.5 mm\/min<\/td><\/tr>
Wire tension sweet spot<\/td>120\u2013150 N<\/td>100\u2013130 N<\/td>100\u2013140 N<\/td><\/tr>
Magnetic debris issue<\/td>Moderate<\/td>High (stronger field)<\/td>Low (weaker field)<\/td><\/tr>
Edge chipping risk<\/td>Medium<\/td>High (microcracking)<\/td>High (spalling)<\/td><\/tr>
Coolant sensitivity<\/td>Moderate<\/td>High (thermal cracking)<\/td>Low<\/td><\/tr>
Oxidation risk<\/td>High (Fe content)<\/td>Low<\/td>None<\/td><\/tr>
Wire life per batch<\/td>~5 days<\/td>~4 days<\/td>~6 days<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

The biggest practical difference: SmCo requires more conservative parameters than NdFeB, not less. Many operators assume that because SmCo is a “high-performance” magnet, it must be tougher. It isn’t. Its fracture behavior is actually worse than most NdFeB grades, especially the Sm1Co5 composition, which has lower fracture toughness than Sm2Co17.<\/p>\n\n\n\n

Ferrite, on the other hand, is the most forgiving of the three for wire cutting. The main risk is exit-edge chipping, which you control by reducing feed speed in the last 2\u20133 mm of the cut. We program a 40% feed reduction for the final approach on thin ferrite tiles \u2014 this alone eliminates about 80% of the edge defects we used to see.<\/p>\n\n\n\n

For a full comparison with NdFeB cutting parameters, see our NdFeB magnet cutting guide<\/a>.<\/p>\n\n\n\n

\"Trapezoidal<\/figure>\n\n\n\n

What About Edge Quality and Crack Prevention?<\/h2>\n\n\n\n

Edge quality is the primary quality metric for both materials. Here’s what we measure and what causes problems.<\/p>\n\n\n\n

SmCo Edge Defects<\/h3>\n\n\n\n

The dominant failure mode in SmCo is subsurface microcracking \u2014 cracks that form 20\u201350 \u03bcm below the cut surface and aren’t visible to the naked eye. These cracks propagate under mechanical stress or thermal cycling in the final application. In aerospace actuators running at 150\u2013200\u00b0C with vibration loading, subsurface cracks from cutting can cause field failure within 6\u201312 months.<\/p>\n\n\n\n

To minimize subsurface damage:<\/p>\n\n\n\n

    \n
  1. Keep feed speed below 2.5 mm\/min.<\/strong> Above this, the diamond grit penetration depth exceeds the material’s critical stress threshold.<\/li>\n\n\n\n
  2. Use 0.42 mm wire or thinner.<\/strong> Thinner wire means lower per-grit cutting force.<\/li>\n\n\n\n
  3. Maintain consistent coolant flow.<\/strong> Any interruption \u2014 even 2\u20133 seconds \u2014 causes a thermal spike that initiates microcracking.<\/li>\n\n\n\n
  4. Avoid clamping pressure on thin sections.<\/strong> SmCo parts under 3 mm thick should be wax-mounted, not mechanically clamped.<\/li>\n<\/ol>\n\n\n\n

    Ferrite Edge Defects<\/h3>\n\n\n\n

    Ferrite’s failure mode is macro chipping \u2014 visible chunks breaking off the cut edge, typically 100\u2013500 \u03bcm in size. This is purely mechanical: the wire exit point creates a tensile stress that exceeds the material’s (very low) tensile strength.<\/p>\n\n\n\n

    Prevention is straightforward:<\/p>\n\n\n\n

      \n
    1. Reduce feed speed for the final 2\u20133 mm<\/strong> of each cut. We use 40\u201350% of the nominal feed rate.<\/li>\n\n\n\n
    2. Support the exit face.<\/strong> If possible, back the workpiece with a sacrificial material (acrylic or wax) on the wire exit side.<\/li>\n\n\n\n
    3. Cut from the larger cross-section toward the smaller one<\/strong> when slicing arc segments or irregular shapes.<\/li>\n\n\n\n
    4. Use higher wire speed (50\u201360 m\/s)<\/strong> for ferrite. The faster the wire moves, the smaller each diamond particle’s bite, and the lower the exit-edge stress.<\/li>\n<\/ol>\n\n\n\n

      In our experience, diamond wire cutting consistently delivers edge chip-out below 50 \u03bcm on both SmCo and ferrite when these guidelines are followed. That’s 3\u20135x better than what we’ve measured from blade saw cuts on the same materials.<\/p>\n\n\n\n

      What Equipment Do You Need for SmCo and Ferrite Magnet Cutting?<\/h2>\n\n\n\n

      For R&D and small-batch production (1\u201350 pieces\/day), a desktop precision wire saw handles both materials well. Our SG20<\/a> is the most common choice \u2014 it covers workpieces up to 20 mm height with \u00b10.03 mm precision and accepts electroplated diamond wire loops<\/a> in 0.35\u20130.50 mm diameters.<\/p>\n\n\n\n

      For arc segment cutting or parts requiring angular cuts, the SG20-R<\/a> adds a rotary axis that lets you orient the magnet for angled slicing without custom fixturing.<\/p>\n\n\n\n

      Key machine features that matter for magnet cutting:<\/p>\n\n\n\n