Editorial Article
Twenty-Six Per Cent of This Market Sets the Price the Other Seventy-Four Per Cent Has to Live With
Twenty-six per cent of this market sets the floor for the other seventy-four
The global market for magnetic mill liners is USD 37.8 million in 2025 on our construction: more than six hundred installed applications at roughly USD 63,000 of annualised liner spend each, spread from the Chilean and Peruvian porphyries to the Chinese iron ore milling base. It reaches USD 66.0 million in 2033 at 7.23 per cent, an increment of USD 28.2 million.
Inside that number sit three supply models that share a relining slot and very little else. The Eriez and JinFa alliance held about 44 per cent of magnetic liner value in 2025, or USD 16.6 million. Chinese independents held 26 per cent, or USD 9.8 million. Metso‘s Orebed held 18 per cent, or USD 6.8 million. Conventional liner houses running magnetic programmes inside a foundry portfolio took the remaining 12 per cent.
| Supplier group | 2025 share of magnetic liner value | Approximate 2025 value | Basis of position |
|---|---|---|---|
| Eriez and JinFa alliance | 44% | USD 16.6m | Metal magnetic liner, more than 600 installations, engineering and selection service |
| Chinese independents | 26% | USD 9.8m | Installed price, delivery speed, willingness to design to undocumented mills |
| Metso (Orebed) | 18% | USD 6.8m | OEM specification at design stage, boltless rubber, narrow duty positioning |
| Conventional houses with magnetic programmes | 12% | USD 4.5m | Existing reline relationships, portfolio bundling |
Pheonix estimate. Shares are of magnetic liner value, not of the mill liner market, and are derived from installation counts, disclosed capacity and installed-price differentials rather than from audited segment reporting, which does not exist in this category. Value columns are our arithmetic on the USD 37.8 million base.
The obvious reading of that table is a share fight inside a small category. It is the wrong reading, and it leads to the wrong commercial decision. The 26 per cent is not primarily taking business from the two technology holders at mills those two already serve. It is setting the price at which a magnetic liner can be sold at all — and in doing so it decides which mills the top end can serve profitably. A supplier that answers a price floor with discounting is answering a question nobody asked.
Three models, three cost bases, one relining slot
The licensor-plus-manufacturer model is an alliance rather than a factory. Eriez, a magnetic separation house based in Erie, Pennsylvania, formed a strategic alliance in 2005 with JinFa Industrial Trading, a subsidiary of China Metallurgical Mining Corporation, which had decades of experience with magnetic mill liners. The division of labour is explicit in the company’s own literature: Eriez brings the magnetics and the customer relationship, JinFa brings the liner. The alliance provides engineering services across selection, design to a specific mill’s features and in-plant installation assistance, and it publishes discrete-element-method simulation work for a 16.5 by 37.5 foot mill. The result is stated plainly — more than 600 applications worldwide, with the longest-serving liner in service for over ten years.
The OEM-integrated model is specification rather than maintenance. Metso developed and patented a magnetic lining system in the 1970s and has designed magnetic linings for more than forty years. Orebed embeds permanent magnets in abrasion-resistant rubber, needs no bolts, and relies on the mill load to build its own wear layer. It is sold through a channel that also supplies the mill, the crushers, the screens and the service contract, which means it can be written into the design of a concentrator rather than sold into a maintenance budget afterwards. Metso’s 2025 numbers show why that channel matters: group sales of EUR 5,240 million, orders received of EUR 5,471 million and adjusted EBITA of EUR 829 million at 15.8 per cent, with aftermarket at 54 per cent of group sales and 60 per cent of the Minerals segment. Wear parts are the recurring half of that business.
The price-led long tail is a set of Chinese manufacturers selling metal magnetic liners and rubber-matrix magnetic liners, frequently through trading intermediaries and increasingly through English-language product sites. They compete on installed price, on delivery time, and on willingness to copy a design for a mill the incumbent does not hold drawings for. Their quality is not uniform and neither is their engineering support, which is precisely the gap the two technology holders sell against.
What separates the three is not capacity. The parent category has capacity to spare: ME Elecmetal runs 10 manufacturing plants and 2 service centres in 7 countries with 140,000 tonnes of global foundry capacity and 600,000 tonnes of grinding media capacity, and Bradken operates five liner sites rated between 10,000 and 30,000 tonnes a year each. None of it is dedicated to magnetic liners. What separates the models is where the engineering cost sits. Our reconstruction of delivered cost for a supplied set puts magnet material at 24 to 30 per cent, alloy casing and casting at 26 to 32 per cent, design and application engineering at 12 to 18 per cent, assembly and freight at 10 to 14 per cent, installation supervision and service at 8 to 12 per cent, and supplier gross margin across an 18 to 26 per cent band. The design link is the one with near-zero marginal cost on a repeat mill and the highest barrier on a new one, because it depends on accumulated mill data. A supplier competing on magnet cost alone has a few points of gross margin to play with. That is why the long tail does not fund application engineering, and why it does not try.
The floor does not take customers. It decides which customers exist
At the bottom of the band, a supplier sells a set of magnet-cased pieces to a drawing. At the top, the technology holders sell a package: mill survey, design, simulation, installation assistance and a wear-life commitment. For a large porphyry copper circuit where a failed reline costs a week of throughput, the package wins. For a 3,000 tonne per day iron ore concentrator in Hebei, or a small polymetallic mill in Inner Mongolia, the package is overhead and the drawing wins.
The consequence is that the top end’s addressable share of the global pool is smaller than the pool itself. The honest measure of the commercial opportunity is not a market size at all. It is the number of mills where a stoppage is expensive — and that number cannot be derived from ore tonnage, regional mining output or a growth rate. It is a count of individual shells.
This also explains an apparent anomaly in the share table. A 44 per cent share of a category this small is worth less, defensively, than it looks, and a 26 per cent share is worth more. Conversion here is sequential inside an account rather than parallel across a market: a mill that has run a magnetic liner for two years without a reline becomes the reference for the next mill in the plant, and then for the sister operation in the same group. The strategic asset is the installation count and the mill data behind it. Six hundred reference mills cannot be bought, and a 44 per cent revenue share does not survive a competitor who can answer the only question a maintenance superintendent asks — what happened at the last mill that looked like mine?
What a supplier does differently with that
- Segment the target list by the cost of a stoppage, not by ore tonnage. The eligible pool is fixed by mineralogy: roughly 28 per cent of the USD 793 million ball-mill liner pool, about USD 222 million. The addressable subset inside that pool is fixed by how much a reline costs the plant in lost tonnes. Those are two different maps, and only the second one can be sold against.
- Sell the survey before the liner. Liner and lifter geometry govern wear more than the material in the liner does. The work in the Journal of the Southern African Institute of Mining and Metallurgy on total primary milling cost records a plant that converted from overflow to an aggressive grate discharge design and saw grinding media wear rise from under 400 g/t to 1,200 g/t — three times — with breakage and spalling of both balls and liners, before converting back. The same body of work records roughly 30 per cent less media wear in overflow than grate discharge for the same alloy. A magnetic liner installed into a badly designed discharge or lifter arrangement will underperform, and the supplier will be blamed for a geometry problem. The survey is not overhead. It is the difference between a reference site and a lost one.
- Do not defend price. Defend the relining window. A reline stops the mill, and if the mill is the bottleneck the cost of the stop is measured in tonnes of concentrate not made. Everything else — price per kilogram, wear life, installation safety — is downstream of how often the mill is stopped and for how long. That is why a thinner lining is a throughput argument, since less shell height is more working volume and a larger effective diameter in the same shell; why pieces around 300 mm long and 17 kg are an availability argument, since they are carried by hand rather than rigged with a crane; and why a boltless rubber system can charge a premium for what is, on paper, a rubber lining. Comminution accounts for roughly 25 per cent of final energy consumption at an average mine site, and possibly up to 1 per cent of total final energy consumption globally, on the CEEC mining energy analysis — which is why anything that raises the efficiency of the mill’s power draw carries a return larger than a wear-part saving.
- Treat the undocumented-mill segment as a qualification decision, not a price war. A large share of the world’s older ball mills have no reliable shell drawings. That is precisely the segment the long tail serves and the technology holders cannot economically survey. Winning it on price means funding an engineering link the price does not carry.
What this analysis does not claim
The shares above are Pheonix estimates, and the value columns are our multiplication of them into the USD 37.8 million base. The cost-stack percentages are reconstructions from installed-price ranges and disclosed foundry economics, not supplier accounts, which are not published. The 17 per cent penetration and the 28 per cent eligibility share are our reconstructions, and changing either moves the headline.
On method, this piece follows our two-layer approach at roughly 70 per cent secondary and 30 per cent primary. The mechanisms it describes — why a price floor reshapes a target list rather than only a price, why a maintenance superintendent budgets a relining interval rather than a mechanism, why the long tail cannot fund application engineering — come from the primary layer behind this edition: our own bottom-up mill-level reconstruction of installed grinding capacity and liner consumption by region and ore type, open-ended interviews with mill maintenance and supply-chain personnel in the secondary and regrind segment, a structured review of comminution conference material, and a willingness-to-pay assessment built around liner replacement intervals rather than liner price. That layer establishes how the market works. It never carries a number. Every figure in this piece is from a supplier document, a public filing, a geological survey, peer-reviewed literature or our own model. No third-party market-research publication is used or cited.
The full segment, share, cost-stack, regional and forecast tables are in Global Magnetic Mill Liners Market, 2026-2033.
Companion reading
The technical half of this argument — Nobody Can Measure the Wear Layer, and That Is What Caps This Market at USD 66 Million — is a different problem from the commercial half, and it constrains the same number: no standard in-situ method exists for measuring a self-renewing wear layer on a running mill, so no supplier can warrant a relining interval — the one figure a shutdown calendar can hold. The rest of the market sits in the Mining segment, where the ore supply that sets milled volume is covered in our reports on lithium, nickel, cobalt, graphite and manganese. None of those is about liners. All of them set the tonnage that becomes liner spend.
Frequently Asked Questions
How big is the magnetic mill liner market?
USD 37.8 million in 2025 on our construction, rising to USD 66.0 million in 2033 at a 7.23 per cent compound annual growth rate — an increment of USD 28.2 million. The 2025 figure is built from the mill upward: more than six hundred installed applications at roughly USD 63,000 of annualised liner spend each. It sits inside a USD 1.76 billion global mill liner market of about 410,000 tonnes at a blended USD 4,300 a tonne, of which secondary and regrind ball mills take about 45 per cent.
Who leads the magnetic mill liner market?
The Eriez and JinFa alliance held about 44 per cent of magnetic liner value in 2025, Chinese independents 26 per cent, Metso's Orebed 18 per cent and conventional liner houses running magnetic programmes 12 per cent. Those are Pheonix estimates derived from installation counts, disclosed capacity and installed-price differentials, not audited segment reporting, which does not exist in this category.
Why does a 44 per cent share not protect the leader?
Because share of the magnetic category is a poor guide to position. The pool of convertible mills is fixed by ore mineralogy, so what matters is share of the mills that could convert and have not. Conversion is also sequential inside an account rather than parallel across a market, which makes the installation count and the mill data behind it the strategic asset — six hundred reference mills cannot be bought.
What should a supplier do differently?
Segment the target list by the cost of a stoppage rather than by ore tonnage, sell the mill survey before the liner, defend the relining window rather than price per kilogram, and treat mills with no reliable shell drawings as a qualification decision rather than a price war. The honest measure of the opportunity is a count of individual mills where a stoppage is expensive, which no growth rate expresses.
