Rare earth markets, grade premiums, grinding time, coating lines, tooling amortization — a neodymium magnet quote is built from parts most buyers never see. This guide opens the box so you can spec smarter and negotiate from knowledge.
A finished custom NdFeB magnet passes through more than a dozen process steps: alloy melting, jet milling, aligned pressing, sintering, machining, coating, magnetizing, inspection, and packaging. Each step contributes to the unit price. The proportions vary by part, but a representative breakdown for a mid-volume custom sintered NdFeB part looks like this:
Two takeaways follow directly from this structure. First, material dominates — which is why grade selection and part volume (in cubic millimeters, not just count) matter so much. Second, machining is the largest controllable cost — which is why tolerances and geometry are where design decisions move the price most.
NdFeB alloy is roughly one-third rare earth metal by weight — primarily neodymium and praseodymium (traded together as "NdPr"), with dysprosium or terbium added to high-temperature grades. These elements trade on commodity markets with real volatility: NdPr oxide has moved more than 2× within single twelve-month windows in the past decade.
Magnet price scales with magnet volume, not footprint. Halving the thickness of a disc roughly halves its material cost. If a lower-profile magnet on a steel backing plate can meet your flux requirement, the material savings are often substantial.
Grade affects price along two independent axes — and buyers routinely underestimate the second one.
Moving from N35 to N52 increases the energy product about 50%, achieved through finer process control and higher alloy purity. The price premium from N35 to N52 is moderate — typically tens of percent, not multiples.
Moving from no suffix to SH, UH, or EH requires adding heavy rare earths (Dy/Tb), the most expensive inputs in the entire magnet. This is where premiums get large:
| Grade class | Max op. temp* | Relative cost index |
|---|---|---|
| N (no suffix) | 80 °C | 1.0 (baseline) |
| M | 100 °C | ~1.05–1.15 |
| H | 120 °C | ~1.1–1.3 |
| SH | 150 °C | ~1.3–1.6 |
| UH | 180 °C | ~1.6–2.2 |
| EH | 200 °C | ~2.0–2.8 |
*Indicative ranges; actual premiums move with heavy rare earth spot prices. Temperature ratings assume a favorable permeance coefficient.
Specifying UH "for margin" when the application never exceeds 100 °C can nearly double material cost for zero functional benefit. Conversely, under-specifying temperature class causes permanent flux loss in the field. Get the real thermal profile, then spec to it — our engineering team can run the load-line check with you.
Sintered NdFeB is too hard and brittle to mill or turn — every surface is diamond-ground or wire-EDM cut, and material removal is slow. Machining cost scales with:
Ask your supplier for their standard "block list" or nearest catalog size. A custom part that can be adapted to an existing pressed blank size skips custom pressing entirely and is machined from stock — often cutting both price and lead time significantly at prototype volumes.
Coating is a smaller slice of unit cost than material or machining, but the differences between options are real — and some coatings carry minimum-lot charges that dominate at low volume:
| Coating | Relative adder | Notes |
|---|---|---|
| Zinc | Lowest | Dry indoor use only |
| NiCuNi | Baseline standard | Default for most applications |
| Epoxy | Comparable to NiCuNi | Better in humidity; bondable surface |
| NiCuNi + Epoxy | Moderate | Two coating operations |
| Parylene | High | Batch vapor process; minimum lot charges significant |
| Gold over NiCu | High | Precious metal + extra plating step |
Coating choice also interacts with scrap: plating very small or very thin parts has higher handling loss, and some geometries can't be barrel-plated and must be racked — a labor adder worth asking about on unusual shapes.
Whether your part needs tooling depends on how it's made:
Every production lot carries fixed costs: furnace scheduling, machine setup, coating-line minimums, first-article inspection. On a 100-piece order those costs may exceed the value of the magnets themselves; on 10,000 pieces they disappear into the unit price. MOQs are the supplier's way of keeping fixed cost per unit sane — and they are often negotiable if you accept a lot charge instead.
For samples and pilot builds, the cheapest path is almost always machining from stock material in a standard grade with NiCuNi coating — even if production will use a pressed, epoxy-coated part. Qualify the design first; optimize the process at volume.
Magnet pricing falls with volume for structural reasons, not goodwill:
A typical price curve for a custom part looks like this — indicative shape, not a quote:
| Quantity | Typical unit price vs. 100 pcs |
|---|---|
| 100 | 1.00× (baseline) |
| 1,000 | ~0.55–0.70× |
| 10,000 | ~0.35–0.50× |
| 100,000 | ~0.25–0.40× |
The practical implication: always request quotes at multiple quantity breaks, including your realistic production volume — even if the first PO is small. It changes sourcing decisions and reveals where the curve flattens for your specific part.
Landed cost is more than the unit price on the quote:
Ranked roughly by impact, here is where real savings come from:
| # | Lever | Typical impact |
|---|---|---|
| 1 | Right-size the temperature class (don't over-spec SH/UH/EH) | Up to 30–50% material cost |
| 2 | Reduce magnet volume (thinner + steel backing, better circuit design) | Proportional to volume removed |
| 3 | Loosen non-critical tolerances to ±0.05 mm | 15–30% on machining-heavy parts |
| 4 | Adapt to a standard blank or catalog size | Eliminates NRE; cuts prototype cost |
| 5 | Consolidate volumes / annual blanket orders | Moves you down the price curve |
| 6 | Choose the standard coating unless environment demands otherwise | Avoids specialty-coating lot charges |
| 7 | Index long-term pricing to NdPr for volume programs | Removes risk padding from quotes |
Send your supplier the requirement — flux at the working point, temperature, environment — alongside the drawing, and ask one question: "Is there a lower-cost way to meet this?" Grade substitutions, geometry tweaks, and circuit suggestions from the people who make magnets daily routinely save more than any negotiation on the quoted number.
Everything in this guide converges on one point: the quality of your quote depends on the quality of your specification. A complete RFQ — dimensions and tolerances, full grade with temperature class, magnetization direction, coating or environment, quantities and breaks, and documentation requirements — gets you firm pricing quickly, with no padding for unknowns.
We've published a step-by-step companion guide with a copy-paste checklist: How to Prepare an RFQ for Custom Magnets. For material selection questions, see NdFeB vs. SmCo vs. Ferrite vs. Alnico, and for fundamentals, Magnets 101.