Magnet Should-Cost Calculator
Build the cost of a sintered NdFeB magnet from the bottom up — alloy content set by the temperature class, material yield, conversion, coating, tooling amortisation and landed cost. The output that matters most is the material share, because it tells you how much of any rare-earth-driven price increase is actually defensible.
interactive — should-cost build-up
A should-cost model is a negotiating instrument and a sanity check, not a quote. It tells you roughly what a part ought to cost from an efficient producer, so you can tell the difference between a price that moved because rare earth moved and a price that moved for other reasons. It cannot see a specific supplier's yields, utilisation, contract alloy position or freight lane. Expect real quotations to land inside the band, and treat a quote far outside it — in either direction — as a question worth asking rather than a conclusion.
what actually drives the number
The single most useful output above is the material share. Once you know what fraction of a part's cost is rare earth, you know how much of a price increase is defensible when the index moves — and how much is not.
| driver | behaviour |
|---|---|
| NdPr price | Roughly 30 % of magnet mass is NdPr. If material is half the piece price, a 20 % index move justifies about a 10 % price move — not 20 %. |
| heavy rare earth content | The steepest cliff in the whole model. Dysprosium and terbium cost several times NdPr, so each step up the temperature classes — H to SH to UH to EH — adds cost disproportionately. Grain boundary diffusion cuts the required content substantially and is worth asking about by name. |
| material yield | Sintered NdFeB is pressed oversize and machined down. Yield of 55–70 % is normal, so you buy appreciably more alloy than the finished part contains. Swarf has recoverable value, which is why the credit appears above. |
| tolerance | Grinding is where the labour is. Loosening a dimension that does not need to be tight is often the cheapest saving available. |
| quantity | Tooling amortisation dominates at low volume and disappears at high volume. Below a few thousand pieces a year, the mould can outweigh the material. |
| coating | Modest in absolute terms for most parts, but medical and specialist coatings are a real step change. |
| size | Very small parts cost more per kilogram — handling, plating racking and inspection are per piece, not per gram. |
The duty field defaults to zero deliberately. Applicable rates depend on classification at the ten-digit statistical level, country of origin, and any measures in force at the time of entry — and they change. Get the rate for your specific part and lane from your customs broker, then enter it here. Background on how the stack is assembled is in tariffs, duties & country of origin.
using this in a negotiation
The productive version of this conversation is not "your price is too high." It is "here is our build-up, here is where we think it differs from yours, which line is wrong?" That reframes the discussion from position to arithmetic, and a supplier operating efficiently will usually engage with it.
| situation | what to ask |
|---|---|
| quote well above the band | Ask which line differs. Common answers are genuine: lower yield on a difficult geometry, a small-lot surcharge, an alloy position bought at a different point in the cycle. |
| quote well below the band | Worth as much scrutiny. Ask what grade is actually being supplied and whether the heavy RE content matches the temperature class on the print. Verify with incoming inspection. |
| price increase citing rare earth | Apply the material share. If material is 45 % of the piece and NdPr rose 20 %, the defensible increase is roughly 9 %, not 20 %. |
| price does not fall when the index falls | Ask for the indexation mechanism in writing — a formula with a stated reference, lag and reset frequency. See rare earth price indexing. |
| cost reduction needed | Attack tolerance, temperature class and geometry before attacking price. Dropping from SH to H where the operating point allows it is usually worth more than a hard negotiation — check with the demagnetization calculator first. |
reference — the build-up
| finished mass | m = V · ρ, ρ ≈ 7,500 kg/m³ sintered NdFeB |
| alloy cost | C_alloy = w_NdPr · P_NdPr + w_HRE · k_gbd · P_HRE + w_Fe · P_Feweights are mass fractions of the alloy; k_gbd reduces heavy RE where grain boundary diffusion is used |
| material into the part | C_mat = (C_alloy / yield) − (1/yield − 1) · C_alloy · recoveryyou buy alloy at the inverse of yield, and recover part of the swarf value |
| conversion | C_conv = m · rate_complexitypressing, sintering, machining, inspection |
| tooling per piece | C_tool = tooling / (annual qty · amortisation years) |
| piece cost | C_piece = [ (C_mat + C_conv) · m + C_coat + C_tool + C_freight ] · (1 + duty) · (1 + margin) |
Composition figures are representative of commercial sintered NdFeB — roughly 29–31 % NdPr by mass with the balance iron, boron and minor additions — and heavy rare earth loadings by temperature class are typical rather than specific to any producer. Actual formulations are proprietary and vary between suppliers and between production lots. Conversion rates are order-of-magnitude figures for established Asian production; Western and defence-qualified supply chains carry materially different cost structures. This page is procurement reference information, not financial advice, and nothing here is an offer or a quotation.
going deeper — the qualitative version of this model, with the sourcing strategy around it, is in should-cost modeling and magnet pricing explained. For structural cost reduction rather than price negotiation, see cost reduction & value engineering. If you would like a real number instead of a modelled one, send us the print.