Printing, casting, or sintering: where is the boundary between LPBF, casting, and MIM

When a metal part is too complex to machine from a block, there are three classic methods: casting, powder injection molding and sintering (MIM), or printing using laser powder bed fusion. These methods are often misleadingly compared because they are almost always compared based on the price of a part, and that's precisely the number that reveals the least about the process.

What truly separates the three is what you do before you make the first part.

The axis that explains almost everything: the tooling

Lost-wax casting requires a wax mold. MIM requires a steel injection mold. Both are investments of thousands to tens of thousands of euros and take weeks to months to complete before you have anything in hand.

The LPBF requires no tooling. The first part and the two hundredth cost virtually the same.

That's where the geometry of the decision comes from, and it's surprisingly simple:

Short or uncertain production run → LPBF (Low-Process Fixed-Function). There's nothing to amortize, and if the design changes, you just update the file.
Long production run and frozen design → mold. Tooling costs are spread across thousands of parts, and the unit cost plummets.
Somewhere in between → depends on the geometry, and that's where you have to sit down and crunch the numbers.

The crossover point falls, very roughly speaking, between a few hundred and a few thousand parts per year for small parts. This isn't a figure that can be given in general: it varies considerably depending on the size, material, and the amount of machining required for each option.

What can each one do that the others cannot

LPBF can create the interior. Closed curved channels, lattices, shaped cavities within the part. Neither casting nor MIM can produce a closed conformal channel without impossible cores or splitting the part in two. This isn't "cheaper": it's that the part doesn't exist any other way. This is the area where LPBF doesn't compete, it wins by default—the clearest example being conformal cooling.

Casting makes it big and cheap. Above a certain size, LPBF (Low-Processed Body Fabrication) quickly ceases to be economically viable because the casting time increases with the volume cast. A twenty-kilo part is cast; it's not printed.

MIM produces small, high-volume parts. Weighing just a few grams, with fine geometry, in quantities of tens of thousands. In its niche, it's unbeatable in unit price, and neither LPBF nor foundry metals come close.

What to look at besides the price

Lead to first part. Days with LPBF; weeks or months with any mold-based option. If you're in development and iterating three times, that lead time triples, negating any unit savings.

Cost of change. Modifying a printed part involves editing the CAD file. Modifying a molded part might require creating a new mold. In a product that is still being defined, this cost outweighs the price.

Properties. A well-made LPBF part achieves the strength values of forged material in many steels and nickel alloys, and typically surpasses casting in strength. However, it is anisotropic and its initial finish is inferior. MIM and casting produce isotropic parts.

Subsequent Machining. All three need functional surface rework. Comparing "price of raw printed part" versus "price of finished cast part" is the most common pitfall in these comparisons, and the one that leads to poor decisions. The honest breakdown is in how much does a metal printed part actually cost.

Redesign. If you're going to print, print a part designed to be printed. Taking the drawing of the cast part and sending it to LPBF as is is the surest way to make the comparison costly: the printed part wins when it is used to consolidate the assembly or to add functionality.

The combination that almost no one considers

There's no need to choose once and for all.

pattern that works very well: print while the design is evolving and volume is low, and move to mold making when the design is finalized and volume increases. You start actual production within weeks, validate with the client, and only invest in tooling when you know the part won't change. The risk of mold making is based on information, not blindly.

And another, in the opposite direction: printing the tooling. A mold with printed conformal cooling, printed inserts for a short run. Here, LPBF doesn't replace casting: it makes it faster.

How to decide in practice

Four questions, in this order:

How many parts per year, really? If the honest answer is "I don't know," you've already made your choice: LPBF, because it's the only one that doesn't charge you for mistakes.
Is the design frozen? If not, same question.
Is there anything in the geometry that can only be done by printing? If so, stop comparing prices: compare what that part allows you to do.
How much does it weigh? Above a few kilos, start seriously considering casting.

If your case falls into the first group, in the quote calculator you can see the order of magnitude of your specific part without sending us anything, and in parts manufacturing we manufacture it for you without you having to buy a machine.

—

The comparison with machining, which is the other frontier, is in print or machine?. The parts you shouldn't print even if it's possible are in five parts you shouldn't manufacture by LPBF. And if you're considering buying the machine or the service, see the honest comparison.