We've had customers come to us with MIM tooling projects after getting burned elsewhere. One automotive supplier spent eight months developing a gear component with a vendor who didn't understand sintering shrinkage compensation. The tooling produced parts 0.4mm undersized across every critical dimension. They had to start over. That experience taught us that most PM failures trace back to fundamentals that someone overlooked or assumed would work out.
How It Works
For conventional press-and-sinter PM, you're dealing with powder around 50 microns. Coarse enough to see if you look closely. Hydraulic presses apply 150-700 MPa to compact this powder in rigid dies. What comes out is called a "green" compact. It holds shape through mechanical interlocking but has no real strength yet. Density sits somewhere between 60-90% of theoretical.
Sintering happens in controlled atmosphere furnaces. Iron-based materials run around 1,120°C. Stainless steels need higher temperatures, typically 1,250-1,400°C. During this stage, atoms diffuse across particle boundaries. The part densifies and shrinks slightly, usually 0.1-0.3%.
MIM is a different animal.
The powder is much finer, around 8 microns, mixed with polymer binders to create feedstock that behaves like plastic during injection. This feedstock gets injected into molds at 50-200 MPa. After molding, you remove the binder through thermal or solvent processes, then sinter. Shrinkage during MIM sintering runs 15-20%. Miss that in your tooling design and you've got scrap.
One thing that surprises procurement teams: MIM powder costs 3-5x conventional PM powder. Gas atomization and carbonyl decomposition processes required for fine spherical particles are expensive. That cost shows up in every quote, and some suppliers don't explain why their MIM quotes run so much higher than press-and-sinter alternatives for seemingly similar parts.
MIM vs. Press-and-Sinter
| Conventional PM | MIM | |
|---|---|---|
| Part weight | 0.68g to 6.8kg | Under 50g typical, 150g max |
| Density | 85-92% | 95-99% |
| Thinnest wall | Around 2mm | 0.3mm possible |
| Tolerance | ±0.3-0.5% | ±0.3% or better |
| Geometry | Mostly 2D | Undercuts, threads, cross-holes |
| Tooling | $2,000-$10,000 | $10,000-$50,000 |
| Volume break-even | 5,000/year | 10,000/year |
Conventional PM's limitation comes from the pressing direction. You're applying axial force, so features perpendicular to that axis are difficult. Internal threads? Not happening. Cross-holes require secondary machining.
MIM removes those constraints but adds new ones. Parts over 150g run into debinding challenges. Wall thickness variation beyond ±40-60% causes warpage that's nearly impossible to fixture out during sintering.
A customer asked us last year about a 200g MIM part with 4:1 wall thickness ratio. We told them it wouldn't work. They went to another supplier anyway, spent $45,000 on tooling, and came back six months later after scrapping the project. Some geometries just don't fit MIM regardless of what a hungry sales team promises.
Against Other Manufacturing Methods
PM wins on material utilization. You're using 95-97% of input material. CNC machining converts maybe half your stock into chips.
A commercial vehicle oil pump gear we quoted against machining came in at 32% of the machined cost. The 68% savings came almost entirely from material and cycle time.
But machining has advantages PM can't touch. Tolerances of ±0.01mm are routine. Any material you want. Prototype quantities that make economic sense. Below about 1,000 units annually, the math usually favors machining even for geometries PM handles well.
Die casting wins for larger simple parts where cycle time drives cost. PM offers better microstructural control and intentional porosity when you need it for bearings or filters.
Forging produces stronger parts. About 26% higher tensile strength for equivalent alloys, near-zero porosity, grain flow alignment that improves fatigue life. We've seen customers spec PM for applications that really needed forging, then deal with field failures.
On mechanical properties: PM steel hits 900 MPa tensile as-sintered, 1,200 MPa after heat treatment. MIM 316L at 97% density runs 520-650 MPa with 35-50% elongation. Close to wrought properties but not identical.
Applications
Automotive transmission components dominate PM volume. GKN supplies parts for Ford F-150 10-speed and GM 9-speed transmissions. MPIF documents over 750 PM automotive applications (mpif.org). Per-vehicle PM content runs around 25 kg for trucks, dropping to roughly 2.3 kg for battery EVs as transmissions simplify.
Medical devices favor MIM. Surgical stapler jaws manufactured as single-piece MIM components have replaced assemblies that used to require 12 machined parts.
Consumer electronics drove MIM's volume growth over the past decade. Apple's Lightning connector ran MIM production at tens of millions weekly. The OPPO Find N2 Flip hinge has 136 parts, many MIM.
Aerospace uses PM for extreme applications. GE's powder-based additive manufacturing for fuel nozzles achieves 25% weight reduction. Turbine discs at Rolls-Royce use PM superalloys for flight-critical rotating components.
What Trips People Up
We had a machinist call us after his shop struggled with a PM part for three weeks. He described it as "machining pure abrasion." The part tested at reasonable hardness but ate through standard tooling. Turns out PM can contain very hard particles dispersed in the matrix that bulk hardness testing doesn't reveal.
His solution: TiN-coated taps, no coolant. Dry chips don't stick to porosity. He'd been running PM parts that way for twenty years with consistent results.
On the quality control side, there's an ongoing debate in industries that use MIM extensively. One machinist we talked to put it bluntly: the process works fine, but quality depends entirely on the shop running it. Good suppliers produce consistent parts. Bad suppliers hide behind "inherent variability" as an excuse.
We've seen this play out. A customer switched MIM vendors to save 12% per piece. Rejection rate went from 0.3% to 4.2%. The savings disappeared into sorting costs and production delays.
When evaluating MIM suppliers, ask for Cpk data on critical dimensions. Ask about in-process monitoring. Verify certifications match your industry. IATF 16949 for automotive. AS9100D for aerospace. ISO 13485 for medical.
Procurement Specifics
Lead times run 6-12 weeks for tooling. Add 1-2 weeks for PPAP on automotive work. First articles need 8-14 weeks from design release. Repeat orders: 4-8 weeks MIM, 2-4 weeks for established PM parts.
Design constraints matter. MIM wall thickness works best at 1.0-3.0mm. Minimum 0.5mm, maximum around 12.5mm. Keep variations within ±40-60%. Corners need 0.5mm radii for powder flow.
Why We're Writing This
We build injection molds. MIM tooling falls within that capability, and the precision requirements mirror what we handle for demanding plastic injection applications. When customers ask whether MIM makes sense for their geometry, we walk through the same considerations covered here.
We've turned down projects that didn't fit the technology. We'd rather lose a quote than watch a customer spend six figures on tooling that can't produce acceptable parts.
If you're evaluating PM for an application, we'll review your design before you commit to tooling. No charge for feasibility assessment. The conversation usually takes thirty minutes and covers whether the geometry works, what tolerances are realistic, and whether volume justifies the investment.
Technical references: MPIF (mpif.org), Mordor Intelligence (mordorintelligence.com)