What Is Sintering? The Furnace Step That Turns Metal Powder into Precision Parts Sintering is the thermal process that bonds compacted metal powder into a solid, high-strength part through solid-state atomic diffusion, entirely below the base metal’s melting point.
For iron-base powder metallurgy, sintering runs between 1120 °C and 1150 °C, roughly 70 to 75% of iron’s absolute melting point. It is the defining stage of press-and-sinter manufacturing: the furnace cycle sets final tensile strength, fixes carbon content, establishes the density band, and produces the predictable 0.2 to 0.5% linear shrinkage that lets press-and-sinter hold ±0.05 mm on critical as-sintered dimensions without machining. Everything a buyer needs to evaluate a sintered part traces to how this one furnace cycle was run. Last updated 1 Sep 2026.
| Process Parameter | Iron-Base Press-and-Sinter Spec | Notes / Control Method |
|---|---|---|
| Sintering temperature | 1120 to 1150 °C (up to 1280 °C for stainless) | Solid-state diffusion range |
| Atmosphere control | Endothermic gas, nitrogen-hydrogen blend, or vacuum | Dew point and carbon potential logged per run |
| Hot zone dwell time | 20 to 45 minutes, continuous mesh belt | Governed by section thickness |
| Volumetric densification | Closes 50 to 65% of initial green void space | Archimedes method per ASTM B962 (Standard Test Methods for Density of Compacted or Sintered Powder Metallurgy Products) |
| Linear shrinkage | 0.2 to 0.5% typical for FC-0208 / FN-0208 | Die cut oversize for compensation |
| Final density | 6.6 to 7.2 g/cm³ (85 to 92% of theoretical) | Meets MPIF Standard 35 material specs |
| MOQ | From 2,000 pcs | Press-and-sinter batch economics |
| Certifications | IATF 16949, ISO 9001; furnace zones mapped per CQI-9 | Annual thermocouple survey on the belt |
How sintering works: solid-state bonding without melting
During compaction, powder particles are pressed together at 400 to 800 MPa and held by mechanical interlock and cold welding at contact points. The result is a green compact strong enough to handle but metallurgically nothing more than a pile of touching particles.
The furnace changes that. Diffusion starts where the compaction pressure left metal touching metal, at those point contacts. Atoms cross the particle boundaries first, and each crossing point widens into a neck. Necks replace the sharp mechanical interfaces with true grain boundaries, and this is where the part picks up most of its strength.
As the necks grow, the voids trapped between particles begin to shrink and lose their sharp corners. Rounded pores concentrate stress far less than angular ones, which is part of why a well-sintered PM part tolerates fatigue loading better than its porosity numbers alone would predict. Because the matrix stays solid through the whole cycle, with the exception of liquid-phase systems like copper infiltration where an added phase melts but the base metal does not, the part keeps the geometry the die generated. There is no melt to slump, flow, or freeze unevenly. That single fact is why press-and-sinter holds shape so well and why our tolerance capability page can quote the numbers it does.
Inside the furnace: the five zones of a continuous sintering cycle
Our mesh-belt continuous atmosphere furnaces at the Shaoxing facility run five functional zones, each with its own control loop. An iron-base structural part passes through all five in one run.
1. Delubrication / dewaxing (up to ~500 °C)
The green part carries press lubricant, zinc stearate or a synthetic wax, typically 0.5 to 1% of the blend. This zone removes it slowly enough that the lubricant does not crack into carbon soot, blister the surface, or burst internal micro-porosity. Most “mystery” as-sintered surface defects we are asked to diagnose originate in a dewaxing cycle run too fast.
2. Oxide reduction (500 to 900 °C)
Every powder particle carries a thin oxide skin from atomization. Hydrogen-bearing atmosphere components reduce these oxides and expose clean metal, which is the precondition for atomic diffusion to work at all. A part sintered through a poor reduction zone has particles pressed together but never truly joined. Strength suffers even though the part looks acceptable.
3. Hot zone (1120 to 1150 °C, 20 to 45 minutes)
Neck growth, pore rounding, and alloy homogenization (Ni, Mo, Cu diffusing to uniform distribution) happen here. Carbon potential in the atmosphere is controlled to the hundredth of a percent: too high and a low-carbon powder picks up carbon it never specified, too low and an FC-0208 part decarburizes at the surface. As our heat treatment page covers, PM carbon is set in this zone, and it is far harder to correct after sintering than to control at the source.
4. Controlled cooling
The cooling rate sets the microstructure the part ships with. On high-hardenability grades, accelerated cooling transforms the structure directly to martensite inside the furnace. This is sinter-hardening, which our heat treatment page covers as the route that eliminates a separate quench entirely.
5. Exit and inspection
Parts leave matte, a few tenths of a percent smaller than the die, dimensionally stable, and bonded. First-line checks are routine: density by Archimedes per ASTM B962, hardness, and, the oldest check on the floor, a drop test. A well-sintered part rings; an under-sintered one thuds. Every furnace operator we run knows the difference.
Sintering atmospheres: the gas does three jobs
The atmosphere in a sintering furnace is not passive protection, and it fails in specific, documentable ways. Bare iron at 1120 °C oxidizes within seconds in air, so the atmosphere has to exclude oxygen outright. Excluding it is not enough on its own, though: hydrogen in the gas mix actively strips the oxide skins left on every powder particle from atomization, and without that reduction step the particles get hot without ever bonding metal to metal. The third job is the one that bites hardest in production. The atmosphere trades carbon with the steel, and the balance point, carbon potential tracked against dew point in our furnace logs, sits at a different setting for each alloy. Run an FC-0208 part under the wrong potential and it either picks up carbon it never specified or decarburizes at the surface, and the hardness response the customer ordered changes with it.
On hardness-critical lots, the atmosphere log (dew point, carbon potential, zone temperatures) ships with the dimensional and density data. A hardness number without the furnace record behind it is a certificate, not a property.
| Atmosphere Type | Composition | Typical Use |
|---|---|---|
| Endothermic gas (Endo) | ~40% N₂, 40% H₂, 20% CO | High-volume iron and steel structural parts; economical carbon balance |
| Nitrogen-hydrogen blends | 90 to 95% N₂ / 5 to 10% H₂ | Automotive gears and structural parts; clean, repeatable |
| Hydrogen / dissociated ammonia | Pure H₂ or 75% H₂ / 25% N₂ | Stainless PM (300/400 series) needing maximum reduction |
| Vacuum | <10⁻² mbar | Reactive alloys, high-alloy stainless, magnetic parts |
What sintering delivers, and where it stops
Well-made iron-base sintered parts reach 80 to 90% of wrought tensile strength, which covers most bracket, gear, and sprocket duty. The honest limits on fatigue are on our porosity and density page. Die-formed geometry survives the furnace with only the compensated shrink, so most surfaces never see a cutting tool. And the 8 to 15% residual interconnected porosity is a design resource rather than a defect: oil retention in bushings, damping in housings, and the network copper infiltration uses all spend that volume deliberately.
To verify the exact density achieved during sintering, the Archimedes water immersion technique (ASTM B962 / MPIF Standard 42) is used on production samples:
Where ma is part weight in air, and mw is part weight submerged in water.
The process also has boundaries a buyer should know before the DFM, not after. Green cracks, density gradients, and delaminations from press setup transfer directly into the sintered part, and the furnace cannot repair a compaction defect, which is why our DFM reviews start at the press. Press-and-sinter lands at 85 to 92% of theoretical density. Applications needing 98%+ move to liquid-phase sintering, MIM, or powder forging, and we will say so when that is the right call. Undercuts, cross-holes, and any true thread are secondary CNC operations, since features perpendicular to the press direction do not form in the die. The shape envelope is documented on our wall thickness and what-is-powder-metallurgy pages.
Three questions to ask a PM supplier about sintering
Instead of a generic audit list, these are the three questions whose answers separate a furnace line from an oven with a conveyor:
- How do you verify zone temperatures across the belt? The right answer is a thermocouple survey cycle per CQI-9, mapped across the full belt width, not a single probe at one spot. If they quote “furnace temperature controlled,” ask what that survey schedule is and how often the mapping is rerun.
- When was the last density check, and where is it recorded? Archimedes density per ASTM B962, sampled per lot, with the numbers printed on the lot report. “Available on request” usually means sampled when someone remembers. MES-traceable density on every lot is the only signal worth taking to a customer.
- How do you catch a carbon profile error before shipment? Surface hardness alone will pass a decarburized part. The answer should include core-versus-surface hardness or microstructure checks by Vickers or Rockwell, plus CMM verification on gear teeth and critical features, with MES traceability back to the furnace run.
Key takeaways
Sintering bonds pressed powder below the melting point by solid-state diffusion: point contacts grow into necks, pores round and shrink, and the part gains its strength in the hot zone at 1120 to 1150 °C without ever melting.
The furnace cycle sets strength, density band, carbon content, and microstructure, which makes atmosphere and zone control the variables with the highest pull on the process, and the ones a buyer should audit first.
Shrinkage of 0.2 to 0.5% is predictable and compensated in the die. What the process does not do (full density, repair of compaction defects, undercuts and cross-holes) routes to the secondary operations and process alternatives covered on our porosity, heat treatment, and wall thickness pages.
The expensive mistake is treating sintering as a black box between the press and the part. Send us the drawing, and we will tell you what the furnace cycle has to do for your part, and where it cannot help, inside 48 hours from a free DFM review.
Frequently Asked Questions
What is sintering in powder metallurgy?
Sintering is the furnace step that bonds compacted metal powder into a solid part at a temperature below the metal’s melting point, typically 1120 to 1150 °C for iron-base parts, around 70 to 75 percent of iron’s melting point. Atoms at each particle-to-particle contact diffuse across the boundary, growing necks that turn a pressed compact into a connected solid. No melting occurs, so the part keeps its pressed shape and shrinks predictably, typically 0.2 to 0.5 percent linearly.
What is the main difference between sintering and casting?
Casting fully melts the metal and pours it into a mold, which introduces liquid shrinkage, gravity segregation, and mold-fill porosity defects. Sintering bonds solid particles below the melting point, so the part retains the die-formed geometry with only a small, predictable shrinkage, which is why as-sintered tolerances of plus or minus 0.05 mm on critical dimensions are practical without machining.
Why do sintered parts retain porosity?
Atomic diffusion stops short of full densification before grain boundaries stabilize the residual pores. Standard iron-base parts hold 8 to 15 percent porosity, mostly open and interconnected. That volume can be left open for oil-impregnated self-lubricating bushings or filled by copper infiltration where near-full density is needed.
What tolerances can be held after sintering?
As-sintered critical features hold plus or minus 0.05 mm. Secondary sizing (coining) tightens selected critical dimensions to plus or minus 0.025 mm, and precision CNC operations on specific features go further where the design requires it. Shrinkage of 0.2 to 0.5 percent is compensated by cutting the die oversize, so the sintered part lands on nominal.
How strong is a sintered part compared to wrought steel?
Static tensile strength of a well-made iron-base sintered part runs about 80 to 90 percent of an equivalently alloyed wrought steel, sufficient for most structural duty. Fatigue strength drops harder because residual pores act as crack starters at high cycle counts. Where a part needs more, the routes are higher compaction density, copper infiltration, or powder forging.