Why Do Gears Need Heat Treated?
Heat treatment changes the hardness, strength, and dimensional stability of a gear before it ever runs in a gearbox. This page explains the metallurgy in plain terms, shows which gear types benefit, and walks through the field symptoms that tell you a heat treat step went wrong. Written for design engineers, manufacturing engineers, and sourcing leads who need to judge a heat treat spec, not just copy one.
Gear heat treatment: symptom, cause, and fix
Use this table when a gear has already failed or when a heat treat spec is being questioned. Each row pairs one field symptom with the likely root cause and the shop-floor action that corrects it.
| Symptom | Likely cause | What to do |
|---|---|---|
| Tooth flank scuffing in first 50 hours | Surface too soft, under 50 HRC | Re-check case depth; re-harden to 58–62 HRC |
| Tooth tip chipping at low load | Case too deep, brittle through tooth | Reduce case to 0.5–0.8 mm, temper 1 h longer |
| Gear whine that grows with speed | Distortion after quench, lead error > 0.02 mm | Press quench or finish-grind after hardening |
| Root crack starting at fillet | Core too hard, no toughness reserve | Temper to 30–36 HRC core, check Jominy curve |
| Bore shrinks 0.03–0.08 mm after quench | Martensitic expansion, no stress relief | Add pre-heat stress relief, size bore after treat |
| Soft spots on one flank only | Uneven quench flow or scale on surface | Clean parts, adjust quench agitation, re-run batch |
| Pitting after 500 h at rated load | Case depth below spec, low residual stress | Shot peen root, verify depth by micro-hardness |
| Crack found only at final inspection | Grinding burn from over-aggressive stock removal | Reduce grind depth per pass, nital etch check |
Why do gears need heat treated at all?
A gear tooth does two jobs that pull in opposite directions. The flank must resist sliding contact and rolling contact without wearing or pitting. The root must bend slightly under load and spring back without cracking. No single hardness value does both well. Heat treatment solves this by giving the tooth a hard skin and a softer, tougher core.
Take a typical 20° pressure angle spur gear in 4140. In the annealed state it sits around 200 HB, roughly 15 HRC. That is soft enough to machine easily but far too soft to run against another steel gear at any real load. Within a few hundred hours the flanks would gall and the pitch line would wear away. Heat treatment raises the surface to 58–62 HRC while keeping the core near 30 HRC.
That hardness gradient is the whole point. Hardness resists wear and contact fatigue. Toughness absorbs shock and misalignment. Put them in the wrong places and the gear fails early, either by wear or by cracking. Heat treatment is how you place them deliberately.
One correction that surprises many engineers: hardness alone does not fix a bad gear. If the lead or profile is wrong before heat treatment, hardening will lock that error in and usually make it worse through distortion. Heat treatment is a finishing step for geometry that is already correct.
- 1Wear resistanceA 58–62 HRC flank resists metal-to-metal sliding far longer than a soft one.
- 2Contact fatigue lifeCase depth of 0.5–1.0 mm pushes pitting past 10 million cycles in many drives.
- 3Core toughnessA 28–42 HRC core bends at the root instead of snapping.
- 4Dimensional stabilityStress relief before final grinding keeps the gear from moving in service.
Which heat treatment fits which gear
Not every gear needs the same treatment. A small plastic gear or a lightly loaded bronze worm wheel needs none at all. A 4,000 mm ring gear for a mill drive needs a different cycle than a 20 mm pinion for a medical pump. The choice follows load, speed, material, and the consequence of failure.
Case carburizing suits low-carbon steels like 8620 or 1018. Carbon diffuses into the surface at 900–930 °C, then the part is quenched and tempered. You get a 0.5–1.5 mm case at 58–62 HRC and a tough core. This is the standard route for automotive transmission gears and many industrial drives.
Nitriding works on alloy steels such as 4140 or 4340 and on nitriding grades. The part stays at 500–550 °C, so distortion is much smaller than carburizing. Case depth is typically 0.2–0.5 mm and surface hardness reaches 60–65 HRC, but the case is thin. Nitriding suits finish-machined gears that cannot be ground after treatment.
Through hardening heats the whole section and quenches it. You get uniform hardness, often 40–55 HRC, but no tough core. It suits small gears, worm wheels, and parts where the whole section must carry load. Induction hardening is a local version of the same idea: the tooth flank is heated in seconds and quenched, leaving the core untouched.
- 1CarburizingLow-carbon steel, deep case, best for high-torque transmission gears.
- 2NitridingLow distortion, thin hard case, good for finished precision gears.
- 3Through hardeningSmall gears and worm wheels where core toughness matters less.
- 4Induction hardeningLarge gears; hardens the flank only, keeps the rest soft.
When gears do not need heat treatment
Heat treatment costs money and time, and it can scrap parts that were already expensive to machine. Sometimes it is the wrong call. A gear that transmits tiny torque, runs slowly, or sees only occasional motion may last its whole design life in a soft state.
Plastic gears are the obvious case. Acetal, nylon, and POM gears run quietly and need no hardening. Bronze worm wheels paired with hardened steel worms are another: the bronze is intentionally softer so it wears instead of the worm. Hardening the bronze would remove that sacrificial behavior.
Some low-load steel gears are simply left as machined, especially in prototypes and low-volume fixtures. If a gear is going into a test rig for a few weeks, a 4140 gear at 30 HRC may be perfectly adequate. The engineering judgment is whether wear or fatigue will ever reach the point where surface hardness changes the outcome.
There is also a practical limit. Very thin gears, gears with sharp internal corners, and gears made from free-machining steels with high sulfur can crack during quenching. In those cases, nitriding or no treatment is safer than a full carburize-and-quench cycle.
- 1Plastic and bronze gearsDesigned to run soft; hardening changes the wear couple.
- 2Low-torque prototypesShort service life does not justify the cost and distortion risk.
- 3Crack-sensitive geometryThin sections and sharp corners fail in the quench.
How heat treatment changes the CNC machining plan
Heat treatment is not a standalone step. It sits between rough machining and finish machining, and it changes what the CNC shop can hold. A gear that is turned and milled to final size before hardening will not stay at final size. The quench moves it.
The usual sequence for a carburized gear is: rough turn and rough hob, stress relieve, semi-finish, carburize and quench, then finish grind the bore and flanks. The grind is what brings the gear back to the drawing. Without that step, the heat treat distortion becomes the final error.
For nitrided gears the sequence is often simpler. Nitriding runs at 500–550 °C, so movement is small, typically under 0.02 mm on a 100 mm gear. Many shops machine to final size, nitride, and inspect without a grind. That is the main reason nitriding is popular for precision gears that cannot be ground after hardening.
At GreatLight we machine gear blanks, housings, and shafts on 3-axis, 4-axis, and 5-axis centers, with tolerances down to ±0.005 mm and surface finish to Ra 0.2–0.8 μm where the drawing calls for it. When a gear needs heat treatment, we plan the rough and finish operations around the treatment so the final geometry lands where the drawing says.
- 1Leave grind stock0.05–0.10 mm per flank if the gear will be ground after hardening.
- 2Stress relieve before finishRemoves machining stress that would otherwise add to quench distortion.
- 3Plan the datumGrind the bore first after hardening; use it as the datum for flank grinding.
Step by step: specifying and verifying a gear heat treatment
Follow this sequence when a new gear design needs a heat treat callout or when an existing one is being re-sourced.
- 11. Fix the load case firstWrite down torque, speed, duty cycle, and expected life. A gear that sees 2 N·m at 100 rpm is a different problem from one at 200 N·m and 3,000 rpm. Do not pick a hardness number before the load is on paper.
- 22. Choose the process from material and geometryLow-carbon steel with a need for a deep case: carburize. Alloy steel with tight tolerance and no post-grind: nitride. Small solid gear: through harden. Large gear that cannot fit a furnace: induction harden the flank.
- 33. Set case depth from moduleA common starting rule is 0.15 to 0.25 times the module, capped around 1.5 mm. For a module 2 gear, that is roughly 0.3–0.5 mm. Too shallow and the case caves in; too deep and the tooth becomes brittle at the tip.
- 44. Specify core hardness with the root in mindAim for 28–42 HRC in the core. Below 28 HRC the root yields under shock. Above 42 HRC the tooth loses the toughness reserve that lets it deflect under misalignment.
- 55. Control distortion before it happensStress relieve before finish machining. Use press quenching or a fixture quench on thin discs. Leave 0.05–0.10 mm of grind stock on flanks that will be ground after hardening, and plan the grind as part of the cycle.
- 66. Verify with the right test, not just a fileCheck case depth by micro-hardness traverse on a sectioned sample, not by a file test. Check core hardness at the tooth center. Check lead and profile after heat treatment, since that is where distortion shows up.
- 77. Inspect for cracks and grinding burnMagnetic particle inspection on the first parts of a batch catches quench cracks. Nital etch on ground flanks catches temper burn. Both are cheap compared to a field failure.
Common questions on gear heat treatment
How hard should a gear tooth be?
Most steel gears run with a case hardness of 58–62 HRC on the flank and a core of 28–42 HRC. The exact numbers depend on the load and the failure mode you are designing against.
Higher is not always better. Above 62 HRC the case becomes more prone to chipping at the tooth tip, especially if the case is deep.
What case depth should I specify?
A common starting range is 0.15 to 0.25 times the module, capped near 1.5 mm. For a module 2 gear that is about 0.3–0.5 mm; for module 6 it is about 0.9–1.5 mm.
Verify by micro-hardness traverse on a sectioned sample. A file test tells you almost nothing about depth.
Does heat treatment always distort the gear?
It always changes dimensions to some degree. Carburizing and quenching move parts the most, often 0.03–0.10 mm on a 100 mm gear. Nitriding moves them much less, usually under 0.02 mm.
You manage distortion with stress relief, fixture quenching, and a planned grind after hardening. You do not eliminate it.
Can I grind a gear after heat treatment?
Yes, and for carburized gears you usually should. Grinding restores the bore, lead, and profile to the drawing after the quench has moved them.
Watch for grinding burn. Light passes and good coolant keep the surface from tempering back below spec. Nital etch checks confirm the result.
Is nitriding better than carburizing?
Neither is universally better. Nitriding gives lower distortion and a very hard but thin case, which suits finished precision gears. Carburizing gives a deeper case and a tougher core, which suits high-torque gears that can be ground after treatment.
Pick based on load, allowable distortion, and whether a post-hardening grind is in the plan.
What steel should a heat treated gear be made from?
For carburizing, low-carbon alloys such as 8620 or 1018 are common. For nitriding, 4140, 4340, or dedicated nitriding grades work well. For through hardening, 4140 and 4340 are standard choices.
Free-machining steels with high sulfur are a poor fit for quenching because they crack more easily. If the gear needs both easy machining and hardening, nitriding is often the safer route.
Send us your gear drawing and heat treat spec
We machine gear blanks, shafts, and housings to ±0.005 mm and plan the rough and finish operations around your heat treatment cycle. Upload a drawing and we will return a quotation and a free DFM analysis within 12 hours.
12-hour quote100% inspectionNo minimum order
More from GreatLight
We publish setup notes, tooling trials and inspection data from the factory floor.
