Koepfer K300 Gear Hobbing: How It Cuts, and Where It Stops

• 1
  How a Koepfer K300 gear hobbing machine removes metal
• 2
  What fits on a Koepfer K300 gear hobbing machine
• 3
  Accuracy, runout and where errors come from
• 4
  Where hobbing sits in a gear production route
• 5
  Hobs, feeds and coolant decisions
• 6
  Hobbing versus other routes for a gear feature

Hobbing is a generating process. A rotating hob with helical gashes and a rack-like tooth profile feeds across a rotating workpiece. The hob and the blank turn in a timed ratio, so each hob tooth cuts one flank of one gap. Nothing is shaped by a form tool pressed straight into the part. The tooth profile appears because the two rotations stay synchronized, which is why the index gear train or the electronic coupling matters more than the cutter shape alone.

On the Koepfer K300 gear hobbing platform, the hob spindle sits on a slide that handles both the axial feed along the workpiece and the tangential shift along the hob axis. Axial feed moves the hob across the face width. Tangential shift slides the hob sideways so a fresh section of the hob enters the cut. That second motion is what lets one hob last through a long batch instead of wearing a single band of teeth.

Cutting speed is set by hob rpm and hob diameter, not by the workpiece. Feed is stated per workpiece revolution. For a typical module 2 steel gear, a hob around 80 mm diameter running at 250–400 rpm gives a surface speed near 60–100 m/min, which sits in the coated carbide range for 4140 or 1045. Push the same hob into a heat-treated blank and the edge life drops fast.

The geometry also sets the limits. A hob cannot cut a shoulder it cannot pass. If a gear sits against a flange, the hob needs clearance equal to roughly the hob radius plus a safe margin, or the flank will be clipped. This single constraint decides more process routing than any speed or feed number.

• 1
  Synchronized rotationTooth form comes from the ratio between hob and workpiece, not from a shaped tool.
• 2
  Tangential shiftSpreads hob wear across the full cutter length to extend batch life.
• 3
  Axial feedControls the number of passes needed across the face width.

The K300 class is a mid-size universal machine. Practical production coverage runs from small pinions up to gears around module 4 with a face width in the low hundreds of millimeters. Below module 0.5 the hob becomes fragile and the chip load per tooth gets hard to control, so micro gears usually move to a different process. Above module 4 the cutting forces and the hob cost climb, and a larger machine frame starts to pay for itself.

Materials that cut well include 1045, 4140, 4340, 20MnCr5 and similar case-hardening steels in the normalized or annealed state, plus 303 and 304 stainless, ductile iron, bronze and brass. Aluminum hobs cleanly but tends to build up on the cutting edge, so coolant flow and sharp, polished flutes matter more than they do in steel. Titanium and Inconel can be hobbed, but speeds drop to a fraction of steel values and tool life becomes the cost driver.

Hardness is the real gate. Hobbing before heat treatment is normal for gears that will be case hardened. Once a gear is through-hardened past roughly 45 HRC, hobbing stops being economical and the finishing operation shifts to gear grinding. Some shops do hard hobbing with carbide hobs on dedicated machines, but that is a different setup than a general-purpose K300 doing soft cutting.

Clamping also decides what fits. A shaft-mounted gear needs a steady and a tailstock center. A thin disc gear needs a rigid arbor and often a support close to the cutting zone, or the blank deflects and the lead error shows up as a wobble in the tooth trace.

• 1
  Best fitModule 0.5–4 spur and helical gears, soft or normalized steel, mid-size batches.
• 2
  Possible but slowTitanium, Inconel, ductile iron with interrupted cuts.
• 3
  Wrong processMicro gears below module 0.5 and hardened gears past 45 HRC.

Hobbing accuracy is usually quoted as a DIN or AGMA grade, and the grade is a system result, not a machine spec alone. Four things dominate: the hob's own accuracy class, the work arbor runout, the thermal state of the machine, and the rigidity of the setup. A grade 6 hob on a perfect arbor can produce a grade 7 gear. The same hob on a worn arbor can produce a grade 9 part.

Runout shows up as cumulative pitch error and as a wobble in the helix. Measure it at the arbor nose before the first cut, not after. A dial indicator on the arbor taper should read within a few micrometers for a finishing pass. If the arbor is out, no amount of hob tuning will fix the part.

Thermal drift is slow and easy to miss. A machine that has run for two hours is not the same machine that started cold. On long batches, the first few parts and the last few parts can differ by a full grade if the spindle and the frame have moved. A warm-up cycle and a mid-batch check on a known master gear catch this early.

For gears that must hold tight lead and profile, hobbing is often the roughing step and a separate finishing pass follows. That is a routing decision, not a machine limitation. The K300 cuts a clean pre-grind gear with stock left for the grinder, and that is usually cheaper than trying to hit final tolerance in one hobbing pass.

• 1
  Check the arbor firstRunout at the arbor nose sets the floor for the whole batch.
• 2
  Warm up before measuringA cold machine and a warm machine cut differently.
• 3
  Leave stock for grindingHobbing to a pre-grind size is often the cheaper route to a tight final gear.

Most gear work is a chain, not a single operation. A bar or a forging is turned to a blank, the blank is hobbed, then chamfered, then heat treated, then finished by grinding or honing if the grade demands it. Each step has to leave the right stock and the right datum for the next one. Skipping the chamfer before heat treatment is a common mistake; the burr hardens and the grinder spends time dressing it off.

For low-volume or prototype work, the economic answer is often not hobbing at all. If you need three gears and the tooth load is modest, a machined gear cut on a 5-axis machining center can be delivered in days without a hob purchase or a fixture build. The tooth form is interpolated instead of generated, which is slower per part but has almost no setup cost.

For high-volume work, the calculus flips. A dedicated hob, a proper arbor and a batch of several hundred parts amortize the setup quickly, and the K300's tangential shift keeps the hob cutting consistently across the run. The break-even point is usually somewhere in the low hundreds of parts, but it depends on module and material more than on part count alone.

One more routing factor: gear inspection capacity. If nobody can measure a lead trace, the grade discussion is theoretical. A functional test on a mating gear pair tells you something, but it will not tell you which of the four error sources is drifting.

• 1
  Blanking and turningSet the bore and the face as the hobbing datum before the gear cut.
• 2
  Chamfer before heat treatSoft burrs are easy to remove; hardened burrs are not.
• 3
  Finish after heat treatGrinding or honing holds the final grade once the part is hard.

Hob material choice follows the workpiece. HSS-Co hobs handle most soft steel and stainless jobs. Powder metallurgy HSS gives longer life on abrasive materials. Carbide hobs make sense where the machine is rigid and the batch is long, but they chip easily on interrupted cuts, so a keyed shaft or a blank with a slot is a risk.

Feed per revolution sets the chip thickness. Too light a feed rubs the edge and work-hardens the flank, especially in stainless. Too heavy a feed overloads the hob tooth and the arbor. A practical starting range for module 1.5–3 steel is 1.5–3.0 mm per workpiece revolution for roughing and 0.8–1.5 mm for finishing, then adjust from the chip and the surface.

Coolant choice depends on the goal. Flood coolant with a high-pressure nozzle clears chips and controls temperature, which matters on deep gaps. Dry or minimum-quantity cutting is used where chip evacuation is not the bottleneck and where the shop wants to skip coolant disposal. Neither is universally better; it depends on module, material and whether the machine has the air blast to clear the gap.

Chip crowding is the most common cause of a sudden bad tooth. If the gap fills, the hob tooth recuts chips instead of cutting metal. The fix is usually more coolant pressure or a change in feed, not a change in hob.

• 1
  HSS-Co for general steelReliable edge on 1045, 4140 and 20MnCr5 in the soft state.
• 2
  Start feeds, then tune1.5–3.0 mm/rev roughing, 0.8–1.5 mm/rev finishing on module 1.5–3.
• 3
  Clear the gapChip crowding, not tool wear, is the usual cause of a bad tooth.