Gear Hobbing Machine

Prove to yourself that a curved tooth can be made by straight cuts.

1. Cut a card disc as a gear blank and mark its centre.
2. Cut a card strip with straight-sided rack teeth — this represents the hob.
3. Roll the rack along the disc's edge without slipping, marking the rack's outline on the disc at each small step.
4. Look at the envelope of all those straight lines.

The curve appears without anyone drawing it. The involute tooth form is the envelope of a straight-sided rack rolled against the blank, which is why a hob with straight-sided teeth can cut a correctly curved gear. This is generation rather than copying, and it is why one hob serves every tooth count of its module: the form comes from the ROLLING RELATIONSHIP, not from the cutter's shape.

A worm with flutes cut across it, so each thread crest becomes a row of teeth.

1. Start from a length of M12 threaded rod — a ready-made single-start worm of known pitch.
2. Cut three or four straight flutes along its length with a hacksaw, at 90 degrees to each other.
3. File a relief behind each cutting edge so only the edge contacts.
4. Mount it on an arbor in two 608 bearings.

Modern build spec (derived). A real hob is a hardened, precisely relieved and ground cutter with a defined pressure angle. Threaded rod flutted by hand will cut soft material and demonstrate the geometry exactly, at a fraction of the work. What it will not do is hold form after a few cuts.

This is the machine. Everything else is a frame to hold it.

1. Mount the blank on a vertical spindle in 608 bearings.
2. Connect the hob spindle to the blank spindle through a train of change gears.
3. For a single-start hob cutting N teeth, the blank must turn once for every N turns of the hob.
4. Fit the change gears to give that ratio and check it by counting turns by hand.
5. Secure everything with M6 × 50 hex bolts × 4, M6 flat washers × 8 and M6 hex nuts × 4.

The ratio is the tooth count. Change the gearing to 1:37 and you cut a 37-tooth gear with the same hob; 1:38 gives 38 teeth. Nothing about the cutter changes. That is the whole economic argument for hobbing — one cutter, every tooth count, and prime numbers are no harder than round ones.

The hob must be tilted, or its teeth cut at an angle to the gear's axis.

1. Measure the hob's lead angle — the helix angle of its thread.
2. Swivel the hob head by that angle so its teeth run parallel to the gear's tooth direction.
3. Feed the hob into the blank to full tooth depth.
4. Feed it slowly along the blank's axis while both rotate, and let the teeth generate.
5. Inspect the finished teeth for the characteristic faceted flanks.

Hermann Pfauter patented a universal gear hobbing machine in 1897 in Chemnitz, and the firm he founded still makes hobbing machines. The same year, Edwin Fellows patented the gear shaper already in this catalogue — two different generating machines arriving simultaneously, which is a good indication that gear cutting was the bottleneck everyone was pushing at.

Generating versus forming, stated plainly. A form cutter is shaped like the space between two teeth, so it copies its own profile and is strictly correct for only one tooth count — which is why form cutters are sold in numbered sets, each covering a range approximately. A generating process rolls a simple cutter against the blank and lets the correct curve emerge from the motion. Generation gives better accuracy from simpler tooling, at the cost of a machine that must hold an exact ratio between two rotations.

What it inherited from this chain: the milling machine's rigid three-axis frame, the dividing head's principle of exact angular relation — here made continuous rather than stepped — and the change-gear train that Maudslay's screw-cutting lathe introduced for connecting rotation to rotation. Hobbing is those three ideas combined and pointed at one job.

Where each still wins: hobbing for spur, helical and worm gears in quantity; shaping for internal gears and for gears with a shoulder close by, where a hob cannot run out; form milling for one-off repairs where setting up a hobber is not worth it. Three approaches to one goal, all still in use — and the choice is about quantity and geometry, not about which is better.