What Should Buyers Check When Matching a Gear and Pinion

Start here — pinion Z1 teeth, gear Z2, that's your basic ratio. But don't stop here. Two gears with identical tooth counts still won't match if the module, pressure angle, profile, or mounting setup differ.

This is the big one. Metric gears use module, imperial systems often use DP. Get this wrong and the teeth won't engage properly no matter how close the outside diameters look.

One trap: don't measure a worn gear's OD with calipers and reverse-engineer from that. Wear eats the tip and the tooth thickness. You're better off combining tooth count, over-pin measurements, span measurements, base circle data, and whatever you can get from the mating gear.

Different pressure angles = different base circles = different involute geometry, even when tooth count and module are the same. The parts might bolt together, but they aren't a correct pair.

And no, you can't tell pressure angle just by looking at it — it needs measuring, or working out from the original mating part.

Helicals add two more: the helix angle and whether it's left or right hand. For a parallel-shaft external helical pair, those have to line up correctly. If you want a decent quote, give us helix angle, hand, normal module, and normal pressure angle.

A photo of one gear alone usually isn't enough to reconstruct the geometry — so for helical jobs, send both parts' info, not one.

This controls the actual mesh. Both gears can be made perfectly, but if the gearbox's real center distance is off from the design, you get shifted backlash and contact. New designs get center distance from the drive layout.

Replacements need shaft positions, bearing positions, housing dimensions, and how the thing was originally assembled. And in old machines — bearings worn, shafts bent, housings warped — the effective center distance can be different from what the drawing says. Sometimes the new gear isn't the problem at all.

The one everyone forgets until it bites. Too tight and thermal expansion or a bit of assembly error causes interference. Too loose and you get rattle, noise, positioning error. And “as tight as possible” is wrong too.

Realistic backlash depends on gear size and accuracy grade, operating temperature, lubrication, load, and how the bearings and housing hold everything. If your customer specifies backlash, get it on the drawing. Otherwise it should come out of standards plus the actual duty and assembly — nobody should just pick a number.

Precision gears often go beyond standard module and pressure angle. Profile shift, tip relief, lead modification — these exist to spread the load, cut noise, or handle a specific working condition.

Which is exactly why copying a worn part is risky: what you see on an old tooth is original geometry + normal wear + uneven wear + years of deformation. Easy to mistake “what the part looks like now” for “what the part was designed to be.”

A lot of buyers assume both gears should be the same material and hardness. Not necessarily. The pinion has fewer teeth, so it cycles more in the same runtime — lots of designs give it a different material or hardness than the gear.

What you actually need to confirm: material, heat treatment, surface hardness, core properties, case depth, and whether the part needs finish grinding. “Same hardness” is not the same thing as “matched.”

Manufacturing accuracy decides how well the pair actually meshes. Check profile deviation, lead deviation, pitch deviation, radial runout. And confirm the standard — ISO, DIN, or AGMA — plus the grade.

Writing “high precision gear” on a drawing without a standard and grade tells the supplier almost nothing. Different shops will read that very differently, so be specific before anyone quotes.

For bevel and other contact-sensitive pairs, individual part dimensions just aren't enough. The contact pattern shows whether load spreads across the tooth as designed or bunches up somewhere.

If contact sits on the tip, the root, or one end of the face, you go check mounting distance, machining parameters, assembly, or the mating part. This is why some bevel sets are made, inspected, and marked as a matched set — not as two independent gears.

Sometimes the new gear is perfectly fine and the machine still runs like garbage. Then stop looking at the gear and look at the system: bearing clearance, shaft stiffness, axial location, center distance, mounting distance, housing deformation, bearing seats.

In an old gearbox, worn bearings alone will wreck a brand-new pair's meshing. Matching is a system problem, full stop.

Same pair, different duty, different requirements — material, accuracy, backlash, heat treatment all shift.

When we evaluate a project we want: input/output speeds, normal and peak torque, continuous vs. intermittent running, shock loads, operating temperature, lubrication method, what the machine is — including agri equipment, excavators, trucks, industrial gearboxes — and how the old parts failed.

That tells us whether the original design still makes sense — and keeps us from blindly copying worn parts.