Home › Guides › Bevel gear strength

Bevel gear strength: ISO 10300 safety factors explained

A bevel gear pair can look right, mesh without interference and still wear out or break in service. ISO 10300 is the standard that says whether it will carry its load: two safety factors, one for the tooth flank and one for the tooth root. This guide explains what they mean, what the recommended minimums are, and - measured on one pair - which design choices move them most.

Two ways a bevel tooth fails

Pitting is fatigue of the flank surface. Where two teeth roll and slide against each other the contact stress is high, and after enough cycles small pieces of the surface break out. It starts slowly and grows. ISO 10300-2 rates it from the contact stress σ_H.

Tooth root breakage is bending fatigue. Each tooth is a short cantilever loaded once per revolution, and the stress peaks in the root fillet on the tension side. A crack there runs through the tooth and the tooth breaks off. ISO 10300-3 rates it from the root stress σ_F.

ISO 10300-1 supplies what both have in common - the forces and the factors for how unevenly the load arrives: K_A for the machine, K_v for the speed and accuracy, K_β for the spread across the face and K_α between teeth. The strength of the material comes from ISO 6336-5.

What a safety factor is

For each gear and each mode, the standard compares the stress the tooth may take for the required life with the stress it actually gets:

A factor of 1.25 means the tooth could take a quarter more stress before reaching its permissible value. The permissible stress already contains the life: it is the material's endurance limit adjusted for the number of load cycles, the lubricant, the surface and the size. The pinion and the wheel are rated separately, and the designer shows the lower of the two against its minimum.

The recommended minimums

These are the values ISO 10300-1:2014 clause 5.2 gives for supplied gears, with the note that the actual values are to be agreed between supplier and customer. A straight bevel here means a mean spiral angle of 5° or less. The root is asked for more than the flank for the usual reason: pitting develops gradually and tends to announce itself with noise, while a broken tooth stops the machine at once.

Why the two respond differently to load

Contact stress grows with the square root of the load; root stress grows in proportion to it. So when the torque goes up, S_F falls much faster than S_H. Measured on a 20/40 straight pair, m_mn 5 mm, face width 25 mm, at 1 500 min^-1 (the example from how to design a bevel gear pair):

Doubling the torque from 250 to 500 Nm roughly halves S_F but takes S_H down only by about √2. The ratios are not exact because the dynamic factor changes a little with the load. Two practical consequences: a large S_F next to a modest S_H is normal for a hardened pair, not a sign of over-design; and a pair that fails on pitting needs more than a little extra size to recover, because the flank gains only with the square root.

What moves them most

The same pair at 250 Nm - K_A 1.25, case hardened steel at 700 HV, quality MQ, accuracy grade B6, 20 000 h - with one input changed at a time. The base gives S_H 1.25 and S_F 2.55.

What the table says:

• Size first. The module moves both factors furthest; the face width helps too, within the limits of 0.30 R_e and 10 m_et.
• Material type and quality next. Case hardening gives the flank the most; a through hardened pair of the same size falls well short on pitting. Within one material the quality grade matters - ML, MQ and ME are the standard's grades of how well the material and its heat treatment are controlled.
• More hardness is not always more strength. Across the 660–800 HV band the designer accepts for case hardened steel, the permissible stresses it reads from ISO 6336-5 are the same, and so are the safety factors.
• Be honest about K_A. It covers the shocks of the driving and the driven machine, and it changes the result as much as a module step. ISO 10300-1 has a guidance table; a uniform motor driving a machine with light shocks is 1.25.
• Accuracy, speed and life matter less here - up to about a tenth each - though at higher speeds the accuracy grade counts for more through the dynamic factor.

A spiral bevel of the same outer size is another lever; on a 20/40 pair of 100 and 200 mm pitch diameters it raised S_H from 1.11 to 1.35 - see straight vs spiral bevel gears.

When a factor is too low

• S_H below 1.0: a larger module or face width, a case hardened material, a better quality grade, or a spiral pair.
• S_F below its minimum: a larger module helps most; a larger root fillet (a larger cutter edge radius) and a profile shift that balances the pinion and wheel roots help too.
• Check the inputs before the geometry: an application factor, a life or a speed that is higher than the machine really needs costs size for nothing.

Questions