The noise level of a planetary gear motor cannot be read off a datasheet as a single product property. It is the outcome of four things working together: how well the teeth share load, how much backlash the gearbox was built with, how precisely the tooth surfaces were finished, and how squarely the motor was mounted. Of those, a buyer can specify the ratio and stage count, the backlash class, the output speed and the mounting tolerance. The rest belongs to the gearbox design, and no catalogue figure compensates for a misaligned installation.

Close-up of a compact planetary gear motor with a smooth anodized housing for quiet low-noise operation, on a white background.
This article breaks the sound down into its sources, separates the variables a buyer controls from the ones the gearbox designer controls, and sets out the fields that belong in a noise clause when sound level is a specification rather than a preference. It also explains why a sound figure quoted without measurement conditions carries almost no information.
Hofon builds a planetary gear motor series with rated ratio, rated torque and backlash stated per unit; sound level is agreed against the application rather than quoted as a catalogue constant.
Three mechanisms dominate. The first is the tooth mesh: as each pair of teeth comes into contact it strikes the following pair, and the energy of that impact depends on load, on pitch-line velocity and on how many teeth are sharing load at that instant. The second is the bearings, where rolling elements passing through the load zone generate a tonal component at a frequency set by ball count and rotational speed. The third is the housing and mounting interface, which either radiate that vibration as sound or damp it.
Contact ratio belongs to the first mechanism. A planetary stage with a high contact ratio keeps more than one tooth pair sharing load at any moment, so each individual pair carries less and load transfer between pairs is gentler. A low contact ratio puts more load on a single pair and makes the handover abrupt. That is a property of how the gears were cut, not something a buyer can change by choosing a different ratio from the same gearbox family.
Backlash is the play between mating teeth, and it affects noise in two ways. A gearbox with generous backlash lets teeth rattle against each other whenever the load reverses or the torque ripple changes sign, which is why a gearbox that is quiet under steady load can become noticeably louder in an application that reverses often. Tight backlash runs quieter but demands better alignment, because misalignment then shows up as binding instead of as play.
Tooth surface finish sets how hard the impact at mesh frequency really is. Ground and lapped profiles present a smoother surface to each other and transfer load more gradually than cut teeth. Lubrication decides how much of the remaining vibration is absorbed before it reaches the housing, so a grease formulated for the speed range and the temperature actually present will damp the mesh better than one chosen for convenience. A gearbox that is quiet when warm is often louder during the first minutes of a cold start.
Sound level rises with output speed. Halving the output speed of the same gearbox lowers both the mesh frequency and the pitch-line velocity, and in most designs that moves the level down appreciably. It also changes the character of the sound: at low speed the dominant component is often the once-per-revolution pattern from a single imperfect tooth, while at higher speed the mesh frequency and its harmonics take over.
Load matters too, but not in the direction people expect. A lightly loaded gearbox deflects less at each tooth and can run more quietly than a heavily loaded one at the same speed, yet a completely unloaded gearbox with generous backlash can rattle and end up louder than one carrying a modest load. A sound figure without speed, load and mounting conditions attached is therefore meaningless.
If the sound level matters, it belongs in the specification with the conditions attached, because a limit that cannot be measured cannot be accepted or rejected. The fields below are the ones that make it measurable. Whatever values are agreed must be agreed for this application: a figure carried over from another project says nothing about this one.
Clause field | What it has to state | Why it matters |
Measurement distance and position | Distance from the gearbox surface, and whether the point is on the output axis or to one side | Sound pressure falls with distance, so distance changes the number more than the gearbox does |
Operating speed | Output speed, or input speed with the ratio, and whether speed is held steady during the measurement | Mesh frequency and pitch-line velocity scale with speed |
Load condition | Torque applied during the measurement, and whether it is steady, varying or absent | Load changes tooth deflection and backlash behaviour |
Mounting and coupling | How the unit is fixed, and coupling type and alignment | Misalignment produces its own low-frequency beat |
Background noise | Ambient level at the measurement point, plus any correction applied | A gearbox measured in a noisy bay can appear to have no sound of its own |
Instrument and weighting | Meter class, weighting curve and averaging time | Two different instruments can return two different numbers for the same unit |
Acceptance limit | The value to be met, and the conditions it applies to | A limit without conditions cannot be tested |
Two of those rows deserve more attention than the rest. Distance and position decide the number more than the gearbox does, so a distance quoted without a position is only half a specification. Background noise decides whether the measurement means anything at all, because a unit tested in a busy workshop will appear quieter than the same unit tested in a quiet room.
Misalignment between the motor shaft and the driven shaft loads the output bearing in a way it was never designed for, and the result is both a shorter life and a louder unit — usually a low-frequency beat rather than a clean mesh tone. A rigid coupling forces the gearbox to absorb whatever angular and parallel error exists, while a flexible coupling absorbs some of it at the cost of introducing its own resonance.
The practical steps are unglamorous. Check that the mounting face is flat and clean, fit a coupling rated for the misalignment present, confirm that the output shaft is not being side-loaded by the belt or gear it drives, and make sure the gearbox is not doubling as a structural member of the machine. A gearbox bolted to a panel that flexes will radiate sound through that panel however well it was built.
Planetary stages are generally quieter than worm gearboxes at comparable output speed, and they reach high ratios without the sliding contact that dominates worm losses. Against a single spur stage, a planetary stage spreads load across several teeth in mesh at once and keeps the load paths symmetric, so each tooth transmits less force and the vibration that follows is lower. The comparison is not absolute, however. A worm gearbox can be quieter in a very low speed range where sliding contact replaces impact as the dominant mechanism, and a helical spur stage spreads contact gradually and can be very quiet at moderate ratios. The choice follows the ratio, the duty and the noise target rather than a general ranking of gearbox types.
The noise level of a planetary gear motor is decided by tooth loading, backlash, surface finish, lubrication, speed and mounting together, and only part of that list sits on the buyer's side of the table. Fix the ratio, the backlash class, the output speed and the mounting tolerance in the specification, put the measurement conditions next to the sound limit, and the requirement becomes something a supplier can design to and you can accept against.
If sound level is a firm requirement in your application, contact us with the ratio, output speed and mounting arrangement so the gearbox can be specified against it.