RV reducers are the dominant cycloidal-pinion reducer class for industrial robots with payloads above 20 kg, single-stage ratios from 30:1 to 240:1, and rated tilting moments typically between 200 N·m and 4,000 N·m depending on frame size [S2].
Selection for a 6-axis arm, a welding positioner, or an AGV drive train follows the same five gates — torque headroom, ratio architecture, backlash class, moment loading, and sealing/lubrication envelope — and failing any one of them forces a different frame or a different reducer family entirely [S2].
Gate 1 — Rated Torque, Peak Torque, and Acceleration Headroom
Spec the rated torque at 2× the continuous RMS load of the joint, then verify the peak (emergency-stop) torque stays below the published peak limit of the selected RV frame — peak limits for industrial-class RV units commonly land between 2× and 3× the rated torque for short transients [S2].
Acceleration headroom is the metric most teams under-spec: an RV reducer driving a 50 kg payload through a 0.3 s acceleration event must be checked against the published allowable peak torque per cycle, not the average joint torque, because cycloidal-pin gearing transmits shock loads through the needle bearings rather than absorbing them [S2].
For background on the cycloidal-pinion geometry that gives RV its shock tolerance, see the cycloidal reducer geometry reference — the difference between an RV and a plain cycloidal unit is the integrated planet gear stage that drops the input speed before the cycloidal disc, which is what lets a single RV stage reach ratios that would otherwise need a two-stage planetary reducer [S2].
Gate 2 — Ratio Architecture, Frame Size, and Single-Stage Density
RV reducers are sold in ratio steps that double roughly logarithmically: 30, 41, 51, 81, 101, 121, 153, 184, 220 are typical catalog values in the small-to-medium frame range, and most builders carry those exact ratios in stock rather than as specials [S2].
Pick the ratio from the joint kinematics first, then back-solve the motor speed: a 6-axis arm with a 120°/s joint requirement and a 3,000 rpm servo motor lands on a 30:1 or 41:1 RV, while a welding positioner that turns once every 6 s ends up in the 121:1 to 184:1 band and pays a proportional cost premium for the extra gear stage inside the housing [S2].
Because the RV is fundamentally a planetary reducer with a cycloidal output stage bolted to it, swapping in a pure planetary for cost-sensitive, lower-ratio joints is a common value-engineering move — see the planetary reducer 2026 spec-gate reference for the ratio map and stage-count rules to apply during that swap [S2].
Gate 3 — Backlash Class, Lost-Motion, and the Hysteresis Loop
Backlash on standard-grade RV reducers is typically quoted in the 1 to 3 arc-minute range; precision-grade units drop to under 0.5 arc-minute and zero-backlash variants with controlled preload on the cycloidal disc land near 0.1 arc-minute, at roughly 1.5× to 3× the cost of the standard-grade equivalent [S2].
Lost motion under reversing load is the more useful figure for path-following applications: a standard RV running at 10% of rated torque typically shows 2 to 4 arc-minutes of lost motion, while a precision-grade unit holds the same test under 0.5 arc-minute, which is the number that actually determines whether a welding torch or a glue nozzle lands where the controller thinks it will [S2].
For applications that need sub-arc-minute hysteresis — surgical robots, semiconductor handlers, high-accuracy CNC indexers — the harmonic drive reducer is the more common alternative, and the harmonic drive selection criteria reference lays out the torque-density penalty that comes with that backlash class [S2].
Gate 4 — Tilting Moment, Radial Load, and Output Bearing Capacity
RV reducers are sold as integrated gearhead-plus-output-bearing packages, and the output bearing's tilting moment rating — not the gear rating — is the first number that fails on a misapplied selection. Typical industrial RV frames carry 200 N·m to 4,000 N·m of allowable tilting moment depending on frame size, and exceeding that number cracks the output bearing race long before the gear teeth show wear [S2].
Check the radial load on the output flange separately: a cam-follower or a chain-sprocket hanging off the output shaft applies a moment that adds to the joint's own inertia load, and the combined vector must be plotted against the published moment-load curve for the selected frame, not just compared to a single number from the catalog header [S2].
Gate 5 — Sealing, Lubrication, and IP Class for the Operating Environment
Standard RV reducers ship as IP65 oil-sealed units suitable for clean factory air, but welding-spatter environments, food-grade washdown, and outdoor AGV service all push the specification to IP67 or higher with food-grade grease rather than oil bath lubrication [S2].
Grease-filled RV units trade thermal capacity for sealing: a continuously-cycling joint running near the rated torque derates by roughly 15% to 25% on grease lubrication because grease cannot reject heat as effectively as a circulating oil sump, and that derating must be applied before Gate 1, not after, otherwise the torque headroom calc is wrong [S2].
For the broad reference on reducer family comparison — including worm gear reducers for low-speed right-angle drives and helical gear reducers for high-speed trunnion service — the RV reducer family map is the starting point to confirm that an RV is even the right topology before any of these five gates are run [S2].
RV vs Harmonic vs Planetary — A 4-Criterion Comparison
On four selection criteria the three families separate cleanly: RV reducers win on shock-load tolerance and single-stage ratio density, harmonic drives win on backlash class and torque-to-weight ratio at the cost of input-speed limit, and planetary reducer units win on cost per newton-meter and on input-speed ceiling but lose on single-stage ratio range and on output bearing integration [S2].
Cost per rated newton-meter for an industrial-grade unit in 2026 typically lands roughly 1.0× for the RV, 1.4× to 2.0× for an equivalently-sized harmonic drive, and 0.5× to 0.8× for a planetary stage of similar torque class — which is why a 6-axis arm almost always mixes the two, with RVs on axes 1, 2, 3 and either harmonic or planetary stages on axes 4, 5, 6 depending on the accuracy budget [S2].
For price-band guidance on the cycloidal-pinion cousin of the RV, the cycloidal reducer 2026 price guide gives a useful baseline, because an RV is essentially a precision cycloidal unit with an integrated planetary input stage and a preloaded output bearing [S2].
Selection Flow — From Joint Spec to RFQ
The locked sequence before any vendor is contacted: (1) compute RMS and peak joint torque, (2) pick a ratio from kinematics, (3) select a frame that meets the moment-load envelope at the output flange, (4) decide backlash class from the application's path-following error budget, and (5) apply the IP and lubrication derating for the real operating environment. Only after those five gates close does the buyer pull a part number and send the RFQ [S2].
Trackable signals for the next 60 to 90 days: vendor lead times on the 30:1 and 81:1 standard ratios in the small frame, and any published update to the tilting-moment curves on the medium frame — both of those are the figures that move first when the supply base shifts, and they are the two numbers worth re-checking before the next RFQ drops.