In precision drive procurement, a datasheet figure becomes a qualification only when three things line up: a model code, a defined measurement basis, and a document that survives the order. Until that alignment exists, an IP rating, a temperature envelope, a lubricant type or an efficiency percentage describes a product family — not the unit that will arrive on the loading dock.

This review examines how the published declarations for ATG precision reducers map back to individual model codes, and what a buyer at the decision stage should demand before treating any of those declarations as a qualification record. ATG is the reducer brand of Qiandeyi Transmission Technology (Suzhou) Co., Ltd., a manufacturer established in 2013 in Huaqiao Town, Kunshan City, Jiangsu Province, China, whose portfolio covers planetary, harmonic and RV reducers alongside hollow rotating platforms, commutators, dividers, joint modules and worm gear reducers.

The framing assumes the technical shortlist is already done. The specification is written, a quotation is under negotiation, and the open question is narrower and more consequential: which declared number applies to the exact code being purchased, and who signs the document that says so.

Family Envelopes Answer a Different Question Than Model Codes

Three declaration types account for most post-order disputes in precision gearing. The first is an envelope quoted as if it were a point value — a 3–15 arcmin backlash range presented as though every unit delivers the best end. The second is a family figure quoted as if it applied to a specific model — an IP rating attached to a catalogue chapter rather than to a code. The third is a published specification quoted as if it were a certification — a number that appears in a brochure and is then assumed to sit behind a test report.

Each of these is a documentation gap rather than a performance gap, which is why it is worth resolving before purchase order issue. A supplier who cannot attach a declared figure to a model code today is unlikely to attach an inspection record to a serial number after delivery.

The Published Declaration Set for ATG Precision Reducers

The table below consolidates what the manufacturer publishes for its precision reducer families and, in the right-hand column, what remains genuinely open at model level. The purpose is not to question the figures — it is to separate the declaration from the trace.

Published declarations for ATG precision reducer families, with the model-level questions each one leaves open.

Two features of this set deserve emphasis. First, the protection class and temperature envelope change with reducer architecture: planetary and RV configurations are declared at IP65, while the harmonic families are declared at IP54 with a correspondingly narrower thermal window. Second, none of these entries is a certification mark. They are published specifications, and the distinction matters throughout the rest of this review.

Why a Range Is Not a Specification

Backlash is the clearest illustration. ATG publishes 3–15 arcmin across the precision planetary range. That band covers single-stage and two-stage configurations, different frame sizes and different precision grades. A positioning axis that requires 3 arcmin must be quoted a code that actually delivers 3 arcmin, and the quotation — not the brochure — should carry the figure. If the supplier cannot state a single value against a single code, the buyer is being asked to accept the best case of a distribution.

Reduction ratio behaves identically. Published category practice shows single-stage planetary units reaching roughly 10:1 and two-stage units reaching roughly 100:1, while one manufacturer lists stage-typical ratios of 3.9:1, 15:1 and 57:1 on its product page. A portfolio described as spanning “3 to 100” is describing reach, not the ratio of the unit in the crate. The same logic applies to torque: one series publishes 14–650 Nm nominal output with 8–10 arcmin backlash, another 16.5–210 Nm with sub-5 arcmin on its precision version, and a third 14–1800 Nm with 1–8 arcmin at portfolio level. These ranges are not comparable to each other without matched frame size, stage count and test basis, and none of them is a substitute for a model-level figure.

Reading the Model Code: Planetary, RV and Harmonic

Planetary: PGH, PEH, PCH, PGF, PGE, PGC and the R variants

ATG planetary model codes include PGH, PEH, PCH, PGF, PGE and PGC, together with the PGHR, PGFR, PEHR, PCHR, PGER and PGCR variants. The letter group is the primary traceability anchor. The quotation, the label on the housing, the dimension drawing and the inspection record should all carry the same code, including any variant suffix. A mismatch between the quoted prefix and the labelled prefix is the single most common reason a declared protection class cannot be honoured at receipt.

RV: RV-C and RV-E

The portfolio also includes RV reducers offered in RV-C and RV-E configurations. Buyers should confirm which of the two is quoted, because the IP65 protection declaration and the -20°C to +80°C operating envelope are attached at that configuration level rather than to “RV” generically. Where an application was sized on the planetary thermal window and later switched to an RV configuration, the upper limit drops by 10°C and the thermal calculation should be re-run.

Harmonic: HF, SF, HSD, SD

Harmonic reducers are offered under the HF, SF, HSD and SD families. These are the configurations for which the published protection declaration is IP54 and the operating range is -20°C to +70°C. That is a materially different enclosure specification from the planetary and RV lines, and it should drive a different enclosure decision rather than being absorbed as an equivalent rating.

Protection Rating: What IP65 and IP54 Actually Cover

On a gear reducer, an IP rating is a property of the assembled unit as delivered. It depends on the housing joint faces, the output shaft seal, the input bore, any breather or plug, and the interface to the motor flange. It is therefore bounded by the tested configuration. Replacing an output shaft, cutting a additional keyway, substituting a seal, or mounting a different motor flange can move the assembly outside the configuration to which the declaration applies — and in most procurement disputes the buyer discovers this only after the modified unit has been in service.

IP65 and IP54 are not simply “better” and “worse”. IP65 indicates dust-tight construction with protection against water jets; IP54 indicates limited dust ingress protection with protection against splashing water. For an indoor robot joint in normal temperature, low dust and non-corrosive atmosphere — a scenario the manufacturer explicitly addresses — IP54 can be entirely adequate. For washdown, heavy coolant exposure or fine conductive dust, neither rating should be accepted without a discussion of the actual exposure.

ATG's own selection guidance points in the same direction: in dusty and humid environments, the protection level of the planetary gearbox should be increased. That guidance is worth taking literally at the specification stage, because it implies the standard declaration is a floor rather than a target.

Operating Temperature: Three Envelopes, Three Reference Points

The published envelopes — -20°C to +90°C for the planetary family, -20°C to +80°C for RV-C and RV-E, and -20°C to +70°C for the harmonic families — differ because the architectures differ. Harmonic units rely on a thin-walled flexspline deformation cycle, which imposes a different heat path and a different sensitivity to grease consistency at the extremes than a planetary gear set. That is an architectural explanation, not a measured comparison, and no independent thermal test data sits behind any of the three figures.

Two practical questions determine whether an envelope is usable. The first is the reference point: is the range ambient air temperature or housing temperature? A reducer mounted inside an enclosure next to a servo motor sees a housing temperature well above the plant ambient. If the published range is ambient and the installation is enclosed, the effective upper limit is lower than the number on the datasheet.

The second is the binding constraint. In most precision gearboxes the lubricant, not the gear tooth, sets the temperature limit. ATG planetary units use synthetic grease, which supports the upper end of the declared window, but the same declaration does not fix a re-lubrication interval. At the cold end, grease channeling raises starting torque, and a drive sized only on running torque can fault on a cold start. Neither point is resolved by the datasheet alone; both should be asked in writing.

Lubrication and Efficiency: Reading “Above 92%” Correctly

ATG publishes a full-load efficiency above 92% for the planetary family, together with synthetic grease lubrication and suitability for any installation method. The installation-method statement has real engineering value: it removes orientation restrictions that constrain mounting in compact cells and on rotating platforms. The efficiency figure requires more care.

Full-load efficiency is a single point on a curve. At partial load, mesh and seal losses represent a larger share of throughput, and measured efficiency falls. Stage count compounds the effect: each additional planetary stage adds a mesh, and category reference data puts single-stage efficiency at roughly 94–97% while multi-stage configurations sit lower. A buyer comparing a single-stage quotation against a two-stage quotation on the same headline percentage is comparing two different operating points. The honest request is to state the load point and stage count at which the 92% figure is claimed.

The same discipline applies to the remaining declarations. Noise at ≤65 dB needs a measurement distance and mounting condition to mean anything. An average lifespan of 20,000 hours depends on load spectrum and duty cycle; it is an average, not a guarantee for a specific axis. None of these figures is unusual in the category — but each is a family-level statement until it is quoted against a code.

Where Specification Ends and Certification Begins

The most important distinction in this review is between a published specification and a certification. Qiandeyi Transmission holds ISO9001 quality management, ISO14001 environmental management and ISO45001 occupational health and safety system certifications. Those certificates attest to the management system under which the product is made. They do not attest to the IP rating, thermal envelope or efficiency of any individual gearbox.

A product-level claim needs a product-level document. For protection class, that means a test report naming the standard applied and the configuration tested. For temperature, it means either a thermal test or a documented derating curve. For efficiency, it means the test condition and load point. For gear accuracy, the reference scale is defined: ISO 1328 sets gear accuracy grades from 0 (highest) to 12 (lowest), so a supplier can state a grade against a standard rather than an adjective.

Where no certificate reference appears on the datasheet, the working assumption should be that no certificate is being offered — and the RFQ should say so explicitly rather than leaving the question to inference. This is a normal state in the category; published portfolios from several manufacturers describe torque, backlash and protection ranges without citing test standards, and the absence is a documentation fact rather than an accusation.

How to Run the Traceability Request

The request is short and does not require technical negotiation. It requires the supplier to convert family statements into model statements.

A seven-item traceability request that converts published declarations into model-level commitments.

Where the Declaration Is Produced

Traceability is ultimately a manufacturing property, which is why the production environment is part of the review. Qiandeyi Transmission operates a 5,000 m² facility in Kunshan with 86 employees, 18 of them in research and development. The workshop is equipped with professional processing equipment and high-precision testing instruments, and the production workflow runs from raw material selection through component processing, assembly and debugging to finished-goods delivery under layer-by-layer inspection.

For a buyer, the practical implication is that a declaration becomes reproducible when it is attached to a point in that workflow. Asking for an inspection record tied to a serial number is not an unusual request in this environment; asking for a brochure restatement is a wasted one.

Application Fit: Which Declarations Carry the Most Weight

ATG precision planetary reducers are applied across automation equipment, robotics, machine tool numerical control, packaging machinery, lithium-ion new energy, printing and textile, and logistics and warehousing. The declarations that matter most shift with the application.

For industrial robot joints in indoor, normal-temperature, low-dust and non-corrosive environments, the reducer amplifies torque, holds precise joint positioning and suppresses transmission clearance. The requirements listed for this scenario are high rigidity, low backlash, fatigue resistance and long-term stability. Here the decisive trace is backlash — the 3 arcmin end of the published band, quoted against a code — plus the protection class under the joint cover, where a sealing change is most likely to invalidate the declared rating.

For packaging lines running continuous start-stop and reciprocating motion under dust, moisture and moderate load fluctuation, the reducer reduces speed, increases torque and controls conveying, film rolling and sealing actions. The listed requirements are compact size, low noise, dust-proof protection and easy maintenance. In this case the protection class and the lubricant service interval matter more than the last arcminute of backlash, because maintenance access is the constraint that determines uptime.

In both cases the same rule holds: the application determines which declared figure has to survive scrutiny, and the RFQ should prioritise accordingly rather than requesting every figure at equal weight.

Comparison with Traditional Solutions: Planetary Against Worm Gear

The planetary architecture is not automatically the correct choice, and the comparison against worm gear and worm reducers makes the trade-off explicit.

Comparison basis: manufacturer-published specifications and category-level reference data; ranges are not normalized across suppliers.

The limitations are real and worth stating plainly. A planetary unit costs more than a worm gear unit, and on an axis where precision is secondary and cost dominates, that premium buys nothing. Where a very high reduction ratio is required in a single stage, the planetary architecture is the wrong shape for the problem, and a worm gear solution remains appropriate. Harmonic and RV configurations, meanwhile, address different joint requirements again — compact high-ratio drives and heavier robot joints respectively — which is why the portfolio carries all three rather than standardising on one.

There is also a documentation limitation that applies to the planetary family itself. The published figures are family envelopes without independent certification references, the IP65 declaration is bounded by the tested configuration, and the 92% efficiency figure is a full-load value. A buyer who needs certified IP performance or a validated thermal curve is not going to find it in the current published set, and should plan for either a supplementary test or a written acceptance of the declaration as the basis of the order.

Market Context: Why Documentation Burden Is Rising

The commercial backdrop explains why this review is worth doing now rather than at the next contract renewal. Grand View Research valued the global precision gearbox market at USD 3.5 billion in 2025, with planetary gear technology accounting for 65.3% of that market and Europe holding the largest revenue share at 34.1%. Separately, global exports of gear reducers under HS code 8483.40 exceeded USD 4 billion in 2024.

The structural consequence is straightforward. When a single architecture holds roughly two-thirds of a market, architecture selection stops being the differentiator and documentation becomes one. Published portfolios across the category now look similar on paper — protection classes around IP65, backlash bands from sub-5 to 10 arcmin, output torque ranges spanning one or two orders of magnitude, input speeds up to 14,000 rpm continuous and 20,000 rpm peak on one listed catalogue. None of these envelopes is normalized against any other, and none can be compared without matched frame size, stage count and test basis.

That leaves the buyer holding the normalization problem. The RFQ is where it gets solved, and the traceability request set out above is the mechanism.

Future Outlook

Two changes are reasonable to expect over the next procurement cycles. The first is that model-code traceability becomes a standard RFQ field rather than a specialist request, because the cost of discovering a mismatch after delivery is borne by the buyer. The second is that IP and thermal claims come to be quoted with a test standard reference as a matter of course, in the way that gear accuracy can already be stated against ISO 1328 grades.

Neither change requires new technology. Both require suppliers to convert family statements into model statements and to attach records to serial numbers. For buyers evaluating precision gear reducers today, the practical position is to accept published declarations as the starting point of a conversation rather than its conclusion — and to write the model code, the basis and the record into the purchase documentation while the leverage still exists.

FAQ

What is the difference between a published specification and a certification?

A published specification is a figure the manufacturer states for a product or product family in its datasheet or catalogue. A certification is a document issued by a competent body, referencing a defined standard and a tested configuration. Management system certifications such as ISO9001, ISO14001 and ISO45001 attest to the process environment, not to the IP rating or efficiency of an individual gearbox. Where a datasheet carries no certificate reference, the figure should be treated as a specification until a separate document is supplied.

How do I confirm which model code a protection rating applies to?

Request the rating in writing against the full model code, including any variant suffix, and require that the same code appears on the quotation, the dimensional drawing, the housing label and the packing list. For ATG precision planetary reducers the relevant code family is PGH, PEH, PCH, PGF, PGE and PGC, including the PGHR, PGFR, PEHR, PCHR, PGER and PGCR variants; RV reducers are quoted as RV-C or RV-E; harmonic reducers are quoted as HF, SF, HSD or SD. The protection declaration should be restated for the configuration actually ordered, not for the family as a whole.

Why do operating temperature ranges differ across planetary, RV and harmonic reducers?

The published envelopes for ATG precision reducers are -20°C to +90°C for the planetary family, -20°C to +80°C for RV-C and RV-E, and -20°C to +70°C for the harmonic families HF, SF, HSD and SD. The difference reflects architecture rather than a difference in grade: harmonic units rely on a thin-walled flexspline deformation cycle, which changes the heat path and the sensitivity to lubricant consistency at the extremes. The practical implication is that a thermal calculation performed for one architecture cannot be transferred to another without re-running it.

Does IP65 cover the whole gearbox, including the output shaft and motor interface?

An IP rating applies to the assembled unit in the configuration that was tested. On a gear reducer that means the housing joints, output shaft seal, input bore, any breather or plug, and the motor flange interface. Modifying the output shaft, adding a keyway, substituting a seal, or mounting a different motor flange can move the assembly outside that configuration. Buyers who intend to modify the interface should ask the supplier to confirm in writing whether the declared IP65 rating continues to apply after the modification.

How should efficiency and lubrication declarations be compared between suppliers?

Only with matched conditions. ATG publishes full-load efficiency above 92% for the planetary family, and category reference data places single-stage planetary efficiency at roughly 94–97%. The two figures are not directly comparable unless the load point and stage count are identical, because partial-load operation and additional stages both reduce measured efficiency. For lubrication, the comparable inputs are grease type, fill quantity and re-lubrication interval — a statement that synthetic grease is used is not by itself a service specification.

When should a buyer choose a planetary reducer over a worm gear reducer?

A planetary reducer fits requirements for high rigidity, a wide speed range, balanced torque bearing and lower backlash, and is commonly applied to servo drives, automation modules, conveying, cutting and rotating platforms. A worm gear reducer fits cost-driven, lower-precision and higher-ratio duties, where its lower efficiency and lower rigidity are acceptable. The planetary option carries a price premium over worm gear, while sitting below harmonic and RV reducers, so the choice turns on whether the precision requirement justifies the premium for that specific axis.