Skip to main content
Omron

Two Industrial Parts Look Identical — The 12 Axes That Decide Whether They Will Actually Interchange

Why two industrial parts that look identical are rarely interchangeable. The 12 axes that decide a real substitution, from mounting and firmware to protocol and lifecycle, and what a written cross-reference cannot certify.

Two Industrial Parts Look Identical — The 12 Axes That Decide Whether They Will Actually Interchange

A substitution only works when every relevant axis — mounting, terminal layout, firmware version, communication protocol, certification status — actually matches between the original and the candidate. Two industrial parts that look identical almost always differ on at least one axis; on most modern control gear those axes include firmware version, engineering-software compatibility, and lifecycle status. For an overseas buyer (panel shop, OEM, MRO, repair shop — including buyers in Russia and the CIS), this page lays out the twelve axes that drive the decision, what a written cross-reference can and cannot certify, and where an independent distributor will refuse to call a candidate safe.

By the AoCtrl Sourcing Desk, procurement editorial team. Data through September 22, 2026.

Why "looks identical" is the wrong starting point

A buyer looking at a control cabinet rarely starts from a clean sheet. The line that needs replacing was chosen years ago by someone who is no longer at the company, the original selection rationale is in a folder no one can find, and the cabinet door is open because the line is down. In that moment the temptation is to find a part that fits the same hole and has the same wires coming out of it. The substitution gets bolted in, the program uploads, and either the line runs or it does not. If it does not, the next twenty minutes are spent on fault-finding that nobody budgeted for.

The problem with this approach is that outline similarity is a single axis among twelve, and it is usually one of the easier ones to satisfy. Two sensors, two relays, two I/O modules, or two compact controllers can share the same housing, the same connector keying, and the same DIN-rail footprint, and still differ on five other axes that determine whether the substitution actually works. The manufacturer did not design the parts to be visually interchangeable; the manufacturer designed them to be functionally interchangeable within a defined series, and a candidate from a different series — even one that looks the same — sits outside that design intent.

A photoelectric sensor pair is a good illustration. The Omron E3ZM family — including the through-beam E3ZM-T81 2M and E3ZM-T61 2M and the diffuse-reflective E3ZM-D62 2M variants — and the SICK W4-3 family both occupy the same compact cubic footprint, both use the same M8 connector family, and both carry an IP67 rating on most variants. A buyer looking for a quick replacement sees four matched axes at a glance and assumes the rest are matched too. The datasheet then shows that the sensing range differs, the default output logic differs, the supply voltage range differs by a few volts, and the EMC class is not equivalent for cabinet-mounted versus field-mounted installations. None of that was visible from the outside. None of it was implied by the housing. Cross-reference inventory and per-line availability are quoted per order against the catalog and confirmed per line on the quote; lead time varies by line and by condition (new surplus, refurbished, used).

A written cross-reference is the document that names which ones.

The 12 axes that decide whether a candidate will actually fit

Every cross-reference we publish is built around twelve axes. Each axis is reported as one of three states: matched, differs, or unknown. Omitting an axis would silently imply it matches, so the framework is deliberately exhaustive — even the axes that look like they should be obvious are written down.

Axis 1 — Form, fit and outline (mounting)

The first axis is the housing and the way it is mounted. DIN-rail versus panel-mount, screw versus snap, single-module width versus double-module width, terminal cover clearance for the wiring duct above the device. A difference here is a hard stop: a panel-mount candidate cannot replace a DIN-rail original without cutting a new hole in the cabinet door, and a wider module may not physically fit in the slot allocated for the original.

Axis 2 — Dimensions

The second axis is dimensional: outer envelope, mounting hole spacing, depth including the wiring space behind the device, and weight as it bears on the DIN rail for a row of devices. Two sensors can have the same outline but a different depth by five millimetres, which is enough to make a pre-wired M12 connector fail to seat or press against the wiring duct.

Axis 3 — Terminal and interface assignment

The third axis is the assignment of terminals and connectors. Pin numbering, supply terminal position (L+ versus M), I/O channel mapping, and front-connector type are not interchangeable across families even when the housing is. A wiring schedule that worked for the original can short the supply if applied to the candidate. This axis is the one where silent swap-ins cause the most expensive field damage.

Axis 4 — Electrical parameters

The fourth axis is the electrical rating: supply voltage range, per-channel current rating, total current, output voltage class (24 VDC versus 230 VAC), surge and short-circuit protection, and power dissipation. A candidate with a wider supply range is not automatically a stronger part; it can also imply a different inrush behaviour that the upstream fuse was not sized for.

Axis 5 — Function and I/O specification

The fifth axis is function: number and direction of channels, signal type (digital, analogue, temperature, counter), resolution and accuracy, isolation between channels, and module-specific functions like high-speed counting, PTO, or PID. A candidate with more points is not a candidate with more features. A candidate that omits high-speed counting cannot replace an original that was using it, regardless of how many spare channels the candidate has.

Axis 6 — Firmware, hardware version and engineering-software compatibility

The sixth axis is firmware and engineering software: hardware revision (FS, version suffix), firmware version, the engineering environment that programs it, the program portability (will the existing program recompile or does it have to be rewritten), and the device description files (GSD, GSDML, EDS, ESI). This is the axis most often reported as unknown, because manufacturers do not always publish the function state of a given revision, and because program portability is decided by the buyer's engineering process, not by the distributor.

When the firmware axis is unknown, the substitution may still be possible — but it requires the buyer's engineering team to validate the program and the configuration, not just the wiring.

Axis 7 — Communication protocol

The seventh axis is communication: bus type (PROFIBUS, PROFINET, EtherCAT, Modbus TCP, DeviceNet, CC-Link, and so on), port count and connector form, baud rate or link speed, maximum station count or topology, master-versus-slave role, and network configuration file compatibility. Two remote I/O stations that both speak PROFINET can still differ on the device-description file revision, on whether they support IRT versus RT, and on the GSDML version that the controller needs to recognise the station. Swapping the station without re-importing the GSDML is a common field failure.

Axis 8 — Mechanical parameters

The eighth axis is mechanical: terminal torque, allowable wire cross-section, mechanical life (number of operations), mating cycles, and latch or locking style. A contactor with a lower mechanical life rating may be acceptable for a panel-mount application but not for a hoist or an elevator duty cycle where the cycle count is the design constraint.

Axis 9 — Environmental class and EMC

The ninth axis is environmental class and EMC: ambient temperature for operation, humidity and condensation, vibration and shock, EMC immunity and emission class, pollution degree, and installation altitude. A commercial-grade sensor cannot be quietly substituted into an industrial-grade assembly — the EMC class alone may change the cabinet wiring practice.

Axis 10 — Ingress protection and temperature range

The tenth axis is ingress protection (the IP rating, with explicit note of whether the rating applies to the device as installed or to the device body), and the operating temperature upper and lower limits. A candidate without an explicit IP rating cannot be substituted for an IP67 original on a washdown line, regardless of how the housing looks. IP ratings are not inferred from the presence of a seal.

Axis 11 — Materials and construction

The eleventh axis is materials: contact plating and base material, housing material and flammability rating (UL94), sealing material (NBR, FKM, silicone), PCB conformal coating, and RoHS/REACH status. A downgrade in contact plating can shorten the life of a low-voltage signal contact that was designed for a gold-plated original.

Axis 12 — Certifications and lifecycle

The twelfth axis is certification and lifecycle: CE/UL/CSA declaration with the actual file number, functional safety rating where relevant, lifecycle status (current, phased out, end-of-life), and the manufacturer's spare-parts and repair support window. This axis also carries the certification requirements of the country the machine will be installed in, which is decided by the importer of record, not by the distributor.

Across all twelve axes, the most common state is unknown rather than matched. The honest cross-reference names which axes are unknown and explains what would be required to make them known — usually a datasheet the manufacturer has not published, a test that the buyer's lab has not run, or a configuration file the buyer's engineering team has not yet ported. The dishonest cross-reference fills the unknown axes with assumed-matched values and ships the document as if it were a conclusion. That is the document an independent distributor does not sign.

What a written cross-reference contains

A cross-reference we issue for a BOM line is a short document, usually one to two pages per line, structured to make the twelve axes easy to read at a glance. The buyer gets a table with one row per axis, three columns for matched, differs, and unknown, and a fourth column with the evidence source — the datasheet section, the revision note, the device-description file version, or the explicit "not stated by manufacturer" tag where the manufacturer does not publish the figure.

The cross-reference also carries the line identity and the substitution class for that specific comparison. We use five classes: similar specification, requires redesign, functionally equivalent, pin compatible, and manufacturer-documented replacement. The vast majority of industrial substitutions land in similar specification or requires redesign. Functionally equivalent and pin compatible are reserved for cases where the buyer can quote specific axes that match and the remaining axes are at-or-above the original. Documented manufacturer-documented replacement is reserved for cases where the manufacturer itself publishes a cross-reference file that says so — and for industrial automation parts, that is almost never.

Each comparison ends with the verification instruction: verify against the original manufacturer datasheet and your own qualification process. The instruction is not a legal disclaimer — it is the actual workflow. The buyer-side validation is what closes the loop; the cross-reference is the input to that validation, not the output of it.

What a written cross-reference cannot certify

A cross-reference cannot certify that the substitution will work in the buyer's specific machine. It cannot certify program portability, because the program is the buyer's. It cannot certify that the certification scope of the buyer's machine still holds, because the certification scope is decided by the importer of record and the notified body for the destination market. It cannot certify functional safety integrity, because SIL and PL conclusions require a documented qualification process the distributor does not perform.

A cross-reference also cannot certify that the manufacturer will continue to support the candidate part for the lifetime of the buyer's machine. Lifecycle is an axis, not a guarantee. The candidate's current lifecycle status is reported, and the buyer is expected to weigh that against the expected remaining service life of the machine.

Finally, a cross-reference cannot make a substitution appropriate for a part that the machine's certification file pins to a specific MPN. The cross-reference can name the relevant axes, but the decision belongs to the buyer's engineering team and the certification body where one is involved, and most rated assemblies are pinned to a specific part number by the machine's own documentation. Substituting there is a redesign, not a replacement, and the cross-reference makes that distinction explicit rather than papering over it.

When an independent distributor walks away from a substitution request

There are four cases where we tell the buyer the substitution is the wrong answer.

The first is when the part sits inside a rated assembly whose certification file names a specific part. UL-listed panel builders, CE-marked machines, and EAC-registered products all carry scopes that depend on the exact parts used; the substitution then has to be a redesign with the certification body involved, not a swap.

The second is when the part is pinned by the machine's own regulatory documentation. Some machines are built under a framework where the documentation names the part, and substituting the part means re-documentation. The right answer is to source the original.

The third is when the application sits inside a certified assembly whose scope would change. A panel builder's UL-listed assembly or a CE-marked machine falls into this category — the substitution has to be a redesign with the certification body involved.

The fourth is when the substitute cannot be verified on the axes the application depends on. If the application depends on EMC class, and the candidate's EMC class is not published, the cross-reference reports the axis as unknown — and the right answer is to keep looking for a candidate that publishes the figure, or to keep the original.

What an independent desk can do, in practice

The independent desk's job is to make the cross-reference visible, per line, before the buyer commits. For a single line, that is a short table. For a 40-line BOM, it is a stack of those tables, one per line, with the rated-assembly lines flagged for engineering review rather than silent substitution. The work happens in the same sourcing flow that produces the quote: each line is sourced independently for price, MOQ, lead time, and condition, and where the original is no longer available, the line is paired with a cross-reference table and the buyer decides whether to accept it, defer it, or rework it.

We screen end users and end uses, classify before quoting, and decline transactions that cannot be screened. Delivery to Russia and the CIS is arranged under the Incoterms stated on the quotation; we do not hold EAC or TR CU and we do not file conformity on the buyer's behalf.

The desk does not certify the substitution. The desk does not issue an EAC or TR CU declaration for the destination market — that work belongs to the importer of record. The desk does not claim to be an authorized distributor for any of the manufacturers whose parts it sources. The desk's accountability is to the line identity, the line condition, and the cross-reference axes it reports as matched, differs, or unknown. The buyer's engineering process is what closes the loop on whether the substitution will actually run the machine.

If you have a line in front of you where the cabinet is open and the production manager is asking how long, send your BOM or the single line to our sourcing desk through the inquiry form linked from the procurement page. Include the part number on the failed unit or the marking if the number is missing, a photo if available, the machine it came from, and any context on the cabinet location and the wiring. The cross-reference for that line comes back with the twelve axes named rather than assumed.

FAQ

Does a cross-reference report mean the substitute is guaranteed to work?

No. A cross-reference report is a documented comparison of the two parts across mounting dimensions, electrical parameters, communication protocol, firmware status, certification scope, and lifecycle. Each axis is named as matched, differs, or unknown. The decision to substitute belongs to the buyer's engineering team, and the report itself ends with the verification instruction verify against the original manufacturer datasheet and your own qualification process.

Why do same-series parts still differ after substitution?

Same-series parts are not necessarily interchangeable. A revision suffix on the same MPN can change connector keying, default output logic, wiring pin assignment, or engineering-software identification. Same-brand, same-series substitutions are a known source of field failures and are treated as requires redesign in the cross-reference until the per-revision datasheet confirms otherwise.

How many axes does a working substitution need to match?

There is no fixed number. A substitution works when every axis the application depends on is matched and the unknown axes are not on the application's critical path. A substitution fails when even one application-critical axis differs or is unknown. The cross-reference table is the document that makes the critical-path list explicit so the buyer's engineer can decide which axes are application-critical for that specific machine.

What does the cross-reference report include when the manufacturer does not publish a figure?

The axis is reported as not stated by manufacturer rather than filled in from a similar-looking product. Where the buyer has the figure in their own records, sending it back to us lets the cross-reference improve. Where the manufacturer genuinely does not publish the figure, the axis stays unknown and the substitution is treated as conditional until the buyer's qualification process closes the gap.

When is substitution the wrong answer entirely?

When the part sits inside a rated assembly whose certification file names a specific part, when the machine's regulatory documentation pins the design to a specific MPN, when the application is inside a certified assembly whose scope would change, or when the candidate cannot be verified on the axes the application depends on. In all four cases the right answer is to keep sourcing the original, or to treat the change as a redesign rather than a replacement.

Does the desk issue an EAC or TR CU declaration for the destination market?

No. EAC and TR CU declarations are issued by the importer of record in the destination market and are not within the scope of what an independent distributor issues. The desk can supply the commercial invoice, the packing list, and the per-line condition disclosure; it cannot issue a conformity certificate on the manufacturer's or the regulator's behalf. The same applies to GOST-R and other regional conformity marks.

What is the fastest way to get a cross-reference for a single line?

Send the line identity, a photo if available, the machine the line came from, and the cabinet location. The cross-reference for a single line is usually issued within one working day, with the twelve axes named and any rated-assembly condition flagged for engineering review.

Data Notes

Cross-reference axes and substitution-class taxonomy above are drawn from this site's published cross-reference practice and from the worked substitution examples in the AO Ctrl sourcing FAQ (questions X01 through X12). Specific MPNs cited (E3ZM-T81 2M, E3ZM-T61 2M, E3ZM-D62 2M) are referenced from the Omron brand hub; per-line condition, price, MOQ and lead time are quoted per order and confirmed per line on the quote, not stated in this article. No external market data sources were used for this article — the buyer question addressed (how to verify a substitute actually fits) is a methodology question whose answer is the site's own twelve-axis framework, not a market news item. This site is an independent industrial automation distributor, not an authorized distributor of any manufacturer named. EAC and TR CU declarations are issued by the importer of record in the destination market and are not within the scope of what an independent desk provides. Russia and the CIS are served under the Incoterms stated on the quotation.

Last updated: September 28, 2026