Connector Industry Roundup — October 5, 2026: ix Industrial Joins EtherNet/IP, Stronger Uplinks & Lunar Power Distribution

Four items this week, and they look unrelated until you notice what each one is really asking: put more signal through a smaller opening. A 10 mm-pitch industrial Ethernet connector that displaces RJ45. A radio uplink that has to carry video from a humanoid. A private 5G core squeezed into a factory cell. And ten kilowatts of solar that has to come down a 20-metre mast on the Moon and reach equipment that is not the lander.

Each is a mechanical problem wearing an electrical costume. That is the part of the story a connector engineer reads first, so that is the order this roundup takes.

ix Industrial Joins EtherNet/IP: A 10 mm Pitch Takes On RJ45

ODVA has officially specified the ix Industrial connector interface for EtherNet/IP network applications (eeNews Europe, 4 October 2026). The significance is not that a new connector appeared — ix Industrial has existed for years — but that it now sits inside five of the major industrial Ethernet ecosystems rather than being a PROFINET-only or vision-only choice.

EtherNet/IP joins PROFINET, CC-Link IE TSN, FL-net and GigE Vision in supporting the interface. For a machine builder, that changes the sourcing conversation: one connector family can now cover a multi-protocol product line instead of forcing a different interface per fieldbus.

The three-supplier detail is the one procurement teams should note. A standardised interface available from Hirose, HARTING and Amphenol means second-sourcing is a commercial option rather than a redesign, and it is why a standard reaching a fifth protocol actually matters: standards with one supplier are a product, standards with three are a market.

Adoption is already visible in products. Siemens, Beckhoff and B&R are named among the industrial equipment suppliers that have incorporated the interface, and transition cables between ix Industrial and RJ45 are available through distributors including Sinbon and Misumi. That last point matters more than it sounds: transition cables are how an installed base migrates one machine at a time instead of one line at a time.

The trade-off is honest and worth stating. ix Industrial is not an RJ45 with a smaller shell — the double-row contact system and reflow shield bond require a different PCB footprint and a different harness build, and the IP65/67 versions depend on the push-pull locking mechanism rather than a bayonet or threaded coupling. If your programme's qualification file is written around RJ45, moving to ix Industrial is a harness and PCB change, not a connector substitution.

Stronger Uplinks: What AI RAN Asks of the RF Interface

Ericsson's Anders Söderlund described AI RAN as a two-sided transformation at RCR Wireless News' Intelligent RAN Forum 2026 (5 October 2026): “AI for networks” and “networks for AI.” The second half is the one with connector consequences. Söderlund said applications such as physical AI, smart glasses and humanoids will demand new network capabilities, naming a stronger uplink specifically.

That is a directional change, not a capacity upgrade. Mobile networks were designed around downlink — content flows to the user. A humanoid robot or a pair of smart glasses streams video, depth and telemetry up, continuously, from a device with a battery and a thermal budget. Ericsson's stated AI-native 5G features give a sense of the scale being targeted: up to 25% higher spectral efficiency, five times more accurate positioning, improved coverage prediction, and up to two times more high-traffic users.

The physical problem is stated plainly in the announcement: with tens or hundreds of thousands of cell sites, there is limited room for additional power and space at individual sites. Adding receive chains and accelerators to a massive MIMO radio does not add a cabinet. It adds boards, connectors and heat inside an existing radome, which is exactly the environment where a connector's insertion loss and PIM performance stop being datasheet footnotes.

The complementary data point is on the coverage side. Kyocera validated a multi-hop mmWave repeater architecture at Qualcomm's San Diego campus (5 October 2026), targeting non-line-of-sight coverage without adding base stations. The stated demand driver is familiar: AI-driven video analytics, machine vision and real-time high-resolution video. Every repeater hop is another RF launch connector, another jumper and another opportunity for a return-loss problem to compound — which is why multi-hop architectures are as much a cable-assembly qualification exercise as a radio design.

Private 5G as Factory Infrastructure: MOSAIC-5G's Three Trial Sites

A UK consortium launched MOSAIC-5G — Multi-vendor Open System Architecture for Integrated Connectivity — an 18-month, £4.3 million programme backed by over £3.1 million in Innovate UK grant funding under the SNPN 5G Assured Integration competition (Bisinfotech, 5 October 2026). Members include Aerix, Antevia Networks, Spry Fox Networks, Noema Signal Labs, Attocore, RANsemi, Liverpool 5G and Bath & Northeast Somerset Council.

The problem statement is an integration one: individual UK-developed 5G components are maturing, but the engineering effort to combine them into a complete, market-ready system has kept deployments too complex for mass adoption. MOSAIC-5G will integrate Antevia's “5G SHIFT” private network, AttoCore's 5G core and Aerix's management platform into a single interoperable 5G Standalone private network that is deployable in hours rather than weeks.

For a connector supplier the interesting number is the 18-cell outdoor network in Bath. A campus-scale private 5G deployment is a real cabling project: every cell needs power and fronthaul, every outdoor run needs a sealed interface, and every installation is a place where an IP-rated connector either survives a British winter or becomes a maintenance ticket. The robotics field trial adds the harder case — a moving work cell where the radio link competes with motor drives for the same electromagnetic environment.

The honest read is that private 5G is being positioned as an alternative to cable runs in exactly the applications where cable is most protected: fixed machinery in a defined cell. A wired fieldbus does not have a scheduling problem or an uplink budget. What wireless buys is reconfigurability, and that is worth a great deal to a factory that changes its line every few months. It is not worth much to a line that has not moved in five years, which is why both technologies will coexist and why hybrid cells — wireless for moving assets, sealed wired connectors for everything fixed — are the realistic outcome.

Ten Kilowatts Down a 20-Metre Mast: Blue Origin's Power Tower

Blue Origin has shown a 20-metre-tall vertically deployed solar array called the Power Tower, designed to provide more than 10 kW of electrical power for lunar surface operations near the Moon's south pole (eeNews Europe, 5 October 2026). It is intended to deploy from the company's Blue Moon MK1 lander and, per Blue Origin, extend power beyond the lander itself to other surface equipment.

The reason for vertical deployment is geometric. At the lunar poles the Sun stays low on the horizon, so ridges, crater walls and small terrain changes cast long shadows. Raising the array above the surface reaches prolonged sunlight and uses less ground area than a large horizontal array. The company describes a modular architecture intended to support future surface power networks.

Consider what “extend power beyond the lander” actually requires. Ten kilowatts at a low distribution voltage is hundreds of amps, and it has to travel across a surface where the thermal cycling is extreme, the dust is abrasive and electrically charged, and there is no maintenance call. Every joint in that path — array to mast, mast to lander, lander to rover or instrument — is a connector that has to hold contact resistance across hundreds of cycles between roughly −170 °C and +120 °C in vacuum.

That is the qualification conversation this announcement sets up, and it is not one that a catalogue part answers. It requires contact resistance stability across thermal vacuum cycling, outgassing data for every non-metallic material in the assembly, and a mating interface that a suited astronaut or a robotic arm can operate without fine alignment. Blue Origin has released terrestrial deployment footage and no flight date; until a mission flies, the useful engineering output of an announcement like this is a requirement list rather than a specification.

The wider context is that lunar power is being treated as infrastructure rather than as a lander feature. The same company is working on Blue Alchemist, intended to turn lunar regolith into silicon solar cells and aluminium conductors. Producing the generation hardware locally and positioning it where the sunlight is are two ends of the same problem — and both ends terminate in connectors.

The Buyer's Read

The pattern in all four stories is the same one that has run through this series all year: the connector is being asked to do more inside a smaller envelope, and the qualification evidence is the product. ix Industrial wins on 75% less volume, but only after someone proves 5,000 mating cycles and a 360° shield bond. An AI RAN radio wins on uplink capacity, but only if the extra receive chain's launch connector holds its return loss at temperature. A private 5G cell wins on deployment speed, but only with a sealed outdoor interface. And a lunar power mast wins on 10 kW, but only if every joint in the distribution path survives a thermal-vacuum profile no catalogue sheet describes.