Week five of 2025. That date code marks an Analog Devices AD9253 analogue-to-digital converter photographed inside a Kometa-M12 satellite-navigation module pulled from a downed Russian Izdelie-30 cruise missile, documented in teardown findings Ukraine's Defence Intelligence (HUR) released on its War & Sanctions portal on 2 March 2026. The identical circuit board mounted an Analog Devices AD9753 converter manufactured in June 2024 and a Swiss u-blox GNSS receiver produced in September 2023. Three years into Western export controls, the guidance pack that determines whether jamming can deflect a Russian missile off course was still relying on American and Swiss silicon under a year old.
Kometa designates the family of controlled-reception-pattern antennas (CRPAs) and companion receivers developed by VNIIR-Progress in Cheboksary, capital of Russia's Chuvash Republic. Russian units bolt these arrays into the strike weapons launched at Ukrainian cities: Geran one-way attack drones, UMPK glide-bomb kits, cruise missiles and the 9M723 Iskander-M. Across Russia's precision-strike inventory, Kometa offers the single component family whose design iterations can be tallied straight from battlefield wreckage. The open record substantiates a narrower case than common commentary suggests, and that disparity matters for planners preparing sanctions or kinetic strikes against the production chain.
Counting Patches: The Variant Ladder
The underlying physics demanding Kometa is detailed in Defence Ukraine's examination of Russia's FAB glide bombs and the UMPK kit: a CRPA projects a reception null toward an electronic jamming emitter rather than trying to filter out interference. What that article did not map out was how rapidly Russian designers have added elements to the antenna face.
The earliest Kometa arrays recovered from Shahed and Geran wreckage through September 2023 operated with four antenna patches. By March 2025, David Hambling reported in Forbes that Russian industry had scaled the system from four elements to eight and then twelve. The component database maintained by HUR now catalogues those subsequent steps by specific product name. It records a Kometa-M8 GM10 inside the UMPB guided munition (published 13 January 2025), a Kometa-M12 aboard the new jet-powered Geran-5 (19 January 2026), a Kometa-M12 in the extended-range UMPK-PD glide kit (2 February 2026) and the Izdelie-30 module described above. For the Iskander-M, the registry details a distinct branch: an ARP-4T antenna for a Kometa-MR4-01 set, published on 1 December 2025, alongside R8 and R12 receiver designations.
A tidy progression from Kometa-4 to Kometa-8, Kometa-12 and Kometa-16 suggests a cleaner lineage than the factory floor delivers. Ukrainian records log VNIIR designations including M8, M12, M16, R8, R12, MR4-01 and A-8 OTK, while older battlefield reporting applied the blanket tag "Kometa-M" to almost any Russian four-, eight-, twelve- or sixteen-element array. Patch count marks only one dimension of this family. Enclosure geometry, receiver architecture and weapon integration form the others.
The sixteen-element tier is the most fragile link in this ladder. Sixteen-element CRPAs surfaced on Shaheds in early 2025, but Militarnyi identified the ones it reported that March as Chinese-made. The particular claim that a sixteen-element Kometa-M flew aboard a 9M728 Iskander-K traces to Defense Express on 17 June 2025, attributed to anonymous sources without a single published photograph of a stamped module. HUR industrial tracking files do list an M16 among VNIIR outputs for the UMPK programme, making the variant very likely real. A combat recovery of that variant, tied back to VNIIR by factory markings, has not appeared in public records.
What Each Step Costs
An array with N elements can steer at most N minus one independent nulls. Four elements provide three, eight supply seven, twelve deliver eleven and sixteen offer fifteen. That remains a theoretical baseline. Calibration drift, jammer geometry, multipath reflection, mutual coupling between neighbouring patches and airframe motion all degrade that figure, and a 2024 Russian academic paper on adaptive arrays found that coherent jamming from two locations proves harder to cancel than two separate sources. Russian sales brochures have claimed 40 to 50 decibels of interference suppression for Kometa-type equipment. No independent test has validated that figure. It is vendor marketing.
The technical trade-offs at each step are substantial, but unclassified sources never measure them directly. Adding patches requires larger aperture surface or closer spacing, more parallel radio-frequency lines and expanded digital signal processing. No open intelligence gives weight, power draw or thermal output for any Kometa variant, and nothing found in public archives shows Russian engineers redesigning a Geran, UMPK or missile airframe to fit an enlarged array. Teardowns supply the sole hard benchmark available: the component bill of materials. The M12 in the UMPK-PD mounts an Altera Cyclone field-programmable gate array (FPGA), an AD9751 digital-to-analogue converter, a Linear Technology LTM4615 power module and a u-blox M10050-KB receiver. Inside the UMPB, the M8 GM10 carries LTM4615V and LTM4613V power units alongside a u-blox MAX-M10S receiver. Each step up this ladder demands more conversion and compute channels, and those are the components Russia must import.
Nothing in the evidence suggests Russia reserves higher-element arrays exclusively for high-cost missiles. Technicians install the M12 into the Geran-5 and into glide-bomb kits priced at a fraction of a cruise missile. Which array mounts to which airframe appears determined by assembly batches, integration timetables and component inventories just as much as the monetary cost of the delivery system.
The solitary price circulating in public, a Defense Express statement asserting that sixteen-element CRPAs cost "tens of thousands of dollars apiece", lacks any published foundation. A firmer price point comes from a Chinese manufacturer, not a Russian state supplier.
Inside the Box: Western Silicon and Chinese Substitutes
Every Kometa record in HUR's database lists the same three Russian contract electronics fabricators alongside VNIIR-Progress: PANTES, PSB Technologies and AVIV Group. The UMPK-PD profile adds Morion, the St Petersburg manufacturer of precision quartz oscillators. Cheboksary is the engineering and integration centre, but plant workers do not produce every printed circuit board. A sanctions or interdiction plan focused purely on that single corporate entity misses the contract assembly tier operating below it.
Imported parts dominate data conversion, field-programmable gate arrays, power management and the GNSS receiver core. The antenna patches came from abroad as well, at least in earlier production runs. In May 2025, researchers from the Ukrainian anti-corruption group NAKO reported that undamaged eight-element Kometa assemblies from UMPK kits carried eight patches manufactured by Irish antenna specialist Taoglas, priced at roughly US$7 each. Trade records cited by The Irish Times on 30 May 2025 placed Taoglas-branded goods imported into Russia between 2022 and 2024 at US$752,928, routed primarily through vendors in China, Hong Kong, India and Turkey. Taoglas said it had cut off Russian buyers in February 2022 and tracked the components NAKO identified back to a commercial sale to a Chinese client in 2017. The 2026 HUR teardowns do not indicate whether engineers have switched antenna brands since, leaving no documented trail of another manufacturer replacing Taoglas patches.
What physical evidence does confirm is full-module substitution. Militarnyi documented a circular eight-element Chinese CRPA on a downed Shahed in January 2025, a twelve-element Chinese array around February and a sixteen-element variant by March. In April 2025, the publication reported an unfamiliar eight-element array recovered near Poltava and raised the possibility that a second factory had begun assembling adaptive arrays. That remains an open question. Analysts have named no second Russian Kometa manufacturer, and no registry filing or production licence links that board to another facility.
The firmest evidence documenting the Chinese pipeline originates from an undercover probe by the investigative outlet Nordsint, released on 8 May 2026. Reporters who contacted Shenzhen antenna maker Harxon through regular sales desks were referred to an affiliate, NavX Technology. NavX quoted seven-, eight- and sixteen-channel CRPAs and drew up an invoice for 200 units: 100 seven-channel, 50 eight-channel and 50 sixteen-channel, totalling ¥13.1 million (about US$1.93 million), settled through Russia's VTB Bank. NavX priced the sixteen-channel model at ¥108,864, roughly US$16,000. Product labels designated the antennas as agricultural-machinery navigation hardware. A NavX representative told the investigators the company supplied Alabuga, the Tatarstan plant that builds Gerans, insisted the partnership remain secret, and noted a 32-channel version under development. Customs logs revealed an Alabuga-based trading firm buying more than US$1 million in Harxon antennas across August and September 2024.
The invoice confirms a supply route and fixed unit pricing. It does not prove handlers delivered the undercover order, and remarks from a sales agent do not constitute an admission by Harxon executives. It does prove that a finished, commercially traded alternative to Kometa sells at a verified unit cost, and that Russian procurement teams were buying it long before heavy strikes hit Cheboksary.
Cheboksary: Six Strikes and No Production Figure
Commentators routinely describe VNIIR-Progress as part of Rostec state corporation's KRET electronics holding. Corporate documentation does not back that up. An EU sanctions measure from July 2026 identifies JSC Sokol Asset as its direct parent and positions the institute within the ABS Electro group; no reviewed filing shows KRET or Rostec owning shares. The plant also manufactures circuit breakers, relays and electrical equipment, including hardware for Russian naval programmes. The US Treasury designated the enterprise on 2 November 2023 under sanctions targeting Russia's electronics sector, and the United Kingdom added it on 13 June 2024.
Ukrainian forces have struck the facility repeatedly. On 9 June 2025, two AN-196 Liutyi drones struck it, and Oleg Nikolayev, head of the Chuvash Republic, said managers had temporarily halted production to protect workers. Commercial imagery following a drone strike on 5 July 2025 revealed a roof breach and fire damage across a recently built workshop. Nikolayev confirmed a major drone attack on Cheboksary on 26 November 2025, which the independent Russian outlet ASTRA geolocated to the factory grounds, and local sources reported another attack on 18 February 2026. On 5 May 2026, satellite photos analysed by the Ukrainian open-source group CyberBoroshno showed an FP-5 Flamingo cruise missile striking the front of the main building and two Liutyi drones hitting a second structure. On 10 June, Nikolayev reported a missile strike against the city, and ASTRA once more mapped the footage to VNIIR. A separate strike on 12 June 2026, occasionally cited in public accounts, finds no backing in the record: reports circulating around that date describe the 10 June attack.
The plant's passive defences thickened after each strike. By May and June 2026, steel masts stringing anti-drone netting encircled the targeted buildings, and later commercial imagery showed one workshop fully enclosed inside a cage. Those local precautions reinforced the Russian air-defence redeployment toward interior industrial sites that was already shielding Cheboksary's defence-electronics facilities. Russia clearly regards the compound as worth protecting.
The true impact on factory output goes unrecorded in open sources. The only Russian public statement touching output remains the June 2025 suspension. No public data reports how many Kometa modules workers produce each month or year, and nothing links any individual strike to declining Kometa recoveries, to older four-element models resurfacing or to degrading Russian strike accuracy. Crews operated a crane beside the main building within days of the May strike, pointing to rubble clearance or repairs rather than a manufacturing restart date.
Three explanations fit available evidence and cannot yet be untangled. Successive strikes might have triggered real production losses that Russian authorities keep secret. Dispersed subcontracting through PANTES, PSB and AVIV, combined with stockpiles of completed modules, could have cushioned assembly lines. Buying Chinese CRPAs may have diminished VNIIR's role across certain Geran production batches. Chinese module substitution began well ahead of the 2026 strikes, so analysts cannot interpret it as a reaction to them.
The Proliferation Claim That Did Not Hold
The Kometa narrative gained an international angle in March 2026. The UK government confirmed that an Iranian drone struck RAF Akrotiri in Cyprus during the night of 2 March, causing no casualties. Within twenty-four hours, video pushed by aggregator account Clash Report claimed to reveal a Kometa-M receiver among the wreckage. Cypriot outlet SigmaLive challenged the report that same week, showing that the photographed component matched footage released in Ukraine during 2024. Cyprus Mail, citing The Times, later mentioned a "Kometa-B" system and said British technicians had received salvaged hardware at a UK laboratory. No British or Cypriot official announcement has confirmed what the laboratory discovered.
The wider transfer claim rests on stronger sourcing but remains unverified at the component level. The Wall Street Journal reported in March that Russia had supplied Kometa-M to Iran, while News24 reported on 23 July that the Kremlin dismissed the report as fake news. The Institute for the Study of War has since cited US officials saying Iranian operators flew Shaheds upgraded with Russian assistance, and separate dispatches describe Russian drone components shipped across the Caspian. None of that reporting identifies a Kometa module by stamped marking, serial code or physical teardown. Across the public record, military drone cooperation between Russia and Iran runs both ways. A Kometa in Iranian hands remains alleged, not shown.
Strategic Implications for Ukraine
Four practical conclusions emerge for allied governments and analysts whose sanctions enforcement, export controls and intelligence operations target the Kometa supply chain through 2027.
- Date codes are the measure that matters. An AD9253 manufactured in January 2025 inside a module fired at Ukraine is a lot-tracing puzzle with an achievable solution: the chip moved through a distributor, and investigators can find that distributor. Broad declarations that export controls have "failed" give customs and sanctions investigators nothing to act against. Routine release of date codes and batch numbers, matching what the HUR portal already publishes, provides Western regulators with concrete shipments to follow, standing as the cheapest intelligence asset Ukraine can deliver to allied governments on this front.
- Sanctions need to cover the network, not the brand. The United States designated VNIIR-Progress in 2023 and the United Kingdom followed in 2024, yet identified Russian contract manufacturers and Chinese rivals draw far less scrutiny on Western sanction rolls. A sanctions policy pursuing only the Cheboksary corporate label overlooks PANTES, PSB and AVIV across the board-assembly tier and misses complete module shipments entering through Harxon and NavX. Dwindling recoveries of VNIIR-marked modules would not prove eroding Russian anti-jam capability; they could just as easily mean substitute suppliers have filled the pipeline.
- Strike results should be judged by production indicators, not imagery. A burned workshop is battle damage. It is not an accounting metric for lost factory output. Hard indicators showing real disruption to Kometa deliveries would comprise verified assembly restart dates, gaps in silicon date codes, a shifting ratio of Russian and Chinese antennas across structured wreckage surveys, and final assembly lines shifting location. Until such evidence emerges, assertions that Cheboksary has been "destroyed" outpace the known facts and create a credibility penalty when recovered Kometas continue reaching Ukrainian soil.
- The Chinese route is a diplomatic file as well as a technical one. The Nordsint invoice ties an identified Chinese company, verified unit pricing and a Russian state bank into one document. Partner governments can take that specific dossier directly to Beijing and pursue action through their respective export-control and sanctions authorities. Such concrete documentation proves far harder to wave off than vague protests over dual-use leakage.
The Kometa progression remains readable because observers can count design changes directly on salvaged circuit boards, with element tallies now reaching twelve elements verified by HUR teardowns and sixteen reported in the field. The next factual benchmark that will alter this assessment is not another image of a damaged workshop in Cheboksary. It is a recovered module bearing a 2026 date code on its converters, or a Harxon array pulled from a Geran with a VNIIR label beside it.
