The ‘Stolen’ Russian Super-Tank Britain Took Apart, And Found Bricks Where The Armour Should Be D
Kurylivka, Kharkiv Oblast. The last week of September 2022, the ground is soft from a wet autumn, and the fields east of the Oskil River are churned into mud the color of old iron. Colonel Pavlo Fedosenko commands the 92nd Separate Mechanized Brigade. He is a hero of Ukraine.
He has been fighting for Kharkiv since the last week of February, through the northern suburbs, through Saltivka, through the counteroffensive that has just collapsed an entire Russian front in 6 days. He has seen a great deal of Russian armor. He has seen it burn to the road wheels. He has seen it turned inside out, turrets thrown 40 m into tree lines.
He knows what a dead tank looks like. This one is not dead. It sits in the open field with its nose pointing east toward the river, toward the direction it was trying to go. The hull is unmarked. The turret is square and welded and unfamiliar, sitting low with slabs of reactive armor laid across the front like roof tiles.
The gunner’s sight is intact, its glass unbroken. The commander’s panoramic sight sits on top of the turret, undamaged, still pointing where the last man to touch it left it. The engine deck is closed. There is no fire damage. There is no penetration. There is no crater.
There is a track lying in the mud behind it, run off the road wheels. That is all. The soldiers approach it the way you approach anything that good, slowly. They assume it is mined. They assume it is bait, that somewhere in the tree line a gun is laid on it and waiting. They assume the optics have been smashed by the crew because that is doctrine, because that is what you do when you leave a tank behind.
They assume the fire control computer has been pulled. They assume the radio has been stripped. None of that is true, either. The crew have simply gone. What is standing in that field is a T-90M Proryv. Proryv means breakthrough. It is the most modern main battle tank in the Russian inventory.
It is the tank the Kremlin has spent 6 years photographing on parade grounds, the tank a former Russian president will stand beside on a factory floor and call the best tank in the world, and a brigade of Ukrainian infantry has just walked up to it in a wet field and found it whole, and the hatches open and nobody home.
This is not a story about a tank that was left behind. This is a story about a country that could no longer build what it was selling. The metal was real. Understand that first, because almost everything you will hear about this vehicle from either direction is wrong. What Western engineers found when they finally opened it up was not a fraud.
It was something more corrosive than a fraud, and it took 2 years and a war to see it. Part 1. To understand what the Ukrainians were looking at, you have to understand that it was not stupid. It was never stupid. The reasoning behind that tank was sound, and the men who made it were professionals.
Go back, not to 2022, but to 1991 and the desert. In February of that year, in about 100 hours of ground fighting, American and British armor destroyed Iraqi tank formations at ranges the Iraqis could not even see, let alone answer. Most of what burned in Iraqi service was Soviet T-55s, T-62s, export model T-72s.
The lesson landed in Moscow like a shell. It was not that the Soviet tank was badly made. It was that the Soviet tank was blind at night and outranged by day, and that the men in the American tanks could see first, shoot first, and hit first. Four years later, the second lesson arrived, closer to home in the streets of Grozny.
Armor without protection on the roof and the flanks, sent into a city without infantry to screen it, was destroyed at contact range by men with rocket launchers standing in basements and on balconies. Russia drew the correct conclusions from both, not the conclusions we would have drawn, but correct ones inside their own frame.
The answer was not to build a Western tank. Russia looked at the Western tank and quite rationally refused it. A Western main battle tank in that period weighed 60 tons and upward. It burned fuel at a rate that required an army of tankers behind it. It could not cross most of the road bridges in the Russian interior.
It could not be moved efficiently by rail across 11 time zones, and rail is how Russia has always moved its army, and it was ruinously expensive. Here is the arithmetic that governed everything the Kremlin did with armor for 20 years. A T-90M in its export configuration was offered at roughly four and a half million dollars a copy.
A Western main battle tank of the same generation cost 10 million dollars and upward before you priced the fuel, the recovery vehicles, and the trained maintainers it demanded. That was the pitch. That was the whole pitch. For the price of one Western tank, you could field two Russian ones and still have change for the ammunition, and it worked.
Russia became the world’s second largest arms exporter on precisely that argument. India bought the T-90S and then license built it at Avadi, hundreds of them, and named it the Bhishma. Algeria bought it. Iraq bought it. Vietnam bought it. This was not a curiosity. This was the mass-market tank of half the planet, and the T-90M was its most advanced expression.
The design followed the money. 48 tons, not 60. A crew of three, not four, because an automatic loader in the turret basket replaced the human loader, which meant a lower turret, a smaller target, and one less man to train, feed, and pay in every tank in the army. A 125-mm gun that could fire a long rod penetrator or launch a guided missile out to 5,000 m, and the protection was genuinely new.
The T-90M carried Relikt, the third generation of Russian explosive reactive armor. Not the old Kontakt-1 bricks of the 1980s, which worked against shaped charge warheads and nothing else. Not even Kontakt-5, which was the first reactive armor in the world that could meaningfully disturb a solid kinetic dart.
Relikt was designed to defeat tandem warheads, the two-stage missiles built specifically to beat the older armor, and to cut the penetration of a kinetic round by something close to half. Add a digital fire control system, a commander’s independent thermal sight so the commander could hunt for the next target while the gunner was still killing the last one, and you had a tank that could see at night, shoot on the move, and survive a hit that would have gutted a T-72.
His assessment was not insane. It was based on the data he had. The problem was that the data was 3 weeks out of date, and the data was about the last war. The model worked until it didn’t. Part two. What the 92nd Brigade recovered from that field was not a trophy. It was a document. The tank belonged to the 27th Separate Guards Motorized Rifle Brigade, part of the First Guards Tank Army, the formation Russia had spent a decade rebuilding as the spearhead it would use against NATO.
It was abandoned near Kurylivka on or about the 25th of September 2022, and photographed with yellow nines painted on it. The recognition mark of the brigade that took it. Understand what an intact vehicle gives you, because this is the whole point of the story. A burned tank tells you it burned.
You can measure the hole, you can guess at the round, you can photograph the wreck, and analysts did thousands of times and learned a great deal. But a burned tank cannot tell you what its armor is made of, because the armor is gone. It cannot tell you what the sight can resolve at 3,000 m, because the sight is slag. It cannot tell you what the fire control computer knows, or how the radio is encrypted, or what the crew were actually trained to do when a warning tone sounded in their headsets.
An intact tank tells you all of it. You can core sample the composite, you can bench test the optics against a known target at a known range and produce a number. You can strip the reactive armor cassette off the glacis, weigh it, cut it open, and find out exactly what is inside. You can read the software.
Andrew Galer of Janes put the value plainly when the capture became public. Examining the vehicle, he said, offered a greater level of insight into its armor composition. That is analyst language for, we can finally stop guessing. The tank was moved west, out of the fighting, and dismantled. Reporting at the time from a security source placed the work at an undisclosed location in western Ukraine or Poland, with the technical reports going back to the United States and the United Kingdom.
That the physical hull ever crossed the Atlantic is not confirmed, and I am not going to tell you that it did, but something did. In the spring of 2023, a Russian tank turned up on a flatbed trailer at a truck stop in Roanoke, Louisiana, because the lorry hauling it broke down. Photographs went around the world in a day.
The shipping label named its destination. Building 358, Lanyard Road, Aberdeen Proving Ground, Maryland, the home of the United States Army’s test center. The consignor was listed as a multinational assessment field team. That vehicle was a T-90A, an older model, not the M. I want to be precise about that because a great deal of what is said about this story confuses the two, and the confusion is doing work that the evidence does not support, but the Pentagon confirmed it.
The spokesperson, Sue Gough, said the tank was being transported to Aberdeen when the truck suffered a mechanical issue, and added one line almost in passing. The tank’s explosive reactive armor, she said, was inert. Pause on that sentence. It will come back. 15 months later, on the 13th of January 2024, near the village of Stepove in the Avdiivka sector, the argument stopped being theoretical.
Two Ukrainian soldiers of the 47th Mechanized Brigade, a commander gunner who goes by Serhiy, and a driver who goes by Oleksandr, were crewing an American M2 Bradley infantry fighting vehicle. Serhiy had come back from training on the vehicle in Germany only the month before. A Bradley is not a tank.
It has a 25-mm chain gun. Against the frontal armor of a T-90M, that gun is, in theory, an irrelevance. It cannot penetrate. Every table in every manual says so. They met one anyway at close range at night in the fog, and they did not try to penetrate it. They hammered the front of that turret with the chain gun, burst after burst, and they walked the fire across the sights, the gunner’s sight, the commander’s sight, the sensors, the vision blocks.
They took the tank’s eyes out. The turret began to traverse wildly hunting and found nothing because there was nothing left to find with. The tank reversed blind and drove itself into a tree. An FPV drone finished it. Seri asked afterward how he had known where to put the rounds. Said that as he used to play video games, he remembered everything, both how to hit them and where.
That is the whole war in one sentence and it is not a joke. It is a statement about knowledge. Somebody had worked out where the weak points were and told the crews and the crews remembered. Men like the crews of the 47th who took a troop carrier into a knife fight with the best tank Russia had deserve to be remembered for it.
A like keeps the story of Steppo visible for a few more people. It costs nothing. It matters more than it should. Now, the record. Part three, the teardown produced three findings. Each of them on its own is embarrassing. Together, they describe something that is not a tank problem at all. The first finding was in the sight.
The Sosna-U gunner’s sight, the eye of the tank, the thing that let it fight at night and shoot on the move and outrange everything it faced in Chechnya and Georgia was built around a thermal imager called the Catherine FC. The Catherine FC is French. It is made by Thales. This was not a secret exactly, but it had never been quantified.
In March 2022, the French investigative outlet Disclose went through the French Parliament’s own annual arms export report and did the sum. Since 2015, after the Crimean sanctions, France had issued 76 licenses for military and dual-use goods to Russia worth something in the order of 152 million euros.
Among them, thermal cameras sufficient to equip more than 1,000 Russian tanks, plus navigation systems and infrared sensors for aircraft. France, by that accounting, was responsible for 44% of all European arms exports to Russia. The chief beneficiaries were Thales and Safran. The French state is the largest shareholder in both.
So, the sovereign Russian supertank, the symbol of a Russia that had freed itself from the West, could not see in the dark without a component from a company part owned by a NATO government. And the day the sanctions bit properly, Russia could not buy any more of them. What happened next is visible in the photographs.
Newer T-90M’s began appearing with a different sight entirely, a domestic substitute. Older tanks came out of storage carrying sights designed in the 1970s because the modern ones had run out. The second finding was in the armor. Go back to Sue Gough’s line at the truck stop. The reactive armor was inert.
There was an innocent reading of that, and for a long time people took it. Of course, it was inert. You would strip the explosive out of a tank before you towed it through Texas on a public road. Nobody argued, but the word kept turning up in places where nobody had demilitarized anything. Ukrainian crews recovering Russian vehicles began levering the reactive armor cassettes off the glacis and the turret roof, and opening them because by then everybody wanted to know.
And the cassettes kept coming up light. Not damaged, not expended, not blown the way a cassette is supposed to blow once outward against the thing trying to come in. Empty. Some of them had been packed with rubber sheeting cut neatly to size so that the weight was about right, and the box did not rattle when you knocked on it.
That was the commonest finding, and it is the one to hold on to because it tells you this was not carelessness. Somebody had measured. Somebody had taken the explosive out and gone to the trouble of replacing it with something that felt like explosive to anyone who did not open the lid. And then, in September 2024, a T-90M was destroyed in Ukraine and photographed where it stopped, and the cassettes on its armor were open to the sky, and what was inside them was bricks.
Construction bricks. The kind you would use to build a garden wall. Read that line again because this is the tank that a Russian president stood beside on a factory floor and called the best in the world. It’s armor, the Relikt array, the third-generation explosive reactive armor that was the single most advanced thing about the vehicle and the reason it was supposed to survive a missile that would gut anything older had been unpacked by somebody somewhere between the plant and the front and refilled with masonry to make up the weight. I want to be careful here and honest because the honest version is worse than the exaggerated one. Not every T-90M carried bricks. Bricks are the extreme instance photographed once and I’m not going to tell you it was universal. Rubber was commoner and plenty of tanks went to war with cassettes that were exactly what they were supposed to be and those tanks survived hits that killed the ones beside them which is precisely how anyone noticed the pattern in the first place. But that is the point and it is the most corrosive thing in this entire story. The crews could not tell which tank they were in.
There was no way to know from the driver’s seat whether the slabs bolted to the front of your vehicle were the most sophisticated armor Russia had ever built or a stack of bricks in a steel box. Somewhere in that chain a man had opened a crate of explosive, sold it and filled the gap with whatever was lying in the yard and then three young men who had been told they were riding inside the best protected tank on earth closed the hatches over their heads and drove it into a war. They did not know. That is the part worth sitting with. Russian soldiers found this out the way the Ukrainians did by opening the cassettes and Russian military bloggers said so out loud angrily on Russian platforms and were not thanked for it. The third finding was the oldest and the one nothing could fix, the autoloader. The mechanism that let Russia field a three-man crew in a low turret is a carousel that sits in the floor of the fighting compartment under the men holding roughly 22 ready rounds with more stowed around the hull. There is no armored bulkhead between the ammunition and the crew and no blowout panel in the
roof to vent the blast upward the way a western tank does. A penetrating hit that reaches the carousel does not damage the tank, it detonates it. The turret which weighs many tons is thrown clear of the hull. The T-90M did not solve this. It could not. It is not a feature you bolt on. It is the architecture of the vehicle chosen in the 1960s, and everything else about the design, the low silhouette, the lightweight, the small crew, the export price, is built on top of it.
And now, the number that the whole argument stands on. In February 2022, when Russia invaded, the International Institute for Strategic Studies counted about 67 T-90 Ms in Russian service. 67. Not 6,700. 67. The tank that anchored Rosoboronexport’s global pitch, the tank that was going to reassure India and Algeria and Egypt that Russian armor was still worth buying, existed in its modern form in numbers you could park in a supermarket car park.
Everything else in that army was a refurbished T-72 or a T-80. Some of them older than the men driving them. The metal was real. There simply was not very much of it, and what there was could not be replaced with the parts Russia was now allowed to buy. And then, the arithmetic that had justified the whole enterprise turned around and ran the other way.
The tank Moscow priced at roughly 4 and 1/2 million dollars a copy began dying on camera to weapons costing a few hundred dollars. The an anti-tank grenade dropped out of a commercial quadcopter. Not an anti-tank missile. A grenade from a drone you could buy with a credit card. The cost advantage was the entire strategic case for the vehicle.
It was the pitch to India, the pitch to Algeria, the pitch to every defense ministry that could not afford a Western tank. And it evaporated because somebody had built a cheaper thing than a cheap tank. If you have served on armor, or worked in the trade, or you follow the vehicle recognition and the loss counts more closely than is healthy, I would be glad to read what you know below.
A great deal of what is now settled about this tank was worked out in public by people doing it for nothing. Part four. The admissions came from both directions, and the Western one came first. In October 2022, days after the capture became public, Sam Cranny-Evans, then a research analyst at the Royal United Services Institute in London, was asked what the West had actually got.
He did not sneer at the tank. He said the T-90M had a much more modern suite of explosive reactive armor designed around the shape of the vehicle and its turret. And then he said the thing that mattered. “This tank,” he said, “would undergo technical exploitation that could be highly valuable in terms of next-generation lethality.
” Translate that. Knowing exactly what the armor is lets you build the round that beats it. Not a better round in general, that round for this array at this angle. The British machinery for doing precisely that was confirmed on the record the following summer. In July 2023, Admiral Sir Tony Radakin, then Chief of the Defence Staff, said publicly that when Britain gets hold of Russian kit, it shares that knowledge, and that Britain has the scientists who unpick the detail to a really forensic level. How does their equipment work? How can we defeat it? In the same period, the Defence Command Paper recorded that the Defence Science and Technology Laboratory had developed bespoke armor solutions to give the Challenger 2 additional protection against specific Russian threats. Nobody has published which threats. Nobody’s going to, but the sequence is on the record in the right order, and you can draw the line yourself. The Russian responses came later, and they were the responses of a system that had understood the diagnosis and could not afford the cure. Cope cages welded over turret roofs,
rubber matting, slat screens. By 2026, the tank had largely gone from the assault role altogether, pulled back to reserve formations and indirect fire. Too valuable to lose and too visible to risk. British defense intelligence had already noted the pattern with the T-14 Armata, which was paraded and never seriously committed, and named the reason, the reputational damage of losing the prestige vehicle, which brings us to the export order book.
And here I have to correct the story that is usually told about it. Russian arms exports did collapse. In March 2025, the Stockholm International Peace Research Institute recorded a fall of 64% between the 5 years to 2019 and the 5 years to 2024. Russia’s share of the global market fell to 7.8% and for the first time it dropped to third place behind France.
The number of countries buying Russian weapons fell from 31 to 12. Independent estimates of the annual figure run from roughly $14 billion in 2021 to something under 1 billion by 2024. But it was not the teardown that did that. It would be a good story and it is not true. Pieter Wezeman of the Stockholm Institute gave the reasons and there were three of them.
Russia needs the weapons on its own battlefield. Sanctions make it harder to produce and to sell them and the United States and its allies have pressured other states not to buy. The largest cancellations, Indonesia’s Sukhoi order, Egypt’s, mostly predate the invasion entirely and were driven by American secondary sanctions.
The teardown did not empty the order book. It explained it. They were right that the tank was cheap. They were right that it could survive the weapons of the last war. What they were wrong about was that any of it was theirs. The optics were French. The machine tools were Western. The electronics were bought through third countries and the reactive armor was whatever the last man in the chain had not yet sold.
As for the tank in the field at Kurylivka, in October 2022, Ukraine’s Ministry of Defense published footage of Colonel Fedosenko driving a captured Russian tank out of a field and captioned it with some enjoyment as lend-lease continuing. Ukraine’s Logistics Forces Command later confirmed through its spokesman Valeriy Shershen that two captured T-90Ms had been formally registered into Ukrainian service.
They did not scrap them. They drove them. Part five. So we come back to the field and to the question the field asked. What was that thing standing whole in the mud that nobody had bothered to destroy before they ran? It was not a fake. I want to be exact about this because the internet has spent 4 years insisting otherwise and it is doing Moscow a favor every time it does.
The T-90M was a real tank, competently designed and in several respects a clever one. Relikt is a genuine advance. The gun is a good gun. The commander’s sight is the right idea. Ukrainian crews who captured them put them into service and used them, which is the only product review that has ever mattered.
What it was was a promise the country behind it could no longer keep. Their metric was the vehicle. The Western metric was the supply chain. Moscow counted tanks and could put 67 of the modern ones on the board. Western analysts counted the chain of custody behind each tank and found a French thermal imager, an imported machine tool and a cassette of reactive armor that had been sold on the way to the factory.
And we should be honest about our own side of the ledger because it is not clean either. The Western commentary overcorrected. For a year the professional consensus was close to contempt and contempt got people killed because a T-90M that hits you first is exactly as lethal as its brochure. Britain sent 14 Challenger 2s to Ukraine and they have been used well at range as snipers.
There is no confirmed instance of one destroying a T-90M. The famous kill was scored by two men in an American troop carrier with a cannon that was not supposed to be able to do it. The loss counts everyone quotes, including the ones I have quoted, are flaws and not totals because they only count what somebody photographed. Even the production figures are contested with the International Institute for Strategic Studies counting low and Ukrainian analysts counting high and neither able to prove it.
The metal was real. That is the last thing to hold on to and it is the thing that makes the story sad rather than funny. Somewhere in the Urals there are engineers who did their jobs properly, who designed a genuinely modern turret and a genuinely modern armor array and who then watched their work go to a war with a garden wall bolted to the front of it driven by three young men who had been told they were safe.
That is the answer to the question the field asked at Kurylovka. The crew did not run because their tank was bad. They ran because they had stopped believing in it, and they were not wrong to stop. To Serhiy and Oleksandr of the 47th, who took a Bradley into the fog at Stepove and shot a super tank’s eyes out. To Pavlo Fedosenko and the 92nd, who walked up to the most advanced tank in Russia in an open field and drove it home.
To the analysts in London and Kyiv who took it apart and told the truth about what they found, including the parts that were inconvenient. The tank was never the secret. The tank was the receipt. If this is the kind of work you want more of, subscribe and stay for the next one. Because the same question is now being asked about a great deal more than one tank.
