China Wouldn't Sell Ukraine Strike Zone - So Kyiv Stole the Blueprints

Whatever it takes. JL

Wes O'Donnell reports in Medium:

On July 16, Ukrainian drones traveled 800 kilometers inside Russia and reached Engels-2 air base, one of the country’s most important strategic bomber facilities. Satellite imagery showed a Tu-95MS strategic bomber with its tail section completely torn away. n the summer of 2023, China tightened its export controls on civilian drones. The Ukrainians accepted an offer for pilot training instead and used their time on the factory floor to quietly document everything they could see. “In parallel with the training, we photographed everything that was there,”. They flew home with eight actual drones and a phone full of pictures of the Chinese production line. China wouldn’t sell them the recipe, so the Ukrainians spent their training time photographing every shelf and cabinet in the kitchen. Once home, Ukraine took the eight aircraft apart and started building its own version. It's now building and firing 6,000 a year

On July 16, Ukrainian drones traveled roughly 800 kilometers inside Russia and reached Engels-2 air base, one of the country’s most important strategic bomber facilities.

Satellite imagery from Planet Labs, examined by the open-source group AviVector, later showed a Tu-95MS strategic bomber with its tail section completely torn away, sitting beside an intact sister aircraft, the kind of clean before-and-after comparison that leaves very little room for argument.

Russia stopped building Tu-95s in 1992. Every one that burns is permanently gone, and Russia’s fleet may now be down to somewhere around 43 airframes, with only about a dozen usually active on any given day.

For three and a half weeks, nobody publicly said what actually hit it.

Now we know.

On August 12, the Ukrainian defense outlet Oboronka published a report, cleared in advance with both the manufacturer and the SBU, identifying the weapon as the MICH 2000, a previously secret black flying-wing attack drone used exclusively by the SBU’s Alpha special operations unit.

Now, for US veterans of the Global War on Terror, the name “MICH 2000” makes us do a double-take. That happens to be the name of the Modular Integrated Communications Helmet (MICH) with “2000” being the first year it was developed.

I can’t find any public reporting to describe what the Ukrainian MICH is named after or whether it’s an acronym for something else.

Press enter or click to view image in full size

The MICH 2000 helmet. DoD

According to that manufacturer’s own account, the aircraft can fly up to 2,000 kilometers, carry a warhead between 25 and 60 kilograms, costs about $48,000 in its base configuration, and rolls off a Ukrainian production line at a rate of roughly 6,000 units a year.

Here’s the story of how they made it…

Back in 2022 and 2023, Ukraine was still scrambling to build a serious domestic deep-strike industry from close to nothing. Ukrainian units had been buying Chinese Mugin-5 PRO civilian drones and converting them into one-way attack aircraft, running around $15,000 apiece, carrying about 25 kilograms, and reaching roughly 500 kilometers.

That was a decent capability, although 25 kilograms is a little on the smaller side for explosives on a deep strike platform.

Then, in the summer of 2023, China tightened its export controls on civilian drones.

Beijing said publicly the restrictions weren’t aimed at any particular country. (Sure, wink wink. Yet they kept exporting junk to Russia during this timeframe.)

Whatever the reason, Ukraine needed another source, and late that year a team of Ukrainian engineers flew to China hoping to buy finished aircraft directly from a manufacturer in-country.

They found a factory near Hong Kong, run by a company called ZTK Drones, building dozens of ZTK-150s; Shahed-shaped flying wings distinguished by a pair of small forward canards.

They tried to buy in bulk. The purchase stalled.

Then the company offered to sell the blueprints.

“They came up with the idea to sell us the technology so we could make everything ourselves,” one team member said. “That’s when they planted the idea that we could really build drones in Ukraine instead of buying and shipping them.”

Ukraine’s military leadership loved it. When the team flew back to China ready to close the deal, the factory’s own management refused to authorize the sale, and the whole thing looked dead.

Then things got wonderfully strange.

The Ukrainians accepted an offer for pilot training instead and used their time on the factory floor to quietly document everything they could see.

“In parallel with the training, we photographed everything that was there,” one unnamed team member later told Oboronka. They flew home with eight actual drones and a phone full of pictures of the Chinese production line.

China wouldn’t sell them the recipe, so the Ukrainians spent their training time photographing every shelf and cabinet in the kitchen.

Once home, Ukraine took the eight aircraft apart and started building its own version. 

Anyways, back to the MICH 2000. It didn’t go well at first.

The first Ukrainian-built prototype flew about one kilometer before crashing, thanks to an off-balance center of gravity.

The second held together for roughly an hour before engine vibration sheared the bolts holding its propeller in place.

The third encountered Russian electronic warfare during testing and lost its control link entirely, proving the original Chinese architecture was never going to survive the electromagnetic soup.

Once Ukraine started producing its own fuselages instead of relying on the Chinese airframes, the crashes kept coming, but were getting fewer and farther between. The Ukrainian manufacturer told Ukrainska Pravda that for roughly six to eight weeks they simply kept destroying test drones while grinding through aerodynamic and structural problems one at a time.

Judge that record by any normal peacetime procurement standard and the obvious conclusion is that the aircraft wasn’t ready. Alpha reached a different conclusion entirely.

Alpha sent it to war before the engineers thought it was ready

According to the manufacturer’s account, Alpha eventually told the company that range testing had gone on long enough. The engineers protested that the aircraft still wasn’t flying properly but Alpha wanted combat results instead of more test data.

The first operational mission flew in October 2024.

Three drones launched.

One got shot down. The other two reached the target area and missed their assigned aim points by roughly ten meters.

Press enter or click to view image in full size

SBU and Oboronka

SBU Alpha approved production anyway…

By any traditional acquisition logic, that first mission reads as a clean failure, a one-for-three combat record with the two survivors both missing the target.

SBU Alpha apparently read the same mission and saw something different: The aircraft launched successfully, it navigated a real combat route through contested airspace, and two out of three penetrated far enough to reach their targets.

What remained broken was accuracy, and accuracy is an engineering problem you can solve, not a fundamental flaw you have to scrap the program over.

They didn’t need a perfect weapon that day. They needed a weapon that was good enough to survive long enough to enter the iteration cycle, and it barely, technically qualified.

Plenty of acquisition organizations would have shelved a weapon after a mission like that. Alpha apparently saw a prototype that had just delivered a very expensive set of test results under the only conditions that were ever going to matter.

The Modern War Institute at West Point has documented this exact pattern across Ukraine’s drone industry more broadly.

Ukrainian manufacturers commonly revise their hardware and software three or four times a year, with frontline combat units increasingly picking their own equipment and feeding performance data straight back to the people building it, no committee required.

Compare that to how major American acquisition programs have traditionally run, often stretching toward twelve years from concept to fielded weapon, with even the deliberately expedited pathways still measured in years rather than months.

Here’s the technical payoff: The manufacturer says almost nothing survives from the original Chinese ZTK-150 beyond its general outer shape.

MICH has grown into a family of eight distinct configurations rather than one fixed aircraft, with customers, (meaning Alpha), reportedly able to select different warheads, fuel arrangements, terminal guidance packages, and analog, digital, or Starlink communications links depending on the mission.

A MICH 2000 spotted in Russia. SBU.

Maximum reported range tops out around 2,000 kilometers, and the warhead swings anywhere from roughly 25 to 60 kilograms.

Not every target needs maximum range though. A mission 400 kilometers out can trade some of that fuel weight for a bigger warhead instead.

A route running through heavy jamming can justify spending money on better navigation gear.

Another target might need dedicated terminal guidance to hit precisely rather than approximately.

Six thousand a year

The unnamed manufacturer reports current production running around 6,000 aircraft a year, with an internal goal of pushing that toward 10,000, and says it’s reinvesting roughly $1 million a month back into research and development.

Worth a real qualifier here. MICH 2000 has exactly one customer, the SBU’s Alpha unit, so it isn’t fair to hold its 6,000-a-year figure up against something like Fire Point’s much larger FP-1 program as though the two are directly competing for the same contract.

FP-1 serves a considerably broader customer base across the whole military. What’s striking about MICH is narrower and, in its own way, more interesting.

A single Ukrainian special-operations organization apparently generated enough sustained internal demand to keep an industrial production line running in the thousands per year, which is a wildly different scale of ambition than what most people picture when they hear the phrase “special operations equipment.”

Back to Engels-2. The manufacturer attributes a whole list of MICH missions to targets including Ust-Luga port, Moscow-area industrial sites, a Wildberries logistics hub in Ryazan Oblast, an electronic-warfare system near Gelendzhik used to jam Starlink, and now Engels-2, along with strikes inside occupied Ukrainian territory.

Ukrainska Pravda’s own reporters independently reviewed dozens of videos of Shahed-shaped drones striking Russian targets and confirmed at least eight showing what they identified as MICH aircraft specifically.

The economics involved are remarkable.

A drone that costs roughly $48,000 in its basic configuration, built by a country under active invasion, potentially took a permanent bite out of a strategic bomber fleet that stopped being manufactured in 1992 and simply cannot be replaced.

Ukraine is trading something engineered to be manufactured by the thousands and destroyed on its very first successful mission against an aircraft Russia inherited whole from the Soviet Union and can never make more of. Chef’s kiss.

MICH+ proves maximum range was never really the point

The successor design, MICH+, reportedly cuts maximum range almost in half, down from 2,000 kilometers to around 600. On its face, that looks like a downgrade for a company that’s spent two years bragging about reach.

It isn’t, once you see what Ukraine traded for it. The shorter-range version reportedly increases the warhead to around 70 kilograms, adds real terminal guidance, drops the ZTK-150’s original forward canards entirely, and lands at a lower target price near $40,000.

That tells you the company was never chasing a single record-distance wonder weapon; it’s building a family of tools instead.

A 2,000-kilometer mission deep into Siberia has a completely different set of requirements than hitting a hardened facility sitting 500 kilometers behind the front line.

Fuel is payload you’re spending.

Range is payload you’re spending.

Guidance costs real money on top of both.

The useful weapon, in the end, is whichever configuration actually fits the specific target in front of you.