Built without a compressor, a hydrogen turbine in Karlsruhe ran for 303 seconds and produced electricity that no machine of its kind had ever generated before
The combustion chamber is roughly the size of a kitchen appliance, ringed with sensors.
Feed it pure hydrogen and the fuel detonates in a wave that moves faster than sound.
Earlier versions survived that violence for only fractions of a second before the walls began to melt.
Because intense heat flux and rapid pressure oscillations repeat many times per second, the test required advanced thermal management and structural reinforcement.
In February 2026, at a test facility in Karlsruhe, Germany, the machine ran for 303 unbroken seconds and put electricity onto instruments for the first time.
What is still missing from every other machine that has tried this is worth understanding before that number sinks in.
Why every gas turbine alive today wastes half its power before it makes any
A conventional gas turbine consumes roughly 50 percent of its output just to compress incoming air to the pressure needed for efficient combustion. That compression energy never reaches the shaft. Strip out the compressor and, in principle, the machine doubles the share of combustion energy available for electricity generation.
Pressure gain combustion replaces that energy intensive step entirely. The detonation wave itself does the squeezing, arriving so fast that it raises chamber pressure before the gas has time to expand and escape. So the combustion event that would destroy a conventional burner becomes, in this design, the mechanism that sustains it.
But detonation exhaust is violent and unsteady, and turbine blades want smooth, steady flow. The Karlsruhe team’s answer was a carefully shaped transition section between the chamber and the turbine wheel that absorbs enough of each pressure pulse to let the blades survive.
What the test stand in Karlsruhe actually recorded on that February morning
Researchers at the Karlsruhe Institute of Technology ran a compressorless hydrogen gas turbine continuously for 303 seconds and confirmed electricity generation under load. That runtime exceeds the previous 250-second record established by a US space agency program for comparable experimental systems.
The electricity came before the runtime record. A conference paper presented the previous September carries the title “Reaching 90 sec runtime, throttling, and turbine operation with a hydrogen based rotating detonation combustor,” putting the same rig at 90 seconds nine months before it reached 303. The team pushed steadily through the window that melted every predecessor.
Because intense heat flux and rapid pressure oscillations repeat many times per second, the test required advanced thermal management and structural reinforcement. Sustained performance across five unbroken minutes is the proof that the chamber and turbine assembly can maintain stable power output, not just survive a momentary burst.
Where the Karlsruhe result sits against the field, and what the 50-percent figure really means
Detonation combustors have been fed into turboshaft engines before, including under a US Department of Energy program. Those engines kept their compressors. A compressorless machine driving a generator is a different animal entirely.
“We are the first to successfully operate such a turbine and generate electricity with it,” says the institute’s director of thermal energy technology. The distinction matters commercially: eliminating a component that accounts for roughly a third of a turbine’s weight and mechanical complexity reduces both manufacturing cost and the lifetime maintenance burden.
Meanwhile, for an iron air battery already linked to a national grid, clean generation without combustion losses remains the harder half of a zero carbon power system. A turbine that skips compression addresses exactly that gap.
The thermal wall that stopped everyone before, and how this design crossed it
Earlier test runs often ended after just a few seconds as thermal stress pushed combustion chambers to their limits. Chamber walls face temperatures that would liquefy most structural alloys within moments, and the detonation wave repeats those spikes many times per second.
Five continuous minutes means surviving millions of those events in sequence. Eliminating the mechanical compressor also reduces parasitic energy consumption and total system mass, making a lighter, simpler and cheaper machine, which matters when the goal is deploying clean power at scale.
The detonation wave, traveling faster than sound inside the loop, generates the pressure gradient that keeps combustion stable without external forcing. That self sustaining quality is what allowed the machine to throttle, a capability the earlier 90-second run proved and the 303-second test confirmed under sustained load.
What still has to happen before a machine like this stands inside a power plant
Five minutes on a test rig is a long way from the thousands of hours a commercial gas turbine runs between major overhauls. A durable blade material that can absorb detonation pulses across tens of thousands of cycles does not yet exist at production scale, and a hydrogen supply chain capable of fueling commercial units remains a separate challenge entirely.
The institute’s director described the runtime record as “an important step toward highly efficient and flexible hydrogen energy for a fossil free energy system.” The phrasing is deliberate. A step is not an arrival, and the real work ahead is not small.
But for ocean based power projects already seeking clean generation partners, a machine that wastes no energy on compression before it makes a single watt is precisely the kind of upstream breakthrough that changes downstream economics. The shaft turned, the instruments registered power, and the melting problem that ended every previous run did not win.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
