Every gas turbine ever built runs the same four steps, and the shop-floor shorthand for them hasn’t changed in decades: suck, squeeze, bang, blow. The squeeze is the one that costs you. Before a turbine can burn anything at a useful efficiency, a compressor has to force air up to pressure, and that compressor draws its power from the very turbine it is feeding.<\/p>\n
Daniel Banuti puts the bill at roughly half. He directs the Institute of Thermal Energy Technology and Safety at the Karlsruhe Institute of Technology, and in KIT’s announcement<\/a> he says a conventional gas turbine, the kind sitting in a power plant or hanging under an airplane wing, spends about 50 percent of its output compressing air. That power never reaches the generator.<\/p>\n So his group built one without a compressor. It ran on hydrogen for 303 seconds, and it made electricity. KIT went public with the run on February 17, and the prototype spent five days in Hall 11 at the Hannover Messe in April.<\/p>\n The technology is called pressure-gain combustion, and the name is the whole idea. In a normal turbine, combustion happens at roughly constant pressure, and the burner actually loses a little of it along the way. Here, the burning itself raises the pressure.<\/p>\n What does the raising is detonation. Instead of a flame front that ambles along below the speed of sound, the chamber sustains detonation waves that travel faster than sound and slam the gas behind them up to pressure. KIT says those waves emerge from a fluid-mechanical instability, the interplay of flow, waves and vortices inside the chamber.<\/p>\n Nothing mechanical is doing the squeezing. No blades, no shaft, no bearings, nothing to stall or surge. Fewer moving parts means less to wear out and less to pay for.<\/p>\n Hydrogen is not a requirement. KIT says the concept works with other fuels, but hydrogen suits it unusually well, because it reacts extremely fast and gives stable pressure rises. It is also, per KIT, the reason for choosing it at all: unlike natural gas, hydrogen can be made with renewable electricity.<\/p>\n None of which is exotic in isolation. Baker Hughes has an industrial turbine certified to push a ship on 100% hydrogen<\/a>, and Rolls-Royce has run a business-jet engine to full take-off power on the stuff<\/a>. Both of those machines still carry a compressor. That is the part Karlsruhe threw away.<\/p>\n KIT frames its 303 seconds against a previous NASA record of 250 seconds, and most of the coverage repeated that without asking which NASA record.<\/p>\n The number lines up with a hot fire NASA ran at Marshall Space Flight Center in the fall of 2023, when a full-scale Rotating Detonation Rocket Engine combustor burned for 251 seconds and produced more than 5,800 pounds of thrust. That hardware is a rocket engine combustor. A rocket has no compressor because it carries its own oxidizer in a tank, so compressorless describes every rocket that has ever flown.<\/p>\n There is also the matter of what NASA has done since. Its Integrated Rotating Detonation Engine System project page<\/a> records a December 2025 test in which a rotating detonation thrust chamber assembly fired for just over 340 seconds. That is a liquid-methane and liquid-oxygen lander engine in the 5,000 to 10,000 pound-force class, and it outlasted the German run by more than half a minute.<\/p>\n Both things are true at once. Karlsruhe holds a runtime record for a compressorless gas turbine. Nobody holds a runtime record for detonation combustors in general, and the 250-second bar was never the ceiling the press release implies.<\/p>\n Strip out the record framing and a genuinely new result is still sitting there. Karlsruhe did not merely run a detonation burner. It bolted a turbine to the back of one and turned the shaft into electricity.<\/p>\n That is harder than it sounds, and the reason is mechanical. Detonation exhaust is violent and unsteady, and turbine blades want smooth, steady flow. KIT’s own framing is that the speed and intensity of the combustion makes stable energy transfer to the turbine difficult.<\/p>\n Banuti’s claim is narrow and he states it plainly: “We are the first to successfully operate such a turbine and generate electricity”<\/a> in the process. Detonation combustors have been fed into turboshaft engines before, including a Rolls-Royce M250 that Purdue researchers worked on under a Department of Energy program, as AIAA’s Aerospace America has documented. Those engines keep their compressors. A compressorless one driving a generator is a different animal.<\/p>\n The timeline also runs backwards from the headline. The electricity came before the record. A conference paper KIT’s Joachim Grune presented last September carries the title “Reaching 90 sec runtime, throttling, and turbine operation with a hydrogen-based rotating detonation combustor,” which puts the same rig at 90 seconds nine months before it reached 303.<\/p>\n The two language versions of KIT’s release do not quite agree, either. The English one dates the electricity milestone to earlier in the year. The German original<\/a> says only that it happened beforehand, with no year attached.<\/p>\n Banuti is an interesting person to be beating a NASA number. Before he took the KIT job he taught at the University of New Mexico, and before that, by the university’s own account<\/a>, he was a postdoctoral scholar at Caltech and NASA’s Jet Propulsion Laboratory, at Stanford’s Center for Turbulence Research, and a staff scientist at the German Aerospace Center.<\/p>\n The 50 percent figure has carried most of the coverage, and it is being read backwards.<\/p>\n That number is what the compressor consumes. It is not what you win by removing it. The high-pressure air still has to exist, the pressure still has to come from somewhere, and thermodynamics does not issue a refund for changing which part of the machine does the squeezing.<\/p>\n The US Department of Energy has actually run the numbers. Its National Energy Technology Laboratory puts the potential efficiency gain from pressure-gain combustion<\/a> at 4 to 6 percentage points for a simple-cycle system and 2 to 4 points in combined cycle. That is a real gain on a machine this mature. It is not half a power plant.<\/p>\n NETL also keeps a list of what still has to be solved: fuel injection, fuel and air mixing, stopping the pressure wave from blowing backwards out of the chamber, initiating the detonation reliably, holding a genuine pressure gain, controlling nitrogen oxides and carbon monoxide, and surviving the unsteady heat loads the hot gas dumps into the turbine.<\/p>\n KIT has published none of those numbers for its own rig. No power output, no efficiency figure, no measured pressure gain. In a field where the pressure gain is the entire point, that is a conspicuous blank.<\/p>\n On June 2, NETL announced its own breakthrough on this class of combustor<\/a>, and it went after a different bottleneck.<\/p>\n Detonation chambers have long struggled with startup instabilities, the waves refusing to form or refusing to stay formed. NETL researchers ran high-fidelity computational fluid dynamics on the injector geometry and produced what they call an aero-strut configuration, which feeds fuel and air into the chamber with a lower pressure drop across the injector.<\/p>\n Then they put it in a water-cooled test rig. According to Justin Weber, who leads the project with the lab’s Advanced Turbines team, the modified injector produced stable, sustained detonation waves across a range of flow rates, air-to-fuel ratios and pressures. Weber is also blunt about why this took so long, saying that “transitioning this concept into reliable hardware has long posed scientific and engineering challenges.”<\/a><\/p>\n Karlsruhe and NETL are working opposite ends of the same machine. One got the thing to run for five minutes and spin a generator. The other got the fuel and air to enter it without tripping over themselves on startup. Neither has a product.<\/p>\n A piston engine already gets its squeeze without a separate compressor, using a piston, and that architecture is not sitting still either. W\u00e4rtsil\u00e4 has run a large piston engine on 100% hydrogen and put the electricity onto Spain’s national grid<\/a>. That is the bar a compressorless turbine has to clear, and it is a moving one.<\/p>\n What Karlsruhe is after is the squeeze with no moving part at all, in a package light enough to eventually hang under a wing. Five minutes of runtime and a working generator is a real step toward that, and a long way from a power plant.<\/p>\n The prototype spent April on a trade show stand. The next thing worth asking KIT for is not a longer runtime. It is a pressure-gain ratio, an efficiency figure, and a NOx number, because those three will decide whether the compressor was ever the expensive part.<\/p>\n Image Credit: Joachim Grune, KIT<\/em><\/strong><\/p>\n","protected":false},"excerpt":{"rendered":" Every gas turbine ever built runs the same four steps, and the shop-floor shorthand for them hasn’t changed in decades: … <\/p>\nThe squeeze happens inside the flame now<\/h2>\n
NASA’s record belongs to a rocket, not a turbine<\/h2>\n
Coupling a turbine to a detonation is the hard part<\/h2>\n
Deleting the compressor does not hand you back half the power<\/h2>\n
The Energy Department is working the same machine from the other end<\/h2>\n
Where this actually stands<\/h2>\n