The Zumwalt Was a Disaster… Until It Became a Hypersonic Warship

Ajouté : 2026-06-03

[00:00:00]In 2023, the US Navy did something that made absolutely no sense. They took the most expensive destroyer ever built and ripped out its only guns. Not upgraded them, removed them. For years, critics called the Zumwalt a disaster. $7 billion warship with weapons so expensive the Navy literally stopped buying ammunition for them. People said the ship had failed and honestly, they were right. Until the Navy suddenly revealed what was replacing those guns.

[00:00:30]Because hidden inside this failed destroyer was something no other warship on Earth had enough power to carry. A weapon so fast that modern missile defense systems can't reliably stop it.

[00:00:44]And the terrifying part is this. The engineers who built the Zumwalt accidentally prepared it for this weapon 25 years before the weapon even existed.

[00:00:55]Let's be precise about what actually failed on the Zumwalt program. Because most coverage collapses three separate failures into one narrative. And that collapse hides the real story. The Navy originally requested 32 Zumwalt class destroyers. They received three. This was a genuine acquisition failure, but not primarily an engineering one. It was a cost accounting disaster triggered by a doctrine change. In 2001, the priority mission of the Zumwalt was naval surface fire support. Replacing the battleship's job of shelling beaches ahead of marine landings. When the Marine Corps shifted doctrine away from opposed amphibious assaults after Iraq, deciding that direct beach landings against defended shores were no longer a core scenario, the justification for 32 expensive shore bombardment ships evaporated overnight.

[00:01:51]The hull count collapsed because the mission changed, not because the engineering failed. The advanced gun system, the 155-mm long-range land attack projectile, is a genuine engineering success that became an economic catastrophe. The gun itself works. It fires guided rounds to 63 nautical miles with GPS precision. In testing, it performed as designed. The problem? The specialized round cost $800,000 each at low rate production. At the planned quantity across 32 ships, that unit cost would have fallen to approximately $50,000, competitive with conventional precision munitions. When the fleet shrank to three ships, the production economics collapsed. $800,000 per round for a destroyer that fires a few hundred rounds a year is not a viable wartime supply chain. The gun failed economically because the fleet was cut. The fleet was cut because the mission changed. The gun itself was not the primary problem. The electromagnetic railgun was defunded in fiscal year 2022 after 15 years and roughly $500 million in development. This one is more straightforward. The technology worked in a laboratory and could not survive the transition to operational hardware.

[00:03:15]Rail erosion, the destruction of the conducting rails after repeated high-current firings, was never solved at operational scale. A weapon that degrades its own barrel after 100 shots is not a fleet weapon. But here is the critical detail. The railgun's failure was independent of the Zumwalt's design.

[00:03:37]The 78-MW integrated power system was built to power the railgun. When the railgun died, the power system remained, more capable than anything else afloat, waiting for the next weapon that needed it. These are three separate failures with three separate causes. The engineering of the hull itself, the power system, the hull geometry, the propulsion, failed at nothing. It simply had no weapon worthy of it for 20 years.

[00:04:09]That ended in November 2025. To understand what 78 megawatts means, you need to understand what a normal destroyer's power architecture prevents.

[00:04:19]An Arleigh Burke class destroyer, the backbone of the US fleet, runs four General Electric LM2500-30 gas turbines producing a combined 108,000 shaft horsepower. Those turbines connect mechanically through reduction gears to two propeller shafts. The system is reliable and proven. [music] Here is its fundamental limitation. The power is mechanically committed to propulsion. The turbines produce rotational energy. That energy goes into the shaft. If you want electricity for anything else, a new radar, a laser weapon, a high-energy weapon system, you must steal it from the propulsion train.

[00:04:58]The Burke generates approximately 7.5 megawatts of electrical power for ship systems. That is what remains after the propellers take their share, 7.5 megawatts. For a modern destroyer packed with Aegis radar arrays, communication systems, combat computers, and sensors, now compare this to the Zumwalt's integrated power system designed by General Electric and Rolls-Royce. Two Rolls-Royce MT30 marine gas turbines, 35.4 megawatts each. Two Rolls-Royce RR4500 auxiliary turbine generators, 3.8 megawatts each. Total installed generation, 78.4 megawatts. But the architecture is fundamentally different.

[00:05:42]None of that power goes directly to a shaft. All of it goes into a common medium-voltage DC electrical bus, essentially the ship-wide power [music] grid. From that grid, 234.6 MW advanced induction motors drive the propeller shafts electrically. The rest of the grid is available for anything else.

[00:06:02]>> [music] >> At cruising speed, 20 knots, the propulsion motors consume roughly 20 MW.

[00:06:08]That leaves approximately 58 MW available for ship systems, weapons, and sensors. 58 MW. The Burke has 7.5. The Conventional Prompt Strike Hypersonic Missile System requires significant thermal management power at the moment of launch. Cooling systems, launch assist, electronics. Exact classified figures are unavailable, but comparison to the Army's land-based CPS launcher suggests requirements measured in MW, not kW. A future high-energy laser weapon for ship defense systems currently in development by Northrop Grumman and Raytheon for the Navy's Helios and Hellcat programs requires between 150 kW and 2 MW, depending on engagement scenario. The Burke cannot spare that margin. The Zumwalt has it waiting. The AN/SPY-6 radar. The system under evaluation for the Zumwalt under the Zeus upgrade draws substantially more power than the current AN/SPY-3.

[00:07:09]The Burke Flight III was specifically redesigned to accommodate the SPY-6's power appetite. The Zumwalt handles it without a redesign. The 58 MW of reserve is not a lucky coincidence.

[00:07:20]>> [music] >> The engineers calculated in the mid-1990s that the next generation of directed energy weapons, high-powered radars, and advanced launch systems would require power in this range. They were right. They were just right 25 years early. The reported acoustic signature of the Zumwalt is one of the most counterintuitive facts in modern naval engineering. A Los Angeles class nuclear attack submarine displaces approximately 6,900 tons submerged. It is specifically designed and built to be acoustically invisible. Every major system, pumps, [music] turbines, generators, is mounted on acoustic isolation systems. Crew members wear soft-soled shoes. Machinery spaces are lined with anechoic tiles. [music] The Zumwalt displaces 16,000 tons fully loaded, more than twice the Los Angeles.

[00:08:08]It sits on the surface. It has propellers turning in water at close to cavitation speed. Physically, it should be one of the louder ships in the fleet.

[00:08:16]It's reported acoustic profile approaches that of a submarine for three reasons. One, electric drive eliminates mechanical noise. A conventional destroyer's reduction gears are significant noise sources. Gear mesh creates vibration at precise frequencies, harmonics that propagate through the hull into the water and are detectable by passive sonar at ranges of tens of nautical miles. The Zumwalt has no reduction gears. The turbines connect to generators. The generators connect to the bus. The bus connects to the advanced induction motors. There is no mechanical gear mesh anywhere in the propulsion chain. The AIM motors are mounted on prairie/masker acoustic isolation systems, the same vibration isolation technology used in submarines.

[00:09:01]The propellers use a pump jet configuration that suppresses cavitation, the bubble formation that creates distinctive broadband noise at high speeds compared to conventional open propellers. Two, the hull geometry reduces flow noise. The tumble home hull with wave-piercing bow changes how water moves around and under the ship. A conventional flared hull creates turbulent water flow at speed, turbulence that transmits acoustic energy into the water. The wave-piercing bow of the Zumwalt slices through waves rather than riding over them, reducing low-frequency hull slap. The overall hydrodynamic profile produces a quieter acoustic wake. Three, the IPS synchronizes power delivery.

[00:09:44]Conventional diesel generators and turbine generators create acoustic spikes when they cycle on and off to meet changing power demands. The IPS maintains a stable balanced load across the electrical bus, reducing the thermal and acoustic variation of the power plant during normal operations. The result, a 16,000 ton surface combatant that passive sonar system struggled to distinguish from background ocean noise at tactically relevant ranges. Why does this matter for the hypersonic mission?

[00:10:14]The conventional prompt strike has a range exceeding 1,500 nautical miles, but range is only tactically [music] useful if the launch platform can reach its launch position without being detected and killed first. A noisy destroyer in the Western Pacific is a target, a quiet one is a threat. The acoustic signature is not a secondary feature. It is a prerequisite for the weapon to reach its launch position. The Tsushima connection is real, but the real lesson is more nuanced than it first appears, and the Navy's decision to use tumble home, despite that history, reflects a calculation that most coverage misses. The Russian Borodino class battleships that sank at Tsushima in 1905 had tumble home hulls.

[00:10:57]They capsized under Japanese shell fire.

[00:10:59]Naval architects cited this for 100 years as proof that tumble home was inherently unsafe. Here's what that analysis gets wrong. The Borodino class ships capsized because Japanese shells struck the upper hull and opened flooding paths above the waterline. The tumble home geometry contributed to the rate of capsize once flooding began. The narrower deck provided less reserve buoyancy to compensate for progressive flooding, but the Zumwalt's tumble home configuration addresses exactly this failure mode through two design features that didn't exist in 1905. Active ride control, the ship carries an automated ballast and stability control system that continuously adjusts internal water ballast to maintain trim and roll stability. The system reacts faster than any sea condition can develop. It is functionally impossible for the ship to capsize in the way the Borodino class did because the stability management system would compensate before the geometry created a dangerous condition.

[00:11:59]Combat damage philosophy, modern warship design accepts that ships will be hit in combat. The question is whether they survive hits and continue fighting. The Zumwalt's outer skin is composite, non-metallic, non-sparking, resistant to radar reflection, and somewhat less susceptible to shrapnel propagation than steel plate. A hit on the outer hull does not automatically create the progressive flooding that doomed the Borodino. Captain Andrew Carlson's account of navigating through sea state six, waves 13 to 20 feet, and having the executive officer in his cabin report conditions as sea state three is not anecdotal color. It is the validation data. The ship encountered 20-ft waves, the crew experienced 6-ft waves. The difference is the hull geometry working exactly as designed. The tumblehome configuration does something else that no conventional hull can match at this scale. It reduces the ship's radar cross-section by approximately an order of magnitude compared to a conventional destroyer of similar displacement. A conventional destroyer, an Arleigh Burke, a Type 45, a Hobart class, presents vertical steel sides to incoming radar energy. Those surfaces are enormous retroreflectors. They bounce radar waves directly back to the emitting antenna. At 610 feet long, a destroyer has roughly 15,000 square feet of above waterline vertical hull surface on each side. The tumblehome hull slips inward. Radar energy hitting those angled surfaces deflects upward and away rather than returning to the emitter.

[00:13:32]The superstructure on the Zumwalt is constructed from carbon fiber composite and fiberglass, materials that absorb and scatter radar energy rather than reflecting it. The reported radar cross-section of the Zumwalt, classified but leaked in various defense analyses, is roughly equivalent to that of a large fishing boat, somewhere in the range of 50 to 100 square meters. A Burke class has an RCS estimated at 5,000 to 10,000 square meters. That is a 100-fold reduction in radar visibility for a hull that is larger. In practical terms, Chinese and Russian over-the-horizon radars, systems like the Type 382 Tombstone naval radar and Russia's Podsolnukh OTH radar, that can track a Burke at 200 nautical miles, cannot reliably acquire the Zumwalt at 50. For a ship that needs to reach a launch position 1,500 miles from its target, those 150 undetected nautical miles may determine whether it survives the approach. The Conventional Prompt Strike (CPS) is a two-component system.

[00:14:36]Understanding why it's so difficult to intercept requires understanding what happens at each stage. Stage one, the boost phase, a two-stage solid fuel rocket booster accelerates the payload from rest to approximately Mach 15 to 20 in the upper atmosphere. At approximately 150 to 200 km altitude, the booster separates and falls back to Earth. The entire boost phase last roughly 3 to 4 minutes. Everyone talks about the Zumwalt's 80 launch cells.

[00:15:06]Nobody talks about the size of those cells. A standard Mark 41 cell, the one on every Burke, every allied destroyer from Japan to Norway, is 22 inches wide and holds 3,500 lb. The Mark 57 is 28 inches wide and holds 9,000 pounds. 68% more volume, 157% more weight capacity. That gap matters because every future weapon the Navy is developing, longer range anti-ship missiles, next generation strike weapons, is being designed around the Mark 41's constraints, built smaller than the physics would prefer. The Zumwalt has no such constraint, and the placement matters, too. The 80 cells sit around the outer edges of the hull. On a Burke, all 96 cells cluster amidships.

[00:16:00]If one cooks off, the blast travels inward toward the engines and the combat center. On the Zumwalt, a hit vents outboard into the ocean. The ship loses four cells. It keeps fighting. Damage tolerance by geometry, the quietest advantage on the ship. The Navy is debating whether to replace the Zumwalt's current radar with the AN/SPY-6, the same system going into the newest Burkes. The SPY-6 detects objects 30 times smaller than its predecessor.

[00:16:33]Same-size target, four times the range.

[00:16:36]There's even a shortcut. Unused radar arrays built for the canceled Constellation class frigate are sitting in a warehouse, compatible with the Zumwalt systems. Using them instead of building new ones could save hundreds of millions of dollars. But, the real debate isn't about the radar. It's about what this ship is supposed to be. With Zeus, the Zumwalt becomes a multi-mission flagship. Hypersonic strikes at 1,500 miles, air defense for the fleet, anti-submarine warfare, all from one hull. Without it, it's a pure strike platform that needs a Burke escort just to defend its own airspace.

[00:17:17]The Pentagon's problem, is it worth the cost for only three ships? The answer shapes what gets built next. Every hypersonic missile the Zumwalt fires is generating data the Pentagon cannot get any other way. How does the power system behave under a live CPS launch? What does igniting a solid fuel rocket 40 feet from sensitive electronics do to the ship's systems? How do crews adapt to the physical reality, not the simulation? This data is being collected because the Zumwalt is not the end of the story. The Trump class guided missile battleship, formerly in the Navy's 2026 shipbuilding plan, is projected at 35,000 tons, nuclear powered with CPS as its primary weapon.

[00:18:06]A ship that can hold an entire theater at risk from a single position. That ship cannot have a test program. It's too expensive to learn lessons on. The Zumwalt is where the lessons come from.

[00:18:18]Three ships, 21 billion dollars, 20 years of criticism, and they're writing the manual for the platform that actually wins the next war. That changes the cost calculus entirely. Right now, a Zumwalt on patrol east of Guam carries 12 missiles that reach Hainan Island, the South China Sea chain, coastal military facilities in Fujian province.

[00:18:43]The ship shows as 50 square meters on radar. The ocean around Guam is 2 million square miles. You cannot target what you cannot find, and a weapon you cannot find cannot be neutralized before it fires. That is deterrence. Not the detonation, the uncertainty, the knowledge that somewhere out there is a hull carrying 12 weapons you cannot stop, and you cannot locate it before it launches. This is what China faces in 2026 that it didn't face in 2016. Not just a new weapon, a new targeting problem with no current solution. The critics were right that the gun failed.

[00:19:24]Right that the fleet was cut. Right [music] that 20 years of waiting was too long. What they missed, if the Navy had built a conventional ship instead and CPS matured on the same timeline, where would the weapon go right now? There would be nothing to put it on. The 20 years of waiting were waste. The hull that survived the wait was not. Three ships, 21 billion dollars, 20 years of being wrong. One hull that cannot be detected. One power grid with room to grow. One magazine sized for missiles not yet built. The engineers weren't geniuses. They were patient.