How to extend the life of a weapon that can't be retested

July 9 2026 3 283 2

The US is modernizing the sea-based and land-based components of its nuclear triad. The engineering challenge here is a strange one: not to build something new, but to test how far the old one can go.

A warhead on combat duty cannot be disassembled for inspection and reassembled, much less tested with a full-scale explosion: the US has not conducted full-scale nuclear tests since 1992. Ground Rocket Minuteman III It's been on duty since 1970, meaning it's older than most of the officers who service it. How long can you repair a product that can't be opened without consequences or tested in action? The United States is currently facing this dilemma simultaneously at sea and on land, and is responding to it in different ways.

"Hybrid" instead of a new product: Trident II D5LE2

Sea-launched ballistic missile Trident II D5 has been in service since 1990. Behind it is a direct ancestor Trident I C4 and the already completed resource extension program DLE (Life Extension), which is supposed to carry the original rocket into the 2040s. Now a second such program is being prepared - D₀LE2In mid-2025, she passed Milestone B, a milestone in the weapons program, after which the project officially enters the engineering and production development phase. The first missiles are promised the fleet in fiscal year 2039.

The word "new" is used here with caution. D₀LE2 — a conscious hybrid: engineers keep what works flawlessly and rework what has physically aged.

They retain: a three-stage solid-fuel power unit and the geometry of the body - a length of 13,42 m, a diameter of 2,11 m, a launch weight of about 59 tons.
They are changing: the electronic “stuffing” – avionics, the guidance system, all the circuitry assembled from components from the 1980s.

The logic is simple and engineeringly sound. The engine is the rocket's riskiest component: it can't be "tweaked" in the software, and thrust deviations can't be fixed with a patch. A proven design, honed over decades of launches, entails less technical risk than a new engine from scratch. The geometry is maintained for a different reason: the rocket must fit seamlessly into the silos of the new Columbia-class submarines, which are replacing the Ohio-class submarines. The hull is left untouched; everything else is tailored to it.

But electronics can't be maintained, no matter how hard one tries. The microchips, sensors, and cable networks of the late 1980s have reached their limits: they're no longer manufactured, and maintaining functionality indefinitely is impossible. Hence the new digital guidance system based on modern radiation-hardened components.

The result is a rocket whose engine operates according to a design that was first tested under Reagan, while the electronics are from the 2020s. The term "state-of-the-art" is a bit of a stretch for such a product. The precise specifications D₀LE2 — range, circular error probable — are not officially disclosed. Secondary estimates of specific accuracy should be taken as estimates, not as facts.

A pointing device that looks at the stars

A sea-launched missile is more difficult to achieve accuracy than a land-based one, and here's why. Minuteman It launches from a point with precisely known coordinates: the concrete shaft isn't moving. The submarine fires from underwater, from an area it only knows approximately, because over long hours underwater, the submarine's inertial navigation accumulates errors.

This is where astrocorrection comes from. After leaving the dense layers of the atmosphere, the rocket "looks" at the stars and uses them to refine its position, compensating for the very same uncertainty of the launch point. The inertial system guides the rocket, and the stars correct it. (This is the same principle that sailors have used for centuries to navigate, only now the rocket needs a celestial reference to hit its target thousands of kilometers away.) It is precisely this combination that has historically allowed SLBMs to approach the accuracy of land-based missiles launched from a fixed point.

В D₀LE2 This combination isn't being abandoned, but rather migrated to a new digital component base with increased noise immunity. The specific figures this provides haven't been officially announced. Only the engineering challenge is clear: preserving the principle while replacing aging hardware. This system guides warheads, which aren't getting any younger either.

W93/Mk7: A "new" warhead based on old physics

With the warhead, the paradox is even more obvious. W93/Mk7 They're calling it the first fundamentally new US nuclear warhead in nearly forty years. Yet, they're building it on proven nuclear designs, and that's not a slip of the tongue, but the essence of their approach.

The reason is the same: a ban on full-scale testing since 1992. There's simply nowhere to test the "new physics" of the charge today, since it's impossible to detonate a prototype. Therefore, the "new" in W93 — is a modern non-nuclear system: components, electronics, and security systems. The emphasis is on resistance to fire, impact, and cyberattacks, ensuring that the warhead detonates only when intended and fails in all other circumstances. The development is being carried out by the Los Alamos National Laboratory under the auspices of the National Nuclear Security Administration (NNSA). It is specifically stated that the program does not increase the size of the arsenal; it is being upgraded in quality.

Updating such a product is expensive: the program W93 The cost is estimated at over $15 billion. This figure is interesting not in itself, but as a measure of the task, showing how much it will cost to bring a charge that cannot be directly tested to a modern standard.

Specific characteristics W93 — the yield, weight, and number of warheads on the missile are all closed, and this is normal for a munition of this class. It is known that the warhead is designed for interfaces D₀LE2, and a related British warhead is being considered for prospective British submarines of the class Dreadnought, as part of long-standing US-UK cooperation on naval nuclear systems.

Sentinel: When it's more expensive to fix than to build

On land, the same dilemma was resolved differently. While at sea, the Air Force chose to maintain its proven force structure, on land, the Air Force decided it had reached its limits and was building from scratch. Minuteman III It has been on duty since 1970, making it the oldest element of the American strategic forces, and further modernization was deemed impractical. It is being replaced by Sentinel (LGM‑35A) — not just a missile, but an entire complex: the missiles themselves, the silos, the command posts, the communications links. The plan is to deploy approximately 400 missiles in silos across five states. The first flight is scheduled for 2027.

The difference in approach is understandable. With a sea-based missile, the platform—the submarine—is upgraded separately and radically (Columbia instead of Ohio), and the missile itself is integrated into the new platform. With a land-based missile, the platform consists of silos, command posts, and cables laid half a century ago. Repairing them individually at some point is more expensive than building a new system all at once.

Curious that Sentinel again a mine, like its ancestors MinutemanIn the 1970s and 1980s, the United States was seriously considering mobile basing: heavy MX (Peacekeeper) they wanted to move between dozens of shelters, but light Midgetman transported by road or rail. Both schemes ran into local resistance and cost: turning entire districts into permanent military zones proved unacceptable and expensive. Ultimately, MX placed in ready-made mines Minuteman, and mobile projects were shut down. Sentinel inherits this choice, a stationary shaft as a compromise, tested half a century ago.

From Minuteman The new missile is distinguished by its payload. Northrop Grumman has released the first photographs of its "integrated forward module," the nose cone section that takes control after the booster stages have fired. It houses the guidance system, flight control unit, thrusters, and warhead. While the booster stages are burning, they merely accelerate the missile; the precise placement of the warhead on target is the responsibility of this module, its sensors, and computers, once outside the atmosphere.

Thunder Test: Why Rockets Are Deafened Before Launch

Before a rocket can fly, it must survive its own launch. Photos of the forward module Sentinel Made in Redondo Beach, California, after acoustic testing.

When launched from a silo, the rocket enters an environment with a monstrous acoustic load: the roar of the engine in a confined concrete chamber creates vibrations capable of destroying the structure before the rocket even leaves the silo. Therefore, the rig reproduces a sound environment as close as possible to a real launch, testing not the flight itself, but survival in the first second. Specialists from specialized nuclear laboratories, the unit responsible for maintaining the arsenal's reliability under the auspices of the NNSA, also participated in the tests.

Test rigs are a separate and easily overlooked part of the modernization program: they're often forgotten amidst all the talk of the missiles themselves. The Strategic Weapons Systems Ashore ground test complex became operational in November 2025, construction began on a new engineering test site, and around three dozen infrastructure projects are planned through 2032. They're updating not only the missiles themselves, but also the things they're tested on.

Both programs answer the same question about resource limits, but the platforms have worn out differently: in some places it’s cheaper to keep what’s been tested, and in others it’s cheaper to build anew. D₀LE2 should go on submarines in the late 2030s, taking up duty next to charges whose physics were confirmed even before 1992. So the electronic “brains” and motor “muscles” of a single missile will differ in age by a good half a century, and for strategic weapons This has been the working norm for a long time.