Through-life defence diagnostics for long-life systems

Defence and other mission-critical platforms are bought to last. Air-defence systems, armoured vehicles, radars and aircraft routinely stay in front-line service for 20 to 40 years, and many are life-extended well beyond that. The capability is amortised over decades because re-procuring it is slow, expensive and politically fraught.

The technology underneath it has no such patience. The microcontrollers, the diagnostic PC, its operating system, the bus interface hardware and the authoring tool used to build the original test sequences all move on a commercial cycle measured in single-digit years. By the time a platform reaches mid-life, the components it was diagnosed with have often been out of production for longer than they were ever sold.

The result is a structural mismatch: a 35-year asset maintained with a 5-year tool chain. Bridging that gap deliberately, rather than improvising it at each crisis, is the whole discipline of through-life diagnostics.

The defence community has a name for the supply side of this problem: DMSMS, Diminishing Manufacturing Sources and Material Shortages. A part goes end-of-life, a foundry closes, a vendor is acquired, and a once-routine component becomes a sourcing project. DMSMS programmes exist precisely because unmanaged obsolescence quietly erodes fleet availability.

Diagnostics suffers a sharper version of the same disease, because three things obsolesce together. The hardware disappears. The tool chain — the authoring environment, the runtime, the licence server, the driver for a discontinued interface — stops being supported on any operating system you are allowed to deploy. And the tribal knowledge walks out the door: the engineers who knew why a particular test tolerance was set the way it was retire, and the rationale was never written down.

Lose any one of these and a fault that should take an hour to find takes a week. Lose all three and the platform becomes effectively undiagnosable in the field while still being expected to fight.

A platform you cannot diagnose is a platform you cannot confidently return to service. Through-life diagnostic capability therefore sits directly on the readiness line: mean time to repair, fleet availability and the ability to certify a system as fit before it goes back out.

It is also a sovereignty question. If your only route to diagnosing a national asset runs through a single vendor's proprietary tool — one that vendor can discontinue, re-licence or decline to support on a hardened operating system — then your operational independence depends on that vendor's commercial decisions. A stable, documented, tool-independent diagnostic capability that the operating nation can own and maintain is what keeps that dependency from becoming a vulnerability.

The defence against decades of churn is to anchor the diagnostic capability to open, slow-moving international standards rather than to any one product. UDS (ISO 14229) defines the diagnostic services themselves and is deliberately independent of vehicle manufacturer and of the physical layer, running over CAN, FlexRay or Ethernet alike. ODX (ISO 22901) describes the ECU diagnostic data in an OEM- and tool-independent XML format, so the knowledge of how to talk to an ECU lives in portable data, not in a binary that needs a 2009 runtime. DoIP (ISO 13400) carries that same UDS communication over IP and Ethernet, and was explicitly designed to provide a long-term-stable interface between the vehicle and external test equipment.

Standards alone are not enough, because the tools that consume them still age. The second half of the answer is architectural: an abstracted, portable tool chain in which the transport, the interface hardware and the user interface are decoupled. When a CAN interface or an operating system reaches end-of-life, you replace one abstracted layer — you do not re-author the diagnostic content. The ODX data, the UDS sequences and the engineering rationale survive the hardware that happened to run them this decade.

Through-life diagnostics is not a phase bolted on at handover; it is a thread that runs from first development to final disposal. During development the diagnostic model, the test sequences and their rationale are captured as documented, version-controlled artefacts. Through long in-service support that same baseline is maintained, audited and traced — with end-to-end traceability and a software bill of materials becoming as important to defence sustainment as it already is to safety-critical automotive work. At each mid-life upgrade the capability is migrated onto current hardware without losing its history.

This is the work Diadrom has done under its Autodefence banner since 1999, applying two decades of off-board diagnostics experience from automotive OEMs to defence platforms with decades-long service lives — including lifecycle diagnostics for systems such as the Saab RBS 70 NG air-defence system and FMV fleet vehicles. The point is not the tool; it is that the operator keeps a diagnosable, documented platform long after the original silicon, software and people have moved on.