Towed array sonar is a sinuous cable of hydrophones trailed astern of warships and submarines, a technology that has taken ASW from speculative hunting at short range to a long-distance pursuit. Here we look at this key sensor and its development. (From: Naval Lookout.)
The acoustic battlefield
Understanding towed array sonar demands a grasp of the ocean’s layered complexity, especially in the High North’s harsh environs. The sea stratifies into zones dictated by temperature, salinity, and pressure, a vertical mosaic that both conceals and reveals submerged threats, refracting sound like light through a prism. The surface mixed layer, churned by winds and waves, facilitates rapid sound travel but scatters signals amid bubbles and turbulence. Below lies the thermocline, a steep temperature drop that bends acoustic paths downward, separating the warmer, mixed surface waters from the colder, more stable deep waters. The thermocline varies considerably depending on the season and different parts of the ocean. Navies measure this directly using expendable bathythermographs (XBTs) or conductivity–temperature–depth (CTD) sensors before beginning ASW operations, since even small changes in the thermocline’s depth can dramatically alter detection ranges.
Below this, the Deep Scattering Layer (DSL) at about 300–500 metres is teeming with marine life, diffuses echoes, especially at dusk and dawn hours. The Deep Sound Channel (DSC), or SOFAR channel (Sound Fixing and Ranging), typically lies between 600 and 1,200 metres in mid-latitudes, shallower in colder polar waters, and much deeper in the tropics. This layer can funnel low-frequency sounds across thousands of kilometres with little loss and is critical in long-range submarine detection.
These layers make ASW a multidimensional challenge as submarines can exploit the thermocline by changing depth to avoid detection. Hull-mounted sonars on warships struggle with self-generated noise, and the propagation losses in layered waters can limit range to a few miles. Towed arrays avoid these problems by extending sensors far astern, free from propeller turbulence, and positioning them below the thermocline for superior clarity. They can be used passively to detect radiated signatures from the target, such as machinery noise or cavitation. As modern submarines are increasingly minimising these emissions, active sonar may be required to achieve either initial detection as well as more precisely locate a target.
Electronic string
Fundamentally, towed array sonar comprises a flexible tube, often more than a kilometre in length, housing hydrophones, piezoelectric devices converting pressure fluctuations into electrical impulses. Usually deployed from surface ships at between 100-500 metres deep via buoyant cables, the array aligns linearly, its hydrophone elements spaced apart to enable beamforming, which allows constructive interference to pinpoint signal directions accurately.
Variants include thin-line arrays (under 50mm diameter) for passive, long-range (over 2 km) low-frequency (10–100 Hz) detection of submarine vibrations, and fat-line types (up to 90mm). Early arrays, like the 1917 ‘Electric Eel’, a rudimentary hydrophone chain towed by destroyers, relied on manual reeling and electrical conductors prone to twisting. By the 1950s, coaxial cables and fibre optics supplanted wires, enabling data telemetry without electromagnetic interference. Today, synthetic fibres like Dyneema ensure tensile strength exceeding 10 tonnes, while embedded neutrally buoyant modules prevent sagging.
Signal processing has advanced from analogue to digital, employing adaptive filters to remove background noise such as marine life, seismic or non-submarine man-made activity, achieving localisations within 1 km at 50 nautical miles. Sound speed in seawater (circa 1,500 m/s) varies with depth; real-time refraction algorithms can be applied to adjust for variable sound speeds in different layers of the water column.
On the surface, vessels like Type 23 frigates can safely deploy the TA at speeds up to 20 knots, but when more than a kilometre of ’string’ is being streamed astern, this severely limits the ship’s manoeuvring envelope. A submarine may be detected at a long distance, and the Merlin helicopter can then be vectored in to precisely localise and potentially neutralise threats.
Surface ship arrays
Developed initially for submarines, Towed Arrays really began to mature in the 1980s, with a big impact on the effectiveness of ASW frigates. The RN’s first surface ship prototype TA (Sonar 2031X) was tested aboard the trials ship, HMS Lowestoft. The RN’s first operational TA was the Sonar 2031I, consisting of a 1800-metre, 5-octave hydrophone array fitted to 4 Leader-class frigates.
This was followed by the much-improved Sonar 2031(Z), which benefited from computer miniaturisation. ‘Waterfall’ displays generated by applying a Fourier transform to incoming acoustic signals were presented on screens in the ops room that provided a clearer visual representation of contacts. The handling system was also redesigned, with a wider winch, low enough in height to be placed in the shelter of the quarterdeck under the flight deck, clear of flight operations. The system was fitted to all the Batch 2 and 3 Type 22 frigates and initially to the Type 23 frigates. 2031 was Critical Angle TA, where the depth of the array is determined by the speed of the towing vessel, which lacked precision.
The successor Sonar 2087 TA was first fitted to HMS Westminster and went to sea in 2005, eventually fitted to 7 other Type 23s and remains the RN’s primary ASW sensor. 2087 is the British variant of the Thales (CAPTAS-4) UMS 4249 Combined Active and Passive Towed Array Sonar series. The active element of the 2087 has 32 projectors that emit low-frequency (0.5–5 kHz) pings that can pierce thermoclines with detection ranges of over 100km in some conditions. It is a ‘depressed’ array, which uses a hydrodynamic body on the far end to precisely control depth. This system is probably amongst the best TAs in the world and according to open source information, in the right conditions, can detect submarines at a range of more than 150km.
As part of the RN’s long-running SPEARHEAD programme to maintain its ASW edge, 2087 underwent a technical refresh in 2017, adding a suite of passive sonar algorithms and Human-Computer Interface (HCI) features originally developed for the Sonar 2076 fitted to its submarines. Both 2076 and 2087 use a common open architecture and have been continually improved through iterative software and processor upgrades. Thales was awarded another £110M upgrade contract in September 2022 – the Design Authority Capability Insertion Project (S2087 DA-CIP) runs until 2026 and includes further improvements. First sea trials were concluded in January 2023, and the upgrades will be gradually rolled out across the fleet.
The TA obviously has major advantages, but a good ASW combatant still needs a hull-mounted sonar. Although detection ranges are considerably less, the fixed array provides a baseline 24/7 capability without the need to deploy the cumbersome TA. The hull-mounted array adds several other important capabilities, such as mine and obstacle avoidance, underwater communications with friendly submarines, automatic torpedo detection, classification/localisation and marine mammal detection. Type 23 frigates have now been fitted with bow-mounted Sonar 2150 made by Ultra, replacing the Thales Sonar 2050 originally fitted. The first ship to receive the new set, HMS Portland, achieved Initial operating capability. S2150 will also be fitted to the Type 26 frigates and is an integral aspect of the combat system, which combines data from both the TA and bow-mounted sonar.
As almost its sole means of operating underwater, submarine sonars are considerably bigger and more complex than those of surface ships. Towed arrays are just one part of this complex integrated suite of sensors. For example, it is believed the RN’s S2076 system has at least 13,000 separate hydrophone elements and the conformal bow array alone weighs around 25 tonnes. The main flank arrays are made up of 20 5m x 1m panels.
The idea of towing a long acoustic array behind a submarine was first explored in the US during the late 1950s by adapting oil-industry seismic sensors for naval use. By the mid-1960s, American SSBNs were routinely fitted with arrays. Retrieving an array is riskier for a submarine than for surface ships as eddies can snag the line. Submarines may have to utilise tenders to assist with clipping on or removing the arrays just outside port at the start or end of a patrol. Once streamed, the boat benefits from all the advantages of a TA and, importantly, it helps the boat avoid being ambushed from the stern, where propulsor noise and turbulence create a blind spot in coverage.
The RN began to adopt TAs about a decade after the US pioneered the technology. The Resolution-class SSBNs were the first British submarines fitted with a towed array, designated Sonar 2023 and supplied by the US. British industry rapidly followed with its own designs.
Sonar 2024, was the result of intensive UK research during the 1970s and equipped the Swiftsure-class attack submarines, establishing the foundation of a highly regarded sovereign sonar development capability. By the early 2000s, Sonar 2074 represented a major advance, combining multiple frequency bands and an integrated suite of bow, flank and towed sensors. Designed for both the Swiftsure and Trafalgar classes, it introduced reelable and clip-on arrays.
Work on a successor to 2074 began in the aftermath of the Walker spy ring revelations, which had exposed key details of Western acoustic performance to the Soviet Union. Development was slowed by industrial restructuring and the post-Cold War ‘peace dividend’, but 2076 was fitted to the later Trafalgar-class boats and the subsequent Astute-class. Reaching full operational capability in 2009, an updated version remains in service today and is very highly regarded.
The Astute-class can deploy both clip-on and reelable towed arrays. When needed, the internally stowed array can be deployed in a matter of minutes through a feed pipe supported by an A-frame above the hull, keeping it clear of the propulsor. A water-flushing system initiates deployment by gently driving the forward end of the array astern. After approximately 100 metres have been streamed, the resulting hydrodynamic drag from the trailing length provides enough force to draw out the rest of the array rapidly. This method avoids the use of mechanical traction equipment such as capstans or linear actuators, reduces the risk of damage to the array, and operates with minimal acoustic signature.
The towed array is connected to the submarine via a 600m towing cable, roughly 50mm in diameter, ensuring the array streams well behind the boat and remains clear of its own acoustic emissions. The vital hydrophone section is about 120m in length, giving an aperture length greater than the submarine itself. To prevent unwanted vibration from distorting the sonar data, vibration-isolating modules are located at both ends of the hydrophone section. These decouple the sensitive hydrophones from any movement in the towing cable or drogue as they pass through the water. A drogue fitted to the end of the array helps keep it straight during deployment. Both the clip-on and internal TAs have cutters to detach the array in case of emergency.
The full array length exceeds one kilometre, and safely streaming and recovering it demands skilled boat handling. Turning is limited to a maximum of about 1.5º per sec and the array must remain straight to provide reliable data, manoeuvres cause it to bend and degrade performance until it resettles. Best results are obtained at low speeds, fast enough to keep it taut, but not so fast that flow noise masks target signals. Performance tends to degrade above about 12 knots, although the array can remain safely deployed up to about 25 knots if required. As a contingency, a secondary clip-on array is also carried, stored in a trough on the casing.
Thinlines
Fat line towed arrays continue to offer a formidable advantage in anti-submarine warfare. Their sensitivity, range, and ability to host low-frequency active sonar make them the backbone of heavyweight ASW platforms. But their size, power demands, and handling requirements mean they remain the preserve of large, crewed vessels with substantial onboard processing and support infrastructure. These systems will remain relevant for decades ahead, but the ASW ‘find’ function is increasingly dispersed across a wider mix of platforms, including USVs and XLUUVs and there’s growing demand for smaller, more agile TAs that can bring similar capabilities to a broader range of vessels.
The SEA Krait Sense is a UK-developed solution built around a (20mm) thin-line towed array, suitable not just for ASW but for broader seabed and infrastructure monitoring missions. Krait can be containerised and deployed from both ships and uncrewed platforms. Each array module is typically 50 metres long, with either centrally-digitised (CD) or locally-digitised (LD) configurations, depending on integration requirements. The system uses the same hydrophones and structural components in both variants, enabling flexible scaling by adding more modules if required. It can be delivered either as a standalone sensor (Krait Array) or as Krait Sense, a complete end-to-end capability with processing, compact winch, tow cable, and control interface.
Krait Sense has already been integrated with a wide range of uncrewed surface and sub-surface vessels and has seen operational use with international navies. Its appeal lies in the low drag and power demands of the array, combined with a high-performance passive sonar processing suite running on a GPU-based console. The same software stack works across crewed and uncrewed configurations, supporting contact detection, target tracking, narrowband and broadband visualisation and basic classification.
As the requiremnet for deployable, autonomous ASW systems grows, the demand for thinline arrays is rapidly increasing.
