USS Thresher disaster still matters

 This article was originally published in the Portsmouth Herald on April 3, 2018 and has photos of the 129 men that died aboard USS Thresher on April 10, 1963. You can see the original article and links to the photos by clicking this link. 

USS Thresher (SSN 593) (Photo from Naval Historical Center)

By Capt. Jim Bryant, USN (Ret.)

On April 10, 1963, the American nuclear submarine USS Thresher (SSN 593), the world’s most advanced hunter-killer submarine crushed at a depth of 2,400 feet killing all 129 onboard during a routine test dive.

     Incredibly, more than a half-century later, details of the Thresher disaster remain poorly understood. Its shattered hull resides at the bottom of 8,400 feet of water east of Cape Cod, Massachusetts.

     The underlying cause of the Thresher sinking 55 years ago and the collisions last summer involving the USS John S. McCain (DDG 56) and USS Fitzgerald (DDG 62) that killed 17 sailors was the failure to effectively integrate emerging technology into the training, procedures, planning and maintenance programs.

     The world situation of then and now are similar as America rushes to maintain naval superiority with new weapons systems like the Littoral Combat Ship, Ford class nuclear aircraft carrier, and the Virginia class nuclear submarine. Insufficient crew training, manning and inadequate operating procedures and shipboard maintenance continue to cause avoidable, recurring at-sea incidents.

     By 1963, Soviet submarines were a serious challenge to America’s national security. Thresher offered innovative improvements over earlier submarine designs. It was faster, quieter, dived deeper, and with advanced sonar and weapons systems, a significant threat to Soviet submarines.

     Built by Portsmouth Naval Shipyard, Thresher was commissioned Aug. 3, 1961, and spent the following year testing weapons and new equipment, measuring radiated sound, shock testing and conducting exercises with other submarines with outstanding results. The ultimate test was to challenge Soviet submarines would have to wait until after a lengthy maintenance period.

     After shock testing using close-aboard explosive charges in July 1962, Thresher returned to Portsmouth Naval Shipyard for a series of upgrades and repairs.

     On April 9, 1963, Thresher departed for sea trials, escorted by the submarine rescue vessel USS Skylark (ASR 20). After a shallow dive in the Gulf of Maine, the ships rendezvoused the following morning in deep water for a two-hour dive to Thresher’s deepest operating or test depth (1,300 feet, nearly twice as deep as previous classes).

     Thresher sank below its crush depth and imploded – raining its shattered hull, nuclear reactor and occupants onto the seabed below. The Navy’s investigation concluded that major flooding from ruptured piping in the engine room was the probable cause.

     The sounds of the Thresher’s death throes were recorded by sound surveillance system (SOSUS) underwater hydrophones located around the world tuned to pluck machinery sounds of submarines out of all the noise in the ocean. SOSUS was a highly secret system designed to track Soviet submarine movements at long ranges. SOSUS hydrophone array Fox was located only 30 nautical miles from the site of Thresher’s sinking.

     Bruce Rule was a top naval acoustic and SOSUS expert who analyzed Thresher’s death sounds and testified at the disaster inquiry. After leaving the Navy in September 1963, Rule spent his next 42 years as the lead acoustic analyst for the Office of Naval Intelligence. Though Rule’s testimony and findings remain classified, Rule recently revealed them in his book, “Why the USS Thresher (SSN 593) was lost,” which helps us understand this mystery beyond the obvious, that Thresher slowed, and uncorrectable negative buoyancy caused it to sink to crush depth.

     Rule is positive there was no flooding because the sounds of high pressure water hitting the inside of the submarine were not detected. Low pressure steams or sprays of seawater (excessive leakage) from multiple sources would be quiet to SOSUS, increase negative buoyancy, and cause concern to the crew trying to isolate them.

     Main coolant pumps (MCPs) moving heat from the reactor core to the steam generators were in fast speed and then stopped. Fast speed MCPs are required to reach maximum speed, but Thresher stayed at slow speed. Running MCPs in slow speed would have been more reliable.

     SOSUS detected compressed air blowing seawater from the main ballast tanks (MBTs) twice. The MBT blow system that should have surfaced Thresher failed because of poor design and the unauthorized installation of strainers with a metal backing plate with a small hole, or orifice, that severely restricted air flow. Ice formed on the strainers as high-pressure air instantly cooled when released into a lower pressure environment through this orifice and strainer. This ice intermittently blocked the compressed air to the MBTs and the strainers, and orifice plates, restricted air flow preventing removal of enough seawater from the MBTs to surface the ship.

     Slow speed MCPs would have been a more reliable lineup as they had an alternative source of power. Fast speed MCPs were run to use the tremendous power of the reactor plant to drive to the surface if there was a problem, but why did Thresher stay at slow speed? There is plausible, circumstantial evidence that Thresher’s stern planes used to control the angle of the ship for depth control likely became stuck in a dive position that required Thresher to stop to prevent a downward angle and depth excursion. Control surface failures were a fleet-wide concern on high-speed nuclear submarines.

     Rule’s analysis of Thresher’s recorded acoustic signature and underwater telephone communications with the escort ship Skylark provides the following timeline of Thresher’s loss.

     At 0853, Thresher descended from 1,000 to 1,300 feet (test depth). Possibly already negatively buoyant from not taking the time to adjust trim as the dive proceeded, increasing sea pressure on Thresher’s seawater systems boosted leakage.

     Somewhere between 0853 and 0909, Thresher experienced the stern plane problem, stopped to counter its effects, and started to sink.

     At 0909, SOSUS detected an electrical bus line-frequency instability, a symptom of an ongoing problem in the engine room, such as crew actions to stop excessive leakage from seawater piping.
Shortly after the electrical bus started to waiver, SOSUS detected the sounds of compressed air blowing into the MBTs. This means the primary means of going shallow, main propulsion was not usable. The blow stopped after 90 seconds due to ice blockage. This MBT blow did not remove enough seawater from the MBTs to reverse Thresher’s descent.

     The submarine’s fate was sealed at 0911 when SOSUS detected main coolant pumps stopping. This caused an automatic reactor shutdown (reactor scram) and by procedure, steam to be isolated to the main propulsion and power-generating turbines in the engine room. Even if the stern planes had become operational, shutting the steam stops prevented steam generated by decay and residual heat in the reactor from being used in the main propulsion turbines to drive to the surface. As Thresher continued to sink below test depth, SOSUS did not detect the sounds expected for the reactor being restarted.

     The Navy’s investigative report describes communications at about 0913 using the conflicting testimony of four witnesses on Skylark, “Experiencing minor difficulties. Have positive up angle. Am attempting to blow up. Will keep you informed.” The “experiencing minor difficulties” phrase is an enigma because Thresher had exceeded test depth, by as much as 600 feet, the reactor had scrammed, main propulsion was lost, the ineffective MBT blow failed to stop the downward acceleration, and the crew could hear the guttural sounds of the pressure hull compressing.

     As the ice blockage dissipated, Skylark and SOSUS detected another 30-second MBT blow before ice reformed and the blow stopped again, all while Thresher’s rate of descent increased.

     The garbled transmission at 0917 was interpreted to contain the phrase “900 North,” understood to mean 900 feet below test depth or a depth of 2,200 feet. This is reasonable given that Thresher was reporting depth relative to test depth in case a Soviet submarine was listening.

     SOSUS and Skylark detected hull collapse 0918.4 at a calculated depth of 2,400 feet with an energy pulse equal to the explosion of 22,500 pounds of TNT.

     The 129 men did not die in vain. Their loss resulted in immediate changes to how the Navy built, maintained and operated its nuclear fleet.

     Justifications for costly safety improvements are written in blood. In this case the Navy created the Submarine Safety (SUBSAFE) program that mandated the redesign of and strict quality control procedures for the manufacture, repair and testing of critical systems on submarines.
New SUBSAFE systems, like a separate emergency MBT blow and emergency, remote, hydraulic seawater hull valve closure systems. On Thresher, SUBSAFE would have prevented the unauthorized installation of the strainers and orifice plates. These critical systems include hull, seawater piping, high pressure air and stern plane. Until a submarine was SUBSAFE certified, it is restricted to operating at half its test depth.

     New reactor plant scram recovery procedures allowed residual and decay heat from the reactor to create steam for main propulsion to drive the ship to the surface and a faster restart of the reactor.
No SUBSAFE-certified submarines have been lost despite terrible accidents like the San Francisco (SSN 711) striking an underwater ridge in January 2005 at top speed that killed one sailor. The only other American nuclear submarine loss was Scorpion (SSN 589) in May 1968, which had not completed SUBSAFE-certification and suffered a main battery explosion before it sank and imploded.


Capt. Jim Bryant served on three Thresher-class submarines, including commanding USS Guardfish (SSN 612) from 1987 to 1990. He was assisted writing this article by a research and editing team of Harold Evans and Nicholas Wulfekuhle. Bryant lives in San Diego.