Author’s Note: This article was prepared with assistance from Steve Walsh, Nick Wulfekuhle, and Captain Fred Best, U.S. Navy Reserve (Retired)
The U.S. nuclear submarine Thresher imploded during a deep dive test 220 miles east of Cape Cod on 10 April 1963, after a nine-month Post Shakedown Availability (PSA) in Portsmouth Naval Shipyard. In response to a Freedom of Information Act lawsuit our team filed, the Navy has released all 1,700 pages of testimony rendered before the Thresher Court of Inquiry (Court) and more than 100 exhibits, with some redactions for public review. What follows is a summary and analysis we have assembled, based on that information.
Apparent Cause Linked to Increased Operating Depth
Our volunteer research team’s consensus after studying the documents is the apparent cause of this disaster was not appreciating the inherent dangers of nearly doubling the Thresher class’s test depth (TD—maximum operating depth) over that of previous submarine designs. Respected open-source literature, such as from authors Norman Polmar and Norman Friedman, state the Thresher’s TD was 1,300 feet, nearly twice the 700-foot depth for other post–World War II Navy sub designs. Our team believes the inherent risks of a deeper-diving submarine were not appreciated in the design process, such that specifications, fabrication, repair, quality assurance, and operational procedures were not improved fast enough or applied with enough rigor to prevent the disaster.
A synergy between technical and human factors contributed to this loss.
No Backup to Nuclear Propulsion
Contract design specifications limited the Thresher’s recovery capabilities from negative buoyancy at TD, such as using its nuclear propulsion to drive the submarine to the surface in case of flooding.1 Common events that could prevent a recovery by propulsion alone were either an automatic reactor emergency shutdown (scram) or failure of the stern plane control surfaces in a dive position (jam dive). The Thresher’s scram procedure secured steam to the engine room, stopping the main engines and steam-driven ship’s service turbine generators (SSTGs), which produced most of the alternating current (AC) electricity. A jam dive requires the submarine to stop—or it will continue going deeper with ever-increasing down angle.2
Attempting to pump seawater from variable ballast tanks and bilges to sea was useless for damage control due to the trim and drain pumps’ extremely low flow rate at TD. The Thresher’s main ballast tank (MBT) blow system could compensate only for relatively minor flooding. Even then, the Thresher’s system did not provide the expected buoyancy because the MBT blow was intermittent.3
Before the Thresher’s loss, MBT blow systems were tested only for a few seconds at shallow depth. Realizing this, the Court ordered an extended pierside MBT blow test on the Thresher’s sister submarine, the USS Tinosa (SSN-606). The test discovered strainers and orifice plates in the inlet piping to high-pressure air-reducing valves. The strainers were found to be crushed, and ice formed on them alternately blocking and reducing MBT blow airflow caused by the Joule-Thompson effect, as air pressure dropped through these restrictions. These strainers and orifices were installed to prevent dirt from scoring valve seats and causing reducer failure; an issue so common that the Thresher verified the strainers and orifice plates were in place prior to PSA sea trials.4
Deep-Dive Test Procedure Was Deficient
The Thresher was scheduled to perform two deep-dive tests: the first (which proved fatal) one for two hours, to be followed by a more controlled six-hour deep dive.
No Court witness explained why the dangerously rapid two-hour hydrostatic test of the pressure hull, seawater systems, and external cabling was not integrated into the controlled deep dive, which included equipment tests and stops at incremental depths while proceeding to test depth. This plan was produced by the new commanding and executive officers, who were not experienced in deep-diving submarines, and it was not reviewed by senior leaders in their chain of command.5
At an April 1961 meeting held only hours before departing for sea trials after new construction, a decision was made to run fast-speed main coolant pumps (MCPs) during the deep-dive test for immediate acceleration to maximum or flank speed, if needed, to get shallow quickly. The Director of the Naval Nuclear Propulsion Program, then–Vice Admiral Hyman Rickover attended this meeting and agreed to run fast speed MCPs.6 He then rode the Thresher on this sea trial. At shallower depths, the MBT blow system and pumping to sea would be more effective, because of decreased sea pressure. This policy was continued for the fatal deep dive after the PSA.
Rickover testified that all aspects of the 1961 deep-dive procedure were unsafe and he had it modified to take more time to do it safely and test equipment. He said the original deep-dive procedure lacked planning and did not take account of the much deeper TD.7
Court testimony revealed that while slow-speed MCPs limited ship’s speed, slow-speed pump operation demanded less AC power and was more reliable for deep-dive tests. However, that information was not provided to the submarine force until the Court recommended it, after the Thresher’s loss.
Fast-speed MCPs demanded a large amount of AC power that required the ship’s service steam generators, while slow-speed MCPs could be run on much less power coming from SSTGs or even the ship’s direct-current battery, converted to AC by ship’s motor-generator sets. If the SSTGs were to stop, fast-speed MCPs would stop, too, causing the reactor to scram. But slow-speed MCPs could keep running on battery power, preventing a scram and maintaining propulsion. The extra speed enabled by fast-speed MCPs gave little recovery advantage.
When the Thresher’s SSTGs and fast-speed MCPs stopped, the reactor scrammed, and the ship lost propulsion and AC power from the SSTGs. The Thresher was negatively buoyant likely because of some combination of leakage and being trimmed too heavy. This negative buoyancy could not be compensated for by either the MBT blow or by pumping to sea. This resulted in an unrecoverable situation. The Thresher sank and eventually imploded, reportedly 450 feet below its specified collapse depth, which is 150 percent of test depth.
Emergency Propulsion Procedure Existed but Was Not Approved for Use
Vice Admiral Rickover testified that Naval Reactors had developed a procedure for emergency propulsion after a scram that allowed steam produced by the residual and decay heat of fission products in the reactor core to be used for a few minutes of emergency propulsion. Rickover acknowledged there was enough residual heat in the reactor after a scram to generate sufficient steam to get the submarine to the surface.10 This procedure was reportedly in effect by ship’s instruction on at least one shallower-diving nuclear-powered submarine.11 But it was not approved for general application until after the Thresher’s loss.
Fast Cruise Was Not Effective
A “fast cruise” prepares the crew for sea trials. The ship remains tied up—made fast—to the pier and simulates being at sea by running drills and shipboard evolutions. The Thresher’s post-PSA fast cruise was not effective, because there were shipyard workers on board and the crew had to repair and write work requests for many equipment failures. The fast cruise was stopped after failure of key equipment, e.g., high-pressure air-reducing, hydraulic hull, and MBT blow valves.12 After these issues were repaired, the fast cruise resumed, and all scheduled drills and evolutions were completed. A drill simulating flooding in the engine room took 20 minutes to isolate the leak, far too long when deep. This shows the crew was not given adequate opportunity to become familiar with the many piping system changes made to The Thresher during the PSA. The Thresher’s chief engineer’s request for more engineering drills was denied.13
Inadequate Testing of Critical Piping Joints
Silver-brazed piping joints were used in critical submarine piping systems for decades. Despite this long-term use, there was a documented history of joint failures in submarine seawater systems that had passed hydrostatic pressure tests.
In early 1962, an ultrasonic test was developed that enabled identification of bad silver-brazed joints. The Thresher started PSA in August 1962, and the shipyard tested all the new silver-brazed joints that were worked during the PSA. All passed.
A testing program for unworked silver-brazed joints (that is, joints that were not touched during the PSA) was ordered, but only 145 (of more than 3,000) were tested to avoid further delays.14 Many senior officers at the Bureau of Ships in Washington, D.C., and at Portsmouth Naval Shipyard—including the admiral commanding the shipyard and Vice Admiral Lawson Ramage, a renowned World War II submarine commander and Medal of Honor recipient, then serving as Deputy Commander of Submarine Forces, Atlantic Fleet—believed that if an original silver-brazed seawater system joint on The Thresher survived more than 40 trips to test depth, the shock testing, and the PSA hydrostatic test, it was a sound joint. Further, it was noted that ultrasonic inspection of such joints would delay The Thresher completing its availability and its subsequent fleet tasking.
Hydrostatic test records for the Thresher’s seawater systems were so poor, it was difficult for the Court to determine which joints had been tested.
Key Officers with Limited Experience May Have Contributed to the Loss
With the rapidly growing submarine fleet, limited officer experience on nuclear submarines was a critical problem. The Navy’s goal was to have a minimum of six nuclear-trained officers that were either submarine-qualified or had a year on board. While The Thresher had seven officers that met this requirement, both the reactor controls and electrical officers missed sea trials because of family emergencies. The Thresher had changed commanding and executive officers three months before sea trials, and the new pair no operational experience on deep-diving, high-performance submarines.15
The Thresher’s weapons and operations department’s division officers’ jobs were split between two non-submarine qualified junior grade lieutenants, both on board for a year, enabling the Thresher to meet the manning requirements—but with only four months sea time. That means that, without experienced officers, the executive officer also served as the navigator/operations officer (the Thresher’s previous executive officer acknowledged this to the Court) and likely the weapons officer. Four non-nuclear-trained officers had just reported on board. Since commanding and executive officers do not stand shipboard watches, the Thresher had only five officers for propulsion-plant watches in port and at sea. The demands for watchstanding and supervision of the nuclear-trained enlisted men operating and maintaining the nuclear propulsion plant left little time for standing diving officer and officer of the deck watches or learning about the Thresher’s weapons, sonar, navigation, and operations.16
Vice Admiral Ramage testified that the nuclear submarine fleet engineering department heads lacked the maturity, judgment, and operational experience to be executive officers and did not have adequate knowledge of weapons, sonar, and submarine operations.17
Vice Admiral William Smedberg III, Chief of Naval Personnel, testified that Rickover’s failure to select well-qualified diesel submarine commanding and executive officers for the nuclear-power program caused this shortage of officers, but he respected Rickover’s need to be selective because of reactor safety concerns.18
***
The testimony before the Thresher Court of Inquiry laid the foundation for the establishment of the Navy’s Submarine Safety (SubSafe) Program that has prevented further disasters. SubSafe is a quality assurance program that dictates material and workmanship standards and procedures for all systems produced for use on board submarines that are exposed to sea pressure or are critical to recovery from flooding.
1. Commander-in-Chief, U.S. Atlantic Fleet, “Record of Proceedings of a COURT OF INQUIRY Convened at U.S. Naval Submarine Base, New London, Connecticut and Portsmouth Naval Shipyard, Portsmouth, New Hampshire,” n.d. [submitted to the U.S. Congress 24 June 1963] https://www.secnav.navy.mil/foia/readingroom/SitePages/Home.aspx under Thresher, hereafter cited as Thresher COI], Release 1, 34, Release 3, 704, 708; Release 4, part 2, 1026–29; Release 8, Findings of Fact, 46–50 1688, 1689
2. Thresher COI, Release 3, 702, Release 8, Opinion 45 1710, 1711.
3. Thresher COI, Release 2, 469, Release 7, 1239–44; Release 8, Opinion 8, 1703.
4. Thresher COI, Release 7, 1239,1240, 1277–79, 1373.
5. Thresher COI, Release 3, 708, Release 8 1527–28, 1553.
6. Thresher COI, Release 8, 1524, Opinion 1 1702, Release 3 549–50, 704.
7. Thresher COI, Release 3, 708.
8. Thresher COI, Release 2, 324, Release 8 1533–34.
9. Thresher COI, Release 3, 702 Release 8 Opinion 45 1710, 1711.
10. Thresher COI, Release 3, 702.
11. James B. Bryant, “Declassify the Thresher Data,” Proceedings, U.S. Naval Institute, July 2018 edition), 65. Personal Communications with CAPT James Collins, USN (Ret.).
12. Thresher COI, Release 10, Exhibit 96.
13. Thresher COI, Release 1, 167–79, 185–88.
14. Thresher COI, Release 4, part 2, 1073–75, Release 7, 1192, 1200-02, 1206, 1338, 1359, 1364, 1373, 1386, 1394, 1395, 1387; Release 8, 1548; Release 10, Exhibit 114, 115.
15. Thresher COI, Release 8, 1568.
16. Thresher COI, Release 8, 1605–9, 1615–16.
17. Thresher COI, Release 8, 1551.
18. Thresher COI, Release 8, 1613–15.
