On 10 April 20 13 it will be the 50th Anniversary of the loss of the USS Thresher.
It is now possible with recently declassified documentation and other supporting data, all in the public domain, to provide this scientific analysis.
A version of this posting has been sent in letter form to ADM K. H. Donald, Director Naval Nuclear Propulsion and Captain L. Bryant Fuller III, Shipyard Commander Portsmouth Naval Shipyard.
Why the USS THRESHER (SSN-593) Was Lost?
In April 1963, the author, then the Analysis Officer at the Sound Surveillance System (SOSUS) Evaluation Center in Norfolk, VA, had just completed the U.S. Nuclear Submarine Acoustic Data Handbook, a comprehensive summary of the low frequency, narrowband acoustic signature characteristics of all U.S. nuclear submarines then operational, including the USS THRESHER (SSN-593). That document was based on analysis of more than 700 acoustic detection events of all nuclear submarines.
Acting in that capacity, and with those technical qualifications, the author, subsequently the lead acoustic analyst at the Office of Naval Intelligence for 42 years and author of WHY THE USS SCORPION (SSN-589) WAS LOST, reviewed on page 151 of the WINTER 2012 issue of THE SUBMARINE REVIEW (WI12TSR), called in acoustic data from all Atlantic SOSUS stations to determine if the loss of the USS THRESHER had been acoustically detected. That analysis identified a signal of extreme amplitude produced by the collapse of the THRESHER pressure hull at 09:18:24R on 10 April 1963. The derived position - a four nautical mile (nm) by eight nm ellipse with a major axis oriented 040-220 - provided the basis for the successful search for the THRESHER wreckage.
That analysis also determined the THRESHER non-vital electrical bus, after two minutes of line-frequency instability, failed for unknown reasons at 0911R while the nuclear reactor coolant pumps (RCPs) were in FAST. (Note: the SSTGs were not acoustically detected; the instability of the non-vital bus was derived from measured instability in the RCP rotational-rates. The non-vital bus line frequency was determined by correcting for the 2.5 percent slip of the RCP drive motors. Also note that the signal strength of the RCP sources at 0911R, at a detection range of about 30 nm, indicated that had the RCPs been shifted to SLOW at 0911R, they should still have been acoustically detected - but no such detection occurred.) The electrical load thrown on the vital bus at 0911R by the failure of the non-vital bus with the RCPs in FAST exceeded the capabilities of the vital bus; the RCPs (initially detected at 0845R in FAST as THRESHER, according to the deep-dive OP-PLAN, was approaching a depth of 1000 feet) went off-line and the reactor scrammed at 0911R. The coincident detection of an acoustic signature component at a fixed ratio relative to the RCP source unique to S5W RCPs confirmed the SOSUS detection was THRESHER. There were no acoustic detections by SOSUS of any THRESHER main propulsion sources as would have been probable had speeds above about 14 knots been employed. The author provided the above assessments of RCP operating mode and loss of signal in testimony before the THRESHER Court of Inquiry (COI) on 18 April 1963 with supporting testimonies by BUSHIPS Code 345 and the David Taylor Naval Ships Research and Development Center personnel, respectively, CAPT Patrick Leahy and Mr. Edwin Savasten.
At 0913R, two minutes AFTER - repeat, AFTER - the reactor scrammed, THRESHER informed her escort ship, the USS SKYLARK (ASR-20), by underwater telephone, that she was (quote) experiencing MINOR difficulty.(end quote) The COI concluded the rupture of a silver-brazed, sea-connected pipe had produced flooding in the engine room that shorted-out electrical systems causing the scram, an assessment still accepted at the highest levels within the Navy; however, that assessment requires that flooding at test-depth that resulted in a reactor scram and a loss of propulsion be described by THRESHER as a (quote) minor difficulty. (end quote)
At 0917R, SKYLARK received a final communication from THRESHER that contained the number 900. That number is assessed to have been the depth in feet (referenced to test-depth as required by the deep-drive OP-PLAN security directive) by which THRESHER had exceeded her test-depth of 1300 feet, or 2200 feet. With an estimated average sink-rate of about 130 feet per minute, the THRESHER pressure-hull collapsed at 09:18:24R at a depth of about 2400 feet, more than 400 feet below her estimated collapse depth. Independent confirmation of that assessment has been provided by a post COI testimony analysis of the collapse event acoustic bubble-pulse frequency which indicated a depth between 2000 and 2400 feet. (See "Technical Comment" page 134 of the WI12TSR.) The author has no information on the change in displacement produced by hull compression at great depth and the extent to which that decrease could have accelerated the THRESHER sink-rate.
There was not in 1963 - nor is there now - any evidence in the specific case of the loss of THRESHER to support the COI conclusion that on 10 April there was a rupture of a silver-brazed, sea-connected pipe that caused a reactor scram. The occurrence of silver-brazing problems earlier with THRESHER, and with other submarine hulls, is NOT conclusive evidence that it occurred during the 10 April deep-dive, especially since THRESHER's 0913R transmission to SKYLARK makes no mention of flooding and because the results of analysis of the SOSUS acoustic data are consistent with failure of the non-vital electrical bus which resulted in a reactor scram at test-depth because the RCPs were operating in FAST. Unable to deballast because of a subsequently confirmed ice-formation condition in the high-pressure air lines, THRESHER sank to collapse at extreme depth without any prior flooding. Both the pressure hull and all sea-connected systems survived well beyond design specifications. As discussed in THE DEATH OF THE USS THRESHER by Norman Polmar, THRESHER had made some 40 dives to test depth prior to April 1963.
To repeat, there was not - as maintained on page 122 of the WI12TSR - any (quote) failure of a silver-brazed fitting in the engine room, with immediate flooding, and subsequent emergency shutdown of the nuclear reactor (scram due to spray on the engine room affecting electrical control panels) (end quote); hence, it is wrongly asserted, also on page 122, that Portsmouth Naval Shipyard personnel were responsible for the loss of the USS THRESHER because of the failure of a silver-brazed fitting.
The acoustic bubble-pulse data indicate the THRESHER pressure hull and all internal compartments were completely destroyed in less than one-tenth of a second (100 milliseconds), significantly less than the minimum time required for human perception of any event: 50 milliseconds for retina integration plus 100 milliseconds for cognitive integration. Measurements made during the lowering and recovery of an instrumented diesel submarine to collapse depth are consistent with the conclusion that the water-ram produced by the initial breaching of the THRESHER pressure hull at 2400 feet traversed the diameter of the pressure hull in about 0.005 seconds (five milliseconds), a velocity of about 4000 mph. That force would have torn the pressure hull longitudinally and vertically as verified by imagery of the THRESHER wreckage. Even allowing for differences in pressure hull design, the extent of the damage to THRESHER, compared to the USS SCORPION (SSN-589), which collapsed at 1530 feet, indicates THRESHER collapsed at significantly greater depth.
The above discussed information on the failure of the non-vital bus, the RCP operating mode and implications for a reactor scram is provided in COI documents available in the public domain.
As of March 2007, the Office of Naval Intelligence still held a photo-copy of the SOSUS paper display (LOFARgram) upon which the above assessments are based. There were no SOSUS recordings of the THRESHER event. The original SOSUS LOFARgram data from all Atlantic stations - except Barbados, which was bathymetrically blocked - were destroyed by SOSUS Evaluation Center personnel because the data was more than five years old and because it was concluded another submarine would not be lost. The date of destruction of the original THRESHER acoustic data was 22 May 1968, coincidently the same day on which it was subsequently determined the USS SCORPION was lost.
With the approach of the 50th anniversary in 2013 of the loss of THRESHER, it would be appropriate for the Navy to officially acknowledge why the USS THRESHER was lost for the benefit of surviving family members and friends of those onboard who may find some solace in the knowledge that the collapse event occurred too fast to be apprehended, and also for the benefit of those surviving Portsmouth Naval Shipyard personnel who continue to be falsely implicated in the loss of the USS THRESHER by assertions that a silver-brazed fitting failed.
As previously stated, the THRESHER pressure hull and all sea-connected systems significantly exceeded design specifications; there was no flooding before collapse of the pressure hull at extreme depth.
Although this analysis advances an understanding of why the USS THRESHER was lost by establishing there is no evidence of the failure a silver-brazed fitting during the 10 April deep-dive, the analysis still leaves the perhaps unanswerable question of why the non-vital bus failed after two minutes of line frequency instability.
The USS THRESHER was lost nearly half a century ago because her nuclear reactor shut down at test depth of 1300 feet and the crew could neither blow ballast nor restart the reactor in the seven minutes during which THRESHER sank to collapse at a depth of about 2400 feet.