PREFACE
In 1974, 76 years after the US Battleship MAINE (ACR-1) sank in Havana harbor, an event then attributed to a Cuban or Spanish mine, ADM H. G. Rickover directed a re-examination of photographs taken of the wreckage when it became available for inspection in 1911. That review determined that “an internally initiated explosion had destroyed the MAINE.” (Reference: 'HOW THE BATTLESIP MAINE WAS DESTROYED,” by H. G. Rickover, 1995 Edition, Naval Institute Press)
In anticipation of a similar hiatus before the Navy may formally review the 1963 Findings of the THRESHER Court of Inquiry (COI), this document provides acoustic-based information related to the event that could otherwise become ”perishable” (lost) and unavailable to support future reviews. Several sections of the following discussions are reprised from earlier articles, especially those articles that provide the basis for rejecting the COI conclusion that flooding occurred before collapse of the THRESHER pressure-hull and thus was the primary cause of the disaster. (See ISSUE A below)
WHY THRESHER WAS LOST
The USS THRESHER (SSN 593) was lost 53 years ago because of the confluence of three circumstances; (1) Loss of Propulsion: a reactor scram (shut down) during the minute prior to 0909R, the proximal cause for which remains unknown; (2) THRESHER was significantly out-of-trim (heavy) at test-depth, and (3), attempts to deballast initiated at 0909.8R and 0913.5R (COI Finding of Fact 18) were unsuccessful because excess moisture in the flowpath from the high-pressure air flasks formed ice on strainers in the system valves: adiabatic cooling as verified during subsequent tests with the USS TINOSA (SSN 606).
THRESHER might have recovered from any two of these circumstances; however, the confluence of all three allowed THRESHER to sink at an average rate of 120 feet a minute from at or near test-depth (1300-feet) at 0909R to collapse at 09:18:24R at a depth of 2400-feet. (See ISSUE B)
SUGGESTIONS FOR FURTHER RESEARCH
Those who may, in the future, investigate the loss of THRESHER should address three critical subject areas: (1) The potential consequences – relative to loss of propulsion - of operating the Main Coolant Pumps (MCPs) in FAST (2-pole mode) during a deep-dive. (See Issues A and B below): (2) Assess why THRESHER could have been significantly out of trim (heavy) at test-depth when there was no flooding (see ISSUES A, B and D below), and (3), If THRESHER's main steam valves were NOT closed immediately subsequent to the S5W reactor scram, assess how long decay-heat could have provided sufficient steam pressure to enable the Ship's Service Turbo-Generators (SSTGs) - with action by the SSTG speed governors - to carry the non-vital bus electrical load – including MCPs in FAST - before the steam pressure dropped too low and the SSTG breakers tripped on under-frequency. Sound Surveillance System (SOSUS) data still held by the Office of Naval Intelligence (ONI) In March 2007 indicates two minutes: 0909R-0911R. (Reference: writer's testimony before the THRESHER COI on 18 April 1963.)
ISSUES AND REFERENCES
ISSUE A: NO FLOODING BEFORE COLLAPSE OF THRESHER's PRESSURE-HULL
Multiple, independent lines of evidence confirm there was no flooding prior to collapse of the THRESHER pressure-hull. Such flooding, conjectured by the COI to have been the initial and primary event responsible for the disaster, would have been a catastrophic event with the water-jet produced by the initial breaching of the pressure-hull or a pipe (OPINION 1a of cited reference estimated the ruptured pipe to have been between two and five inches in diameter) expanding into the interior of the submarine with a velocity of about 1800 mph at THRESHER's depth of 1300-feet. Such an event would not have been reported by THRESHER to her escort ship, the USS SKYLARK (ASE-20) at 0913R as (quote) experiencing minor difficulties (end quote). Additionally, the conjectured flooding and the associated water-jet, upon impacting structures within the hull, would have generated extreme noise levels not only within the breached compartment but also in adjacent compartments, and at reduced but still high levels throughout the submarine. Such noise levels would have made it extremely difficult for THRESHER to have communicated with SKYLARK. No such noise levels were evident during communications with SKYLARK nor were they detected by SOSUS. (Reference: Loss of the USS THRESHER: http://www.jag.navy.mil/library/jagman_investigations.htm http://www.jag.navy.mil/library/investigations/USS%20THRESHER%20PT%201.pdf, 202.pdf, 203.pdf, 204.pdf
Another compelling reason for rejecting the COI conjecture that flooding occurred at test-depth of 1300-feet (580 psi) is that water expanding into the relative vacuum (15 psi) within the THRESHER pressure-hull would have instantly atomized into a dense vapor (fog) upon impacting solid surfaces and/or vaporized in the low-pressure environment making it difficult to see within those spaces, yet immediately subsequent to telling SKYLARK at 0913 that she was (quote) experiencing minor difficulties (end quote), THRESHER transmitted (quote) will keep you advised (end quote). Neither of those transmissions are consistent with the disaster that flooding at test-depth (1300-feet) would have represented.
There is, in the public domain, a report by a former Commandng Officer (CO) of the USS SHARK (SSN 591) who claimed to have listened to a tape recording of UQC exchanges between THRESHER and SKYLARK that contained the following statement made by CO, THRESHER: (quote) Stand clear, emergency surfacing from test depth, flooding in the engine room. (end quote) Unfortunately for the accuracy of that memory and many other assertions made by the former CO, there was no tape recording to which he could have listened for - during Congressional hearings on the loss of THRESHER held on Thursday, 27 June 1963 - RADM John Maurer, Director, Submarine Warfare Division, in response to a question from Representative David Bates, NH, stated: (quote) All of the ASRs are equipped with recorders now. This is since the (THRESHER) incident. At that time, they did not have recorders. (end quote) (Reference: Page 51 Congressional Record for 27 June 1963.)
A rebuttal to all the misinformation provided by the former CO SHARK is available at http://www.iusscaa.org/articles/brucerule/misinformation_about_the_loss_of_thresher_and_the_sosus_detection_thereof.htm
These assessments are independently confirmed by THRESHER COI Finding of Fact (FoF) 153, i.e., provides compelling support of the conclusion there was no flooding before collapse. FoF 153 states: (quote) That during the course of proceedings, a test demonstration for the Court of Inquiry was held in Drydock No. 2 at the Portsmouth Naval Shipyard. A stream of water was released to atmosphere at Thresher’s test-depth pressure against a piece of electronic equipment. The stream produced tremendous force, spray, fog and noise (end quote), conditions – to repeat - that would not have been reported by THRESHER to SKYLARK as (quote) experiencing minor difficulties (end quote).
Russell Preble, CDR USN (ret), who was in Portsmouth, NH, in April 1963, and who actually observed that test, made the following statement: (quote) A memory that stands out in my mind was watching one of the Board of Inquiry’s tests. An old SS radar console was placed on the floor of one of the empty dry docks and a high pressure stream of water was directed against the console. The noise was overwhelming. I remember thinking that nothing could be heard over the noise of the water smashing up against the radar casing and how at deep submergence no orders could be heard over the roar of water striking anything in its way. (end quote) (Reference: "USS Thresher, Lest We Forget", Burke Consortium, Inc, 2013, p. 5; file name: ThresherBooklet_printv2.pdf)
The above described THRESHER COI test confirmed that high-velocity water-jets entering a submarine pressure-hull at great depth will generate extreme levels of acoustic energy.
As of March 2007, ONI still held acoustic detections of an event during which a water-jet with a velocity in excess of 2000 mph entered an otherwise intact submarine compartment at great depth. The acoustic energy produced during that event consisted of more that 100 individual resonances in the low- to mid-frequency spectrum. Those acoustic sources were detected by a SOSUS array at a range in excess of 700 nautical miles.
The frequencies of such resonances are a function of the dimensions of objects directly or indirectly impacted by the high velocity water-jet. Such a water-jet – essentialy a water-ram - will excite not only the pressure-hull as it passes through a breach but, upon impacting internal structures, will act like a pneumatic jackhammer.
Essentially, the submarine pressure-hull responds like an enormous bell being continuously struck by an equally outsized hydrostatically-driven clapper.
Impacted internal structures connected to the pressure-hull, bulkheads or decks respond like drumheads which resonate - as noted above - at frequencies that are a function of their dimensions - especially the thicknesses - of the component members of the excited structures.
The presence/absence of such resonances before the collapse of a submarine pressure-hull at great depth is an unambiguous indication of whether or not flooding occurred before collapse. No such signals were detected when THRESHER was lost only 30 nautical miles from the nearest SOSUS array.
And finally – immediately below - the most compelling reason of all for rejecting the COI conclusion that the reactor scam was the result of an electrical short-circuit caused by the rupture of a sea-connected pipe while THRESHER was at test-depth (1300-feet).
The references cited below provide values for the rates at which flooding would have occurred aboard THRESHER had a pipe with a diameter between two and five inches actually ruptured at a depth of 1300-feet as concluded by the COI. Those values are: (1) for a two-inch diameter pipe: 1800 gallons per minute (GPM), and (2), for a five-inch diameter pipe: 11,000 GPM. Note: one US gallon of water weighs 8.34 pounds.
In the four minutes that elapsed between loss of propulsion (the reactor scram) by 0909R - which was based on analysis of SOSUS data, and which the COI attributed to the rupture of a sea-connected pipe - and the 0913R time of the (quote) experiencing minor difficulties (end quote) UQC transmission by THRESHER, we have the following values for flooding: (1) for a two-inch diameter pipe, flooding would have added 60,000 pounds, and (2), for a five-inch diameter pipe, flooding would have added 367,000 pounds.
What makes the COI pipe-rupture-flooding assessment equally untenable are the volumes associated with these flooding rates: (1), for the two-inch pipe, a water volume of 965 cubic feet in four minutes, and (2), for the five-inch pipe, a water volume of 5,900 cubic feet in four minutes. Note: the volume of one US gallon is 0.134 cubic feet.
References:
(http://www.translatorscafe.com/cafe/EN/units-converter/pressure/38-59/psi-foot_sea_water_%2815%C2%B0%D0%A1%29/)
convert PSI (seapressure) and PSI (internal) and hole diameter (inches)
http://www.tlv.com/global/TI/calculator/water-flow-rate-through-orifice.html
SUMMARY ASSESSMENT of ISSUE A: For the multiple reasons discussed above, the THRESHER COI conclusion that flooding occurred before collapse of the THRESHER pressure-hull and was the direct cause of the reactor scram should be rejected.
ISSUE B: DERIVATION OF THRESHER PRESSURE-HULL COLLAPSE DEPTH
The THRESHER pressure-hull collapse depth calculation of 2400-feet - made for the frst time in 2008 - was based on the empiric relationship that exists among three values: the depth at which collapse occurs, the volume of the collapsing structure, and the frequency of the bubble-pulse acoustic signal produced by the collapse: 3.4 Hz for THRESHER. (Reference: Page C4 of USS SCORPION (SSN 589) RESULTS OF NOL DATA ANALYSIS (U) (NOL Ltr 69-160 of 29 January 1970), Robert Price and Ermine Christian).
The calculated collapse depth of 2400-feet at 09:18:24R also is consistent with the conclusion that the number “900” transmitted by THRESHER to SKYLARK at 0917R via UQC was an indirect reference to test-depth, i.e., THRESHER was 900 feet below test-depth or at 2200-feet about 90 seconds before collapse of the pressure-hull 200 feet deeper.
Survival of the THRESHER pressure-hull and all sea-connected systems to a depth 450-feet greater than the calculated collapse depth of 1950-feet (test-depth plus 50%) exculpates all Portsmouth Naval Shipyard personnel of any complicity in the loss of THRESHER. Indeed, they built better than they knew.
A corollary of the depth derivation and measurement of the bubble-pulse frequency (BPF) is determination of the energy release – expressed in pounds of TNT – required to produce the observed BPF at the derived depth. In the case of THRESHER, the value was 22,500 pounds of TNT, produced by the nearly instantaneous conversion of potential energy (1070 psi sea pressure) to kinetic energy, the motion of the water-ram which entered the THRESHER pressure-hull at a velocity of about 2600 mph at the collapse depth.
THRESHER was destroyed in less than 0.15 seconds, the duration of the compression phase of the collapse event as determined by half the reciprocal of the bubble-pulse frequency: 1/3.4 Hz = 0.30s/2 = 0.15s (15 milliseconds), slightly less than the combined retinal and cognitive intergration times of those onboard, i.e., they were not aware of the event; it occurred too fast to be apprehended.
Note: acoustic data provided by the writer to ONI in October 2009 - a recording of the SCORPION collapse signals - suggested collapse events may not be temporally symmetrical – that the destructive collapse phase may be shorter in duration than the subsequent expansion (rebound) phase because the duration of the compression phase is truncated when the intruding water encounters the compacting wreckage whereas the expansion phase terminates more gradually – and probably asymmetrically with respect to aspect - until the pressure/velocity of the expanding bubble of compressed air (from within the hull) is overcome by the local (ambient) sea-pressure. In the case of THRESHER, the total destruction (fragmentation) of the pressure-hull and internal structures may have occurred in less than half the reciprocal of the bubble-pulse frequency or in as little as six milliseconds (0.006s).
If the “focus” of the compression wave front is taken to be about half the diameter of the THRESHER pressure-hull - or about 15-feet - the involved time of 0.006s equates to a velocity of about 1700 mph. Because this calculation does not consider the time for the shock wave to propagate to sections of the hull beyond the initial breach point - at the speed of sound in steel (19,000 f/s) - the derived velocity of the compression wave front within the fragmented hull is necessarily very approximate but clearly slower than the 2600 mph at initial entry.
ISSUE C: DERIVATION OF ESTIMATED THRESHER's SINK-RATE OF 120 FEET A MINUTE
THRESHER was at or near test-depth (1300-feet) at 0909R by which time propulsion had been lost. THRESHER's pressure-hull collapsed at 09:18:24R at a depth of 2400-feet.
THRESHER's depth increased 1100-feet (1300 to 2400) in 9.5 minutes (0909R-09:18:24R) for an average sink-rate of 120-feet a minute (ft/min).
The 120 ft/min value for THRESHER compares with 52 ft/min for the USS SCORPION which sank in 21 mins and 50 secs from an estimated transit depth not greater than 400-feet at the time of the internally destructive battery-explosion (18:20:44Z, 22 May 1968) to collapse at a depth of 1530-feet at 18:42:34Z. The calculated atmospheric over-pressure created by the battery explosion was 150-200 psi at the site of the event, three times the fatal value; hence, it is improbable SCORPION's sink-rate was slowed by actions of the crew who were either killed or functionally incapacitated by the battery event. ((Chapter SIX of 'WHY THE USS SCORPION (SSN-589) WAS LOST, ISBN 978-1-60888-120-8, Nimble Books, 31 October 2011, reviewed/commented upon by (1), Letter to the Editor from George W. Jackson, CAPT USN (ret) (Fall 2011 Issue of THE SUBMARINE REVIEW, pp 149-150), (2), The Loss of SCORPION, a Book Review by CAPT Jim Patton (Winter 2012 Issue of THE SUBMARINE REVIEW, pp 151-152), and (3), Letter to the Editor from Norman Polmar (Summer 2012 Issue of THE SUBMARINE REVIEW, pp 141)
It is concluded THRESHER sank at an average of 120 ft/min subsequent to loss of propulsion at test-depth because the submarine was significantly out-of-trim (heavy) and unable to deballast, and not because of flooding. The significantly lower sink-rate indicates SCORPION was closer to being in trim (neutrally buoyant) than was THRESHER.
ISSUE D: ACOUSTIC DETECTIONS OF THRESHER's MAIN COOLANT PUMPS AND TIME OF THE REACTOR SCRAM.
While passing a depth of 1000 feet during the scheduled 10 April 1963 deep-dive to test-depth (1300-feet), rotational-rates of THRESHER's Main Coolant Pumps (MCPs) were initially detected acoustically at 0845R in FAST (2-pole mode). Identification of that detection as an MCP source was based on the coincident detection of an associated acoustic source at a ratio of 0.485:1 unique to S5W reactor system MCPs. That detection – and all subsequent detections of MCP activity and the two unsuccessful attempts to deballast - were made at a range of 30 nautical miles by SOSUS hydrophone array 1411 which terminated at HMCS Shelburne, Nova Scotia.
The line-frequency of the non-vital electrical bus that powered the MCPs in the 2-pole mode was established by correcting the frequency of the MCP rotational-rates for the 2.5 percent slip-rate of the MCP drive-motors. The implied non-vital bus line-frequency and stability were normal based on comparisons with analyses of several thousand hours of previous detections of S5W SSTG systems – including THRESHER from pre-overhaul detections - published by Commander Oceanographic System, Atlantic on 1 April 1963 as THE US NUCLEAR SUBMARINE ACOUSTIC DATA HANDBOOK.”.
The results of analyses of the HMCS Shelburne acoustic data were provided to the THRESHER COI on 18 April 1963 and are summarized in Part one of the U.S. Navy Judge Advocate General's Report of the loss of the USS THRESHER available at http://www.jag.navy.mil/library/jagman_investigations.htm.
At 0909R, the MCP rotational rates – still in FAST – became unstable, varying randomly plus/minus about 18 rpm until 0911R when the signal was abruptly lost while in FAST. As of March 2007, the Office of Naval Intelligence still held a photocopy of the original graphic display (lofargram) upon which this 1963 analysis was based. There were no magnetic tape recordings of any SOSUS detection of the THRESHER event.
Because of the highly anomalous instability of the MCP acoustic signal – which was a reflection of instability of the same magnitude in the line-frequency of the non-vital bus powering the MCPs - and initiation of the first falied attempt to deballast at 0909.8R - it is assessed the reactor scram occurred during the minute prior to 0909R.
Had THRESHER retained a propulsion capability after 0909R, the attempt to deballast at 0909.8R should not have been necessary; THRESHER could have used propulsion to drive to the surface, an evolution that should have been detected by SOSUS had a speed greater than 12 knots been employed. No such detection occurred.
Bruce,
In review of a recent article (http://ussnautilus.org/blog/the-loss-of-uss-thresher-ssn-593/) concerning the anniversary of the sinking of the Thresher, I would greatly appreciate your evaluation of their findings. Secondly, could you clarify to me why an MCP would be considered part of a non-critical bus versus critical? Your insight would be greatly appreciated.
Best,
Stephen Lechowicz
FIRST QUESTION: What is the writer's evaluation of the cited 2014 SUBMARINE FORCE MUSEUM article that states flooding was the cause of the THRESHER disaster?
RESPONSE: There was not in 1963 – nor is there now – any evidence to support the Court of Inquiry (COI) conclusion - reprised by the article - that flooding occurred onboard THRESHER before collapse of the pressure-hull at great depth. As discussed in detail by the posting “WHY THE USS THRESHER (SSN 593) WAS LOST,” multiple, independent lines of evidence make it indisputable there was no flooding.
COMMENT: Attributing the loss to flooding allowed the COI to avoid the essential question which was then – and remains now: How did THRESHER arrive at test-depth significantly out-of-trim (heavy) if there was no flooding? The single answer to that question can be provided by anyone who qualified on submarines.
SECOND QUESTION: Why would Main Coolant Pumps (MCPs) “be considered part of a non-critical bus versus critical?”
RESPONSE: Under normal operating conditions for submarines with an S5W nuclear reactor, all ac electrical power requirements were produced by two steam-turbine-driven Ship's Service Turbo-Generators (SSTGs) which - operating in a twin-bus distribution configuration – could accommodate all loads including operation of MCPs in either SLOW (4-pole mode) or FAST (2-pole mode). MCPs in FAST were needed to support operation at speeds above about 20 knots and FAST required ac power several multiples of that required for SLOW. The SSTGs also drive Ship's Service Motor Generators (SSMGs) which are ac/dc machines that, when driven by the output of the SSTGs, function as dc generators. Basically, this is the non-vital bus.
If steam for propulsion and the SSTGs is lost for any reason, the SSMG flips function and becomes a dc motor powered by the battery to produce limited ac power, sufficient to operate the MCPs in SLOW – but NOT in FAST. This is the vital bus. For a limited period and under certain conditions, it can provide the ac power needed to retain propulsion. This is a critical back-up capability and is why the term “vital bus” is used.
One such condition is that if you are running the MCPs in FAST and there is a propulsion plant casualty, you MUST immediately shift the MCPs to SLOW so the vital bus can pick up the MCP electrical load. If the MCP speed is not shifted to SLOW, the SSMGs cannot accept the electrical load of the MCPs in FAST and the reactor will scram.
The SOSUS acoustic data previously discussed confirms a propulsion plant problem occurred during the minute prior to 0909R; however, the MCPs continued to operate in FAST - although powered by an unstable SSTG line-frequency - until the MCPs went off-line (strong signal abruptly lost) at 0911R. Thus the acoustic data confirms the MCPs were NOT shifted to SLOW when the propulsion plant problem occurred.
(Note: the strength of the MCP in FAST acoustic source relative to the known detectability of the MCP source in SLOW indicates that had the THRESHER MCPs been shifted to SLOW at 0911R, they would still have been detected at THRESHER's range to the nearest SOSUS hydrophone array: then 30 nautical miles.)
COMMENT: Attributing the loss to flooding allowed the COI to avoid another question which was: When the propulsion problem occurred, why were the MCPs not immediately shifted to SLOW? An ancillary issue is why were the MCPs operated in FAST during the deep-dive when operation in SLOW would have accommodated speeds up to about 20 knots and also would have provided the electrical system flexibility needed to possibly avoid a reactor shut-down?
Stephen: Thank you for your perceptive questions which strike at the two most basic issues involved in the loss of the USS THRESHER which - to date - have never been addressed by the Navy.
FINAL COMMENT: Much of the above information is already available on the internet. Example: scroll to the bottom of the linked site.