A Steel Ship and Iron Men – Catastrophic Failure and Courage on USS IWO JIMA (LPH 2)
The application of sound engineering practices and common sense should have prevented this accident.”
“There were many acts of bravery and good judgment under extremely stressful and dangerous conditions. Concern for shipmates was a common thread present throughout the tragedy.”
– Record of Proceedings of the Inquiry into USS IWO JIMA (LPH 2) Engineering Casualty.]
In August 1990 an international coalition led by the United States began preparations for war. Called Operation Desert Shield, it involved the coordination and prepositioning of military forces that would be deployed in Desert Storm. One warship included in the buildup was USS IWO JIMA (LPH 2). She was a 30-year old amphibious assault ship with a crew of over 700 and capable of transporting a Marine aviation squadron and landing team of over 1,500. She was dependable and deployable. Just weeks after the start of Desert Shield, she participated in maneuvers meant to confuse the Iraqi army. In late October, the ship transmitted a maintenance request to the Navy’s repair unit in the region, Ship Repair Unit Detachment Bahrain (SRU), for repairs to components in her propulsion plant which required repairs before returning to the operational theater. An item added later, which did not have to be repaired before returning to sea, was main steam valve 2MS-7, which provided steam generated from #2 boiler to a turbo service generator. IWO JIMA requested it be added to “to maximize valve maintenance” and “correct a myriad of small packing and flexitallic gasket leaks”. SRU requested additional information about 2MS-7 and ship’s force personnel responded with information, a part of which indicated it was a six-inch globe valve. In reality it was a four-inch gate valve.
Because failures in main steam systems pose great danger to the crew and its proper operation is vital to a ship’s mission, the Navy categorized them as having the highest “level of essentiality”, referring to the degree of regulation and control required to assure reliable repair and maintenance. Because maximum confidence was required, the strictest oversight – “Level I control” – applied to maintenance and repairs conducted on 2MS-7 valve and adherence to numerous operating, maintenance, material, inspection and quality assurance (QA) procedures were mandatory.
SRU personnel were not aware 2MS-7 was in a Level I system subject to Level I controls. They thought they were managing repairs on a less critical Level III system, but even so, the surveyor chosen to define and have oversight of the work informed the resident marine surveyor that he had never developed repair specifications for valve repairs. Nonetheless, he was directed to create them, for despite the recent augmentation to the unit increasing its staff from nine to seventy-six people, he was the only surveyor available. SRU’s workload had increased from seven ships to thirty-one. The SRU acting officer-in-charge later testified he was under pressure from Naval Logistics Support Force to complete IWO JIMA repairs quickly to free up the power barge supplying electricity to the vessel for use by USS LA SALLE and the resident marine surveyor, who supervised all surveyors, was under the impression repairs needed to be completed quickly for operational reasons. Neither impression was accurate.
On Thursday, the 25th of October she arrived in the Port of Manama in Bahrain, an island less than 400 square miles in size 20-miles off the coast of Saudi Arabia. A conference was held on the ship that included the SRU acting officer-in-charge and resident marine surveyor, representatives from fleet- or Navy-wide units and the IWO JIMA Chief Engineer. During the conference the ship’s Chief Engineer, who knew 2MS-7 was a Level III valve by design but Level I by application, stated the ship had no replacement parts for 2MS-7. Neither he nor any other conference participant asked about or mentioned Level I requirements on any of the scheduled work. The Chief Engineer was under the impression the surveyor was familiar with Level I procedures, that his written work specification called for Level I controls, and the contractor knew how to do Level I work. His thinking was contrary to a Naval Surface Forces Atlantic instruction which stated “The engineer of a ship shall ensure that controlled material requirements are included in work requests where a determination has been made regarding the requirements of Level I controlled material.” On Level I taskings that required material, a ship would normally provide what was needed because detachment level SRUs would not maintain an inventory of Level I material.
This error of initially communicating that 2MS-7 was a globe valve was discovered. The surveyor knew his specifications should be corrected but he did not do so due to perceived time constraints and how the work on 2MS-7 had been characterized as a Level III item (due to ship’s force never identifying it as Level I). He wasn’t knowledgeable enough to have caught the error on his own, but a surveyor’s work package was supposed to be reviewed by the resident marine surveyor, whose position required knowledge of Level I ship repairs.
Following the conference, inspection of the listed items was made. Because the plant had been on line recently and was still hot, 2MS-7 remained covered in insulation preventing examination. The work specification was expedited so the contract could be awarded to a civilian contractor before noon. That was when the Bahrainian weekend began and only a few hours after the end of the conference. Even so, work would not begin until the start of the new work week. The next day, the oversight of the repairs were reassigned and the surveyor turned his documentation and work specifications over to another surveyor, who was familiar with Level I requirements on diesel and oxygen generation systems, but not on steam systems. As he transferred the work to the second surveyor he said the ship was gathering technical documentation for SRU and what parts the ship was providing. Both surveyors later testified neither of them received the requested information.
Due to miscommunication, misunderstandings and mistaken assumptions, a correct understanding of quality assurance requirements, and who was responsible for ensuring the required QA was indeed performed, was thwarted. Where procedures, policy and practical common sense should have made for a clear development and execution of QA requirements, a rough patchwork of vastly different interpretations emerged:
- Ship’s force personnel thought responsibility for all work specification check points were shared by SRU and the civilian contractor. This is the case when work is performed at larger U.S. based repair units, but not in foreign ports.
- The SRU officer in charge and one surveyor thought IWO JIMA personnel was responsible for QA for the contractor’s work, and indeed, the Chief Engineer and Commanding Officer ultimately were. Had they reviewed the specification they should have discovered QA requirements that would have ensured progress inspections and check points were absent.
- Another surveyor thought the contractor had responsibility to provide a QA inspector to be on board during the repair work.
- Other SRU personnel thought that surveyors were responsible only for completion of a job in accordance with the specifications, but this was not the same as the QA that was required.
- The contractor’s perspective was the surveyor is primarily responsible for confirming the work was adequately completed.
- The Sixth Fleet’s maintenance officer said QA in repair work is a shared responsibility between the ship’s force and the surveyor.
It was the responsibility of the Chief Engineer to establish a team of qualified personnel to monitor progress and inspect the work being performed by commercial contractors, witness quality control tests, and assure correct completion of the work. He had a standing requirement that his personnel monitored all work in the plant, no matter what entity was doing the work, and such an organization existed when he reported aboard nineteen-months earlier but it waned. In July 1989, IWO JIMA promulgated a ship’s quality assurance instruction but in October 1990, the engineering department was not utilizing it.
On Sunday, October 28th, a civilian employee of the shipyard contractor boarded the ship and began disassembling of 2MS-7. He was not expected to read the repair specification but to take “general repair guidance” from his foreman, who did have a copy of the specification, which SRU routinely issued in a simple form because of language barriers, and because the contractor wouldn’t have the reference documentation cite therein. After he resurfaced the portions of the gate and valve body that controlled steam flow, IWO JIMA’s Chief Engineer and a Chief Boiler Technician inspected the work and the Chief Engineer directed the surveyor to reassemble 2MS-7. The contractor foreman was required to inspect the work but did not do so; he was confident in the pipefitter’s ability to work on Level I systems even though neither one spoke the other’s language. An amendment to the “Master Agreement for Repair and Alteration of Vessels” between the Navy and the contractor stipulated at least one English speaking employee be on board when work was being done. Although the Master Agreement had been in force for six years, SRU did not have a copy.
Not aware of the stringent material control on the work he was performing, the pipefitter approached a crewmember for new parts. The ship’s force had not been instructed to not provide any parts for the work, so, despite the language barrier, he was shown the parts bin from which he selected four bolts, eight studs and 20 3/4 -inch nuts. He possessed ten years of experience but hadn’t noticed that some of the nuts he chose were brass. Because those fasteners were covered with a manufacturer-applied black coating, at a glance, they could be mistaken for the correct grade 4 steel nuts. The pipefitter knew silver nuts in the bin were steel, but was unsure of the dark colored ones. Closer examination by way of a scratch or magnetic test, would have revealed their metal content. Brass nuts were not appropriate for use on any system exceeding 400° F. IWO JIMA’s propulsion plant operated in excess of 800° F. Past the maximum rating of 400 ° F, the tensile strength of this copper and zinc alloy was lost and when IWO JIMA left port two days later, the nuts would be subjected to a system temperature 116% above their limit. Without “general repair guidance” from his foreman, he reassembled the valve, mixing eight studs with four bolts. The valve should have been reassembled using only B-16 steel studs; doing otherwise was a violation of good engineering practice. He also unwittingly started a battle with physics that could not be won when he placed at least one brass nut on each stud or bolt.
The non-English speaking pipefitter indicated to ship personnel he had finished the work and disembarked IWO JIMA. The next day the valve was reinsulated with lagging, covering it up again. The foreman had never looked at 2MS-7 and now no one would be able to. When the plant was lighted off, every sailor in the fireroom then would be at extreme risk, and the likelihood the ship’s mission commitments could be missed was significant. Navy propulsion plants were dangerous places. From the late 1800’s to mid-20th century, no less than 52 Medals of Honor were issued for heroism in engineering spaces.
On October 30th in preparation to get underway and proceed to her operating area, fires were lighted in the boilers at 0218 (#1) and 0556 (#2). At 0353, one side of 2MS-7 was initially pressurized with steam generated from #1 boiler. Between 0630 and 0720, valve 2MS-7 was opened to supply steam to the generator that supplied electrical power to the vessel. The passage of superheated steam – a powerful and invisible gas with immense energy and now at a searing 865° F – pressurized the bonnet and heated the bolts, studs, steel and brass nuts. The brass nuts were coming under a strain for which they were never designed – and were incapable of holding – and they rapidly softened. At 7:50 am, a Boiler Technician Third Class saw the lagging insulation covering 2MS-7 smoldering and discussed this with two other petty officers. Four minutes later the wheelhouse received the report from main control the plant was ready to “answer all bells”. At 7:56 am, IWO JIMA was underway, headed for her part in the largest military buildup since World War II. She would quickly reach a catastrophic destination made inevitable due to improper repair specifications, inadequate work procedures, use of non-certified material and no quality assurance merging with the lack of proper supervision, missing inspections and check points.
At 0811, main control requested permission to shut down #2 boiler because of a steam leak reported by the Boiler Technician of the Watch. The officer of the deck approved the request. One minute later “a major steam leak” was reported to the bridge along with the request to sound general quarters. Seconds before the valve’s failure, one sailor, who had just entered the fireroom, saw four men on the upper level looking at the failing valve beneath them. Immediately he was waved off by the main propulsion assistant. As he turned around and began exiting the space he heard the loud bang of the valve bonnet explosively detaching. Instantly, the entire space within the fireroom was overcome by 640 p.s.i. superheated steam. He was already on the ladder leading up but because of being suddenly enveloped in extreme heat, he questioned himself “Will I be able to make it out?” Because he had literally just set one foot in the fireroom when the lieutenant motioned for him to leave, he escaped with no injuries and survived. Just a few feet more and he would have received irremediable injuries along with ten watch standers in the space.
Moving between eight- and nine-knots, the ship’s steering was lost for about a minute and when the vessel had slowed to a safe speed of five-knots the OOD ordered both anchors dropped to slow, and then halt, the ship. Bold professionalism, augmented by training in major steam leak casualty control procedures in June and August, was evident in many locations. In main control, quick actions were taken to mitigate further impact. Like the watch standers on duty in the fireroom, other individuals moved with rapid purpose, making lightning-fast assessments in main control, on the mess decks and the bridge saving the ship from more damage.
The four men who were looking at 2MS-7 initially did not attempt to escape, rather, they ran away from the exit toward the lower level to undertake casualty control measures, a disciplined, professional and sacrificing move because the decision would unquestionably doom them. All ten of the sailors standing watch that morning knew the procedures for a major steam leak stated personnel should attempt, as time permits, to locate and isolate rupture and secure equipment including securing boiler. Without hesitation, three of the men standing above 2MS-7 rushed farther into apocalyptic space but quickly realized there was nothing they could accomplish. Gravely injured in the few seconds that had passed since the release, they made their way to the fireroom exit and ascended the ladder to the mess deck where they were treated by other crew members until the medical staff arrived. The other sailor who had been above the valve climbed through the escape hatch and walked to sick bay. Despite their own fatal injuries, the four men who escaped pleaded to those who were caring for the injuries that those who hadn’t escaped had to be helped. In time, they were transferred to the USNS Comfort to receive appropriate medical care for their devastating thermal injuries. All of them, these sailors who had responded courageously to be immediately vanquished by the horror that enveloped them, died later that day on the hospital ship. In the moments following the accident, one thought was on the minds of the rest of crew: Six shipmates were still in the space.
The remaining sailors in the fireless inferno went about implementing procedures for controlling a major steam leak. The fuel oil pump for #2 boiler was manually tripped to shut down the fires and attempts were made to trip the quick closing valves on the fuel oil supply to #1 boiler. If the fires in either boiler were not extinguished significant damage to the ship probably would have occurred. The sailor performing this casualty control procedure also tried to time it so steering wouldn’t be lost for a longer period that would pose an even greater hazard to the ship leaving port and to prevent causing a major brownout of the electrical system. Another, forgoing a good chance to escape, stayed to switch control of the turbo generator to main control. On the mess decks another sailor who had not been in the fireroom at all, shut the main and auxiliary main steam stop valves on both boilers by using remote operators. The main steam stop valve for #2 boiler, 2MS-1, did not shut because the air supply line to the valve motor had been severed by the explosion. The hellish leak from 2MS-7 could not be isolated. Upon recommendation of the Machinist’s Mate of the Watch, the Chief Engineer ordered throttles to full open to pull as much steam from the system as possible and ordered positive air flow into the fireroom to clear the steam and cool the space. The other turbo generator was left on line to continue its use of the steam supply. A Boiler Technician First Class attempted to enter the fireroom to trip the fuel oil pumps but the forceful heat was impenetrable.
Twenty-three minutes after the explosion, and now on their second attempt, two petty officers entered the menacing fireroom to search for their shipmates. They too had been turned back earlier, but as scared as they both were, what kept them going forward was their shared thought, a hope really, that someone had made it through. While approving their aggressive and courageous request, the Chief Engineer instructed them to assist anyone in the fireroom as needed. Otherwise, they were told to identify the sailors still in the space.
Once a space filled with engineering wonders – where man created and harnessed 16 megawatts of energy that moved the 18,400 ton vessel around the world – was now a furnace of devastation. Still exceptionally hazardous because of the incredible heat, the fireroom presented a stunning landscape: Everything was white. The explosion obliterated insulation covering pipes, valves and fittings and dusted everything within the space with fine particles. Working quickly for the sake of the six as well as themselves, the two sailors found the lifeless bodies their shipmates. Where they found them told the story: The delay of all ten, and the inability of the six who never escaped, was due to their attempts to shut down the steam plant and save their ship.
A harbor tug maneuvers the amphibious assault ship USS IWO JIMA (LPH-2) into port following a boiler room accident aboard the vessel which killed ten crew members. U.S. Navy photograph.
Reminiscent of five Marines and one sailor raising the flag on Mt. Suribachi – the highest point on Iwo Jima – ten sailors in one of the lowest locations on USS IWO JIMA, with many shipmates in other locations of the warship, acted in the highest traditions of the United States Naval Service the day 2MS-7 surrendered and hell came calling.
In Memory Of:
“Fair winds and following seas”.
The explosion on IWO JIMA rumbled through the entire non-nuclear surface fleet. The Judge Advocate General (JAG) and Naval Investigative Service (NIS) investigations determined that missteps in the required quality assurance procedures were not isolated to SRU or USS IWO JIMA. The Court’s proceedings found “the failure of USS IWO JIMA to maintain a viable QA program is a tragic example of a greater QA deficiency in the non-nuclear” Navy. Concurrent with the investigations a significant fleet-wide review was undertaken and it was discovered a vast number of ships were not consistently embracing required quality assurance processes. Even IWO JIMA’s Chief Engineer, who had initiated or resumed certain QA measures since reporting aboard, wasn’t following his department’s engineering organization and regulations manual. In response to the major engineering casualty, quality assurance manuals were simplified and QA responsibilities of ship personnel clarified throughout the fleet. Pocket-size guides to QA was provided for every sailor in the fleet and proper training and program reviews were instituted along with increased attention to QA by the leadership of the Navy’s surface fleets.
The direct cause of the accident was the improper use of black coated brass nuts. The potential for an engineering casualty anywhere in the fleet was identified by the Navy in 1975, but because an all-out purge never took place, they remained an inventory item in 1990. The Court called for the removal of all such fasteners, fully and finally rectifying the problem that was first acknowledged fifteen-years earlier.
After the accident IWO JIMA spent six weeks in port for repairs before getting underway and serving in Operation Desert Storm. In 1992 she participated in the humanitarian mission Operation Provide Promise as a search and rescue resource in the Adriatic Sea for aircraft delivering aid supplies to Sarajevo. In 1993, weeks short of thirty two years of active service, the USS IWO JIMA (LPH 2) – the first vessel built from the keel up to deliver both air and ground forces – was decommissioned and cut up for scrap three years later.
What’s your plant’s SITREP (situation report for civilians)?
- Is your operation founded on properly trained and qualified personnel?
- Is it established on proven and proper procedures?
- Is it free of assumptions and shortcuts?
- Are honest and professional questions and suggestions genuinely encouraged?
- Are there any real or perceived limitations on anyone stopping or reporting an unsafe situation?
A former Sailor’s outreach to Shipmates he never knew.
Timothy C. Cummings, a former Navy 1,200 p.s.i. boiler technician, is a Certified Professional Maintenance Manager and antiterrorism specialist. His original case study of the USS IWO JIMA accident was selected to be the keynote speech at a national human performance/root cause conference in 2013.
Postscript: In May 2014, this piece was published as the cover story of a plant engineering trade magazine. Ironically, the publisher misspelled “catastrophe”. After being annoyed at their amateurish oversight, I came to view it as an ironic way to underscore what went wrong on 30 OCT 90: People not paying attention. – TCC