By Geoff Fein
Three days after acknowledging a weld problem with its Virginia-class submarines, the Navy yesterday said the issue was not as serious as first thought and all three boats should be ready to go this week.
“We have come through the process of clearing the North Carolina (SSN-777) for alpha trials and the other two ships who might have been underway,” Rear Adm. William Hilarides, program executive officer (PEO) submarines, told Defense Daily yesterday in a telephone interview.
“All three ships should be back in ready to go sea status and they will be ready to go or not depending upon their schedules,” he added.
Last Friday, the Navy, General Dynamics [GD] and Northrop Grumman [NOC] began inspecting welds on all Virginia-class submarines after discovering a weakness in the process associated with non-nuclear piping welds (Defense Daily, Dec. 7).
“The quality of our work is something we take very seriously,” Northrop Grumman said in a statement yesterday. “We have a rigorous program in place that includes inspecting and evaluating our work to ensure it adheres to the Navy’s strict requirements. When issues arise, it’s something we address in an immediate and methodical way. We are working with the Navy and Electric Boat to perform additional inspections and engineering studies.”
The company added it is also taking the following corrective actions:
- Made changes in weld filler metal control processes that aid in the elimination of mistakenly using the wrong filler metal in non-nuclear welds.
- Will continue to conduct briefings with all construction management, general foremen, foremen and welders to ensure everyone is aware of inspection findings, process changes and the corrective actions going forward.
- Briefing production managers to ensure everyone is aware of each finding and course of action going forward.
- Providing a mandatory eight-hour specialized briefing to each welder and welding foreman in the next few weeks. The session will include discussion and practical application training followed by an examination.
- Taking appropriate actions with individuals involved in the recent occurrences of using the wrong weld filler metal.
The first event occurred in August during testing of the USS New Hampshire (SSN-778) at General Dynamics’ Electric Boat. During testing, pieces of the New Hampshire built by Northrop Grumman Newport News were found to have a weld problem. Two months later, a second weld problem led to a joint coming apart, Hilarides said.
“It was that second problem in October that alerted us that our process might have a problem,” he added. “[The first problem] wasn’t that big of a deal. The second one really alerted us [that] this is bigger than we thought.”
Electric Boat was conducting an air test of the New Hampshire‘s hull. During the test, the hull is pressurized to see if air is leaking out. It was during the air test that a joint came apart, Hilarides said.
“When they looked at it, it was due to poisoning of the weld material. It was a stainless steel weld and it was poisoned with copper,” he said. “That was what weakened the joint to the point where it couldn’t even hold the low air pressure, let alone the high pressure hydrostatic test you would have done before the system was certified to be ready.”
One of the things the Navy found out in its investigation is that copper contamination causes cracking and weakens the weld. “So the question out of that test failure was how did that copper get in there,” Hilarides said.
Between 2000 and now, the Navy discovered 15 instances where there had been some contamination of one type or another in a weld where it weakened a weld. The Navy knew about those issues and the welds were fixed, Hilarides said.
“We used all of those in our analysis of determining how big the problem was,” he added. “It’s relatively low numbers. I think we are talking .2 to .3 percent potentially. Because we identified a weakness in a process, the actual number is hard to determine.”
The root causes at Northrop Grumman Newport News (NGNN) was that their welders were allowed to carry two types of weld wire on them, Hilarides said, and they could be from different processes. So one could be from corrosion resistant steel (CRES) and the other could be nickel cooper which would have copper in it.
“That had been their processes for a very long time. The weakness in the process is a relatively new finding,” he said.
A second issue was that a welder is allowed, if he knows he made a mistake, if he knows he put copper on CRES, to grind it off, etch it and fix it without telling the weld engineering folks, Hilarides said. “So there really is no record of the times that he did that.”
As a result of forensics and testing of joints, the Navy discovered that copper contamination, particularly copper in CRES, is a very bad combination, Hilarides explained. “We needed to be more aggressive in making sure there wasn’t copper in there.”
However, Hilarides pointed out that while a large amount of copper in CRES is bad and weakens the weld, in this case, the Navy found only a trace amount of copper in CRES.
“Once we found the weakness we then said we have to go look at all Virginia-class submarines. In fact we went through a process, we did a bunch of welds, we tested [them] to see what it meant. What we found was that no matter how much contamination you put in, if it made it through the initial testing, that the weld would leak before it would break,” Hilarides said.
Second, the Navy found that if welders put too much copper in, the weld would fail fairly early. That finding enabled the Navy to whittle down the number of valves and joints that would have to be inspected on each boat, Hilarides said.
Out of that subset of valves, inspectors found just a few with contamination: one on the USS Virginia (SSN-774) and one on the USS Texas (SSN-775). None were found on the USS Hawaii (SSN-776) or the North Carolina, Hilarides added.
“Once the copper is in there, it leaches into the CRES, so once you find the copper you just get it out…replace the valve or joint,” Hilarides said.
When the valves were further diagnosed, Hilarides said inspectors found that the cracking was not as aggressive as what had been modeled.
The repairs to Virginia will delay the boat’s operational evaluation by just a few days, Hilarides noted. The North Carolina was also delayed a few days for the beginning of its alpha trials.
The Texas is undergoing post shakedown availability and is not scheduled to take to the sea again until February, he added.
“The leadership of NGNN clearly recognizes that this is a loss of confidence in their ability to put submarines together, so they have been very aggressive in attacking this, in doing the forensics, in retraining their workforce, at stopping welding where it made sense and doing that training before they went on,” Hilarides said. “I don’t want to speak for the company but they have put in place better controls of their weld wire that we find acceptable in this light.”
NGNN has also instituted better reporting of welding problems so that things that were just allowed to pass without people knowing about them won’t happen, Hilarides added.
While in the short term the issue was all about allowing North Carolina to go to alpha trials and Virginia, Hawaii, and Texas to continue their operations, Hilarides said there is a longer term price to the weld issue too.
“Some of these welds will be weakened by the potential impurity and that over the life of a ship they may not meet full specifications,” he said. “The corrective actions in the long term are still due out. Our goal is to have those fully bounded and fleshed out by the spring, and they may involve checks of certain joints in the ship throughout its life.”