March 12, 2016: by Stockton Rush
When OceanGate began the Cyclops project there were three main areas of innovation we planned to pursue:
The creation of a carbon fiber hull has always been the most technically challenging area of innovation. As we began this project, there were two main questions we needed to answer:
The results of the fiber placed carbon fiber hull design project we conducted with Boeing in 2014 gave us confidence that carbon fiber can be used in a simple cylindrical hull. However, achieving the second goal, accurately detecting when we are approaching hull failure, has been a major question as it is critical that we have this ability in order to build and operate a safe and useful hull out of a carbon fiber. As our vessels will be operated on ships at sea and will be transported over land and sea they may see extreme temperatures as well as high shock loads and impacts that may not be observed, reported, or detected prior to a dive. We must have certainty that a degraded hull can be discovered before it poses a risk to its crew.
Five months ago we decided to attempt to use a Filament Wound Carbon Fiber (FWCF) structure for the hemispherical endcaps. If we were successful this capability would have huge value for a number of military applications and could lead to greatly reduced manufacturing costs for OceanGate. During our first pressure test, the first set of hemispheres failed at 4,300psi (the same pressure at almost 3,000 meters depth). We then tested just the carbon fiber cylinder using thick flat aluminum end caps in January and succeeded with that structure taking it to 6,000psi (4,100m).
On Friday evening we tried another test with a new set of FWCF end caps. The goal of this test was not just to test whether we could make FWCF work in a hemisphere, but, more importantly, to test acoustic and strain gauge monitoring to see if we could detect an impending failure well before it happened. We contracted with Boeing Research and Technology to assist in the test as they had for our second test, but with many more sensors this time. We had 9 acoustic transducers “listening” for the pop of carbon and/or resin failing. In addition, we had 16 traditional mechanical strain gauges.
The first noticeable acoustic event occurred at 3,000 psi, a few more at 3,500psi and then, after about 30 seconds holding at 4,000 psi, we saw a cascade of events and then a loud “bang” as one of the endcaps imploded. As this endcap had been improved from the prior ones that failed in December, we were surprised that it failed at a lower pressure – that is the bad news.
The good news is we collected excellent data from both the Boeing sensors as well as a system we are developing internally for use on production hulls. Both of these systems saw activity well before failure. This is the result we wanted (and needed). The end cap failure merely means we will temporarily abandon the FWCF endcap project and continue with metal hemispheres that are well understood and predictable.
Our next step is to commence the build of the full size pressure vessel. We now have confidence in the main cylindrical hull and in our ability to detect a failure well before it happens. As this full-sized hull construction progresses we will continue to test the 1/3rd scale cylindrical hull. We want to obtain data from many cycles of loading and unloading to assess potential life cycle issues and collect more validating data from our new acoustic monitoring system. Not surprisingly, the strain gauges were of little use as they only showed signs of imminent failure about 2 seconds before it happened.
While we prepared for the test, a film crew interviewed me and asked me who I admired. I mentioned Burt Rutan who is one of the greatest aircraft designers of our time. His low profile persistence and ability to make “out of the box” designs for very unique needs – like Spaceship One or the round the world Voyager aircraft is impressive. This morning I came across a quote from Burt:
“Testing leads to failure. Failure leads to understanding”
As I mentioned to one of our supporters – any test that yields actionable data and expands understanding is a good test.
On Friday night we had a very good test.
(425) 595-5014 email@example.com
Weight: 275 lbs.
Displacement: 8.2 cu. ft.
The pressure vessel was designed and fabricated in collaboration with the University of Washington Applied Physics Lab Collaboratory, and Spencer Composites.
This pressure vessel model is approximately 1/3 of the size of the full sized vessel. The walls of the scale model are made of multiple layers of carbon fiber material and resin. The model is assembled from three main components: a center cylinder, and two 6" thick aluminum end plates. The full sized vessel will include a carbon fiber center cylinder and a hemisphere with viewport on each end of the cylinder.