The engineer is human - lessons from Columbia

Materials World magazine
6 Feb 2012
space shuttle

Civil engineer Matt Melis of the Glenn Research Center talks in-depth about the findings of NASA’s post-Columbia Accident Investigation and the resulting lessons for industry. With 28 years of modelling and aeronautics experience, Melis still finds himself reciting to doe-eyed engineers that ‘it's not the weight of the material but the speed it’s travelling at’. Ledetta Asfa-Wossen learns more.



Q: What are the environmental conditions in space?

Space is as hard as it gets. In orbit, it’s essentially a vacuum, there is not much going on in terms of atmosphere. However, we have weather to consider on any given mission for both ascent and descent. Consequently, we have a dedicated meteorology team for shuttle launches, which maintains strict weather requirements for launch. Once you are in space, you don’t have any atmospherics to deal with but what you do have is radiation, microscopic dust, and micro-meteoroids the size of a paint chip or a grain of salt. These particles can impact your vehicle at extraordinary speed. When the vehicles return from space, we have a very extensive inspection process to see if any impact damage to their thermal protection materials has taken place and, if so, evaluate and replace those materials. When I do talks to engineers explaining impact dynamics, I emphasise that it really is not about how much something weighs, but how fast it’s travelling. Some micro-meteroid particles may strike an orbiter at 15-20,000mph so it doesn’t take too much thought to say, “yeah, that would cause some damage”.

Q: You were a key ballistics researcher in the post-Columbia STS-107 Accident Investigation. What exactly triggered the shuttle’s failure?

The root cause of the accident was a failure in Columbia’s wing leading edge thermal protection system (TPS), brought on by the impact of a large piece of insulating foam (debris), which separated from the shuttle’s external fuel tank during ascent.

Q: How common is debris and what are the associated problems?

Debris is a concern for both launch and return to Earth, so some shedding is expected in launching any space vehicle. Liquid hydrogen and liquid oxygen propellants are extremely cold and require thermal insulation prior to launch and then, of course, thermal protection materials are needed to withstand the expected aerodynamic heating on re-entry. These materials are prone to shedding as a consequence of launch loads. At the same time, you have a lot of ice forming and that adds weight. Regardless of the thermal insulation, ice formation can still occur, which can pose a serious debris threat. Debris associated with shuttle launches is a serious concern throughout the entirety of every mission.

When debris shedding occurs of a size or type outside the expected parameters, as was the case with the large foam separation on Columbia, there can be tragic consequences. The Columbia foam debris was about 1-1.5lbs, which separated near the bipod strut, a connection that joins the chin of the orbiter to the external tank. The foam was installed at that connection to prevent ice build up resulting from the extremely cold liquid hydrogen at the tank. With liquid hydrogen, you can get down to below zero, and it doesn’t take much imagination to know if you don’t insulate that, you are going to have an enormous boil off and loose your propellant. That’s going to cost a lot of money to replace.

Q: Foam is so light that it’s hard to imagine what damage an impact could have.

True, foam is very lightweight, but it’s the speed it was travelling at that made it dangerous to the shuttle. One quantifies kinetic energy, or energy in motion as one half of the mass times the velocity squared. The velocity component of this equation is what drives the energy up very quickly, not the mass. Of course, the higher the kinetic energy, the higher the risk of impact damage as that energy has to be absorbed or dissipated during the impact event. 

Q: Tell me more about what your role involved and what specific projects you were enlisted for post-Columbia.

I served as a technical lead of the NASA Glenn Ballistic Impact Team during both the Accident investigation and the Return to Flight programme, which followed for the two years after the investigation. Our primary responsibility was to characterise material debris threats to the shuttle and develop predictive computational models that could accurately predict and prevent future threats. The Columbia Accident Investigation Board’s responsibility was to establish the most likely cause of the accident and make recommendations to avoid a similar tragedy happening. The investigation lasted about seven months.

Q: How critical is the TPS in a shuttle design?

The shuttle is a unique spacecraft that has the ability to go up to orbit like a rocket and come back like an aeroplane and land on a traditional runway. As a consequence of this the vehicle is reusable, a lot of space vehicles are designed to be used for one mission only, but the orbiters were designed to be reusable for many missions. Consequently, the TPS had to be resilient and largely reusable on the shuttle, unlike the ablators used on the Apollo missions. When we use ablators as a TPS, they erode and burn off during re-entry, and hence need to be replaced after every flight should the vehicle fly again. The orbiter’s TPS don’t have ablators, but use refractory tiles, thermal blankets, and reinforced carbon-carbon (RCC) as its primary TPS. The tiles largely cover the belly of the orbiter, RCC protects the leading edge and nosecaps, and the thermal blankets cover the remaining areas. RCC sees the highest vehicle temperatures on re-entry, tiles are the mid-level temperatures, and thermal blankets see the lowest thermal loads. When you get on top of the orbiter, the thermal blankets are the coolest part. If you get up close to it, the blankets look like a heated quilt. The closer you get the more handmade it appears.

A round 40-inch aluminum storage tank from space shuttle Columbia’s power reactant and storage distribution system rests on the edge of Lake Nacogdoches in Texas. Lower lake water levels due to a local drought allowed the debris to become exposed. Approximately 38 to 40% of Columbia was recovered following the accident in a half-million-acre search area which extended from eastern Texas and to western Louisiana

Q: What components and materials used in the space shuttle did you look at during the investigation?

We looked at every type of material that might liberate as debris during launch and potentially impact the orbiter. We also looked at ice and the behaviour of ice as it commonly forms on the tank. You don’t think of ice warming up but it really does, so we had a very controlled process for testing the ice and its impact dynamics at tightly controlled temperatures.

Q: What is the main factor to consider when looking at materials for space shuttle design?

Weight. We have to maintain safety, integrity and durability while keeping our structural weight to a minimum. Weight is absolutely critical as you are compromising payload every time you bump up the weight to meet other requirements. The driving force for materials innovation in this discipline is about maintaining durability, integrity and flight safety while at the same time optimising your structural vehicle weight.

Q: What advances have there been in materials used in space shuttle design?

We are seeing tangible improvements in materials to be excited about in the current era, particularly in high-temperature composite materials. The shuttle was built using 1970s technology and many of those materials, such as the wing leading edges, remained in use for the full 30 years of the programme. One of the things you have to realise is that certifying a composite for space flight, or any new material for that matter, is an extensive, rigorous, timeconsuming process. The more critical the material, the more thorough the investigation and certification requirements.

Since shuttle has been flying, NASA has been looking at new materials over the last 20 years to reduce weight while maintaining or even improving safety. Many modifications have been made over the years however, few, if any, radical changes have been made with materials that would affect the shuttles original design, such as the leading edges for example. Since the original shuttle design, materials superior to RCC have come about, but the flight certification and development process is too prohibitive to employ them on shuttle as a new leading edge thermal protection system. As our next generation of vehicles are now being designed, we will begin to see many new exciting materials being used in those applications.

Q: What is the key lesson to take away from Columbia?

It must be pointed out that Columbia and her crew were lost to a design flaw that was embedded in the vehicle in the late 1970’s and it was flown for nearly 25 years before it stung us.

There are dozens of lessons that came out of Columbia but the most important one is that when you are involved in an engineering operation of complex systems and this doesn’t just apply to aerospace engineering, but any complex engineering endeavour, design flaws can remain hidden for years and suddenly surface, causing critical failure. This is a painful reminder that we always need to stay sharp. That’s the key lesson that came out of Columbia for myself and many others. To look further into this, I suggest reading the Columbia Accident Investigation Report, which is an excellent resource for the interested reader.

A great book to read is The Engineer is Human by Henry Petroski. It’s a bit old now but it’s about a civil engineer commenting on the lessons learned from the most notable engineering disasters in history. It was written before Columbia but covers Challenger and disasters such as the collapse of the famous Tacoma Narrows bridge which occurred because the resonant response of bridges under wind loading was not fully understood at the time. Petroski states in the book that the lessons learned from these significant engineering tragedies have done more to advance engineering knowledge than all the successful machines and structures built. I am inclined to agree with him. Our greatest disasters and tragedies are not positive events but they have provided an opportunity to understand and gain knowledge as how to avoid repeating such events in the future.

Further information