The challenge of making sound engineering decisionsby Eric R. Hedman
|
| Given the fact that the specifications for the Orion spacecraft are fairly clear, what is the best approach for selecting a design for the launch vehicle to carry it into orbit? |
The basic goals for the Ares 1 and the Orion capsule are rather clearly defined. It is, at first, to carry up to six astronauts to the ISS and return them after a long stay. Later evolved versions are to carry four astronauts to rendezvous with the Lunar Surface Access Module (LSAM) on the Earth Departure Stage (EDS), go to lunar orbit, stay in orbit until the astronauts return from the surface, and then return directly to Earth. These are clearly defined goals that make a great starting point. With the size of the Orion capsule and service module along with the orbit it needs to be placed in, the performance specifications for the Ares 1 fit a well-defined envelope.
Given the fact that the specifications for the Orion spacecraft are fairly clear, what is the best approach for selecting a design for the launch vehicle to carry it into orbit? From what has been made publicly available it sounds like the plans started with the “stick” configuration that has evolved somewhat since it first announced. It has since leaked out that other configurations are being looked at as backups in case the stick design is determined to be unworkable. The options include a configuration with two three segment SRB variants, possible man-rated versions of the Atlas 5 and Delta 4, and a concept known as Direct Launch.
Starting a new project is the time for the first major brainstorming session. No matter what the project is, no idea should be off the table. This normally will and should produce a wide variety of concepts that range from the tried-and-true to the outlandish. In a project of this size and scope, any credible source for these ideas should be included. The “not invented here” prejudice that is quite common in many engineering departments really should have no place in an agency where all citizens are stockholders in the adventure. No idea should be proclaimed the frontrunner until some discussion and analysis has been done.
The first cuts of engineering designs often have nothing to do with what is technically the best option. They usually have to do with parameters given to the engineering staff at the start of a project. These usually include the available budget, the required timeline, and how much risk in development is acceptable. Political factors always come into play in both government and private projects and they usually start from the beginning of the project. People and organizations almost always have agendas that have nothing to do with the main goals of the program. The success of a program may depend on how close these agendas have parallel goals with the major program goals.
Management style is also very important to the process for a number of reasons. A manager dismissing other people’s ideas without explanation is a recipe for creating disgruntled employees. I’ve discovered that when making decisions for a team, it is important to explain the reasons behind them whenever possible for at least the major decisions. In most cases it defuses problems caused by team members not agreeing with it. It helps the team buy into the decision and more effectively carry it out.
Ego and ownership of an idea can seriously cloud one’s judgment as how to proceed with a project. If a key individual was the one who came up with the current option, they are usually very reluctant to allow an alternative, even when it’s a better choice. They won’t go down in history as the key innovator. When the whole nation owns all the stock in this project, ego needs to be pushed aside for the good of the taxpaying public and the success of the project.
From the start the Ares 1 development is different then almost any other project on the planet. Not only is it being designed to carry the Orion capsule into orbit, it is part of a much larger, less clearly defined project. It is also being developed in a metaphoric fishbowl, being observed from all sides by people with radically varying interests and backgrounds. This has only been magnified by the emergence of the Internet and people (like me) who feel compelled to offer our two cents worth about it. This also means that NASA needs to be aware of this new environment as they move forward.
NASA has been criticized for their choice of the stick design for the Ares 1. Some of this criticism is because NASA has not been good at explaining why they have been making various design decisions, although I have to admit that they have been getting a little better at explaining as of late. Communicating the reasoning for their major decisions to the public is every bit as important explaining them to their own engineering teams. The public, after all, is paying for it.
| Communicating the reasoning for their major decisions to the public is every bit as important explaining them to their own engineering teams. The public, after all, is paying for it. |
I recently read the Congressional Budget Office report comparing the possible options for both the Ares 1 and the Ares 5 Cargo Launch Vehicle. The report compares the options with respect to cost and development time. One of the possible options for launching the Orion spacecraft has been the Atlas 5 family of launchers. Like the other options it has both good and bad features, creating a balance sheet for ranking it as a possibility. On the plus side it has relatively new technology. It has a perfect record of successful launches. On the flip side, the current configurations don’t have the lift capacity to lift the Orion capsule and its launch profile would put extra loads on the crew. A variant would have to be developed to enhance its lift capacity. A new launch pad may need to be built or an existing pad at KSC would need to be heavily modified to support a manned capsule and provide an escape system for the crew in a pad emergency. While the report doesn’t recommend a specific choice, it does make it obvious that there is no solution using EELVs that would radically speed up development or reduce costs by an order of magnitude.
Lockheed recently posted on their web site information on how the Atlas 5 can evolve into a human-rated launch system. This information was not in the CBO report, and is definitely worth readying. It opens up new possibilities for a wider range of options for human access to space.
The Ares 1 design has its pros and cons. It does make use of some of the technology in the Shuttle, most notably the solid-rocket boosters, albeit a new five-segment design. It is being designed upfront with the lift capacity to handle the Orion capsule and service module. However, it requires all new avionics and a roll control system that is being added at the interstage area. It also requires a new mobile launch platform, major modifications to one of the Shuttle launch pads, and modifications to the work platforms in the Vehicle Assembly Building (VAB). The matrix of factors to come up with decisions for a myriad of choices is incredibly complex. Making the best decisions when the choices are so complex comes down to experience and wisdom including listening to a variety of opinions within the engineering team and outside opinions that are backed up by rational arguments.
A few articles have stated that the alternative incorporating two three-segment SRBs is being studied in case there is a showstopper with the stick design. There are several streams of action that are dependent on the choice of launcher so a final decision has to be made in a timely fashion. Once a choice is “final” several actions that are hard to reverse will begin, including the modifications to the launch pads and the VAB.
So are there showstoppers with the stick design? There are a few possibilities. The stick is called the stick because it is so tall and thin. There are reasons why a tall and thin rocket is a problem, starting with the dynamics of such a shape. The ratio of height to body diameter is 18 to 1. Control of a rocket is done by gimballing the nozzle of the rocket motor to keep it on the correct path, adjusting for deviations that are caused by crosswinds and other forces during launch. For those of you that have not studied motion control, consider this analogy. If you set a tall glass or a wine bottle on the outstretched flat palm of your hand without gripping it, you should not have much of a problem balancing it vertically. Now try it with a taller, thinner item, such as the extension tube from a vacuum cleaner. You’ll find it significantly more difficult. You’ll notice that you’ll have to move your hand around to try to keep the tube vertical. This is exactly what a rocket’s control system is doing to keep the vehicle on course. To complicate things, try balancing the tube on the palm of your hand outside in a windstorm. That is what a tall, thin rocket is facing as it climbs through the atmosphere.
Two more issues come into play to a greater extent with a rocket of this design. Before staging, the center of gravity of the vehicle moves upward as the fuel is burned away in the first stage, adding to the control problem. In addition, the longer and thinner a vehicle is, the more it will flex as forces are applied to it. Instead of a vacuum cleaner extension tube, try balancing something more flexible while you’re standing in the same windstorm.
Roll control is also a bigger problem with vehicles of this type. The solution that seems to be on the drawing board is to add a roll control system at the interstage section between the first and second stages. It has thrusters that can twist the vehicle around the vertical axis as needed. While this is a workable solution, it adds weight and complexity while reducing payload capabilities.
The last, and possibly biggest, problem with the stick design is its lift capacity. According to some rumors—but denied by NASA—it may not be able to carry the Orion capsule to the desired orbit. This problem may not even have a bad workaround. This could definitely be a showstopper for the Ares 1.
| If NASA management won’t seriously look at the Direct Launch proposal, I’m asking Congress to do their job as the “board of directors” of our government. This decision is crucial for the future of the US manned space program. |
The two three-segment SRB alternative, which is shorter and squatter, has some advantages when it comes to these issues. Being shorter and wider reduces some of the control issues. With two boosters set wide apart there is no need to add a separate roll control system. The engines can be twisted in opposite directions to roll the vehicle. Being set so wide apart, only a slight twist of the engine nozzles can apply enough torque to control the roll of the vehicle. This alternative also has advantages in that it doesn’t need as many modifications to the launch pad or the VAB. I’m sure this option also has disadvantages that I’m not aware of, possibly in recurring costs per mission.
The real interesting alternative to me is the Direct Launch concept, proposed by Ross Tierney. It is a proposal to develop only one vehicle that would be capable of launching the CEV, the LSAM, and the EDS on two or three launches. It truly makes use of elements of the Shuttle to significantly reduce the development time and costs for creating a capability for going to the Moon. Details can be found on the proposal’s web site. I would encourage anyone to take a close look at it. If this is a feasible concept (and I am not an expert at making this judgment), it should be seriously looked at. I understand that this concept is the favored concept of a number of engineers and mid-level managers within NASA. If it works, it would free up over the next decade or so up to $35 billion that could be shifted to speed up development of the EDS and the LSAM.
The direct launcher has several advantages. It requires the development of only one launcher, not two. It truly makes use of Shuttle technology. It does not require massive changes to infrastructure. It can be developed faster. It does not have the performance problems the stick does. It could also launch new modules to the ISS. It could reduce the gap between the retirement of the Shuttle and the resumption of manned US flights. It will shorten the time until humans put another footprint on the Moon. It would save the taxpayer money. It could be used to launch a wide variety of new heavy payloads if the need arises for them.
If NASA management won’t seriously look at this proposal, I’m asking Congress to do their job as the “board of directors” of our government. This decision is crucial for the future of the US manned space program. Don’t let the design be finalized before know that a potentially much better option wasn’t considered. If this proposal is dismissed without serious consideration, NASA may lose the support and confidence of the many space enthusiasts that pester their representatives in Congress who, in turn, help keep NASA funded. I can’t say if the Direct Launch concept is the best ultimate choice, but I do think the concept need a fair hearing before irreversible changes to NASA’s infrastructure are started.
I’m also a little puzzled as to why the Ares 1 was down selected to only one option before all possible technical showstoppers were eliminated. I’m sure the CEV competition between Lockheed Martin and the Northrop Grumman-Boeing team spurred greater creativity before the final selection. For the public to trust that the decision is being made with wisdom, it needs to be as open and transparent as possible. The American public often disbelieves that government decisions are made in the best interest of the public and the goals of the programs. The Army Corps of Engineers’ work on the levees in New Orleans comes to mind. Also recall the old joke in defense spending that the perfect weapons system is one that has parts built in all 435 congressional districts. It didn’t matter nearly as much how well the final system works.
For the public to trust that sound engineering decisions are being made, it is absolutely necessary that NASA effectively communicates with the interested members of the public. They periodically need the decisionmakers at all levels to be available to the trade and general press to explain what they are doing, how they make their decisions, and why they have made them. Having to regularly explain to the public the rationale for their decisions would, in my opinion, help ensure that NASA personnel make sound engineering decisions and increase the likelihood of success. I, like most Americans, really am impressed with many of the things that NASA accomplishes and do want to see them succeed spectacularly.