“C’mon, we’re not interpreting 733t-speak here!”
w00t!
I doubt that this was a phrase in the 1950’s. Perhaps the 1970’s.
When I studied engineering, somebody once put it this way: engineers use pre-prepared tables of numbers to choose pre-prepared products and build new things. Scientists do new things without the pre-prepared lists"
He was talking about civil engineering design–calculating weight loads on a roof and choosing the right structural steel beam to support it. The loads are routine calculations based on columns of numbers printed in approved tables*.And the steel beams are made in pre-prepared sizes by the manufacturers. The engineers job is to match the numbers and make it work.
A scientists job is to create new numbers (non-standard loads and non-standard shapes of steel) and make it work.
(* this was back in the days of slide rules and logarithms. Gee, I’m old )
I remember a Far Side cartoon showing the doctors standing around a missile laying on the operating table, and saying “C’mon guys, this isn’t brain surgery”.
The problem with this notion is that “engineering” encompasses such a broad range of activities, that using “pre-prepared tables of numbers” is not necessarily the norm. Research and development, for example; the study of polymers, for example; the study of hydrology, hydrogeology … the list can go on and on.
Absolutely, some areas of engineering such as structural engineering are all about designing structures with materials posessing measurable qualities.
Not all engineering is that. 
Personally, I use established equations and methods of analysis to predict the results of rainfall/runoff (to put it in a nutshell). Do I consider myself both an engineer and a scientist? Yes, I do.
I’m an engineer working in the rocket propulsion/aerospace field. My take on the difference between rocket science and aerospace engineering is that the former focuses on fundamental understanding of what goes into making a rocket work; this would include not only the functioning of rocket propulsion engines and relateded phenomena but also guidance, control and navigation (GNC), linear and nonlinear control feedback theory, et cetera. An engineer, on the other hand, is more concerned with the details of making a particular system work. There’s certainly overlap–I am (or was before funding ran out) collaborating on some work regarding slag accumulation in solid rocket motors and modeling its influence on the flight dynamics of a booster–but most of my work involves pretty mundane stuff like assessing ground test and support equipment, structural analysis of motor interfaces, looking (often literally) at the nuts and bolts that comprise an actual rocket, and making slides for an unconsciounible amount of PowerPoint presentations. While I’m versed in the fundamentals of what makes a rocket go and why you can only squeeze so much performance out of any given fuel/motor/nozzle configuation, I am in no way advancing the basic understanding of rocketry.
Rocket science is hard because there is so much more to it than just stuffing propellent out the back, and even with advanced fluid modeling software and processing power available today it’s very difficult to get a clear idea of the actual conditions inside of a rocket motor; the conditions are so extreme and difficult to monitor that we have to make a lot of inferences. There is a lot of phenomenolgy–that is, knowledge of the effects of different areas of study influencing the final result–in propulsion science, and so it takes someone who is really good at learning and integrating knowledge of different fields.
On the whole, I’d say that neuroscience is harder, though, or at least, more difficult to become accomplished. Understanding how the brain works and what the effects of different illnesses and trauma are is an intimidatingly difficult area of study.
Stranger
Still engineering, just research engineering.
As a neutral bystander–I’m neither a scientist nor an engineer, so take this with a grain of salt–in my opinion engineers are indeed scientists, but applied scientists. Isn’t that different from someone who is doing original research in a scientific discipline?
I guess I see them the other way. Rocketry has been solved. Yes the tolerances are small, but the solution is known.
Orbital mechanics is largely unsolved. There are solutions for the simplest cases, but anything involving multiple planets requires a combination of wise guesses and plenty of computation time. Especially if you want to optimize for fuel or time.
Of course, once the orbit has been determined, it comes down to rocketry to execute…
In my feeble way, this is what I am trying to express - there are engineers in research who develop new ways of looking at old problems - for example, the work in the late 80’s in materials science; research engineers who came up with new materials, composites etc.
There is definately a branch of engineering which is out there building roads etc. “by the book”. There are also a lot of engineers working behind the scenes in a more scientific mode. Engineering to me does not appear to be a static thing; university professors and private researchers constantly come up with new materials, radical ways of looking at old problems, breaking through barriers, coming up with new ways to tackle problems.
IMHO, Science and Engineering are inexorably linked. YMMV. 
I don’t think I know what you mean by either of these statemetns. “Rocketry” has not been “solved”. The art and science of rocket propulsion is in continuous development; while the motors and hardware we have are considerably more advanced in materials and structural capability than those sixty years ago, we have great strides to make before we can even consider chemical rocket engines a mature, reliable technology, much less more advanced (electric/ion, nuclear steam, magnetoplasmadynamic, et cetera) propulsion systems.
Orbital mechanics, on the other hand, is a pretty mature field, such that we can (assuming the booster functions per mission parameters) place a satellite in a precise orbit and make a reliable estimate on how long that orbit will remain stable. Heck, we can even plot orbital maneuvers with multiple swingbys around Venus and Mars (see the Cassini-Huygens mission) with high reliability. It’s true that there are no general solutions to the N-body problem, but that’s not because there’s a lack of understanding about the underlying theory of celestial dynamics, but because the problem just doesn’t allow for closed-form solutions. With a modern desktop computer and commercial software like STK/Astrogator you can plot whole families of trajectories in minutes and optimize for any given set of non-contradictory parameters.
Stranger
I thought rocket science was botanical engineering…
We’ll just have to disagree on this. Developing new types of rocket engines or improving the reliability of old may be difficult, precise work, but it doesn’t change the fact that the basics have been solved.
I suppose the same could be said for orbital mechanics, since gravitation is well understood. We can compute any given orbit to high precision and reliably send a vehicle along that solution.
But there’s a near-infinite number of ways of getting that vehicle from here to there and, as you said, no closed-form solution for finding the “best” one. The parametric trade studies done by off-the-shelf software makes many simplifying assumptions. Active research is still being done in this area, since even a tiny improvement in the orbit (and I’m talking about interplanetary orbits here) saves significant amounts of fuel. (For those not familiar, fuel/mass is one of the major costs and constraints of spacecraft.)
Full disclosure: my spouse does this kind of optimization research for JPL, so some bias may be slipping through as to its importance.
Disclosure statement: I call my self a rocket scientist, because it is easier than explaining what I do and it always draws a disclaimer. My Ph. D. dissertation was in physics - General Relativity, so I’m biased towards being a scientist.
But. I’m not really. Kimera’s dad summed up the view of the physicists I knew or know. I help build things, therefore my job is engineering. When I was in grad school, I was helping to figure out how the universe works, therefore I was a scientist. Vocationally, emotionally, or intrinsically, however you want to put it, I am a scientist. My brother the engineer really cares how to build things. I just build things to provide my family with the kinds of things my scientist dad could not.
If your spouse works for JPL, he is probably a scientist, but on the broad fuzzy line between hardcore civil engineers and string theorists. He is helping to build things in order to understand how the universe works. Whereas, I help build things in order for someone else to use it.
I have a friend who says anything that calls itself a science, isn’t. Rocket science and computer science being prime examples.
heh, that was supposed to be a “chuckle”, not a “disclaimer”. Durn, coworkers and their actually expecting me to do something.
I can’t disagree more. “The basics” of propulsion can be described as well understood only at the most general level. Specific areas of phenomena like combustion instability, dynamic feedback, active unstable control systems, variable thrust chamber and nozzle geometry, transitional exhaust dynamics, et cetera are all active areas of essentially basic research, often distantly separated from operational hardware. This doesn’t even get into advanced propellant research and non-chemical propulsion systems. Saying that “the basics have been solved” is like dismissing gravitational research based upon the fact that we have Newton’s laws to work from.
Nor will there ever be a general closed form solution; that’s the nature of the problem. This alone doesn’t make it “science”, though; most engineering problems don’t have discrete, analytical solutions. Generally speaking, if you’re performing activities under the moniker of “trade studies” and “optimization”, you’re well into engineering (that is, science applied to the economical solution of specific technical problems) rather than research science, regardless of what the title on your door placard says.
Stranger
Stranger, I think we’re agreeing on the facts and disagreeing on interpretation.
“The basics” of propulsion can be described as well understood only at the most general level.
And I have been describing it at the most general level. Rockets are the things that kick mass to produce thrust. Modern rockets do this so reliably that the details of how don’t need to be considered when computing orbits. Your basic mature technology.
Specific areas of phenomena like combustion instability, … are all active areas of essentially basic research, often distantly separated from operational hardware.
Yes, that research will improve rocket engines. And I would tend to categorize them as separate fields of endeavor (unless they are only applicable to rocketry).
This alone doesn’t make it “science”, though; most engineering problems don’t have discrete, analytical solutions.
Yes! I did call orbital mechanics “more art than science”. That is, engineering.
I’m staying out of the endless engineering vs science argument, it’s more semantics than substance, and the boundary is so fuzzy as to be non-existent. (Me and my spouse are both scientists by training (PhDs in physics) and engineers by profession.) It’s a fun pissing contest, but nothing more.
I’m torn between being amused and aghast by this statement. Rocket boosters are about as far from being “your basic mature technology” as an abacus is from being “an advanced desktop computer”. 95% reliability from a particular booster design is considered excellent reliability for an unmanned system. (If you had a 1 in 20 chance of surviving a commercial airline flight, or even sending a parcel by air, I guarantee that you wouldn’t consider the technology to be “mature”.) In this [thread=390356]thread[/thread] Doper Jurph succintly outlines the basic reliability issues with solid and liquid fuel boosters. As it happens, I’m currently reading Space Systems Failures, in which it is clearly illustrated how unreliable rocket systems are on a mission-by-mission basis even after a long development and launch history.
As for not needing to know “the details…to be considered when computing orbits,” this is so far from reality I’m not sure where to even begin addressing this. Achiving a particular orbit for a specified payload requires careful attention to the performance characteristics and aerodynamics along the entire range of flight; you don’t simply figure out how much overall thrust and average specific impulse you need and then pluck a matching booster off the shelf. The selection of engines, aero loads on the fairing, dynamic flight loads, plume effects, et cetera are all very important and have to be considered in order to reliably make orbit with a functional payload.
Yes, these areas are specifically applicable to rocket boosters. The environment inside of a rocket engine and controls necessary to correct and guide a booster to orbit are well and beyond anything found in transportation equipment found in eveyrday life (i.e. automobiles and jet aircraft). There is plenty of basic research science in materials, fuel and propellant, chamber and plume aerothermoelastic modeling, et cetera that is way beyond mere refinement of the state of the art.
Sorry, I don’t regard it as a pissing contest; however, when you make factually incorrect statements of the reliability of existing rocket systems, I feel compelled to correct this. While there is certainly a grey area between basic science and engineering (falling into the spectrum from “applied science” to “development engineering”) the distinctions are more than just semantics; when people complain about how NASA can’t perform the simple task of keeping the Shuttle running, it’s because they don’t appreciate the fact that the STS turned from what should have been a developmental applied science problem to a strictly operational engineering endeavor, with a different and often inappropriate approach to problems and issues. When the public questions the value of interplanetary missions on the basis of a fiscal return, they’re failing to understand the value of “pure” science, i.e. that it gives us a better fundamental understanding of the natural world upon which to build technological applications. There is a worthwhile and necessary distinction to be made here.
Stranger
Thanks for the links, I’ve skimmed through them. I was thinking more of orbital transfer systems rather than the launch systems (which is apparently the focus of everyone else). You are correct about the reliability of launches.
Can you recall any failures in interplanetary orbit transfers?
Here’s a select few (quotes from the linked article unless otherwise noted):[ul][li]Surveyor 2: “A mid-course correction failure resulted in the spacecraft tumbling and losing control. The spacecraft was targeted at Sinus Medii, but crashed near Copernicus crater.”[/li][li]Surveyor 4: “This spacecraft crashed after an otherwise flawless mission; telemetry contact was lost 2.5 minutes before touchdown.” (Harland mentions that, “It is possible that the motor had some kind of structural fault that cuased the solid propellant to explode,” but I don’t see any official findings to that effect.)[/li][li]Viking 1: Although unmentioned in the Wikipedia article, the Viking 1 spacecraft had a major propulsion system anomoly, resulting in an overpressure condition due to a leak in the regulator that controlled the ullage system (which pressurizes the fuel and oxidizer, making it flow). The problem was resolved with an ad hoc solution in which helium as slowly bled into in the tanks to control pressure spikes. The mission completed successfully, but there was a high risk of mission failure.[/li][li]Mars Observer*: “Contact with Mars Observer was lost on August 21, 1993, three days before scheduled orbit insertion, for unknown reasons and has not been re-established…It was speculated that there may have been an explosion in a propellant line during pressurisation procedures just before the orbital insertion engine burn. It is believed that hypergolic fuel may have leaked past valves in the system during cruise to Mars, allowing the fuel and oxidiser to combine prematurely before reaching the combustion chamber.”[/li][li]Galileo: Although unmentioned along with other anomolies in the Wikipedia article, from Harland: “…the main vehicle made a manoeuvre on 27 July…In preparing the engine for this deflection burn, the flight controllers were puzzled by a telemetry indication of a difference in pressure between the fuel and oxidiser tanks…As the risk was of temprature changes in the propulsion system forcing nitrogen tetroxide into the pressurisation system, the flight controllers initiated a programme of thermal management to avoid such thermal pumping by using small electrical heaters to minimise temperature differences ascross the propulsion system.”[/li][li]NEAR*: From The Final Report of the NEAR Rendezvous Burn Anomoly Board, Applied Physics Laboratory, Johns Hopkins University, Baltimore, November 1999: “Almost immediately after the main engine ignited, the burn aborted, demoting the spacecraft into safe mode. Less than a minute later the spacecraft began an anomalous series of attitude motions, and communications were lost for the next 27 hours. Onboard autonomy eventually recovered and stabilised the spacecraft in its lowest safe mode (Sun-safe mode). However, in the process NEAR had performed 15 autnomous momentum dumps, fired its thrusters thousands of times, and consumed 29 kilograms of fuel (equivilent to about 96 meteres per second in lost delta-V capability.)…After reacquisition, NEAR was commanded to a contingency plan…The make-up burn placed NEAR on a trajectory to rendezvous with Eros on 14 February 2000, thirteen months later than originally planned. The remaining fuel is sufficient to carry out the original NEAR mission, but with little to no margin.”[/ul]The list of failures and near failures from orbital propulsion systems goes on and on. The environments and long-term hibernative durations of onboard propusion systems of interplanetary craft are so far out of normal experience and the ability to perform integrated testing that it’s often impossible to foresee potential problems, and it’s great kudos to the engineers in being able to come up with innovative and successful recovery schemes on a machine that they can’t examine or test beyond very limited onboard diagnostic systems. If anything, the reliablity of interplanetary propulsion systems is far worse than that of orbital boosters. Neither should be considered a mature technology, even in the way advanced commercial aircraft are. [/li]
Understand, I’m not denigrating people who do orbital ballistics and trajectorty plotting, which is a difficult and no doubt stressful task. But it is not, for the most part, basic science; it’s the integration and application of known principles to maximize mission capability while mitigating risk. There are many areas of rocket propulsion and controls research, however, that are essentially basic research. One is not “better” than the other (and I think playing around with orbital mechanics sounds like a lot more fun that trying to formulate analytical models for turbulent flow in a rocket engine) but they do have distinctly different approaches and general methodology.
Stranger