Need a rocket scientist...

mrmorris · Jul 25, 2004

Except for the fact that they always have been -- why are current boosters generally designed 'tall and thin'? I ask this because I was recently pondering the Falcon I and thinking that if it were shorter and squatter, then less material would be required to enclose the same volume of propellant. I started to calculate it out, but had to make lots of assumptions since they didn't post engineering diagrams on their site for my convenience. Given my plethora of WAGs, however -- I worked out that increasing the diameter of Falcon I from 1.7 meters to 2.5 meters would reduce its height by ~8 meters and drop ~388 kg of aluminum required for the outer shells of the two stages. Since this is a significant fraction of the payload weight, it would seem that if this could be done, it should be done. Except nobody does it. Which leads me to think there's a significant engineering gotcha that I'm missing. My best guess is that it's flight stability. While having engines at the bottom is inherently unstable (hence Goddard's "engine at the top" rockets), the longer and thinner the rocket, the larger the percentage of thrust that runs through the center of mass and the less unstable it is. However, if this is the *only* reason, then it would seem to be an obsolete one. While the computer power that existed at the dawn of spaceflight might have been incapable of keeping larger-diameter boosters stable in flight -- that shouldn't be the case today. Given the current state of processing power -- if flight stability were the only issue, I would expect at least somebody to be experimenting with a boosters whose shape more closely approached a sphere (i.e. the minimum surface-area to volume ratio). So -- are there other reasons that I'm missing?

nacnud · Jul 25, 2004

I think you have missed out aerodynamics. Idealy rockets sould be long and thin to reduce the frontal area and therefore the drag. If you look at model rockets that are designed for maximum height from a relatively small engine they have much higher aspect radio (lenght to breath) than orbital capable rockets. This is due to the increased importance of aerodynamics in the smaller rockets as their entire flight is within the atmosphere.

drwayne · Jul 25, 2004

"While having engines at the bottom is inherently unstable" Actually, it not inherently unstable, its actually what is called neutrally stable. Many people are thrown off by the balancing a pencil example, which is not at all a good representation. Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

nacnud · Jul 25, 2004

I am not a rocket scientist, IANARS new acronym <img src="/images/icons/smile.gif" /> but another advantage of tall and thin is that it actualy should be easier to balance. Try balancing a pencil on your finger compared to a broom. The longer broom has a slower rate of rotation and you therefore there is more time to correct the instablity.

drwayne · Jul 25, 2004

Warning: The example of balancing a pencil or a broom leads to some bad conclusions. Why? Because in that example, the force you apply is only aligned with the axis of the object when it is straight up and down. If it topples at all, you have an imbalance. In the case of a rocket, the thrust is always aligned with the vehicle axis (give or take a few degrees of gimballing of the nozzle.) Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

nacnud · Jul 25, 2004

good point, knew I should have kept my mouth shut

mrmorris · Jul 25, 2004

"... to reduce the frontal area and therefore the drag. If you look at model rockets ..." I'm not an aeronautical engineer, and I've certainly had no experience with drag (I can't even imagine learning to walk in heels). However -- I don't believe that your arguments will hold up to mathematical analysis. First of all -- the increase in cross-sectional area that I calculated is fairly minimal -- 1.7 meters to 2.5 meters. The rocket is still aerodynamically shaped. Also -- there is still frictional drag along the sides of the rocket -- not only the nose. The shortened length of a stubbier rocket would have a reduction there to counter somewhat the increased amount at the leading edge. Also -- the ideal shape to minimize drag is *not* a long-thin bullet-nosed cylinder. It's actually a teardrop shape -- with the blunt portion of the teardop leading. The stubbier rocket I've proposed is actually closer to the ideal than the traditional long-thin rocket. That's not to say that I think that short & stubby would have a lower drag, but it's certainly not a foregone conclusion that it has a lot more. Model rockets have no place in this argument, as they have zero relationship to the principles involved. 1. The vast majority of a model rocket body is dead space. Increasing the volume of the model rocket doesn't affect the propellent capabilities *at all* and therefore any increase in the diameter of the rocket body beyond the diameter of the engine is just adding more dead space. 2. Model rocket engines aren't mounted on gimbals and can't act to stabilize the flight of the rocket. Ergo they must have the most stable flight configuration possible, which is long and thin -- as I said in the first post. A short and stubby model rocket tends to end up on Mr. McGillicutty's roof -- and you know what an ogre he is.

davf · Jul 25, 2004

Excellent example... I'm going to pocket that one and keep it for a rainy day conversation.

drwayne · Jul 25, 2004

Gee, I know what IRBM stands for, does that mean I'm old? Of course, I still have been known to call it the Lunar Excursion Module... Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

najab · Jul 25, 2004

I'm not old but I know what an IRBM is too....maybe I'm old at heart?

propforce · Jul 25, 2004

Why don't we make a rocket in the shape of a sphere This way we can even save more weight as the sphere has the minimum surface area <img src="/images/icons/smile.gif" /> Obviously it's not that simple. But it could if it is ONLY a satellite !! <img src="/images/icons/wink.gif" /> .... well, not really neither. -- the increase in cross-sectional area that I calculated is fairly minimal -- 1.7 meters to 2.5 meters. ..." That's a lot of increase in frontal area, which is a lot higher drag than skin friction. The difference is 3.36 m^2, or 36.167 ft^2, more than double. Simplistically, the frontal drag is equivalent to the air dynamic pressure, q, times the front area, S, then with typically rocket goes as high as 500~750q (lbf/ft^2) that's an additional/i> ~27,000 lbf your engine needs to overcome and additional stiffening one needs on your skin. A torpedo shape actually makes a lot of sense (I believe Goddard's first rocket is torpedo shape, so is the German V-2?), it reduces base drag as well, but it is difficult to fabricate and costly. It is much easier to manufacture cylindrical shape rocket and cheaper. For missile-derived system, e.g., IRM, IRBM, MRM, ICBM and just your daily variety of SAMs, you must fit into the launch-tube diameter constraint. So that pretty much defines the maximum diameter of your rocket. Oh one more thing - steering. Engine gimbaling angle requirement is less (shallower) with a long vehicle than it is with a shorter vehicle. Your hydraulic or EM actuator will not be as 'beefy' and vehicle steering is inherently more stability and is much more controllable. <div class="Discussion_UserSignature"> </div>

mrmorris · Jul 26, 2004

"Thor had no size constraints and was designed for efficiency." Actually, no. According to Astronautix, one of the core Thor requirements was: 'requiring the missile to be air-transportable by C-124 Globemaster transport aircraft, the basic design and overall dimensions of the Thor were quickly determined' Your post seems to indicate that you feel the Thor was the more efficient vehicle but nothing that I can find seem to indicate this is so. In fact -- from the Astonautix specs, the Jupiter seemed to have a better performance with a higher ceiling, and longer range: Data for PGM-19A (Jupiter): Length 18.3 m (60 ft) Diameter 2.67 m (8 ft 9 in) Weight 49800 kg (110000 lb) Speed 16100 km/h (10000 mph) Ceiling 610 km (380 miles) Range 2980 km (1850 miles) Propulsion Rocketdyne LR79-NA (Model S-3D); 666 kN (150000 lb) Data for PGM-17A (Thor): Length 19.8 m (65 ft) Diameter 2.44 m (8 ft) Weight 49800 kg (110000 lb) Speed 16100 km/h (10000 mph) Ceiling 480 km (300 miles) Range 2400 km (1500 miles) Propulsion Main: Rocketdyne LR79-NA-9 (Model S-3D); 666 kN (150000 lb) Vernier: 2x Rocketdyne LR101-NA; 4.5 kN (1000 lb) each Also from Astronautix: The SM-78 was a single-stage rocket, powered by a single Rocketdyne S-3D engine fueled by kerosene and liquid oxygen. This was the same engine as in the SM-75/PGM-17 Thor. However, the SM-78 was a more effective IRBM than the SM-75, because of its mobility. Although a Jupiter launch site was far from easy to move, requiring more than 20 vehicles, it did significantly increase the missile's survivability in a pre-emptive attack, because the location of the Jupiters could not be pre-targeted by the enemy. Also, the SM-78's ablative reentry vehicle flew through the atmosphere at much higher speed than the SM-75's Mk.2 RV, making i

mrmorris · Jul 26, 2004

"... the sphere has the minimum surface area..." Yes -- I seem to vaguely recall mentioning that in the initial post. "The difference is 3.36 m^2, or 36.167 ft^2, more than double." I can't figure out how you made that calculation. If I treat the frontal area as a flat circle, I get a figure that doubles SA over the 1.7 meter diameter but isn't close to your 3.36 number. If I treat the frontal area as a cone with a height of 2.5 meters, I get a number approaching your 3.36 but it's nowhere close to double the area of the 1.7 meter figure. 1.7 m diameter pi r^2 3.1412 * .85^2 2.27 sqm 2.5 m diameter pi r^2 3.1412 * 1.25^2 4.91 sqm Difference of 2.64 sqm if treated as a circle. Teated as a cone (using calulator here

1.7 -- SA= 13.4 sqm 2.5 -- SA = 19.6 sqm 3.2 sqm difference. "Simplistically, the frontal drag is equivalent to the air dynamic pressure" If decreasing the frontal drag was the only thing that mattered, then rain would fall in spaghetti-like strands. That's not the case. In fact -- rain falls with a very blunt leading edge. 'Simplistic' doesn't work when dealing with airflow and drag calculations -- this is why wind tunnels and computer modelling are required when designing high-velocity vehicles. "...vehicle steering is inherently more stability and is much more controllable. " From the initial post, I acknowleged that a longer, thinner rocket is easier to steer. However -- our capabilities (both mechanical and electronic) for providing accurate steering has *vastly* improved in the past several decades. If making the rocket le

propforce · Jul 26, 2004

<blockquote>In reply to:<hr />"... the sphere has the minimum surface area..." Yes -- I seem to vaguely recall mentioning that in the initial post. <hr /></blockquote> Sorry I must've miss it. <img src="/images/icons/smile.gif" /> <blockquote>In reply to:<hr />"The difference is 3.36 m^2, or 36.167 ft^2, more than double." Difference of 2.64 sqm if treated as a circle. <hr /></blockquote> You're correct. I screwed up in math <img src="/images/icons/blush.gif" /> <blockquote>In reply to:<hr />"Simplistically, the frontal drag is equivalent to the air dynamic pressure" If decreasing the frontal drag was the only thing that mattered, then rain would fall in spaghetti-like strands. That's not the case. In fact -- rain falls with a very blunt leading edge. 'Simplistic' doesn't work when dealing with airflow and drag calculations -- this is why wind tunnels and computer modelling are required when designing high-velocity vehicles. <hr /></blockquote> Nothing is the ONLY thing that matters in rocket design (with perhaps the except of budget & schedule), it is a MULTI-VARIABLE OPTIMIZATION problem. Airplane design today is also a "multi-mission" design. Rain drop shape is a poor analogy to rocket shape. First the rocket is a RIGID structure (well, relatively speaking), where as raindrop is not. Second, raindrop falls in a speed way below the speed of sound, whereas rocket quickly goes into SUPERSONIC speed. The flow physics changes from subsonic to supersonic and onto hypersonic flow. ".....'Simplistic' doesn't work when dealing with airflow and drag calculations...." -- but it does when you do a qualitative comparison on the first order of magnitude. It does pretty good in the hands <div class="Discussion_UserSignature"> </div>

najab · Jul 26, 2004

Right to left - at least on the plans.

mrmorris · Jul 26, 2004

"Rain drop shape is a poor analogy to rocket shape." I didn't claim the two were analagous. What I indicated was that it is non-intuitive as a drag-minimizing shape. There are lots of variables in play that are non-obvious. "but it does when you do a qualitative comparison on the first order of magnitude. It does pretty good in the hands of those who can do hand-calculation actually." Not that I don't beleive you, but this is from the same source that miscalculated the area of a circle. <img src="/images/icons/smile.gif" /> You're claiming that an off-the-cuff approximation (what I like to call voodoo math) is accurate enough that I should accept conclusions based on it. Is anyone else using this approximation? A quick Google search didn't turn up anything, but then I don't really have good search terms. All of the sites I found calculating drag essentially said one of two things: 1. Don't do it, and mention that your results neglect drag. 2. Here's the basic formula, but you need a wind tunnel to determine the coefficient of drag. "But that's NOT necessary so.... by requiring a higher gimbaling angle you also making the vehicle turning at a higher "S" curve thus wasting more propellant. I said 'if', not 'because'. We're still hypothesizing here. I agree that if the rocket 'slithers' up into the sky, this is definitely non-optimal. However -- part of the assumption in the improved steering technology is that computers today can make corrections that occur with great speed and precision. If the computer is making several corrections a second to minimize gimballing angles required -- the thrust losses due to such should be insignificant. "(diameter - for example for launch box constraint, tooling, transportation, etc.)"

nacnud · Jul 26, 2004

OK here is my hand waving and non maths reasoning behind the shape of your first example of a modern rocket, the falcon I. First off it comes in two parts the first and second stages, for the sake of argument I’m lumping in the payload with the second stage. The first stage needs to be powerful enough to lift itself and the second stage above the atmosphere the second stage needs to be efficient enough to achieve orbital velocity. Initially let’s look at the second stage, this operates above the atmosphere but has to pass through the atmosphere to get there. Crucially weight is the major factor in the design of this stage as every pound saved increases the payload by a pound. As we have seen already the most weight saving shape for the fuel tanks is for them to be spherical. Looking at the Falcon I design this seems to be roughly the case, the LOX tank in the second stage is spherical, the PR-1 tank that has a common bulkhead with it makes the tank look slightly elongated. Now line everything else up in a line i.e. engine, tanks, payload and the widest point will probably be the LOX tank. Now let’s look at the first stage, this doesn’t have to reach orbit only to accelerate the second stage to an altitude and speed from which it can comfortably reach orbit. Also a significant proportion on the flight will be in the atmosphere so aerodynamics is increasingly important. Frontal area increases with the square of the diameter of the vehicle, and drag increases with the square of the velocity, well for subsonic velocities I don’t know what happens at supersonic velocities but I imagine it only gets worse. I feel that because drag increases so rapidly with increasing diameter in order to minimize the size, and therefore the cost of the first stage the frontal cross section should be kept as small as possible. So what is the smallest practicable cross section, well it is going to be the second stage spherical LOX tank. The length of the

davf · Jul 26, 2004

Frontal area (form drag) will kill you most every time. In aircraft, at least, it is a greater component than skin friction (which is relevant to the wetted area). Of course, induced drag can likely be excluded unless the vehicle is maneuvering... and in that case, I'm guessing the higher aspect ratio vehicle could generate much greater aerodynamic forces for trajectory correction. There is also another component you are neglecting: hypersonic stability. One of the greatest challenges with the V2 was finding a shape that was stable in hypersonic flight and why the Wasserfall SAM borrowed the exact same shape despite being much smaller. Simplistically, D = qScd where c d (should be c sub d) is the coefficient of drag, S is the area, and q is the dynamic pressure (q = 1/2 rho V^2). This can be used both on the frontal area and the skin friction except the coefficient of drag will change. In the case of form drag, this will be dependant on the shape of the vehicle (ie: shape of the nose, aspect ratio of the body, any treatment at the aft end, etc). I would think that one of the issues you are going to have with a thicker, shorter rocket is going to that around the aft end of the rocket. The N1 had some aerodynamic treatments to minimize this but I'm guessing it still wouldn't have been as efficient as a more slender aft body. I would guess that the skin friction is probably minimal in most cases... compared to the form drag (frontal area related drag). A quick look-up online will give you some coefficient of drag figures for sample shapes (flat plate = 1.0, etc). Of course, this should be measured for each design but you can get an idea of the impact of some of the different shapes. BTW, IIRC these will change for supersonic and hypersonic flow... my guess is any search online will likely only turn up values for incompressible flow (low speed aerodynamics). Anyway, there's a few things to think about.

propforce · Jul 26, 2004

Not that I don't beleive you, but this is from the same source that miscalculated the area of a circle. /i> *sigh* Believe it or not I used to be very good with aerodynamics, but now I have wiper-snaper young kids like you to do them. Plus, I get CRS on Sundays now <img src="/images/icons/smile.gif" /> Sometimes it comes down to the maximum diameter tank you can put on open-bed truck and still go under-neath interstate bridges and over-passes, that determines the maximum diameter for your rocket. Of course SpaceX has a long way to go before worrying about this problem. The difference of 2.64 sqm frontal area is still substantial, multiplying that with a dynamic pressure upto 700 lbf/ft^2 you still get subtantial frontal aerodrag. Notice how smart I am today by not giving you the final number? <img src="/images/icons/laugh.gif" /> <div class="Discussion_UserSignature"> </div>

propforce · Jul 26, 2004

".....Right to left - at least on the plans...." It is dificult to educate some people that air, in fact, flows from left to right, so ALL airplanes fly from right to left !! <img src="/images/icons/laugh.gif" /> In rocket design, we discourage people use the term exploded viewwhen making briefing charts <img src="/images/icons/smile.gif" /> <div class="Discussion_UserSignature"> </div>

davf · Jul 26, 2004

<blockquote>In reply to:<hr /> In rocket design, we discourage people use the term exploded view when making briefing charts <hr /></blockquote> LMAO!!!!!

propforce · Jul 26, 2004

Another factor that can't be ignored, is you are not pushing a solid shape through a stable medium. While it is burning, that candle is changing the local flow around the rear end quite a bit! Yes... the effect of rocket plume and base pressure drag. Velocity losses constitute as much as 15~18% of total propellant used to get to orbit (LEO) for ground-launch vehicles. A majority of the loss is due to gravity well, but both drag & steering losses are not insignificant --- when you consider payload weight is effectively 2~4% of total lift-off weight only. <div class="Discussion_UserSignature"> </div>

drwayne · Jul 26, 2004

Interesting exchange with folks using terms like detonation, explosion, and uncontrolled pressure release: http://yarchive.net/space/rocket/hybrids.html Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

drwayne · Jul 26, 2004

I seem to remember at some point, when people still pronounced it Lem, when they said the words, they called it the Lunar Module. There was that whole Redstone, Jupiter C, Juno, Saturn thing going on in naming at one time...it is easy to lose track of who was what... Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

drwayne · Jul 26, 2004

There is another sig line: "The good old days weren't" Wayne <div class="Discussion_UserSignature"> "1) Give no quarter; 2) Take no prisoners; 3) Sink everything." Admiral Jackie Fisher </div>

Need a rocket scientist...

mrmorris

Guest

nacnud

Guest

drwayne

Guest

nacnud

Guest

drwayne

Guest

nacnud

Guest

mrmorris

Guest

davf

Guest

drwayne

Guest

najab

Guest

propforce

Guest

mrmorris

Guest

mrmorris

Guest

propforce

Guest

najab

Guest

mrmorris

Guest

nacnud

Guest

davf

Guest

propforce

Guest

propforce

Guest

davf

Guest

propforce

Guest

drwayne

Guest

drwayne

Guest

drwayne

Guest

Similar threads

TRENDING THREADS

Latest posts

Share this page