You are not logged in.
Kbd512 has described the same bearing assembly as found in the knuckle of the automobile steering assembly wheel bearing.
A radioisotope thermo-electric generator (RTG, RITEG) is a type of nuclear battery that uses an array of thermo-couples to convert the heat released by the decay of a suitable radioactive material into electricity by the Seebeck effect.
http://large.stanford.edu/courses/2013/ph241/jiang1/
An Overview of Radioisotope Thermoelectric Generators
https://ntrs.nasa.gov/api/citations/196 … 025340.pdf
RTG/SCIENCE INSTRUMENT RADIATION INTERACTIONS STUDY FOR DEEP SPACE PROBES
These power units were handled on the moon by astronauts but the scale of those needed for this ship are going to need more protection.
https://nuke.fas.org/space/safety.pdf
SAFETY STATUS OF SPACE RADIOISOTOPE AND REACTOR POWER SOURCES
https://solarsystem.nasa.gov/missions/c … generator/
https://www.osti.gov/servlets/purl/1459307
Radioisotope Power Source Dose Estimation Tool
https://www2.dsa.no/filer/e45324ac29.pdf
Threat Assessment of Radioisotope Thermoelectric Generators (RTG) Management Radiation Protection and Safety Regulations
https://bellona.org/news/nuclear-issues … nerators-2
Nuclear issues in ex-soviet republics, Radioactive waste and spent nuclear fuel
The cleanup of RTG units that have out lived the 10 year use cycle.
Offline
This calculation will be rather crude since I am not integrating the results to account for mass distribution inside and outside of the radius of revolution. Since I intend to cant the habitation rings outwards, away from each other, we need to first determine the force being applied using a force diagram. We know that the intended radius of rotation for the habitation rings is 27.75m.
Mass: 350,000kg (not precisely correct because some of that mass is not on the radius of rotation; outer half of habitation ring, for example)
Radius of Rotation: 27.75m
Angular Velocity: 5.675rpm (I kept using 5.37rpm because I forgot about "shrinking" the habitation ring to reduce mass)
Centrifugal Acceleration: 9.8m/s
Force: 3,430kN
Now we need a force diagram and a diagram for the ship design. Let's say each habitation ring is angled 30° away from each other. The forward habitation ring is 30° forward and the aft habitation ring is 30° aft, so if you were to bisect the plane perpendicular to the axial direction (fore and aft the ship), then there is 60° between each torus.
[nose]*>*|*<*[tail] ← Axial direction / direction of travel
* slewing ring bearings (one on each side of the habitation ring, between the nose and tail sections of the ship
[nose] - forward part of the ship
[tail] - aft part of the ship
> - forward habitation ring
| - plane bisecting the two habitation rings, perpendicular to the axial direction / direction of travel
< - aft habitation ring
Forward habitation ring rotates ↑ this way (doesn't really matter, so long as they rotate in opposite directions)
Aft habitation ring rotates ↓ this way (doesn't really matter, so long as they rotate in opposite directions)
Please forgive my exceptionally crude ASCII character "force diagram"
* ← slewing ring bearing
|. 30°
|60° .
| . ← periods represents the "spokes" connecting the habitation ring to the hub module
-------------.( ) ← "( )" the habitation ring
* ← slewing ring bearing
R
R (radius of revolution) = 27.75m (also the "base" / "adjacent" side of the triangle)
We need to use a common Trig function to help us find the lengths of the "height" / "opposite" side and "hypotenuse" side of the triangle.
Everyone remember SOHCAHTOA?
We're using TOA here:
tan30° = Opposite/Adjacent
tan30° = o/27.75
o/27.75 = tan30°
o = 27.75 tan30°
o / "opposite side" / "height" = 16.02m
opposite^2 + adjacent^2 = hypotenuse^2
(16.02)^2 + (27.75)^2 = hypotenuse^2
256.6875 + 770.0625 = 1026.75
sqrt(1026.75) = 32.04
hypotenuse = 32.04m
Earlier in this thread there was some commentary on Newton's Laws of Motion, as it related to the plasma beam acceleration of the ship.
Well... We're going to need to use those here:
ax = (ΣFx) / m
ay = (ΣFy) / m
We have forces in two different directions. I will probably screw this up because I haven't seen this stuff in more than a decade. Consider yourself lucky I still remember anything from Trig, much less physics class.
We're going to calculate force in the "y" direction first:
m = 350,000kg
g = (9.8m/s)^2
0 = (-mg + (T * cos60°)) / m
T = mg / cos60°
T = (350,000kg * (9.8m/s)^2) / cos60°
T = 3,430,000 / cos60°
T = 6,860,000N / 6,860kN / 699,525kgf / 1,542,189lbf
So...
V^2 / m = Tsin60° / m
V^2 / (32.04 * sin60°) = Tsin60° / m
V^2 / (32.04 * sin60°) = Tsin60° / 350,000
V^2 = Tsin60° * (32.04 * sin60°) / 350,000
V = sqrt(Tsin60° * (32.04 * sin60°) / 350,000)
V = sqrt(((6,860,000N * sin60°) * (32.04 * sin60°)) / 350,000)
V = sqrt((5940934.269961249116799 * 27.747453937253414) / 350,000)
V = sqrt(164845799.999999999999996 / 350,000)
V = sqrt(470.988)
V = 21.7m/s <- this is obviously wrong
V = 16.49m/s <- this is what the value should be and I arrive at this value by using force in the y-direction of
Notes:
1. I know I should use the cosine of 60° for the first equation solving for T / Tension / tangential force applied to the slewing ring bearing, but then I get an incorrect tangential velocity.
2. I know that 16.49m/s is the correct tangential velocity for a 27.75m radius or revolution. That means I have the applied force value wrong.
3. This is why an engineer should be doing this work, rather than someone who studied computer science who barely remembers Trig or Calc or basic physics.
4. If I derive an incorrect force value of 3,960,622N by using cosine of 30°, then when I plug that value back into the second equation with a correct sine value for 60°, then I get the correct tangential velocity of 16.49m/s. I can also plug an incorrect sine of 30° into the second equation and get the correct tangential velocity result.
To hell with it. I'll figure it out tomorrow.
Offline
For kbd512 re #102
Thanks for showing your work here!
Just curious ... why angle the rings with respect to each other? I'm assuming you must be going for stability?
Now we need a force diagram and a diagram for the ship design. Let's say each habitation ring is angled 30° away from each other. The forward habitation ring is 30° forward and the aft habitation ring is 30° aft, so if you were to bisect the plane perpendicular to the axial direction (fore and aft the ship), then there is 60° between each torus.
I was expecting the rings to be side by side with the same shaft for axis of rotation in opposite directions.
If you do that, then the rings can operate as linear motors with respect to each other, and there would be no need for separate linear motors.
***
Are you familiar with Blender? It is a free package that runs on practically anything.
There are (by now) a large number of books to help with the (somewhat steep) learning curve, but both RobertDyck and I are (potentially) available to toss in suggestions or even work with models you might create.
You can see some examples of Blender output in RobertDyck's Large Ship topic.
A search for Blender in the archive should turn up some examples. Be sure to use Author tahanson43206 to get a manageable list.
There are many other 3D animation/modeling packages. Blender is intended to make movies, but it works well enough for fixed object design for 3D printing (for example). It has a physics engine for "real universe" simulation. There is an example of a (simple) animation in the Large Ship topic.
http://newmars.com/forums/viewtopic.php … 67#p172067
Above is link to post with link to animation.
(th)
Offline
tahanson43206,
The angle applied to the habitation rings is about directing some part of a force imbalance in the z-direction... Now I know what's wrong with my force diagram. I tried to over-simplify it. I want independent motion, thus the use of motors dedicated to each habitation ring.
Edit:
This was what I needed to apply: Adding and Scaling Force Vectors
Edit #2:
Here we go: Real World Physics Problems - Gyroscope Physics
Edit #3 (and finally, a YouTube video of what I'm actually after):
Angular Momentum of a Symmetric Rigid Body Rotating About an Axis of Symmetry
Last edited by kbd512 (2022-01-18 09:21:25)
Offline
For kbd512 re #104 and topic, and specifically use of reaction wheels to orient a space craft
Your configuration of two reaction wheels at 30 degrees is interesting, and I'm glad to learn about it.
The "traditional" design for spacecraft orientation using reaction wheels is to align the wheels on the three axes, x, y and z
Not intending to be anything more than curious, why stop at two reaction wheels since three gives you total control of spacecraft orientation for thrust maneuvers or for orientation to protect against radiation, and other purposes I'm not aware of?
Is there a "real world" example of using just two reaction wheels to achieve spacecraft orientation objectives?
There must be, or (I'm pretty sure) you wouldn't have decided to go that route.
In composing your reply, please think about the future high school aged student who will (hopefully) be reading this topic.
It contains already content that shows the evolution of thought that will lead to a working spacecraft in the not-too-distant future.
You've decided to take on challenges that RobertDyck was attempting to avoid by using a single reaction wheel for the habitat (and smaller ones for orientation internal to the central shaft).
Tags indicating what each post is attempting to achieve would be helpful as you build this sequence.
I understand that when you start a post (like any author) you cannot be sure where you will end up, so a brief summary of what was achieved would be helpful.
I also understand that brevity does not come naturally to everyone, so please make a special effort at the end of each post to report what it is about.
I am guardedly optimistic that you might be able to earn funding if you stay on this track long enough.
I will continue to attempt to persuade SpaceNut to contact Executive Director James Burk to see if there is any chance the Mars Society can help with recruiting of retired professionals who will be willing to pitch in.
A key factor for possible success is your (apparent) willingness to accept input from people who know more than you do (because of decades of education and experience in their fields). Your role as Project Visionary (should you decide to accept it) is to inspire the work force we may be able to enlist.
(th)
Offline
tahanson43206,
There's nothing magical about choosing 30°. Maybe 15° provides more than enough offset. These are flexible fabric wheels, though, and I definitely don't want them flexing and accidentally contacting each other. They're only about half as rigid at 8psi as 14.7psi. The hubs that the habitation rings are connected to need to be as small and rigid as is practical, because resisting more than 350t of force trying to tear the hub apart requires a very strong and stiff and therefore heavy structure. 30° was a totally arbitrary value, only selected to provide offset / separation between two very large and heavy spinning masses and to illustrate the basic concept. It provides 32m of separation between the two reaction wheels.
All of this stuff is ever-so-slowly coming back to me. Last night I couldn't even remember what formulas I needed to use. This morning, after sleeping, I was able to quickly Google that YouTube video that has the formulas that should be used and that mechanical engineering site that shows formulas for application of force vectors. I learn what I need to know to perform a given task, like taking a math or physics test or writing a software program, but unless I use it on a regular basis, it's gone inside of a few months while some new task occupies my mind. Once again, I'm not a professional engineer and don't claim to know anything specific about engineering. I am able to teach myself what I think I need to know about a topic, use it, then forget about it if I never have to use it again. If I know what the formulas are that I need to apply, then I can usually apply them.
As to the question of whether you could add more habitation rings, the answer is yes, but only at the expense of added mass and complexity. This ship is a 1,000t object, or roughy twice as heavy as ISS, that must break orbit using a small quantity of fuel, cruise for tens of millions of miles through deep space, and then be stopped in orbit around Mars using MagBeam stations orbiting that planet. It uses a series of impulse bits delivered over 2 weeks to accomplish that, always starting from circular low orbits around Earth or Mars, moving through highly elliptical orbits as successive impulse bits are added at a single point in the ship's orbit, and finally achieving escape velocity. Most of the fuel stays in orbit around Earth and Mars, so nearly everything being transported is useful payload. The bulk of the fuel only has to be delivered to low orbit, where all of the MagBeam satellites will remain.
Since the specific impulse provided by the plasma beam is so high and the ship does not carry massive quantities of fuel with it, fuel being reserved for minor course corrections / adjustments, the cost of maintaining the ships is a matter of keeping the hulls air-tight, the interiors clean, and stocked with consumables and colonists. The MagBeam stations must be supplied with Hydrogen, but the majority of the delivered tonnage is useful payload. Starships that stay on Mars and never leave, will ascend to orbit carrying fuel to restock the MagBeam stations, as well as any colonists or crew to ferry back to Earth. Most of the ship's payload (food and water rations for a 2 year round trip) will be offloaded at Mars with the colonists.
Now that I think about it, the tonnage devoted to the habitation rings is relatively minor, so all vessels could be built the same way, but cargo freighters would only require a small crew, similar to our Merchant Marine / USNS ships. The cargo vessels would be heavily loaded with additional consumables for the colonists to use on Mars- additional food and water / bedding / toiletries / medical supplies / computers / machine tools / raw materials like metals and plastics / etc. 700t is still quite a bit of cargo, even if a purpose-built vessel could carry more. If all vessels are crewed and equipped with the same power and life support, then the possibility exists for deep space rescue of crews aboard damaged ships. All vessels will be operated as part of convoys to provide safety in numbers. They're not designed to be capable of large delta-V changes on their own, but they have enough excess power for course corrections. All ships will be equipped with their own plasma thrusters to provide attitude control and minor velocity changes to "catch up to" or "lag behind" other ships in the convoy.
We will eventually determine how to provide supplemental electrical power using a special type of robotic "MagBeam ship" that only deploys its photovoltaic arrays in deep space and serves as a portable external power and propulsion source for the ships in the convoy.
Each convoy would be a squadron of ships. With 500 colonists per ship, we need to launch 80 passenger carrying vessels per year to populate the colony of 1 million people over 25 years. We probably need an equal number of cargo ships, along with a smaller number of MagBeam ships that supply power and propulsion to a squadron of passenger liners and cargo ships in deep space. We could reduce the number of MagBeam satellites required to "stop" the squadron of ships in orbit by initially slowing some of them down or speeding them up so that there's a steady stream of arrival and departures. The potential exists for these MagBeam ships (that do not require massive quantities of batteries because they're providing continuous power over time and are always in full sunlight) to act as true "cyclers", that simply orbit through "shipping lanes", shepherding squadrons of ships through deep space. The fuel they carry could be replenished by the cargo ships in the convoy, en-route to Earth and Mars- the space-based version of "underway replenishment" if you will. This is a future / aspirational concept of operations to permit "anytime" arrivals and departures (continuously scheduled passenger and freight service). After escape velocity is achieved, these "MagBeam ships" add or subtract delta-V, as required, to permit a continuous stream of arriving and departing ships that do not wait for launch windows, instead launching whenever they're ready to depart.
A pair of reasonably well-balanced counter-rotating wheels will intrinsically counter-act gyroscopic precession. This is an actual problem that will manifest itself with any spinning top / single gyro design. A spinning top ship will do 360s in space without counteracting-torque. It is possible to counteract precession using sets of reaction / moment wheels located within or otherwise attached to a spinning top ship, but the forces involved will be substantial and that requires massive spinning momentum wheels. My method simplifies propulsion and attitude control, as the expense of the complexity of using bearings and motors to provide artificial gravity. However, a counter-rotating design also means that you have fine control over artificial gravity and do not require propellant to generate gravity or huge gyros to counteract precession.
There are no "perfect" solutions to be had here, merely a series of trade-offs. I traded off overall vehicle design simplicity for, what will hopefully prove to be lower total mass and assured control over the spin gravity provided by using motors and counter-rotation. I'm perfectly willing to accept input from others, with or without specific expertise, and any valid criticisms. I'm not particularly interested in pleasing any specific party, because pleasing vendors or special interests is not how you create a highly accomplished space program. Pleasing vendors is why we haven't already sent people to Mars. During the Apollo era, there was far less focus on pleasing people and considerable focus on simply getting the job done in a practical way, whether that was the absolute "best way" in some specific person's mind or not.
Offline
For kbd512 re topic .... This topic would make a (literal) quantum leap forward if you were to master use of imgur.com to hold a hand drawn sketch of your vision. There is ** something ** in your brain, but it is not getting out when the only tool you have to work with is this (text) medium.
If you go back to look at the topic of RobertDyck, you will see that in the early period, he used a hand drawn sketch to illustrate his vision of the cabins for his vessel.
I have created a mental image of your vessel, based upon the words you've provided, and it looks absolutely ** awful **. I would appreciate your providing a sketch to help to put these wildly careening rotating habitats into some sort of order.
(th)
Offline
For kbd512 re topic ....
Your competitor, RobertDyck, is going to hold steady in his vision of a massive unitary rotating passenger space craft. He has accepted the consequences of his decision, and is confronting the issues that flow from the decision as they show up.
On the other hand, from the outset, you have reported willingness to consider counter rotating habitats as a way to manage gyroscopic forces in a large vessel. Today, I'd like to offer a thought for your consideration.
As you began to develop your vision of a passenger vessel you started with the vision of RobertDyck, of a rotating habitat that is exposed to space.
In your early thinking (as I understand it) you've been thinking of two counter rotating habitats, both of which are exposed to space.
Because you (may) still be flexible in your thinking at this early point in development of your vision, I'd like to offer an alternative that came to me recently.
Years ago, I "met" (through the Internet) a gent who patented a magnetic warehouse movement system. His last name was Barber, so I expect you could find the patent by searching for "magnet" and "barger'. If there is more than one patent with that combination, I'd be surprised.
However, the details of that patent are not the focus of this suggestion. The key concept was that permanent magnets were configured to support a moving mass.
The problem of dealing with two large rotating masses in a space vessel received some attention in your topic here, recently.
I'd like to toss Mr. Barber's idea into the ring, to see if any aspect of it appeals to you.
If you mount your rotating rings inside a housing that is fixed to the fixed body of the space craft, and if you separate them from the fixed structure of the ship using permanent magnets, then (presumably) there would relatively little viscocity to deal with. There might be ** some ** hysteresis to deal with, but I'm betting it will be less than any chemical bearing lubrication system.
The rings could be spun up and down (with respect to each other), and stopped entirely when it is time for a force maneuver, such as Earth departure or Mars arrival.
(th)
Offline
tahanson43206,
In Post #98 in this thread, I provided a link to pictures of two different offset hand wheel designs. Try your best to imagine a pair of those facing away from each other and rotating in opposite directions. Unfortunately, this design was never intended to be aesthetically pleasing to anyone. It's a way to counteract the basic physics of a gyroscope. Form follows function, not the other way around.
Lots of pilots think Cessna 172s are ugly little planes. Heck, even I know that it's not the most aesthetically pleasing bird on the tarmac. However, they were designed to be easier to fly than other designs intended to train new pilots, and they are in many respects, which is why there are more copies of that "ugly little plane" (strut-braced high-wing / tricycle landing gear / tractor rather than pusher powerplant / conventional tail / two door cabin entry) than any other aircraft design that ever existed (it's not "the best" at anything, yet it serves most users exceptionally well for learning to fly and transporting small numbers of people and some cargo). Moreover, there are more copies of that basic design than all other types of airframe designs in existence. There have been well over 100,000 152s / 172s / 182s / 205s / 206s / 207s / 210s produced and they keep building more, more than half a century later, because so many people keep buying them. There are only so many ways to do something that work at all, and far fewer ways to do something well enough to merit doing it again and again. No other basic design has proven quite so successful over the past 75 years of powered flight. There's a very simple explanation for that level of success. If you want to be a pilot, then you have to start somewhere. You may as well start in a plane that's docile to fly, very stable in flight, and has performance very similar to other airframes intended for the same purpose.
Similarly, that's why we're still using "ancient" air-cooled / opposed-cylinder / carbureted / twin spark plug and magneto ignition Continental and Lycoming piston engines for power requirements of 450hp or less. Pilots have a "love-hate" relationship with these seemingly "backwards" engines. We all learned to fly using them. We know all of their problems from 50+ years of continuous production and operation. We know exactly how and why they fail. We know that they're a constant maintenance headache (25hr oil changes, $15K to $60K overhauls- if the engine makes it to TBO). However, we also know that nothing except a gas turbine has proven more reliable over time. Every experiment known to man has been conducted using "current technology" automotive engines, to no avail. By the time you add in the cost of all the engineering / flight testing / design features necessary to improve reliability to the point where they match or approach that of those old "dinosaur engines", they're every bit as expensive and even heavier to boot (weight being the mortal enemy of anything that flies). Unfortunately, gas turbines that could get the job done reliably, with less weight, are even more absurdly expensive because they're made in even more limited quantities than "ye olde piston engines". If that wasn't enough, they're far more sensitive to improper maintenance and at lower altitudes they guzzle fuel like it's going out of style. That's why you only see independently wealthy owner-operators and businesses flying them. The terms "simpler" or "easier to operate" seldomly translates into a less complex or expensive product, especially as it relates to power and propulsion. Sure, that gas turbine has 1 spinning component, but that 1 component is comprised of hundreds of finely machined parts made from super alloys. It is utterly intolerant of abuse from a novice pilot. If you "forget something" with a gas turbine or other computer-controlled engine, there can be fires and explosions. Every machine has operating limitations. The higher the performance, the more constrained your operating environment becomes. Things that would be acceptable with a piston engine will never be acceptable with a gas turbine engine.
So... If you're going to start building giant ships to take people to other planets, then you may as well start with a "stable in flight" design, working your way up to more advanced designs that require far more sophisticated control schemes, as technology and experience operating that technology permits. The F-22 is a spectacularly nimble high performance jet aircraft, but the only reason it doesn't tumble end-over-end is exceptionally sophisticated computerized flight control systems. No human pilot would be capable of keeping the nose pointed where intended without it's flight control software- as proven by that fatal crash during a test flight after the avionics suite failed to properly control the jet and all redundant systems simultaneously failed because they all suffered from the same issue.
Due to that simple fact of life, all F-22 pilots "earned their wings" flying that "ugly little Cessna 172", or the US Air Force's version of it. From there, they learn how to operate high performance piston-engined airframes like the Cessna 182 before making the quantum leap to turbine power in a T-6 Texan II. That covers basic instruction in higher performance aircraft. Subsequent flight training is conducted in the twin turbojet engine T-38. After all training has been completed to the satisfaction of the instructors, which many prospective fighter pilots fail to accomplish (take a look at the washout rates), then and only then do they have the combination of skills and experience to not immediately kill themselves flying a fine-tuned supersonic speed machine like the F-22. No, you don't have to fiddle with a prop pitch control or carb heat in a F-22, because you have dozens of other potentially fatal problems to contend with, merely to keep your jet in the air. When things go wrong, you have to immediately recognize and fix them. You may have had a few seconds or even minutes to deal with them while flying the 172, but no such luck with a F-22- because it can and will climb and descend 60 times faster than a 172. If you're not fully capable of recognizing and resolving the issue immediately, then there's only one likely outcome.
If a pair of spinning modules connected to 4 slewing ring bearings that permit independent counter-rotation is too complicated to conceptualize and build, then... How are we going to build a ship 50 times more massive that requires continuous computer control over multiple gyros, that must provide redundant capability if you wish to avoid losing attitude control- a fatal problem if left uncorrected? This alternative ship design has less severe intrinsic flight control issues. It's not a "superior design" in any other way. It's an admission that cost is a real thing, control software complexity is real complexity, and excessive mass kills performance.
Offline
For kbd512 re computer image creation and rendering ...
This is ** not ** beyond your capability.
Just download Blender and buy one of the vast number of beginner's books and both RobertDyck and I are (potentially) available to help.
Some folks are gifted with talents that lean one way or another in the capability spectrum.
Visualization may not be in your strong suit tool box, but modern computer software makes up for what might otherwise be impossible.
You can find numerous examples of Blender renderings in the Large ship topic.
Most of those models are available for you to download and run on your computer to get a feeling for how they work.
More words are not always the answer. Sometimes a picture ** really ** is worth 1000 words.
I suspect that what you are seeing in your mind's eye is not a workable solution to the problem of transporting passengers safely and comfortably from Earth to Mars, but by investing your time and energy in learning how to create compelling images for a tough audience, you may be able to prove otherwise.
(th)
Offline
Moment of inertia is different in a dual opposite rotating structure so long as the movement is the same and equal amount for both sections.
re-posting from zoom
tahanson43206,
$3B for the development program, based upon a similar tonnage military warship, the LCS program. I will, of course, look for ways to pare down the development budget by incorporating as many off-the-shelf components as we can repurpose. We need distributed propulsion technology development here. We don't need much development of any other technology. Every other technology is perfectly adequate to the task demanded of it. We're not going to reinvent any wheels here.
Examples:
1. Power - scale-up of thin-film ROSA technology currently being deployed aboard ISS
2. Propulsion - scale-up of laser ablation of solid propellants
3. Life Support - scale-up of NASA's CAMRAS and IWP
4. Navigation - repurpose Lockheed-Martin's Orion Program navigation computers and software
5. Communications - repurpose of the NASA / Harris Software-Defined Radio (SDR)Software Defined Radios (SDR) for NASA Spaceflight Applications
Space Software Defined Radio Characterization to Enable Reuse
SDR/STRS Flight Experiment and the Role of SDR-Based Communication and Navigation Systems
6. Radiation Protection - repurpose of NASA's hand-portable "water bricks" (flexible plastic bags containing potable water)
7. Food - use of commercial shelf-stable freeze-dried food sold to the general public$250M to $350M to build each ship, presuming SpaceX is accurate in their estimate about $2M Starship Super Heavy launch costs.
The 2019 list prices for Boeing 787s range between $248.3M and $338.4M.
The Lockheed-Martin Freedom-class Littoral Combat Ship costs $362M and weighs 3,500t at full load.
The ship I want to create would weigh about 1,000t. ISS weighs 420t for comparison purposes. We require at least 1,000N of thrust to achieve acceptable acceleration rates the produce escape velocity in about a month. I think we need around 23MWe to generate 1,000N of thrust using laser ablation / thermal acceleration of H2 propellant. NASA's X3 ion engines would only require about 9.5MWe, which looks pretty good until you realize that the propellant used (Xenon) is almost unobtanium and the engine generates a maximum specific impulse of 2,650s. At that point, it doesn't look so great compared to the 86,308kg / 1,233m^3 of LH2 (something is wrong with my math there, I think it should be about 200t of propellant for 8.9km/s of dV, which is about 0.9km/s more than you actually need) that the laser ablation propulsion would required, at 4,077s. It's hard to say exactly how much laser power is required because lasers can generate thrusts between 100N/MW and 10,000N/MW, but the higher thrust levels also sacrifice specific impulse and require metallic propellants such as Aluminum.
Either way, we need to de-couple the power generation facility from the ship itself. A laser power / propulsion station that sits in high orbit can direct the laser onto mirrors used by ships to achieve escape velocity. The power station's mass and weight does not need to be endlessly moved about the solar system. These stations can be deployed using conventional electric propulsion at very low acceleration rates.
Even at the expense of specific impulse, Aluminum is a much more desirable propellant than Hydrogen because it's so dense and indefinitely storable. If spin-launch can cheaply hurl cannon balls into orbit, then Hydrogen is probably a technological dead-end, except for chemical propulsion to get off the surface of a planet.
Offline
For SpaceNut .... this topic, and the Large Ship topic of RobertDyck, are potential sources of inspiration for folks who might want to prove all these theories, one way or the other, in space. Theoretical mental models are the first step. Computer models are next, and at least two members of the forum are capable of generating those. beyond those digital models are the real-Universe flying demonstrations.
It is past time for the members of this forum to try to get beyond an endless flow of words to actual demonstrations.
If we do not have the needed expertise in house, then we have the opportunity to increase our recruiting efforts.
We need engineers of every kind, who are willing to donate their expertise in a good cause, which is helping humans to set up shop on Mars.
(th)
Offline
https://ntrs.nasa.gov/api/citations/200 … 001008.pdf
PHYSICS OF ARTIFICIAL GRAVITY
https://www.reddit.com/r/KerbalSpacePro … tat_rings/
https://www.quora.com/Why-doesnt-the-In … -occupants
https://www.quora.com/Would-a-space-sta … stribution
http://spacearchitect.org/pubs/IAC-16-D3.3.4.pdf
Artificial Gravity Conceptual Orbiting Station Design
https://www.reddit.com/r/space/comments … em_of_the/
ree-posting from zoom
I keep going back-and-forth on this idea of distributed power and propulsion, but I don't think we can get away from it using technology we actually know how to make and use. The alternatives are unaffordable propellant costs, thus unaffordable ticket costs, and "habitat modules" that closely resemble the upper stages of rockets. To be fair, this ship is still a giant tin can, but at least it's not almost entirely propellant.
Using the Alibaba price of $610/t for rather cheap 1026 carbon steel, which is still sufficiently strong, even if it's nowhere near the strength of the 2X more expensive C250 maraging steel, or the 3X more expensive 304L stainless steel- which is still weaker than 1026, our total steel bill to enclose 29,608m^3 of interior volume in 2mm thick steel, providing an opulent 59.216m^3 per person, is less than the price of most houses. NASA's long duration habitability studies says 25m^3 of space per person is the "Gold Standard".
The surface area for the pair of 25m inner radius / 5m torus diameter / 5,922m^2 surface area per torus, totals to 11,844m^2 of surface area. At 15.68kg/m^2 for 2mm thick 1026 plate, that works out to 185,713.92kg or 186t. 186t * $610/t for 1026 sheet/plate = $113,460 for enough steel to provide more than double the habitable volume that NASA says is required for crew comfort and privacy. The price of steel could double or triple or quadruple, but that's still "noise" when compared to the total cost of rocket transport for fabrication / construction, consumables, and propellant.
If we were sensible about our total habitation ring volume allocated to crew use, then we only need 93t of steel for that purpose, with the other 93t reserved for the connected spokes and center barrel sections. Some additional steel will also be required for internal reinforcement / compartmentalization. I'm budgeting around 250t for structures. The interior will be rather plain in appearance, because there's not enough mass that can be devoted to outfitting the vessel like a luxury cruise liner.
The crew will sleep in hemp "sail cloth" hammocks, similar to the crews of sailing vessels. Each man / woman / child will have a small pillow, a wool blanket, personal clothing items, a towel, a washcloth, toiletries (soap / shampoo / toothbrush / toothpaste / comb / razor / nail clippers and file), a single pair of sneakers, a wristwatch / personal communicator, and a tablet for learning or entertainment. There needs to be a cap at 20kg for all personal effects, which will include silverware / plate / cup.
After more thought on how much food the ship is carrying, I don't think it's realistic to carry more than 9 months of food. There's no practical way to send it back to Earth without first stopping in orbit at Mars. At 1.25kg of food per person per day, that's still almost 169t of food. If we carry 200L of water per person, then we're add 100t of water.
By the time we include the hull structures, food, water, atmosphere, crew, and personal effects, we're rapidly closing in on 600t. We still haven't included the mass of a 500kW solar array, communications equipment, avionics, life support equipment, etc. All that stuff adds up really fast.
I think we need at least 4 Starship flights to load all consumables, so we're already at $16,000 per person simply to get to Mars. If the ticket prices get much more expensive, then there simply won't be a sufficiently large pool of applicants who can afford to pay their own way, which means it won't succeed as a business proposition. Can we find at least a million people who can pay that kind of money? Probably, but they also have to be mentally suitable, pass physical health screenings, pass a rigorous training program, and then be gainfully employed on Mars. In short, we're asking for a lot. We have to prove that we can make a large ship concept work well enough to dissuade people like Elon Musk from attempting to use much smaller ships where vastly more money has been paid for propellant vs people and useful cargo. It's a tough sell, but I think there's an economics-based case to be made.
Offline
Blender tutorial. I downloaded Blender 2.93 which was the latest at that time, so I watched the tutorial for Blender 2.8. As of last December there's an updated version of the tutorial, now designed for Blender 3.0. The video is produced by a guy called "Blender Guru", and he's known for his tutorial to produce a donut. It's YouTube videos, starting with part 1. Each part leads you to the next one. He also has more advanced tutorials, but trust me you will need to start with the Beginner Tutorial.
Online
This quote is from one of the links SpaceNut found, above in the topic.
https://www.reddit.com/r/KerbalSpacePro … tat_rings/
level 2
NonstandardDeviation
·
8 yr. ago
The only one I can think of personally is that with two rings of the same mass counter-rotating, the overall angular momentum of the station will be zero. So if both were braked to a stop, the entire station would not end up rotating. That might make maintenance/construction work easier, as it would avoid needing to burn RCS to spin-up or spin-down. (However, if you have reaction wheels or control moment gyroscopes, you can store the angular momentum of a ring in them while it's stopped and dump the spin back into it once you're done.)Additionally, without any net angular momentum, the station wouldn't precess at all. With a one-ring station, a torque perpendicular to the axis of rotation would cause the station to gyroscope and start spinning in the third axis (i.e. use the right-hand-rule to find the cross product of the angular momentum X the torque). For example, the ISS deals with quite a bit of torque from the gravity gradient, as there's a nonzero difference in gravity's pull on the bits of the station closer and further from Earth. It also sees some torque from atmospheric drag, Earth's magnetic field, and solar pressure. If a one-ring station rotating along the roll axis (axis of ring goes prograde-retrograde) were to experience a torque in the pitch axis, it would counterintuitively start rotating in the yaw. This station would have that effect canceled out between its two rings, and overall wouldn't see any effect, although the middle of the hub between the two rings would see a bit of bending force.
NonstandardDeviation has done a good job of explaining the rationale for two counter-rotating habitats.
(th)
Offline
Here is another image of a counter rotating ring
center hub isolated
Offline
For SpaceNut re #116
Thank you for both graphics in Post #116
Thank you in particular for showing the animated drawing! That is a digital graphic, and it has the distinct advantage of not having to deal with real-Universe concerns. However, it ** does ** provide an opportunity for viewers to think about the physical properties of the two cylinders that carry force from the non-rotating central shaft to the rotating habitats.
The entire mass of the rotating habitats will be brought to bear on the welds at the base of those cylinders, when thrust is applied to the stern by those impressive looking rocket engines.
My expectation is that with the first burn of those engines, the center shaft will rip free of the habitats, drive through them, and continue to the destination without them.
The space station is more robust, with four connectors between central shaft and the habitats, and it is NOT subject to massive thrust along the Z axis. However, the station ** will ** need to be able to change orbit in response to drag or other forces present in LEO.
Since this topic was created by kbd512, I am confident he will eventually display engineering drawings showing that his rotating habitats will be mounted to the central shaft in a way that allows them to remain attached when thrust is applied in the Z axis.
(th)
Offline
SpaceNut,
The second drawing from your post #116 is pretty much what I had in mind. The narrow center barrel connection between the two frustum-shaped center barrel sections that the habitation rings are attached to doesn't make much sense from an engineering perspective, and seems to be stylistic in nature rather than practical. If this was a practical design, then the center barrel connection would be the same diameter as the center barrel sections that the two habitation rings are attached to, greatly increasing the stiffness of the structure to mitigate any compression loads that resulted from asymmetric loading generated by a mass imbalance inside the habitation rings.
tahanson43206,
I've already indicated that the thrust applied will be fairly low. This is where I differ from you and RobertDyck. Given the mass of the ships in question, low thrust applied over time negates the requirement for stronger components to withstand the force generated by a high thrust propulsion system. Most of the load from the spinning habitation rings to the center barrel sections is radial, in the X-Y plane. The load in the Z plane / axis is mostly a function of the propulsion system used.
Offline
For kbd512 re topic .... This topic is of great interest (to me for sure and hopefully to others)...
It has been a couple of weeks since your last post. The ship is not yet (quite) ready for construction. There may be a few details to add, so I'm hoping you will keep your interest in this important topic.
A working model of the counter-rotating concept is a worthy goal, and it should be achievable, with a bit of sustained effort.
A figure of $2,000,000 (USD) for construction and deployment of a 200 Kg test article was tossed about in the forum recently. A person with deep pockets and a strong interest in a demonstration of the validity of your idea could afford the entire venture.
Afterthought: The ability of the Captain of a counter-rotating ship to point accurately is a potential advantage of your design.
Please pursue it so that orientation control can be studied, and (hopefully) demonstrated.
A satellite with three momentum wheels aligned on the 3 major axis can point itself accurately in any desired direction.
It's not clear (to me for sure) what control the captain of a ship with counter rotating momentum wheels might have.
By advancing or retarding the momentum of one wheel with respect to the other, the Captain can change the orientation of the central axis.
What changes are possible would (potentially) be of interest to future readers of your topic.
It ** should ** even be possible to use the physics engine of Blender to produce animation that accurately reproduces the physical system in digital form.
Blender runs and Linux and Windows, and ** surely ** it runs on Mac.
(th)
Offline
tahanson43206,
I'm trying to read enough of what I can on new propulsion technologies to determine whether this entire concept is feasible at any level, because it requires a tremendous amount of power and capital investment in the infrastructure (orbital power transfer satellites). The ships themselves are the least costly part of the entire endeavor. I'm also trying to read enough about acceptable strain values for flexible / fiber-based structures (without inducing permanent or potentially catastrophic deformation) to know how feasible the inflatable hull mass figures happen to be. This started as a half-baked idea, like all others, and now I'm working out the details. Rather than post endless revisions, I'd prefer to nail down the details and then post what I've come up with. I said this would be a learning process for myself as well as others not familiar with the various engineering topics, and while I can't speak for others, it's certainly been just that on my part. Since I have near-zero assistance and budget, that means I have to derive what I can from various online sources, and then try to cobble that together into simple models that both make sense to me and are valid from an engineering perspective.
Beyond that, I have a recent addition to our family, a new German Shepherd named Kilo, who has taken up much of my free time as of late. Oddly enough, he came to us about two weeks ago.
Offline
First things first! Welcome to Kilo!
Please consider introducing Kilo to the next Zoom.
I will try to do a better job of contacting prospective attendees to be sure there is someone in the queue when you open the session.
OK! On to the substance of what I interpret your post to mean for the forum ...
You have embarked upon a course of study in a field that humans have very little experience to even think about ....
I would imagine the technical team behind the Movie Passengers included a physicist or two, and an engineering team to support the animation staff. Those folks would have been thinking hard about the environment of the sublight speed vessel they were designing.
In your studies, you might have imagined yourself as a sole operator, bobbing about in an ocean of information, surrounded by books and papers with instant access to electronic resources of all kinds.
In contrast, ** I ** see a scholar (pro tem perhaps) taking in a flood of information and trying to make sense of it, in order to (try to) arrive at a conclusion in the far off future.
Because you are one of our younger members, there is a chance you might have the energy to take the next step, and document your research as you go. This doesn't need to be a burden.
In fact, if you get feedback from even ** one ** reader, that may be enough encouragement to keep going.
For 20 years, this forum has been an ad hoc learning environment, with absolutely no discipline (that I can discern) other than the occasion banning, which most certainly ** did ** happen.
I see in your work an opportunity to create a flow of ideas/facts/images that a student might follow to try to prepare for a career building or perhaps evening designing ** real ** space vessels.
You don't need to make a major production out of this, and (since you appear to be as practical as your topic), I'm hoping you will toss out links to such resources as you may have found useful online.
SpaceNut is a remarkable achiever in the Internet Search space, and no one would expect you (or anyone else in the forum) to achieve at ** that ** level, but what you ** do ** find useful to others could appear in your posts, along with such conclusions as you may have reached.
Finally (as i run out of steam) please consider helping me help our producing members to create a permanent repository of hard won knowledge. GW Johnson suggested an Engineering Notebook, and I would very much like to see that happen with DropBox. Unfortunately, I am ** not ** an engineer, and have no idea what that might look like.
So far, despite requests, GW Johnson has not published anything (that I have seen, which admittedly is much less than all) that I can work with.
If you would like to set up an Engineering Notebook for your version of Large Ship, you can start the process by asking GW Johnson what he has in mind.
We can (hopefully) then work together to make it happen.
(th)
Offline
I keep coming back to the propulsion aspect of this ship because it requires so much power. So long as we're going to seriously radiation-harden this ship, then we may as well take our time transiting through the Van Allen belts. I don't think we need to invoke magnetized plasma beaming or nuclear isotope power. We can propel this ship using onboard solar electric power alone, in conjunction with conventional MPD thrusters. That brings the ship's power and propulsion scheme back into the realm of real physical hardware that we've either flown or tested extensively. We can also eliminate the giant batteries on the solar power satellites by only thrusting during periods when the ship is in direct sunlight. The mass of the ship goes up because its carrying the propellant with it, but the total mass on-orbit goes down and the headaches of maintaining a constellation of plasma beaming satellites is entirely eliminated.
If we can supply about 8MW of power 50% of the time, then we should be able to achieve escape velocity in about 16 weeks or so. Since the thrust will be quite low, the solar array sizes can be relatively large. At 200W/kg, 8MW of power translates into about 40,000kg for MegaFlex arrays, excluding support structures and PMAD. The PMAD mass should be around 14,400kg. Given a demonstrated (demonstrated as recently as last year during MPD thruster developmental experiments by ESA in Germany) 5,000s Isp for 250kW to 1MW MPD thrusters, 600t of propellant provides ~23km/s of delta-V (enough to complete a single round-trip). That 5,000s figure was achieved using storable propellants like Ammonia and Argon, rather than Hydrogen. If the propellant is LNH3, then a 6m diameter spherical tank provides more than sufficient volume. That seems pretty doable to me.
Edit: The 1MW MPD thrusters are about the same size as a 5 gallon bucket, meaning tiny considering how much power is going through them. The entire thruster literally glows white hot in operation. Anyway, since this new thruster is being developed for ESA, I figure NASA can lean on one of their more reliable partners to develop and supply the engines for these ships.
Last edited by kbd512 (2022-03-16 02:53:06)
Offline
After reviewing NASA's Rapid Cycle Amine Swing Bed (RCA 3.0), I think we've found a suitable candidate for the sort of massively redundant and distributed CO2 removal technology that I wanted for my ship design.
Rapid Cycle Amine (RCA 2.0) System Development
Although RCA 3.0 was intended to remove CO2 from Mars space suits and operates on a very rapid cycle time, it should be adaptable as a hand-portable shipboard life support subsystem that plugs into ship's power. I'm interested in hand-portable life support subsystems that crew members can affix to wall mounts and quickly move to different locations within the ship, so that if some portion of the ship starts leaking, experiences a power failure, or a fire (worst case scenario), any operable life support equipment can be redistributed to other parts of the ship, preferably without the use of special tools, while damage control activities are undertaken in the affected portion of the ship.
Total mass for 500 crew would be 6,365kg and total volume would be 370ft^3. Each unit weighs 12.73kg. Peak power consumption for the prior generation RCA 2.0 was 12W, standby was 3W, and time-averaged power consumption over 8 hours was 3.1W. At peak power consumption 6kW would have to be supplied by the onboard solar array. Average power consumption is only 1.55kW. This is a drastic reduction in power consumption over prior generations of CO2 scrubbers, but doesn't account for the significant fan power that would be required to circulate cabin air into the CO2 removal assemblies. Shipboard duct work and fans would be required to push cabin air into the CO2 removal apparatus.
I'm currently reading about US Navy submarine ventilation / air circulation requirements, and intend to use that as a proxy for the air circulation fan power requirements, based upon equivalent interior volume. I know that the fan power requirements will be significant, as they are for the CO2 scrubbers installed aboard ISS.
Current ISS upgrades include, I believe, CDRA-4EU and 4BCO2.
4-Bed CO2 Scrubber – From Design to Build
CDRA-4EU Testing to Assess Increased Number of ISS Crew
CO2 Removal Onboard the International Space Station – Material Selection and System Design
NASA-STD-3001 Technical Brief - Carbon Dioxide
Amine Swingbed Payload Testing on ISS
Parametric Analysis of Life Support Systems for Future Space Exploration Missions
Carbon Dioxide Removal Technologies for U.S. Space Vehicles: Past, Present, and Future
ECLSS reliability analysis tool for long duration spaceflight
4BCO2 was sized to provide CO2 scrubbing for 4 crew members, so we have to scale up the mass and power by a factor of 125 for 500 crew / colonists. 4BCO2 weighs 400lbs. Peak power consumption is 1.35kW. Average power consumption is 0.975kW. ISS legacy CDRA peak power consumption is 1.5kW and average consumption is 1.1kW. After 4BCO2 has been scaled up to provide adequate CO2 removal for 500 people, we're looking at 168.75kW peak consumption and 121.875kW average consumption if power consumption scales linearly with capacity. There's little reason to think it wouldn't since we're opting for massively redundant life support. Pumping power and electrical heating accounts for nearly all of the power consumption. Total weight would be 50,000lbs / 22,686kg. The individual modules are quite heavy, but still hand-portable.
Paragon SDC's Ionomer Water Processor (IWP) requires about 93kWh of power over 9 days of operation, or 10.33kWh/day. It was sized to process urine / brine for a crew of 6, but operates about 3X faster than the required processing rate. That's okay, because we want generous margins / excess system capacity in case a given unit fails. IWP's Constant power requirement is therefore 430W per 6 crew. Therefore, when scaled up to provide water processing for 500 crew, we're looking at 36.12kW of constant input power. The filtered water product requires some polishing for optimum taste and to remove the faint Ammonia odor, but is also potable according to standards, as-is.
ACPH = Air Changes Per Hour
ACPH = 3.6Q / Vol
Q= Volumetric flow rate of air in liters per second
1m^3 per minute = 16.67L/s
Vol = 5,053m^3 for a torus with a 12m inner radius / 20m outer radius / 4m tube radius
1L/s = 2.1188799728
Vol = 10,106m^3
10 ACPH / 1 air change every 6 minutes = 101,060m^3/hour, or a little less than 60,000cfm
A typical belt-drive 10,000cfm fan is 3/4hp, so 7.5795hp to meet the mass flow requirement. We're rounding up to 10hp for inefficiency.
Our constant power requirements for life support are thus:
6kW for the RCA 3.0 CO2 scrubbers
4.5kW fan power to circulate cabin air into the CO2 scrubbers; potetnially 9kW if half the ship is unusable after a fire
36.12kW of power for the IWP waste water / brine filtration system
No idea what cooking / cleaning / washing power consumption will be, but again, probably considerable
No idea how much power for lighting, but I'll figure this out
51.12kW constant power requirement for minimal life support. That is drastically lower than I originally thought it would be, but about 1/2 of ISS power generating capabilities. We haven't accounted for O2 generation, waste water polishing, stabilization gyros, avionics, or personal electronics. If we budget 40Wh per person for personal electronics (more than the total battery capacity of an iPad Pro plus an iWatch), then that's 20kWh/day, or 0.834W constant power.
Hysata's reverse fuel cell consumes 41.5kWh of electricity to produce 1kg of H2, and 8kg of O2 is produced as a result. NASA's ISS OGA produces 1.775lbs / 0.805kg of O2 per person per day. Total requirement for 500 crew is 403kg per day. H2:O2 mass ratio is 1:8, so 51 * 41.5kWh is 2,116.5kWh/day or 88.1875kW constant power. That takes us up to 139.3075kW constant output for basic life support.
I still don't have my numbers for waste heat removal, but IIRC, human resting metabolic rate is about 300W/hr. I'm not sure if I have that correct or not, but waste heat removal also requires considerable electrical power for the Ammonia-based waste heat removal system, as used aboard ISS.
If we can supply about 150kW of constant power via whatever system (solar / batteries / fuel cell / nuclear), then that would probably cover all crew uses. That figure does NOT include the electric propulsion system. Since ISS generates about 160kW of power using its various solar arrays, that should let you how much power is devoted to supplying the myriad of science experiments.
Offline
Thanks for the post and lots of links to read.
Is the removed co2 being dumped or processed such as to keep it or fed into a greenhouse or other process to reclaim the oxygen?
Offline
SpaceNut,
CO2 is dumped overboard in this design solution. That's why an OGA is required. The alternative is stripping Carbon off of CO2, which is more energy-intensive than stripping O2 from H2O at the present time. The CO2 could be collected, but over the course of 180 days, 500 people could be expected to generate a minimum of 93,920kg, given that the average daily human CO2 exhalation rate is 2.3lbs, or just over 1kg. About 72.7% of that is recoverable O2, or 68,280kg. It would be highly desirable to recover that amount of O2, which would not have to be replaced using onboard water supplies, but the energy cost of doing so would be extreme.
You'd need 393.5kJ to split 1 moles of CO2 into Carbon and O2. Assuming near 100% efficiency, which is never a good assumption, there are 22.73 moles of CO2 per kg, so 8943.18kJ per kg, so 2,484.22Wh/kg of CO2 split. You need to split about 521.78kg/day (379.33kg of recoverable O2), so 1,296,213Wh/day, which equates to a constant power output of 54,009W.
Hysata's reverse fuel cell requires 41.5kWh to produce 8kg of O2 and 1kg of H2. The theoretical minimum energy required is about 39kWh, so 1,967,795Wh/day or constant power of 82kW to replace lost O2. That means they're pretty close to theoretical energy efficiency. I can't think of any similarly efficient CO2 splitting process, but that could easily be my lack of knowledge. SOFC's that are hot enough to strip C from CO2 are about 80% efficient, so 67.5kW constant power is required. There also needs to be a way to get rid of the waste heat. Hysata's fuel cell operates at room temperature, which is why it's so efficient.
If you ignore the EV cheerleading squad's utter BS about "theoretical energy density", since theory has so very little in common with achievable results, then a Li-CO2 battery could feasibly be "charged" / "recharged" using human-generated CO2:
Li-CO2 Batteries Promise 7 Times The Energy Density Of Lithium-ion
Offline