Date: Wed, 18 Jan 1989 01:00-EST From: space-tech-request@cs.cmu.edu To: "~/st/lists/stdigest" Subject: Space-tech Digest #22 Contents: Marc Ringuette Space-plane model: good or bad idea? Bill Higgins Address correction + lunar probe activities Dale Amon Water at Lunar Poles Henry Spencer Re: Satellite Servicer Marc Ringuette Craig and Joe on Modular satellites, Mars Observer Marc Ringuette Re: Modular satellites (questions posed) Joe Beckenbach Re: Modular satellites (some questions answered) ------------------------------------------------------------ Date: Wed, 18 Jan 1989 00:07-EST From: Marc Ringuette Subject: Space-plane model: good or bad idea? Here's a summary of a long discussion between Ray Collins and me on whether or not a space-plane model is a good idea. The conclusions are: 1. Ray thinks it's a good idea, and is scalable to a sea-level, 1/10 scale model of plane plus propulsion. A smaller-scale plane would use reduced speeds to give similar conditions to actual flight. 2. I think it's a bad idea, because the operating speeds for the model are certain to differ from the operating speeds for a scaled-down SCRAMjet. I still push for separate testing of airframe and propulsion. 3. Both of us agree that $100-250M is the cost of a real space-plane project, but Ray believes that $100k to $1M is sufficient to fund a volunteer- designed model, especially if you count volunteer designers' time as "free". I still hope to find answers to the questions I posed earlier. I'll duplicate them here, with some partial answers paraphrased from Ray's mail: - What are the characteristics of an appropriate scramjet? What is burned, at what temperature, and who has done the development? What kind of materials are used? - How does a scramjet compare to conventional rockets for this application? Particularly, how about engine complexity, fuel tank size, re-usability? SCRAM-jets have no moving parts and rely on the force of the air entering the engine to provide the compression needed for ignition. This provides a certain limitation in that the engine must be propelled through the air at a significant speed before it can provide thrust--specifically 500-600 mph. Ceramics are probably the best material to construct SCRAM-jets from; they have sufficient strength to withstand the forces involved and they can withstand high temperatures. There are a number of companies doing SCRAM-jet research, but the only one I am familiar with is Marquot Industries, a California based company which has a wind tunnel capable of providing hypersonic test capabilities. They have developed a SCRAM-jet with a flameout of about 5,500 mph, which is significantly higher than the fasted conventional jet. The specific impulse of a SCRAMjet is around 3500 seconds, ten times higher than most rockets. Exit velocities are probably similar to rockets, but you don't have to carry the oxidizer, and there's extra nitrogen in the air to use as reaction mass. - What kind of skin will the craft need, and what kind of similar craft have been built in testing supersonic flight? Hypersonic velocities require skin materials capable of withstanding temperatures in the thousands of degrees. This does not mean there are not materials available for hypersonic skins. There are a number of ceramics (somewhat similar to fiberglas boat construction) which are applicable. The best sample of this is a material which looks something like cardboard and has temperature resistance approaching 2,000 degrees. It is extremely light and strong; I think it would be an ideal skin for hypersonic craft. I hope to hear more on these topics if any of you can find anything out. -- Marc Ringuette [ Ray Collins's address is: ] ------------------------------ Date: Mon, 16 Jan 89 11:03 CST From: Bill Higgins-- Beam Jockey Subject: Oops! My address was wrong To: SPACE-TECH@CS.CMU.EDU Oops! On Friday the 13th, Dale Amon posted an invalid address for me in Space-Tech Digest #21. Our system people have shuffled the machines in our Vax cluster around, and FNALCDF is no longer one of 'em. Furthermore, I believe WISCVM no longer operates as a Bitnet/Internet gateway. The following addresses should work for me, Bill Higgins: HIGGINS@FNALB.BITNET HIGGINS@FNALC.BITNET HIGGINS@FNALF.BITNET You'll have to figure out your favorite gateway into Bitnet. HIGGINS%FNAL.BITNET@UICVM.UIC.EDU seems to be one. If you happen to be on the SPAN/Hepnet DECnet, try 43011::HIGGINS. *Why* would you want to talk to me? I'm vice chairman of the next Space Development Conference, in Chicago the weekend of 26-29 May. And I am in touch with various people who are trying to get a lunar polar orbiter built. Yes, Dale, I was in the corner of the Lunar Bases hotel when Gordon Woodcock, Greg Maryniak, Jim Burke, Rob Stahele, and somebody from AMSAT were discussing amateur lunar orbiter possiblities. You, Rick Tumlinson, and I were among the quiet kibitzers. It was my first inkling that this might be a pretty serious project. Who knows whether it can really be pulled off-- but, given that OSCAR/AMSAT/UoSat projects have nearly 30 years of history behind them, it's not impossible. Finding ice on the Moon is a long shot with a high payoff, but a well-designed probe would also map distributions of plenty of elements. So there would be very good science payoff even if the Moon is dry. ______meson Bill Higgins _-~ ____________-~______neutrino Fermi National Accelerator Laboratory - - ~-_ / \ ~----- proton Bitnet: HIGGINS@FNALB.BITNET | | \ / NEW! IMPROVED! SPAN/Hepnet/Physnet: 43011::HIGGINS - - Now comes with ~ Free Nobel Prizewinner Inside! ------------------------------ Date: Tue, 17 Jan 1989 14:44-EST From: Dale.Amon@H.GP.CS.CMU.EDU To: space-tech@CS.CMU.EDU Subject: Water at Lunar Poles I recently helped a friend of mine get some small funding from SSI to do an experiment in this area. I will try to give some details from memory: forgive me if I screw it up. His pre-print is at home and I am not! The expected source of lunar polar H2O would be cometary/asteroidal ices. Now this water comes mixed with other volatiles, some of which are less volatile than the water itself. Now besides the Umbral areas at the pole, there are also Penumbral areas. That is, areas that are usually in the dark but are sometimes lit and sometimes visible from Earth. It was his contention that these areas would retain some amount of the other volatiles (I think Na may have been the one, I'd have to refer to the paper). I believe that Na has a spectral line in UV that he was interested in. He compared a penumbral region to a similarly lit region elsewhere to subtract out the background. He found no Na at the limits of his resolution, and he felt that this bodes ill for the presence of polar water ice deposits. It was a fairly elegant shoestring piece of research. If you want to get more information, you can contact him directly: Graham, Francis (Assistant Professor, Physics) WorkPhone: 216-385-3805 WorkMail: Kent State University 400 East 4th St PO Box 769 East Liverpool, OH 43920 ------------------------------ Date: Tue, 17 Jan 89 00:26:07 EST From: attcan!utzoo!henry@uunet.UU.NET To: space-tech@cs.cmu.edu Subject: Re: Satellite servicer > Is it reasonable to have an solar ion engine propelled craft spiral up > through the radiation belts? Is the idea reasonable? Yes; the JPL proposal for a cheap (NASA meaning of the word) ion-propelled lunar polar orbiter was to do this. Does it create complications? Yes. Note that some of those complications will have to be faced anyway if the target is Clarke (geostationary) orbit, since that orbit is in the fringes of the outer Van Allen belt. Henry Spencer at U of Toronto Zoology uunet!attcan!utzoo!henry henry@zoo.toronto.edu ------------------------------ Date: Tue, 17 Jan 1989 12:27-EST From: Marc.Ringuette@DAISY.LEARNING.CS.CMU.EDU To: space-tech@cs.cmu.edu Subject: Modular satellites, Mars Observer Camera Here's some mail exchanged between Craig Keithley and Joe Beckenbach about designing a modular satellite, and the example of the Mars Observer Camera project that Joe has been involved with. Craig forwarded it to me, and I pulled out what I thought were the most interesting parts. I'm posting with their permission. From: C43CJK%ENG4.gm@hac2arpa.hac.com ____________________________________________________________________________ >From Joe (joe@csvax.caltech.edu, in response to Craig's Core module posting) ---------------------------------------------------------------------------- Scott Brylow [scott@moc.jpl.nasa.gov] was my supervisor when I took a working vacation with the Mars Observer Camera project. To summarize the most salient points: Mars Observer depends on the Mars Observer remaining stable relative to its orbital velocity vector: it uses line CCDs on its three optics subsystems to sweep the ground path. The experiment package itself will be within the volume of 1m by .5m by .5m, and will fly chips never before considered for flight; JPL approval bureaucracy is getting extremely hide-bound for flight hardware. MOC wants 12 megabits internal-- and the 'available chips' stopped at 8 kilobits per package. Subsequent testing of available chips netted one 1 megabit chip which had substantially better radiation resistance in the normal commercial version than any 'available' chip in its rad-hard form! Needless to say, MOC isn't just pushing technological envelopes with its quarter cubic meter! [[These comments pertain to the situation as of about 6 months ago. Some changes almost certainly have occurred since.]] MOC interfaces to the Mars Observer data/command delivery system through Manchester-encoded serial lines [MIL-STD-1553!] and has all its electronics sitting on the space craft end of the the main telescope. All Observer experiments communicate with the up/downlink via MIL-STD-1553 high speed serial interfaces in a simple twisted-pair net. To save weight, MOC is made from a graphite-epoxy frame, and [if I remember correctly] has minimal backup power. Some of the rad-hard area needed will be composed of gate-array, again a first for JPL scientific flight, and standard circuit board technology. Scott is the Ground Support Equipment engineer. In other words, he's the one responsible for getting it assembled and tested correctly. I developed a prototype system for tracking the necessary testing; the flight software was being developed on a similar system. The heavy thought stuff was on microVaxen running Ultrix; heavy designing was done on Mac IIs with CAD packages and a pen plotter; the paperwork was on Macs; circuit board design was done on an IBM clone with a fab-line's proprietary package, and an assist from the microVaxen. Your Core module would be equivalent to the Mars Observer itself, with the experiments in add-on modules. The problem would mainly be worries with radiation, though Scott Brylow should be able to direct you to suppliers of rad-hard chips and to good shielding ideas. As long as the design material is at hand [compiler, chip source, programming support tools] then it shouldn't make too much different what serves as the CPU. The small-experiment subsystem might benefit from talking to AmSat: if it's cubical, the AmSat 'Microsat' series would nestle right in. Small biological experiments would fit easily, and many physical-proporties experiments as well. Propulsion experiments would need more room, as would most imaging and sensing experiments, I would think. Now, nothing prevents you from letting an experiment span two of three contiguous small-experiment volumes? This flexibility would be a great plus for experiments which would benefit from more volume and the regularity and surety of design that a modular system provides. ______________________________________________________________________________ To Joe, from Craig ______________________________________________________________________________ What I am currently envisioning is a 50cm x 50cm x 25cm box with four openings that are 20cm x 20cm. The 20cm x 20cm openings would each accept a 20cm x 20cm x 20cm "subsystem". For discussion purposes, imagine that the box is placed so that "subsystems" slide "down" (there is no down in outer space) into the box. The bottom of each "subsystem" would have at least one "standardized" connector, thru which the power, communication, and some test points would be available to be distributed to other subsystems. Obviously, some subsystems would have special connectors. The R/F subsystem would have to connect an antenna, the power subsystem to solar panels, and so forth. After place the subsystems in the module, the module itself would be covered on one or more sides with standard solar panels. It was after the "core module" posting that I gave some greater thought to the actual physical layout. During this re-thinking, I discarded the idea of a 20000 cubic cm experiment subsystem. I decided that the modular effort would be better served by creating a generic module (box, whatever) that a number of subsystems would plug into. The choice of 20cm x 20cm x 20cm is not set in concrete. As I understand it, the AMSATS are approx. 7 or 8 inches on each side. I'll use 20cm (7.87 inches) until I get some more specific information. The 20cm x 20cm x 20cm will also hold 8 Gespac cards (If that turns out to be a viable choice of CPU/BUS). I am only operating on a gut level instinct when I suggest that building an amateur satellite is a choice between: 1. Custom frame, electronics, R/F, experiments, CPU, etc. The benefits are probably less weight, size, power consumption, and a few other factors that I am blissfully ignorant of. The draw backs are that only a select few have the engineering experience to manage the project, design the space qual'd frame, electronics, etc. Other less tangible things relate to the manufacturing, testing, and higher cost due to buying lower numbers of parts, etc. OR 2. Standard frame, standard electronics, standard all of the above. The benefits of the modular method stem from the ability of designing a standard box and getting (limited) economies of scale. Aside from lower costs of manufacturing and testing, an engineer might (hypothetically) be able to "order" a frame, some subsystems, etc from a cental source. Having ordered the parts, a complete satellite could be built, tested, and ready to fly in a month, instead of months (years?). The negatives would probably be tied to any optimization in size and mass. Given these two choices, I prefer the second. I am more likely to be successful with a modular concept than with designing a "custom" satellite. Why is this? Well, for one I don't have the experience in earthly mechanical designs, let alone space qualified stuff. Sure, we ultimately might have an advanced CAD system which allows the whole kit and kaboodle to be designed in a matter of weeks (months?), but the manufacturing and testing would be very difficult to say the least. None the less, I haven't even convinced myself that the trade off between ease of design and size/mass is (with absolute certainty) modular. It "seems" like it is. The important test of this is going to be "who stands up and is counted". If this conference returns to different discussions (read Dreams) of a half dozen different missions/satellites, then obviously no one is interested in the modular concept. If my gut level instinct can be found to have merit, and the other posters out there respond to my requests for engineering talent, then maybe we can develop a methodology that allows individuals, companies, universities, etc to create a satellite for perhaps $10,000 to $50,000 a shot. Oh, of course if no one wants this, then building one modular satellite is not much better, cheaper etc, then building a custom one. Can we find the engineering talent needed? Space frame, power system, mechanical, communication, electrical, and software engineers are required for the design of the modules and subsystems. Thanks, Craig Keithley ------------------------------------------------------------------------------- To Joe (this includes his some of his comments to Craig, in >'s) ------------------------------------------------------------------------------- The concept of the core module would be to provide a satellite to those who don't need much. Some people would not need star/sun finders, gyro, attitude adjustment, etc. Those who need more, would buy other modules. Like a "500 Watt power module", a "Navigation/attitude control module", or a "8 experiment module", etc. Structural and Heat dissipation analysis required. The "catalog" of parts might have a variety of standard "cages" (or "modules" or "boxes") available. Single high, double high, etc. Provide for removable partitions between subsystems -- or provide several different versions from the "factory" > The big question, of course, is whether the full assembly can stand >being launched, or whether the aggregate would be better off with being >assembled in space. In fact, an interesting exercise might be to aggregate a >robot/remote-piloted base for robot/remote-piloted satellite assemblers, with >an attached or nearby depot for manufacture and assembly of modular parts. >[Having the assemblers be modular as well (and the base!) would be an elegant >bootstrapping method; when the remotes get good enough, bad designs can be >stripped for parts or sent down to Terra for rework. Space On-Site Inspections >Company. 1/2 serious 1/2 :-] ------------------------------------------------------------------------------- In several postings I have indicated the need for experts in satellite design. I would be greatly interested in you recommending reading and also in your efforts to attract engineering help. The basic direction of the modular satellite concept is "To reduce the engineering overhead" in going from concept to finished product. /--------------------+---------------------------------+---------------------\ | Craig Keithley | C43CJK@ENG1.GM.HAC.COM | (805) 968-5981 | | GM DSO-SBO | C43CJK%ENG1.GM@HAC2ARPA.HAC.COM | | \--------------------+---------------------------------+---------------------/ These ideas which have been hatched out of my fertile and vivid imagination SHOULD NOT IN ANY WAY be construed to represent the opinions or policies or plans of General Motors or Delco Electronics or other subsidiaries of General Motors. ------------------------------ Date: Tue, 17 Jan 1989 00:27-EST From: Marc.Ringuette@DAISY.LEARNING.CS.CMU.EDU To: space-tech@cs.cmu.edu Subject: Re: Modular Satellites I think Craig's and Randall's earlier suggestions for modular satellites are just fine, but they haven't mentioned the questions I'm most concerned about. I think the computing should be the *last* part of the design, because until the rest of the setup is sketched out, you don't know what computing you need. How about these as some start-up questions we can try to answer? Most of them could be answered by anybody who's actually made a satellite and has a feel for the problem, but I'd like to get a handle on them. 1. What are the pros and cons of rotating vs. nonrotating satellites? I'm biased towards nonrotating, allowing better pointing of instruments. Can a sat be gyro-stabilized effectively? (I have in my imagination a little box with power leads and a couple of attachment bolts coming out of it, that magically doesn't rotate) What kind of maneuvering thrusters are appropriate? How steady can you expect a sat to be? Good enough to aim a laser consistently? What if you have the ability to track stars to maintain your heading? 2. What's a typical solar cell setup, in terms of size, weight, operating temperature and voltage, power conditioning that needs to be done, etc? 3. What are some means of communication that have been used, and what are the sizes and characteristics of an appropriate version of each? Can they send back near-real-time video, or 300 baud? How much power? 4. What are the sizes and specs of relevant thrusters that have been used (both chemical and ion thrusters)? 5. What kind of radiation is experienced in different orbits? Does shielding help, or must your parts just be radiation-resistant? Just what are the Van Allen belts, anyway? 6. How much of a problem is thermal control? Can you just wrap the thing in foil, or do you need some heat-conducting parts and perhaps even a little electric heater to make sure everything comes out right? 7. What hope is there for getting more watts of power? Sizes and weights of deployable solar cells, research on large flexible mirrors? Can solar cells take advantage of 10x the sunlight concentration? 100x? 8. Hard vacuum can cause plastic parts to stiffen up through 'outgassing'. Is this a problem for, say, insulators on wiring? Plastic connectors? Chips and circuit boards? I hope some of you are interested in heading out to the library on some of these. I'll try. Also, if you think of some questions I've missed that seem just as important, mail them to me or post them. Hi ho. -- Marc Ringuette ------------------------------ Date: Tue, 17 Jan 89 09:12:39 -0800 From: Joe Beckenbach (joe@cit-vax.caltech.edu) To: Marc.Ringuette@daisy.learning.CS.CMU.EDU Cc: space-tech@cs.cmu.edu Subject: a few quick answers A few quick answers to Marc Ringuette's questions, based on my summer with the Mars Observer Camera: Computing was worked on most of the way through the project, but the device-dependent stuff waited until the interfaces to the Mars Observer were known and until the chips to be used were pretty well selected; the image-processing algorithms were worked on a full year before I got there by a friend of mine in CS who hired out to the Geology and Planetary Sciences Division. As for startup questions: 1. [rotating vs. nonrotating] Because Mars Observer is an observational mission, nonrotation is crucial. Hardware to compensate for rotation causes extra weight, and rotation itself adds extra stresses, need for maintainence fuel [and its mass and structures], and forces most observational instruments onto the axial faces ["top", "bottom"]. Mars Observer Camera has some prime nadir area on Mars Observer. It is expected to get two 140-degree views [one in red wavelengths, one in blue] and one two-degree(?) footprint as the linear CCD arrays sweep the surface. The actual 'sweep' is done by orbital motion; the target surface resolution for MOC narrow- angle is 1.5m pixels (raw). In-flight, I would guess that the Mars Observer will not rotate, since one of the possibilities in the in-flight check-out of the instrumentation was some astronomical observations with the MOC. I'm not sure if this was viable, but I know that such would be the first truly astronomical observations done outside a plantary gravity well. [Unless Pioneers beat us to it.] 3. [baud rates] Mars Observer has several different data-return rates; I believe that tops is about 144K baud burst for image dumps and the like. I could try asking Scott Brylow, my 'supervisor' at MOC. 5. [rad shielding] MOC is bound for Mars, therefore shielding was necessary for most parts; for the rad-hard it was 'belt-and-suspenders', since those were vital pieces. Any exploration to or past Luna's orbit much seriously consider this-- I don't remember the maximal extent of the different protection zones in near-Earth space. 6. [thermal control] MOC's optics must be precisely placed, and variations in the structure precisely located and calibrated. Therefore, MOC has two or three small heaters on board to compensate for the cold sections of the trip and orbit. Thermal blanketing around the electronics helps. The MOC itself is framed by graphite- epoxy and is very sensitive to water, so calibrations must be done in a low-humidity clean room [now in place on the Caltech campus]. 8. [plastics outgassing and components] I'm not sure that this has been really heavily considered for the MOC; the optics will have no plastics around, if I remember correctly, and most of the electronics should be in a separate box, minimizing outgassing problems. I assume that the necessary solutions have been in place for a good while concerning chips and circuit boards: take the outgoing Pioneer probes. [Though God knows that the JPL people on all the encounters have been absolute miracle-workers with remote jury-rigging!] If I remember the snippets I heard while working with the Camera, the team started with five people in 1986, grew slowly to ten until summer 1988, burst to 18 for the summer, and is back to ten until MOC goes off for integration in 1991. I don't remember the price tag, but extenuating circumstances prevented making both a flight model and an engineering/backup model. The work is being done under the auspices of Caltech's Division of Geology and Planetary Sciences. ---------------------------------------------------------------------------- Joe Beckenbach Caltech CS Department ------------------------------ End of Space-tech Digest #22 *******************