Gemini: We can rebuild it, we have the technology

[scottb50]

It just seems to me that Hydrogen is the very best fuel, in orbit, or anywhere else. Carbon uses a lot of physical space, Carbon is also, heavy, why take it along? <br /><br />Water is also the simples,t and safest, way to get Oxygen and Hydrogen to orbit. <br /><br />If you can provide cryogenic refriguration in orbit, Hydrogen and Oxygen are the only way to go. The technology is available. Transport and store as water, until solar power breaks them down. That means launching water, to be used as propellant, on a regular basis, as well as water that will be needed by visitors. Not a bad thing, the more launches and advances discovered, that add even more launches.<br /><br /><br /><br /> Luckily, whenever we are in open Space, solar energy can provide forcrogenics, and operating power, well beyond Jupiter. Deliver the liquid propellants and keep them cold. Not a major leap with unlimited power. Simple, thermal blanketing, will greatly reduce refrigeration electrical requirements. <br /> <br /><br />I for one would want multiple redundancy. I have more than 20,000 hour of flying time, but Hawaii to the mainland, with less than three engines, scares me. Sorry. <br /><br />If you look at the cheapest way to go then fine, and that seems to be the trend. <div class="Discussion_UserSignature"> </div>

 

[mrmorris]

I'm still working on some of the technical portions of the document -- but "Appendix A" keeps running around in my head, so I took some time to write it down. A rough draft, anyway.<br /><br /><br />Appendix A: The Case for Gemini-X3<br /><br />Why Use the Gemini-X3 Design?<br /><br />Getting mass into space is expensive. That's been a truism since the beginning of mankind's space efforts, and it will be one for the foreseeable future. Without revolutionary technological advances -- no craft with the same capabilities as that proposed for Gemini-X3 can be built with significantly less mass. The capsule allows for the lowest mass of spacecraft designs and the spacecraft itself contains nothing more than the essential hardware to get passengers to orbit, dock with a space station, and return to earth safely. Modest reductions in mass are possible given future advances in electronics, power storage, thermal protection systems, and propulsion systems. The use of composites in the structure would allow for modest reductions in the end mass as well. However, these are evolutionary improvements improvements and can be rolled into later models of Gemini-X3. <br /><br />If Gemini-X3 is built -- it will not be possible for a competitor to enter the market with a different spacecraft that has the same capabilities and a significantly lower mass. At best -- a competing craft would have comparable capabilities and an equivalent or slightly lower mass, or increased capabilities and greater mass. The intended goal is to provide manned ferry operations from Earth to objects in LEO, and the initial design has the ability to perform that task easily. The driving factor for the consumers of this service is likely to be price. Any additional capabilities will have to be extremely valuable to warrant choosing another vehicle over the G-X3. In addition, it's unlikely that a craft could be designed with additional capabilities that could not be added to Gemini-X3. Appendix B will dis

 

[mrmorris]

I've been pondering options with the Launch Escape System for some time. The SRMs used for de-orbit thrust don't have the delta-v to do it by themselves. Or put another way -- the amount of delta-v required for the LES (at least for the lower portions of the launch) is much higher than that required for the de-orbit. Increasing the number of SRMs to provide sufficient reserve for abort in the early stages of the launch makes the number ridiculously oversized for the de-orbit. I'd considered using a combination of a LE tower with the DO thrusters. The LET rocket mass could be reduced because of the SRM thrust. However -- I hate the inherent waste of thrust of a launch tower that goes unused. I understand the Kliper is supposed to use the tower rocket as a booster in the later stages of the launch -- but from all the mass calculations so far -- the Falcon-V is still going to have plenty of excess lift as it is -- so there's little need for this.<br /><br />I considered having extra SRMs and using them both for de-orbit *and* for the delta-v to match orbits with the space station. However -- that's a precision burn that wouldn't be a good match for a solid motor.<br /><br />However -- I finally thought of a perfect use for extra SRMs. Once Gemini-X3 docks at the station -- all of the SRMs which are in excess of that required for the de-orbit burn can be used to lift the station orbit. This is not a burn requiring great precision and it's an excellent means to pay back the cost of lifting the SRM mass to orbit. The DO section of Gemini-X3 then can be packed with sufficient SRMs to provide the LES, and the mass will not be wasted.

 

[mikejz]

Umm....Since you where talking about carrying a secondary payload (microsats, etc)...why not use the remaining SRMs to as a kick stage to get them to a higher orbit? Just a quick thought. Or maybe deorbit the 2nd stage and be a nice space citizen and not litter space <img src="/images/icons/smile.gif" />

 

[mrmorris]

<font color="yellow">"...microsats, etc)...why not use the remaining SRMs to as a kick stage to get them to a higher orbit?... Or maybe deorbit the 2nd stage..."</font><br /><br />Each presents the problem that the SRMs would need to be attached to something *other* than G-X3. Orientation of the SRMs would also become a major issue. Spend a few minutes seriously thinking how the SRMs could be attached to GX-3 (and later disattached from) -- but also provide thrusts to either a microsat or the 2nd stage to perform the tasks you're talking about. It's not a trivial problem.<br /><br />By contrast -- placing the SRMs in a simple, <b>fixed</b> mounting in the DO Module can provide thrust for either LES or orbit-raising. There's either no or minimal added complexity (possibly additional SRM firing control programming). It simply provides a secondary purpose for a safety system that has outlived its primary mission.

 

[najab]

The only potential problem I see with this idea is that the sudden impulse provided by lighting a SRM may damage the station or disturb experiments onboard.

 

[mrmorris]

<font color="yellow">"...sudden impulse provided by lighting a SRM..."</font><br /><br />It depends on the SRMs, of course, and how much thrust each provides. The original Gemini had four SRMs. For launch aborts (above 40,000 feet IIRC the SRMs becamse a valid abort option), all four were salvo-fired to provide a big kick. When used for de-orbit -- they were fired sequentially.<br /><br />I would assume a similar system for G-X3. Firing SRMs sequentially to lift the station orbit shouldn't provide excessive thrust levels. I believe I read the sequential firing of the SRMs provided a little under 1G of acceleration during de-orbit. When the G-X3 is pushing a station several times its mass -- a similar thrust would provide a small fraction of a G. Still might cause problems, I suppose. There's too many unknowable factors involved to state with any certainty.

 

[Peter the Dane]

I cant se anny problem using the g-x3 for lifting the station.<br /><br />the station are being lifted rutinely during each mission and they just stop their experiments while the lift are ongoing. <br /><br />otherwise it meigth be possibel to create a system to pump the fuel aboard the station, were it could be used at a better time?<br /><br />Peter Gotthardsen

 

[mrmorris]

OK -- forget ceramic tiles and metallic TPS using superalloys. Gemini-X3 is going back to ablatives. The original Gemini used Dow-Corning DC-325 as an ablative material. It ablated fairly rapidly and was therefore thick and heavy. Today DOW Corning has a COTS product: "93-104 Ablative Material". It's used to line the insides of rocket nozzles, launch structures, and rocket nosecones. It has the capability of handling the heat loads required and ablates more slowly.<br /><br />Using an ablative TPS with a replaceable heat shield is likely to be the least expensive method for handling the highest heat loads of re-entry. I Googled one source indicating a price of $750/gallon for 93-104. At a Heat flux of 850 Watts/cm², the penetration rate is .035 mm/second. I don't know if the heat flux is valid, but we'll assume so for the moment. Using that for a base -- the Gemini re-entry times from retrofire to splashdown were about 16 minutes. Only about a quarter of that time would be in an environment where the shield would be ablating, but we'll use 8 minutes to be conservative. <br /><br />8 min * 60 sec * .035 mm/sec = 16 mm.<br /><br />To be even more conservative, I'll use 1 inch as a worst-case scenario, and assuming a shield diameter of 10 feet, the volume of coverage required would be ~49 gallons or ~$37,000 of ablative material. More likely the thickness required would be a fraction of this, and in any event would thin towards the edges of the heat shield. An average thickness of 10mm is more likely -- with a cost of approximately $15,000. This is chicken feed.

 

[mrmorris]

OK -- in the previous post, I was working under the assumption that the 850 Watts/cm² was a reasonable number for the heating that the shield would encounter on re-entry. I went back to patent number 5,064,151 for the ACRV since they calculated this figure for a comparable craft and flight path: <i>"A nominal atmospheric entry velocity of approximately 26,000 fps and flight path angle of -1.3 deg produced a maximum stagnation point convective heating rate of 61 British thermal units (Btu)/ft2/sec."</i> Converting that to the units used by Dow:<br /><br />61 British thermal units (Btu)/ft2/sec<br />= 64,358 Watts/ft2<br />= 69.3 Watts/cm<br /><br />This is one heckuva lot less than 850 Watts/cm²! Ablation rates are unlikely to be linear with temperature. I doubt it ablates at 1/12th the rate of the tested values. However -- we can be sure that it ablates <b>somewhat</b> slower. Still staying conservative -- I'll assume it ablates at 50% of the tested rate, or .0175 mm/s. This will mean that given the same overstated re-entry assumptions above, that the ablative at the stagnation point would need to be 8mm thick. We'll use an overall average for the heat shield of 6mm thick and this provides the following results:<br /><br />~11.5 gallons of ablative.<br />~$8,700 cost<br /><br />The relative density is listed as 1.54, which equates to 1.54 g/cm3.<br />Which means the weight of the ablative would be ~64kg. <br />This is comparable with the weight of a TPS using advanced ceramic tiles for a fraction of the cost.<br /><br /><br />The ablative can be applied as a spray or trowled on. Considering the task -- I see it being applied as a spray by a ridiculously simple robotic system. Picture a pole mounted vertically to the ceiling of a spray shed. A crossbar at the bottom makes it into an inverted 'T'. The sprayer is attached to the crossbar on a sliding mount. The heat shield base (a parabola of aluminum -- or possibly stainless steel for higher temperature tolorance) is

 

[najab]

Could I suggest adding a ultrasonic transducer to the spraying rig, to test the shield to make sure there are no bubbles or voids in the ablative coating? Your method of applying it in layers should reduce the likelihood of this happening, but it never hurts to be careful.

 

[mrmorris]

<font color="yellow">"Could I suggest adding a ultrasonic transducer to the spraying rig"</font><br /><br />I was planning to have Spock go over it with his tricorder after the final layer, but I suppose we could use an ultracosmic transformer thingamajigy. <img src="/images/icons/smile.gif" />

 

[mikejz]

When This thread is complete I suggest a follow on thread: Gemini X3--Lets take it to the moon!

 

[mrmorris]

<font color="yellow">"...follow on thread..."</font><br /><br />While I'd certainly read the posts on such a thread, and probably contribute -- it's certain I won't be starting or leading one. Unlike all the other "Let's design/redesign a spaceship" threads I've seen on SDC, I'm trying to use real technology and verifiable math in G-X3, and keep it 100% free of both Unobtanium and Hardtogetium.<br /><br />Even with very narrow and extremely modest mission parameters -- this is really hard to do. Trying to maintain the same standards when expanding the mission goals to include lunar flybys or landings is just not practical. One of the aspects of the current G-X3 mission profile that makes it possible for me to feel reasonably confident in the numbers I'm crunching is the extremely short mission duration. Since the round trip time on the craft is only intended to be about 12-16 hours of flight time -- anything that I'm off on (oxygen consumption, power usage, etc.) can't get <b>very</b> far off -- because there's simply not enough time for it to do so. A lunar mission would entail a minimum of 3 days and that increases any errors in calculations 6x. By the same token, for the current mission, I can easily double my figures to be conservative, and not hugely impact the mass being lifted to orbit. This wouldn't be possible on longer missions.<br /><br />Besides that -- heckifIknow when this thread will be <b>complete</b>. I'm still writing up a document on the craft, and including a distillation of much of what I post here in it. Unfortunately -- it's slow going, and as I research the options -- I continue to swap out items as I determine a better alternative, or find that what I had initially suggested isn't practical for some reason. This thread is my sounding board for new ideas on the craft as that process continues. Because of the heavy dependence this craft has on the FalconV -- I had really hoped to see the FalconI launch this month. Its success

 

[mrmorris]

Since the heat shield will be a one-shot affair -- we need not treat it with kid gloves. The plan of course at this point is a ground-landing, so the heat shield was going to be damaged in any event. However, the length of the heat shield assembly could be extended and made into a crumple-zone to absorb some of the landing shock and reduce the G-forces at impact. Crumple Zones are cheap, the engineering is mature, and there are no failure modes. Properly engineered, the CZ could reduce the impact G forces by 50% or more.

 

[tap_sa]

<font color="yellow">"... The original Gemini had four SRMs. For launch aborts (above 40,000 feet IIRC the SRMs becamse a valid abort option), all four were salvo-fired to provide a big kick. When used for de-orbit -- they were fired sequentially. .... the sequential firing of the SRMs provided a little under 1G of acceleration during de-orbit ..."</font><br /><br />Compared to something like Soyuz the Gemini LES appears a little underpowered. If one SRM gives 1G then four salvo-fired gives 4G acceleration, correct? I've read that Soyuz LES gives <i>20G</i> jolt. Rough but survivable and you really want a lot of distance from that exploding launch vehicle in a split second.<br /><br />If G-X3 wants to use same SRMs for LES and launch assist/de-orbit/station push then it looks like the number of individual SRMs has to be a lot higher than 4. If we take the 20G as a baseline what is required for LES use and, say, 2G for other operations that would mean 10 individual SRMs, provided that we align the thrust of each of them through crafts center of gravity (otherwise the craft would spin). Other option could be to fan out SRMs and for normal operations fire them in pairs but then the the required number would double. That might not be such a bad thing at all, a lot of simple but reliable relativelty low thrust SRMs might be much cheaper to buy than just a few very powerful.

 

[mrmorris]

<font color="yellow">"If G-X3 wants to use same SRMs for LES and launch assist/de-orbit/station push then it looks like the number of individual SRMs has to be a lot higher than 4. "</font><br /><br />The size and number of the SRMs used is one of the greyest areas of the ship. As I mentioned earlier in the thread -- most SRMs are much larger than what is required for G-X3. In my CAD drawing 2/3 of the way back through the thread, I show six engines, but that was just a baseline assumption using motors of the same thrust as the original Gemini.<br /><br />I will note that your reply is making the assumption that all of the SRMs must be the same. That's not the case. The LES-specific SRMs will only be used for escape, and to (presumably) lift the orbit of the space station. They can be considerably more powerful than the DO SRMs.<br /><br />For example -- you might have 4 LES SRMs that provide 4G of thrust apiece and 6 DO SRMs that provide 1G of thrust apiece. Salvo fired -- this would provide 22G of thrust in the event of a launch failure. <br /><br />I believe the Nautilus is supposed to require a Proton to launch to orbit, and it will require a control module similar to (I always forget if it's Zvezda or Zarya and I'm too lazy to look it up right this second). This also requires a Proton launch. The Proton-M can take a bit under 20,000 kg to a 400km orbit, so we'll guesstimate the minimum BSS mass at about 35,000 kg. This is about 7 times the mass of Gemini 3x. Therefore -- if the LES thrusters provide about 4G to Gemini itself -- when used to lift the station, they'd provide a <b>maximum</b> of about 1/2 G. This number would drop steadily as additional segments are added to the BSS.<br />

 

[tap_sa]

<font color="yellow">"I will note that your reply is making the assumption that all of the SRMs must be the same. That's not the case. The LES-specific SRMs will only be used for escape, and to (presumably) lift the orbit of the space station. They can be considerably more powerful than the DO SRMs."</font><br /><br />Yes, I was still keeping the launch assist as on option (where you'd use LES thrust to achieve orbit) and max G quite bening for the rich Hollywood sissies traveling to Bigelow's hotel. But now that I think of it again, G-3X's destination is practically always a space station so the way you are designing to use LES is the most natural one.<br /><br />Even with 4G LES SRMs you might use them as a plan B for achieving orbit if the launch vehicle underperfoms (like Delta IVH!). You'd sacrifice boosting the station (that wouldn't be mission critical) and some passenger comfort but main mission would be saved, getting those people to the station.<br /><br />

 

[mrmorris]

Cool! Elon has announced he's going for it. I can see that I need to finish my G-X3 doc up and send it to him. Possibly he can find something in it that is of use.

 

[nacnud]

Cool <img src="/images/icons/smile.gif" />, good luck in getting it finnished.

 

[arobie]

That is awesome! It is great to hear of a company going for America's Space Prize. <br /><br />Good Luck MrMorris! You've spent alot of work on this, I hope it's considered!

 

[rancamp]

Just a real quick note as I'm playing severe "catch-up" with this whole thread...<br /><br />About the Gemini landings. Has anyone pointed out that it always landed horizontally?<br />It's not a 'base-landing' design as was Mercury or Apollo. After the mains came out a "second-leg" riser was released from a slot between the cockpits and the craft landed on it's side with the hatches in the upright position. The same as it would have under a para-foil landing.<br />There are a couple of websites with pics of the mock-up for the para-foil Gemini landing gear, but these could easily be replaced with air-bags.<br /><br />Anyway, let me finish reading up and catching up and I'll try and input more later.<br /><br />Randy

 

[najab]

Yes. This has been noted. I <i>believe</i> the idea was that we would be landing on skids.

 

[mrmorris]

<font color="yellow">"...we would be landing on skids. "</font><br /><br />Not for the first version, no. I really like the parafoil, and I'd like to incorporate it into the system. However -- for G-X3, model 1, S/N 0000001, I see Apollo-style 'two-drogue/three main' parachutes and a bottom-first landing. Five years isn't much time for a private company to develop a spacecraft and 50 million isn't much money for same. For these reasons, 'simple' trumps 'optimal'.

 

[mrmorris]

<font color="yellow">"Did you ever get a chance to look into hydrogen peroxide as a monopropellant? "</font><br /><br />OK, najaB -- I finally got around to looking into H2O2 engines. Yep -- they definitely look interesting. Safer than hydrazine, less complex than LOX/Ethanol. The isp bites (~140), but you can't have everything. Running up some numbers for G-X3:<br /><br />Dv = Isp * g * ln (W0/Wf) <br /><br />300 m/s = 140s * 9.8 m/s2 * ln (Wo/Wf) <br />300 m/s = 1372 m/s * ln (Wo/Wf) <br />0.218658892 = ln (Wo/Wf) <br />1.244406727 = Wo/Wf<br />Wf/Wo = 80.3% (i.e. ~20% propellant mass required)<br /><br /><br />Figure 4200 kg dry weight for Gemini-X3 <br />20% propellant would be 840 kg <br /><br />This is comparable to 8.9% by mass of Ethanol or about 374 kg. If Gemini-X3 truly needs 300 m/s of thrust, it will have a mass penalty of ~466 kg. However -- I was using 300 m/s as a worst-case scenario (the original Gemini only had 222 m/s of RCS dv), so it's probably not that bad.<br /><br />In addition, there's another factor involved. Given the mass range I expect G-X3 to fall into, the Falcon-V actually has the lift capacity to take it all the way to a 400km orbit. If the FalconV SSE can actually provide all of the thrust to station rendevous -- possibly even docking, then the RCS system needs <b>much</b> less dv. I contacted SpaceX and asked if this capability was planned, and Gwynne Shotwell replied indicating that it is planned for 2008. Of course given the delays in launching the I (and now the announced delay in launching the V) means that they might well miss <b>that</b> date too. However, since G-X3 <b>already</b> pretty much depends on the F-V becoming a reality to be successfull -- this adds little additional uncertainty. Given that the isp of the Merlin will be much higher -- this is definitely a more mass-optimal solution.<br /><br />If we assume that the Falcon-V will be providing curbside service for the G-X3, then the RCS system will only be used f