On-orbit techniques for strapping on fuel tanks/boosters

[tdamskov]

I know the ULV studies aim towards being able to boost very heavy loads into space, but they will be very, very expensive. Hauling heavy payloads into space is important for human exploration where large exploration craft can be pre-fabricated on Earth instead of assembled on orbit (where human workforce is exceedingly expensive and difficult). But for deep space probes something else might be needed to keep budgets and mission times reasonable.<br /><br />My idea is to haul fuel tanks into space, strapping them on to the probe thereby allowing very high deltas and short mission times. After all, the assembled craft will be have to boost a 1-2 tonne probe and the fuel along, with no atmospheric drag.<br /><br />I have been wondering if any studies on automatic or human remote assisted assembly of such a craft have been conducted? I know the recent automatic docking systems explored by the US military were only a partial success.<br /><br />Can anyone calculate what mission times / speeds would be available if one were to launch a 2 tonne probe with x tonnes worth of fuel from LEO? (let's just assume that's 6 cheap launches with e.g. the Zenit booster). Cost? What's the smartest and cheapest way be to strap on fuel tanks to an orbiting spacecraft? What shape would they be? How would they connect to the primary craft? Solid or liquid fuel? Engines?<br /><br />I'm also assuming that if techniques for automatic docking/assembly are developed, they would be useful for sending fuel tanks to e.g. Mars orbit for rendezvouz with a space probe or a human expedition?<br />

 

[qso1]

To the extent possible, payloads should be assembled and launched from Earth. It is more expensive and complex to develop craft components designed for on orbit assembly. ISS has been pretty successful as an assemble on orbit station for the most part. It is however, very expensive.<br /><br />Deep space probes are currently relatively small enough to accomplish their missions without strap on tankage. More complex, heavier missions may require probes with such tankage.<br /><br />tdamskov:<br />Can anyone calculate what mission times / speeds would be available if one were to launch a 2 tonne probe with x tonnes worth of fuel from LEO?<br /><br />Me:<br />This would be dependant on the mission. The destination which will determine the trajectory which in turn will determine the mass and propellant requirements of the mission. I would think just off the top of my head that you could launch the propellant tanks in as few as 3 Zenit flights. However, the main limitation is the Zenits payload capacity which is approximately 30,000 lbs to low orbit. If you go with cryogenic LH2 and LOX, the LOX tends to be weighty but this is probably not that big an issue.<br /><br />tdamskov:<br />Cost? What's the smartest and cheapest way be to strap on fuel tanks to an orbiting spacecraft?<br /><br />Me:<br />The quickest way to me would be probe and drogue systems similar to Apollo CSM/LM systems and QDs for the tank to spacecraft plumbing interfaces, the QD type interface rigged to connect to the spacecraft upon docking.<br /><br />tdamskov:<br />What shape would they be?<br /><br />Me:<br />The most efficient shapes for tanks are spherical or pill shaped. Simplest to manufacture. A pill shaped tank with spherical halves at both ends would have the advantage of having the spherical portions be the same as a spherical tank. That is, a spherical tank is made from two half spheres and pill tanks are the same type half spheres with a cylindrical middle.<br /><br />tdamskov:<br />Solid or liquid fuel? Engine <div class="Discussion_UserSignature"> <p><strong>My borrowed quote for the time being:</strong></p><p><em>There are three kinds of people in life. Those who make it happen, those who watch it happen...and those who do not know what happened.</em></p> </div>

 

[tdamskov]

Very interesting points.<br /><br />It is true that most deep space probes are small and efficient in terms of weight and fuel consumption. I would say that is partly because of fuel limitations. Consider anything like return missions to Jupiter moons and beyond, or missions requiring a nuclear reactor (say, a high power radar) which is high on scientist wishlists. The extra fuel would leave more flexibility in choosing targets - no need to wait for celestial mechanics to provide gravity assists, just go for a direct boost. Or take the efficient route, leaving loads of fuel on arrival. Think of a mission with a craft like Cassini, imagine the freedom to match orbits with individual moons instead of having to coast around the system, waiting for flybys. I haven't seen calculations on this, but I think the time between launch and arrival of a space probe requires a sizable workforce of scientists employed and infastructure going. Shorten the travel time, do quicker missions. Especially for deep space missions the shortened travel time would have advantages politically.<br /><br />Cost-wise I suspect the lower cost of separate cheap launches would very quickly repay the cost of developing the technology to do automated fuel tank docking. There is also the advantages of being able to launch the valuable equipment on a safer vehicle than the fuel itself.<br /><br />You are probably right that most complexity comes from automating the work beyond LEO. Is it currently out of reach technologically to develop a standard "tug" for moving cargo/tanks? Or is the incentive missing? It seems to me this "tug" would be the logical next step after learning the basics of automated LEO assembly.<br /><br />Interesting that the Apollo/LM system might provide a system off-hand. Docking would have to be serially (one after another) and not as much "strapping on" to the sides of the craft? Would there be a limit to the number of tanks you could attach this way? I'm also worried about storage of c

 

[mrmorris]

<font color="yellow">"My idea is to haul fuel tanks into space, strapping them on to the probe thereby..."</font><br /><br />Mind you, I'm not going to claim it's *much* more feasible than what you suggested, but rather than actually re-configuring your probe in orbit, it'd be simpler to have on-orbit refueling of the tanks the probe ran dry getting *into* orbit. <br /><br />Mind you -- this isn't easy either. However -- both proposals assume a spacecraft with propellant/oxidizer in orbit for said probe and a rendezvous of the two. Given that -- it's *much* easier to engineer a refueling operation than one which actually makes structural changes of the sort required to phycically connect the new tank to the existing probe.<br /><br />Oribital refueling, however, if one of the many 'Holy Grails' of space travel. As with locating the Grail -- it's easier said than done.

 

[tdamskov]

<blockquote><font class="small">In reply to:</font><hr /><p>Given that -- it's *much* easier to engineer a refueling operation than one which actually makes structural changes of the sort required to phycically connect the new tank to the existing probe.<p><hr /></p></p></blockquote><br /><br />Do you have any technical explanations (or studies pointing to) why refuelling is easier? Airplanes did in fact have drop tanks long before refuelling in air became available.<br /><br />The reason I'm interested in strapping on tanks is for flexibility and standardisation. That means creating a modularised "package" enabling any craft to attach a number of snap-in-tanks using simple yet reliable standardised docking mechanisms and astrogation systems. It would be practical not just for probes but any craft that needs to boost out of LEO.<br /><br />I'm trying to understand exactly what the limitations are. Is it the engineering complexity of handling the flow of cryogenic propellants from an arbitrary number of tanks? Is it mechanical; making a secure structural connection between two orbiting objects with minimum fuzz?<br /><br />Would it better to manufacture and launch solid fuel booster packages instead? Although wasteful to some degree on both weight and cost (each package contains an engine) could they still be effective for driving a craft out of LEO? Attaching them to a spacecraft would require very little else than pure mechanical fixing - no valves, no liquids, just a software problem of center of mass. And no cooling of cryogenics. But less flexible than liquid propellants.<br /><br />Glad to hear this is one of the "holy grails" - but I can find surprisingly little information about the subject..

 

[mrmorris]

<font color="yellow">"Airplanes did in fact have drop tanks long before refuelling in air became available. "</font><br /><br />Granted. However -- those drop tanks got strapped on *before* the plane left the ground... by people... on the ground. It was not a matter of the plane rendezvousing with the tanks in mid-air and having them automagically attach themselves to the aircraft structure and hook up fuel lines.<br /><br /><br /><font color="yellow">"Do you have any technical explanations (or studies pointing to) why refuelling is easier?"</font><br /><br />Studies -- nope. Not something I've investigated. I'm basing my statement on a logical progression. The two methods would each have to do much the same things in the early stages:<br /><br />- Launch probe into orbit<br />- Launch fuel tanks into orbit.<br />- Match orbits and rendezvous<br />- 'Dock' with each other.<br /><br />The two plans start to diverge here. <br /><br />For refueling, the docking must be in the nature of something similar to a refueling boom -- which itself has some similarities to the probe & drogue docking system of Soyuz/Progress. The 'tank' craft would pump propellant and oxidizer into the existing tanks of the probe, then undock and back away. There are no structural requirements for the tank craft beyond the ability to have the requisite tanks and perform the rendezvous/docking.<br /><br />For integrating tanks -- the docking has to be *much* more robust. For refueling, the connection has to be secure enough to last a few minutes/hours while propellant is transferred, and will never undergo acceleration. For integrating the tanks, the connection must be extremely solid. The connection will also change the cg and performance characteristics of the craft. The avionics must already know how the probe/engines/etc. perform with the original tanks. Adding strap ons means that there must be a second set of handling instructructions/etc. in order to navigate properly i

 

[j05h]

on orbit tank swapping is extremely complex, as described. There are people working on it, though propellant depots are a far more popular, flexible and less challenging solution. <br /><br />http://selenianboondocks.blogspot.com/2006/11/generic-tankers.html<br /><br />Also, there was a company that made news recently about doing a tank-swapping architecture, but I don't have my links in front of me. I think Buzz Aldrin was involved, but don't remember details. <br /><br />Prop transfer is quite a bit easier and an essential technology.<br /><br />Josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[mrmorris]

<font color="yellow">"Also, there was a company that made news recently about doing a tank-swapping architecture..."</font><br /><br />Ayup -- I certainly don't want to imply that tank swapping is <b>impossible</b> -- just more complex than refueling. Nothing is actually impossible except ideas that violate basic laws of physics. Even then -- if you can find a cooperative theoretical physicist and slip him a C-note or two, you can probably get him to write you a letter of quantum absolution. <img src="/images/icons/smile.gif" />

 

[tdamskov]

Thanks for your explanations, they illustrate much one takes for granted when it comes to the difference between human hands working on the ground and robotics working in free fall.<br /><br />I see how refueling would probably be the easier first step until human workforce becomes available for on orbit assembly. A comment on flight avionics - since they already have to compensate for the amount of fuel available in the tank (center of mass changes even during flight, propellant slushes around etc) I still think it possible to design a snap-in system which allows the geometry of the craft to stay the same in all dimensions but one, that is, length. That should make things a lot easier. Think Apollo/LEM modules. Add fuel tanks to the nose of the craft?

 

[j05h]

ROMBUS would have fulfilled both requirements - it would have modular droptanks and propellant transfer. The kicker: human serviced. The craft would have been able to land 225tons on the moon with 9 fuelling flights to LEO. <br /><br />http://www.astronautix.com/lvs/rombus.htm <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[tdamskov]

Thanks, that's an interesting link!<br /><br />The ROMBUS advantages in many ways replicate those argued for the Space Shuttle program. Reusability and low cost. Even if real life cost ballooned.<br /><br />It seems to me heavy lift vehicles won't become cost effective until we have some industry on-orbit which requires regular lofting of large payloads?<br />

 

[qso1]

tdamskov:<br />It seems to me heavy lift vehicles won't become cost effective until we have some industry on-orbit which requires regular lofting of large payloads?<br /><br />Me:<br />This is the real problem facing human spaceflight. Short of some magic bullet concept, various proposals have been thought of over the years but not developed due to cost. I haven't seen a proposal such as this one but with most of the cost of human spaceflight tied up in personnel and spent on just getting to low orbit, were probably going to have to wait till this cost comes down.<br /><br />For nearly three decades people proposed building structures out of the shuttle external fuel tank as a cheap way to a space station. The fact this has yet to happen suggests to me that theres really more expense to it than meets the eye. <div class="Discussion_UserSignature"> <p><strong>My borrowed quote for the time being:</strong></p><p><em>There are three kinds of people in life. Those who make it happen, those who watch it happen...and those who do not know what happened.</em></p> </div>