Le vendredi 30 septembre 2016 03:57:11 UTC+2, nu...@bid.nes a écrit :   
   > On Thursday, September 29, 2016 at 3:20:46 AM UTC-7, elie....@gmail.com   
   wrote:   
   > > > > Bulk is a drawback, but not such a big one I suspect. By far the most   
   > > > > energy received is from the Sun, and this is taken care of with the   
   > > > > solar-thermal engine. For the rest, you end up with a long, thin cone,   
   > > > > but this craft doesn't have to manoeuvre anyway.    
   > >    
   > > > Why not? You cannot aim your launches from several AU away to guarantee a   
   > > > hit without course corrections just from stray gusts of solar wind, never   
   > > > mind magnetic and gravitational perturbations or enemy action.   
   > >    
   > > I meant high-thrust manoeuvres, like a chemical or even nuclear rocket   
   would   
   > > give. It already has manoeuvring capabilities akin to a ion drive. But the   
   > > minimal acceleration allows for the less thrust-optimised shape of a long,   
   > > thin cone.   
   >    
   >   Remember you're constantly accelerating to get maximum bang on impact and   
   to minimize chance of interception. That means you're going to build velocity   
   rather more quickly than today's burn-a-little-then-coast-a-lot space travel.   
   You're going to need    
   a lot of lateral deltavee to do even the most minor course corrections the   
   closer you get to a target, especially for precision targeting like   
   bunker-busting.   
   >    
   >   Your assumption that the spaceframe only needs to handle lengthwise thrust   
   needs re-examining. Below you mention a point that makes it worse, turnaround   
   before impact. I think turnaround is a very bad idea, and I'll go into why   
   below.   
      
   It is constantly accelerating in an arbitrary direction, but at extremely low   
   accelerations - micro-g or probably less. This is not quite freefall, but ou   
   can have some pretty outlandish designs with that as a constraint.   
   The idea is that you already know where the target is - probably a fixed   
   installation (including orbital stations), so the entire trajectory can be   
   calculated in advance. It would include some astrogation system, for example   
   tiny sensors locked on a few    
   bright stars, planets and (obviously) the Sun, for course-correction, but   
   those can be incredibly small.   
   The point of a stealth system is that the target doesn't see it coming, and   
   thus won't be able to take counter-measures like moving a station out of the   
   way.   
   Even if the terminal phase imply detection (ideally it shouldn't), this should   
   be short enough that no meaningful counter-measure can be taken, including   
   shooting it down.   
      
   As for vertical acceleration, this is only for the catapult launch. This would   
   have tremendous acceleration for a very short time.   
      
   > > > I mentioned earlier that there will be components you can't allow to get   
   > > > very cold at any time or that have a significant required warmup time, so   
   > > > they can be buried in the warmer parts of the heat sink.    
   > >    
   > > > OTOH having different parts of it at different temps may also be   
   > > > problematic, if it's even possible. A multi-stage heat sink arrangement   
   > > > could mitigate that at the expense of more tankage and plumbing.   
   > >    
   > > With expandable coolant like here, (as described above), I would use the   
   > > opposite design: use the expandable coolant to first cool the coldest   
   parts,   
   > > then the warmer ones in order. This way, it is always colder than what it   
   is   
   > > supposed to cool down, until the solar-thermal drive part where it is   
   heated   
   > > up to possibly thousands of K.   
   >    
   > (Include correction of "expandable" to "expendable"- I wondered about that   
   too)   
   >    
   >   Simple is good but that's too simplistic to work. Just because a subsystem   
   has to run at a specific *temperature* during cruise doesn't mean it's   
   emitting more *heat* than another system that can run at a lower temperature,   
   or that either one will    
   remain constant throughout the mission profile. Consider a small transmitter   
   pointing aft (to let the launchers know where it is, I assume there's no stray   
   signal going forward to give it away to defenders) that has to run at 300 K   
   but only emits 1 watt    
   of waste heat, and a nuke sleeping at 100 K but emitting 10,000 W. It's   
   possible to equalize heat flow rates by carefully sizing the plumbing actually   
   in contact with the heat sources *for constant heat flow rate* but when the   
   rates change...   
      
   But can't the design be made so components all produce more or less constant   
   (or at least known) heat, so the cooling system can manage it? Also, I would   
   design as many systems as possible to run at 3K. Is there a reason   
   specially-designed electronics (   
   that is, not our silicon chips) can't work at those temperatures?   
   Similarly, even if the payload can't be lowered down at 3K, it should at least   
   be at a constant temperature and heat generation.   
   The point is not to make it move arbitrary payload, but to put adapted payload   
   for specific mission.   
      
   >    
   >   See, before impact the radio can be shut down and allowed to freeze solid   
   but the nuke has to come up to ~300 K so its explosives and timers will work.   
   That's going to change the total heat throughput, changing the available   
   thrust when the coolant    
   is chucked out of the exhaust. You can mitigate that somewhat with a variable   
   nozzle, which adds complexity and a potential breakdown point.    
   >    
   >   (Remember, the ship as a system has only one heat sink, the exhaust   
   nozzle.)   
      
   Then we should keep the nuke at 300K at all time (and it should give constant   
   heat). Though, can we design a nuke for 3K temperature?   
   More on terminal phase below.   
      
   Also, if we really have no choice and there is changing heat production, we   
   can vary propellant flow with clever design. If acceleration becomes too   
   great, we can spiral its direction to have a lower constant acceleration.   
   (Of course, we put bigger reserves on it.)   
   If it becomes too low, then we should have a bypass that gets more propellant   
   directly from the tank to get higher flow.   
      
   >   Internal heat budgeting is a lot more complex than just overall solar   
   collector/hull coolant-input-to-exhaust-output budgeting. That latter isn't   
   constant either- the closer to Sol you get the more heat you collect. Unless   
   your collector area is    
   variable too...   
      
   It is more complex, but it is still solvable. Engineering will be... a fun   
   problem, but there is nothing impossible there.   
      
   >   Seems to me parallel, switchable *mutually insulated* cooling subsystems   
   are unavoidable.   
      
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)