Slow Stealth
Jack Tingle
work (at least once). It would depend on a lot of misdirection, really
good security, and an inattentive, though not really incompetent victim.
The key is a mass driver, driven by a fission reactor, and the fact that
most of the viewpoints humans can reach easily are in or near the plane
of the ecliptic.
For example, suppose the First Marshen Republick launches an expedition
to visit various uninhabited asteroids, a-to-b-to-c-to-d. The carefully
selected crew (who all happen to have high security clearances and tight
lips) visit the asteroids and publish their papers and all is right with
the worlds. Ten years later, an unmapped giant asteroid wipes out the
Terran Pacific Rim with giant tsunamis and earthquakes, and severely
inconveniences the rest of the earth with several nuclear winters.
It just so happens, that this asteroid was one that the Republick's
survey visited. And that the Republick has a long-standing bone to pick
with Terra.
Mass drivers probably have the lowest radiated signature of any
propulsion system. If you keep a very well designed radiator panel
aligned parallel to the ecliptic (hard, when it's a few acres, but not
impossible), very few sensors will ever see it.
Now the whole deal can be queered if someone happens to look at the
asteroid during the initial movement. Whether they see the heat
signature, or the odd whatever signature that the mass drivers will
cause, the sight of an asteroid actively changing course surely gives
the game away. You have to do it when the asteroid is obscured by the
sun from your target, and from anyone else who might rat. You can
probably also only do this once.
Still, it is a stealthy attack in space. There should be some kind of
extra credit for that, other than the massive Terran space fleet that 20
years later non-stealthily nukes the Republick into near-extinction.
Regards,
Jack Tingle
SolomonW
wjti...@hotmail.com says...
Such asteroid bombs have been discussed for at least 40 years. I suspect
that a space faring society which Earth would have to be for this
scenario to work would be able to detect such an asteroid long before it
hit.
Jack Tingle
> In article <re6dnUOVT59lgF3a...@comcast.com>,
> wjti...@hotmail.com says...
>> I occurred to me that there is one way a stealthy attack in space might
>> work (at least once). It would depend on a lot of misdirection, really
>> good security, and an inattentive, though not really incompetent victim.
> Such asteroid bombs have been discussed for at least 40 years. I suspect
> that a space faring society which Earth would have to be for this
> scenario to work would be able to detect such an asteroid long before it
> hit.
That's certainly the major risk of failure, but 'able to' and 'does'
aren't identical in public policy and budgeting.
I'm sure the Republick would have a lame, face-saving, story ready: "Oh,
that? Uh, that's our test site for our revolutionary, new, mass driver
cargo vessel... Huh? Oh. Yes, it's secret, heh! Uh, we didn't want
anyone to get the jump on our market... Really? Well, I'm sure we'll
make sure an accident like that won't happen. It's probably just a
coincidence."
As I noted above, the target has to be somewhat inattentive. The first
guy to actually try it has the advantage of surprise, if he's careful
about security. I doubt there'd be a second guy, just for the reasons
you suggest.
Regards,
Jack Tingle
dwight...@gmail.com
Sigh. The anti-stealth people were wrong. Period. They
misunderstood a bit of basic physical law.
IsaacKuo
> For example, suppose the First Marshen Republick launches an expedition
> to visit various uninhabited asteroids, a-to-b-to-c-to-d. The carefully
> selected crew (who all happen to have high security clearances and tight
> lips) visit the asteroids and publish their papers and all is right with
> the worlds. Ten years later, an unmapped giant asteroid wipes out the
> Terran Pacific Rim with giant tsunamis and earthquakes, and severely
> inconveniences the rest of the earth with several nuclear winters.
> It just so happens, that this asteroid was one that the Republick's
> survey visited. And that the Republick has a long-standing bone to pick
> with Terra.
> Mass drivers probably have the lowest radiated signature of any
> propulsion system. If you keep a very well designed radiator panel
> aligned parallel to the ecliptic (hard, when it's a few acres, but not
> impossible), very few sensors will ever see it.
> Now the whole deal can be queered if someone happens to look at the
> asteroid during the initial movement.
Or any time after the initial movement. Everyone is going to have
asteroid detectors which do NOT rely upon detecting the signature
of a rocket drive because asteroids. None of the current asteroid
detection programs rely upon detecting the signature of an
asteroid's rocket drive. Why? BECAUSE ASTEROIDS DON'T
HAVE ROCKET DRIVES.
> Still, it is a stealthy attack in space. There should be some kind of
> extra credit for that, other than the massive Terran space fleet that 20
> years later non-stealthily nukes the Republick into near-extinction.
No, it's not stealthy, because asteroids aren't stealthy.
Isaac Kuo
Bryan Derksen
> None of the current asteroid
> detection programs rely upon detecting the signature of an
> asteroid's rocket drive. Why? BECAUSE ASTEROIDS DON'T
> HAVE ROCKET DRIVES.
Clearly, the key step in this plan is to install rocket drives on
_every_ asteroid, so that the one you're moving won't stand out.
Jack Tingle
>> Now the whole deal can be queered if someone happens to look at the
>> asteroid during the initial movement.
>
> Or any time after the initial movement. Everyone is going to have
> asteroid detectors which do NOT rely upon detecting the signature
> of a rocket drive because asteroids. None of the current asteroid
> detection programs rely upon detecting the signature of an
> asteroid's rocket drive.
Detecting an asteroid requires you to be actually looking for it.
"Everyone" may or may not have detectors. "Everyone" in the Indian Ocean
should have had a tsunami network, but they didn't. If most, or all of
the hazardous ones have been mapped, I doubt there'd be much of a
program for new detection. There might likely be spot checks to update
the orbits now and then.
I also assume that the nefarious bastards would darken the albedo or
whatever it takes to make it more difficult for any program that _is_ in
place to detect it. An accelerating asteroid is clearly odd, and points
to skulduggery. An asteroid you aren't looking for, detected in an
intersecting orbit, doesn't immediately say 'smoking gun', although it
does queer the deal (if there's time).
Regards,
Jack Tingle
John Schilling
wrote:
>I occurred to me that there is one way a stealthy attack in space might
>work (at least once). It would depend on a lot of misdirection, really
>good security, and an inattentive, though not really incompetent victim.
>The key is a mass driver, driven by a fission reactor, and the fact that
>most of the viewpoints humans can reach easily are in or near the plane
>of the ecliptic.
Actually, that's false. If you're not in a great hurry, a viewpoint say
10 AU above or below the ecliptic is not significantly harder to reach
than one 10 AU out from Sol (or Earth) on the ecliptic.
[Evil Martians push an asteroid onto an impact trajectory; nobody notices]
>It just so happens, that this asteroid was one that the Republick's
>survey visited. And that the Republick has a long-standing bone to pick
>with Terra.
Which means that they should have launched a barrage of nuclear missiles;
that would have been faster, cheaper, more efficient, and slightly less
obvious.
>Mass drivers probably have the lowest radiated signature of any
>propulsion system.
Care to make a quantitative estimate for the RF and microwave signature
of a decent mass driver?
Just because the energy doesn't come out as IR, doesn't mean it isn't
a signature.
>If you keep a very well designed radiator panel aligned parallel to the
>ecliptic (hard, when it's a few acres, but not impossible), very few
>sensors will ever see it.
Except for the off-ecliptic sensors that will be deployed at negligible
cost by any spacefaring power that ever considers the possibility that
someone might try to launch a stealthy attack against them.
>Now the whole deal can be queered if someone happens to look at the
>asteroid during the initial movement.
Which they will do. Part of the problem with Stealth In Spaaaaace!!!
is that, if you're looking for high-energy propulsion systems, it is
pretty easy to look at *everything, simultaneously*. Or close enough
as makes no difference.
Worse, this particular scheme requires not only that nobody happen to
look at your asteroid, not just during the initial movement, but at
any time thereafter. Or look at where the asteroid used to be and
is supposed to still be.
Asteroids are very hard to move (yes, really, no matter how easy it
is for an SF author to describe an asteroid being moved). They are
moderately easy to find and very easy to track.
So by the time anyone can even seriously consider actually doing this
sort of thing, everyone else who matters will have mapped and charted
*all* the asteroids, and will be periodically updating those maps. A
routine update of the solar cartographical database intended to pin
down perturbations in the sixth decimal place of asteroid orbits, will
notice that there's an asteroid gone completely missing, long before
it reaches its target.
>Whether they see the heat signature, or the odd whatever signature that
>the mass drivers will cause, the sight of an asteroid actively changing
>course surely gives the game away. You have to do it when the asteroid
>is obscured by the sun from your target, and from anyone else who might
>rat.
An object can only be obscured by the sun from one direction. Well, two,
really. But your target will employ professional rats with vantage
points in many directions.
>You can probably also only do this once.
You can't do this even once. The waste heat from the nuclear reactor will
be detected by off-ecliptic sensors shortly after it goes critical. The
RF and/or microwave emissions from the extremely powerful mass driver will
be detected the moment it begins operation. And the asteroid's absence
from its normal orbit will be noticed within a few months at most.
>Still, it is a stealthy attack in space.
An attack that remains stealthy only if the enemy is incompetent and/or
impotent. We've seen plenty of proposals for that sort of thing, all of
them assuming the attacker has extensive and highly sophisticated space
infrastructure and the target is the United States of America of roughly
2000 AD. Whee.
--
*John Schilling * "Anything worth doing, *
*Member:AIAA,NRA,ACLU,SAS,LP * is worth doing for money" *
*Chief Scientist & General Partner * -13th Rule of Acquisition *
*White Elephant Research, LLC * "There is no substitute *
*John.S...@alumni.usc.edu * for success" *
*661-718-0955 or 661-275-6795 * -58th Rule of Acquisition *
IsaacKuo
> IsaacKuo wrote:
> > Or any time after the initial movement. Everyone is going to have
> > asteroid detectors which do NOT rely upon detecting the signature
> > of a rocket drive because asteroids. None of the current asteroid
> > detection programs rely upon detecting the signature of an
> > asteroid's rocket drive.
> Detecting an asteroid requires you to be actually looking for it.
Asteroids are enough of a threat that every planet and most
transport vessels will be keeping a keen eye out for them.
> "Everyone" may or may not have detectors. "Everyone" in the Indian Ocean
> should have had a tsunami network, but they didn't.
A tsunami warning network is expensive, and the relevant
countries were poor. Now, it's entirely possible that an
entire planet could have a horribly backwards economy,
and the population can't even afford backyard reflector
telescopes. They're pretty much screwed if anyone
wants to lob big rocks at them, since they won't see it
coming, stealth or no stealth.
But if a planet has enough of an economy that it's any
sort of player in the interplanetary scheme of things,
it's going to have enough of an economy for a decent
asteroid detection program (one capable of detecting
low albedo asteroids).
> If most, or all of
> the hazardous ones have been mapped, I doubt there'd be much of a
> program for new detection. There might likely be spot checks to update
> the orbits now and then.
Asteroids don't just orbit in perfect periodic orbits. Their
orbits jitter and drift due to outgassing and gravitational
interactions. Also, according to current theories there's
a constant supply of new objects from the Oort cloud.
> I also assume that the nefarious bastards would darken the albedo or
> whatever it takes to make it more difficult for any program that _is_ in
> place to detect it.
There are enough asteroids which have naturally low
albedo that any asteroid detection system will already
take that into consideration.
Isaac Kuo
dwight...@gmail.com
> On Sat, 23 Feb 2008 08:21:43 -0500, Jack Tingle <wjtin...@hotmail.com>
It's been pointed out to you multiple times that your reasoning is
based upon a factually false premise, something I think you heard but
did not understand. But given the debacle where you claimed that you
'weren't convinced' that there was no link between Aluminum and
Alzheimer's, despite being corrected repeatedly, I'm not guessing
you're going to publicly change your mind.
Luke Campbell
> Mass drivers probably have the lowest radiated signature of any
> propulsion system.
Nah, because you need to power the mass driver with something else.
Fission? You have a waste heat of about twice your usable electric
work that you can pump into the mass driver (maybe only about equal
to, but slightly greater than, your electric work if you use an
advanced helium cooled high temperature fast reactor). In this case a
nuke-thermal drive is actually stealthier than a mass driver, because
the nuke thermal exhaust comes out hot, and the exhaust cools the
reactor so you don't need radiators. Hot means a higher peak black
body spectrum, which means fewer photons radiated per unit energy
produced. Fewer photons means a lower signal to noise ratio. You are
also able to turn maybe 80% to 90% of the thermal power into useful
work, so only 10% to 20% of the total thermal power is radiated away
(mostly in the drive plume which, as I mentioned, is hot).
Now the bit about keeping the radiators edge on to your observer is
clever, but as others have noted it is not particularly effective
assuming a civilization with an extensive presence in space. And if
they don't have an extensive presence in space while you are capable
of moving asteroids, you probably out-class them so badly that there
is no need to be stealthy.
Luke
dwight...@gmail.com
I imagine this is a game where people try to keep tabs on everyone
else's sensors. They themselves radiate, after all, more-so if they
have maneuvering capability. If not, keep track of the ships that
might place them in strategic orbits. Also, no matter how far-flung
your net is, it can easily be foiled by maneuvering outside the convex
hull formed by the individual units.
Of course, coff, in real-life, the notion of any sort of manned space
war seems very unlikely. War-craft might very well be at ambient
temperature and use something like a mag-sail for propulsion. Pretty
hard to detect one of those, at least in comparison with anything else
discussed so far.
IsaacKuo
<dwight.thi...@gmail.com> wrote:
> Sigh. The anti-stealth people were wrong. Period. They
> misunderstood a bit of basic physical law.
Exactly what basic physical law is misunderstood?
Isaac Kuo
dwight...@gmail.com
The one where Schilling insisted that the radiator had to scale in
inverse proportion to directionality. It doesn't. I gave a proof
(two actually) of why this wasn't so, and consulted with someone in
the physics department who actually knows a little thermodynamics. He
said it was a new one on him. In fact, I had a thread about just this
not too long ago.
IsaacKuo
> > Now the bit about keeping the radiators edge on to your observer is
> > clever, but as others have noted it is not particularly effective
> > assuming a civilization with an extensive presence in space. And if
> > they don't have an extensive presence in space while you are capable
> > of moving asteroids, you probably out-class them so badly that there
> > is no need to be stealthy.
> I imagine this is a game where people try to keep tabs on everyone
> else's sensors. They themselves radiate, after all, more-so if they
> have maneuvering capability. If not, keep track of the ships that
> might place them in strategic orbits.
Knowing where the sensors are doesn't automatically give you
the magical ability to have them conveniently all lined up in a
plane so you can "hide" a radiator edge on with all of them
simultaneously.
You can design a radiator to radiate into a cone a bit narrower
than an entire hemisphere, but it is HARD to get a radiator to
radiate into a small cone. The basic design would be a large
actively cooled polished parabolic mirror reflecting light from
a small hot blackbody radiator. Unfortunately, some light
will be absorbed by the mirror, which is why it must be actively
cooled. You're going to be consuming a lot of energy pumping
heat from this large 3K mirror into the small hot radiator.
And that energy consumed adds to the waste heat generated.
> Also, no matter how far-flung
> your net is, it can easily be foiled by maneuvering outside the convex
> hull formed by the individual units.
This depends on how patient you are. It could take decades to
get outside the network if you use a highly visible drive to get
outside the net quickly (and thus it's something suspicious to
the enemy). Or if you lob the thing slowly with something like
a planetary mass driver, it could take centuries to get outside
the network.
This assumes it's even possible to get outside the sensor
net at all. The easiest way to launch the sensor drones is
to just send them out with just enough fuel to accelerate
outward. No deceleration burn when reaching a particular
desired radius from the Sun. Thus, the sensor network just
continuously gets bigger and bigger. Instead of wasting
resources on deceleration burns, you simply periodically
launch more sensor drones to "replace" the ones that get
too far out to be particularly useful.
Essentially, the older less sophisticated sensor drones
naturally retire themselves by cruising off into interstellar
space. As an incidental side benefit, they can scan around
interstellar space, which may be interesting on a scientific
level even if there's no significant military threat out there.
The Oort cloud is huge and retired sensor drones could
usefully do good science with a flyby. There may even
be huge Earth sized ice worlds out there in interstellar
space:
http://news.bbc.co.uk/2/hi/science/nature/7249884.stm
Isaac Kuo
dwight...@gmail.com
What is this HARD you speak of? Some numbers, please? In fact, an
arrangement that covers 1% of the sky or less should be assumed to be
doable. Assume the mirror is 99.99% reflective. You seem to want to
imply that in doing this calculation the sums converge slowly, but I
don't see you showing your reasoning as to why this is so.
> > Also, no matter how far-flung
> > your net is, it can easily be foiled by maneuvering outside the convex
> > hull formed by the individual units.
>
> This depends on how patient you are. It could take decades to
> get outside the network if you use a highly visible drive to get
> outside the net quickly (and thus it's something suspicious to
> the enemy). Or if you lob the thing slowly with something like
> a planetary mass driver, it could take centuries to get outside
> the network.
Again, you're assuming stuff without using any numbers. But so what?
Why shouldn't these sorts of manuveurings take decades?
> This assumes it's even possible to get outside the sensor
> net at all. The easiest way to launch the sensor drones is
> to just send them out with just enough fuel to accelerate
> outward. No deceleration burn when reaching a particular
> desired radius from the Sun. Thus, the sensor network just
> continuously gets bigger and bigger. Instead of wasting
> resources on deceleration burns, you simply periodically
> launch more sensor drones to "replace" the ones that get
> too far out to be particularly useful.
But, uh, if you're doing that, aren't you kinda showing where those
sensors are? The rules are most definitely, 'assume sensors can
always be perfectly stealthed and never found, while spaceships have
to obey the laws of physics.'
> Essentially, the older less sophisticated sensor drones
> naturally retire themselves by cruising off into interstellar
> space. As an incidental side benefit, they can scan around
> interstellar space, which may be interesting on a scientific
> level even if there's no significant military threat out there.
> The Oort cloud is huge and retired sensor drones could
> usefully do good science with a flyby. There may even
> be huge Earth sized ice worlds out there in interstellar
> space:
>
> http://news.bbc.co.uk/2/hi/science/nature/7249884.stm
>
> Isaac Kuo
Perhaps. But it doesn't detract from the fact that the paradigm is
that stealth is an active campaign, for _all_ components. Assuming
otherwise, or assuming more passivity is simply not realistic.
IsaacKuo
> > Exactly what basic physical law is misunderstood?
> The one where Schilling insisted that the radiator had to scale in
> inverse proportion to directionality. It doesn't. I gave a proof
> (two actually) of why this wasn't so, and consulted with someone in
> the physics department who actually knows a little thermodynamics. He
> said it was a new one on him. In fact, I had a thread about just this
> not too long ago.
What specific design of radiator do you have in mind for a
highly directional radiator?
I don't know what exactly you mean by "scale in inverse
proportion to directionality". Obviously, you can make any
radiator any size you want. If you restrict the angle it's
radiating in, you'll reduce the amount that's radiated. So,
if you're required to radiate a particular power level, then
a more directional radiator will need to be larger...but
this in and of itself is no big deal. So it would need to be
larger...so what? Just make it larger.
The real problem, as I see it, is the technical difficulty
of a highly directional radiator. If you have a great
directional radiator design whose size merely scales
inversely proportional to the directionality, then isn't
that good enough for stealth purposes? Space is
really really big.
Of course, this is assuming you ONLY depend upon
positive energy passive radiation sensors. For John
Schilling, that's fine because these sensors alone are
sufficient to detect everything. Well, everything that
matters. But I'm not satisfied with "good enough".
Why use these sensors alone when there may be
some superior alternative?
For instance, I really like the idea of the occultation
sensor network. This is a bunch of widely dispersed
passive sensor drones which stare at the stars for
anything that winks them out. Besides the obvious
military application, this system can be used to scan
for Oort objects and rogue planets that would be
invisible to any other sensors.
Isaac Kuo
IsaacKuo
> > You can design a radiator to radiate into a cone a bit narrower
> > than an entire hemisphere, but it is HARD to get a radiator to
> > radiate into a small cone. The basic design would be a large
> > actively cooled polished parabolic mirror reflecting light from
> > a small hot blackbody radiator. Unfortunately, some light
> > will be absorbed by the mirror, which is why it must be actively
> > cooled. You're going to be consuming a lot of energy pumping
> > heat from this large 3K mirror into the small hot radiator.
> > And that energy consumed adds to the waste heat generated.
> What is this HARD you speak of? Some numbers, please? In fact, an
> arrangement that covers 1% of the sky or less should be assumed to be
> doable. Assume the mirror is 99.99% reflective.
Why would I assume that? The only mirrors we have which are
that reflective are only so reflective at extremely narrow
bandwidths.
> You seem to want to
> imply that in doing this calculation the sums converge slowly, but I
> don't see you showing your reasoning as to why this is so.
I've done numbers on various "stealth" radiators before, usually
coming from the perspective of trying to design one (i.e. I was
"pro-stealth"). I don't remember the specifics, but basically I
settled on a design with a 60 degree radiation cone. I wanted
to design one with a 15 degree radiation cone, but the numbers
never came anywhere close to adding up.
If you have a better design in mind, I'm all ears.
> > This depends on how patient you are. It could take decades to
> > get outside the network if you use a highly visible drive to get
> > outside the net quickly (and thus it's something suspicious to
> > the enemy). Or if you lob the thing slowly with something like
> > a planetary mass driver, it could take centuries to get outside
> > the network.
> Again, you're assuming stuff without using any numbers.
Just using basic intuition about how long it takes to get around in
the outer solar system. Even with 300km/s class drives, it takes
decades to get around.
> But so what?
> Why shouldn't these sorts of manuveurings take decades?
Well, if you don't mind the enemy knowing exactly where you
are at all times, because he's suspicious of this rocketship
zooming out beyond the sensor network at high speed and he
tracks it with active sensors...then fine. You could very well
do some "stealthy" maneuver with a heat signature he can't
detect because you're beyond the passive sensor network.
But he's tracking you with active sensors anyway, so your
patience is a wasted effort.
On the other hand, if you'd rather the enemy not noticed,
and stealthily lobbed the ship using a planetary mass
launcher, then it could easily take centuries to get outside
the outer solar system.
> > This assumes it's even possible to get outside the sensor
> > net at all. The easiest way to launch the sensor drones is
> > to just send them out with just enough fuel to accelerate
> > outward. No deceleration burn when reaching a particular
> > desired radius from the Sun. Thus, the sensor network just
> > continuously gets bigger and bigger. Instead of wasting
> > resources on deceleration burns, you simply periodically
> > launch more sensor drones to "replace" the ones that get
> > too far out to be particularly useful.
> But, uh, if you're doing that, aren't you kinda showing where those
> sensors are? The rules are most definitely, 'assume sensors can
> always be perfectly stealthed and never found, while spaceships have
> to obey the laws of physics.'
Yes, the enemy knows where the sensor drones are. So what?
It doesn't give the enemy any particular capability to do anything
about it.
Isaac Kuo
dwight...@gmail.com
> On Feb 23, 7:56 pm, "dwight.thi...@gmail.com"
>
> <dwight.thi...@gmail.com> wrote:
> > On Feb 23, 7:41 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > > You can design a radiator to radiate into a cone a bit narrower
> > > than an entire hemisphere, but it is HARD to get a radiator to
> > > radiate into a small cone. The basic design would be a large
> > > actively cooled polished parabolic mirror reflecting light from
> > > a small hot blackbody radiator. Unfortunately, some light
> > > will be absorbed by the mirror, which is why it must be actively
> > > cooled. You're going to be consuming a lot of energy pumping
> > > heat from this large 3K mirror into the small hot radiator.
> > > And that energy consumed adds to the waste heat generated.
> > What is this HARD you speak of? Some numbers, please? In fact, an
> > arrangement that covers 1% of the sky or less should be assumed to be
> > doable. Assume the mirror is 99.99% reflective.
>
> Why would I assume that? The only mirrors we have which are
> that reflective are only so reflective at extremely narrow
> bandwidths.
Uh-huh. You've an economic power, or powers that can support a huge
space industry. Possibly parts of the solar system off Earth are
permanently manned if not outright colonized. You've got all sort of
toys like nuclear thermal rockets, at the least, but at any rate,
extremely advanced technologies beyond anything we can achieve today.
Yet mirror technology improves not a whit. Even though there is
nothing in the laws of physics that forbid such reflectivities, and
even though such reflectivities have been achieved for certain
wavelengths.
And you think this is realistic? Or do you think, as I do, that this
is swallowing camels and straining at gnats? If not, why suppose all
these other advances, but not advances in optics(and rather
pedestrian advances at that.)
>> > You seem to want to
> > imply that in doing this calculation the sums converge slowly, but I
> > don't see you showing your reasoning as to why this is so.
>
> I've done numbers on various "stealth" radiators before, usually
> coming from the perspective of trying to design one (i.e. I was
> "pro-stealth"). I don't remember the specifics, but basically I
> settled on a design with a 60 degree radiation cone. I wanted
> to design one with a 15 degree radiation cone, but the numbers
> never came anywhere close to adding up.
Instead of saying this, why don't you just show the numbers?
> If you have a better design in mind, I'm all ears.
What? A radiator whose output is redirected by an advanced, actively
cooled optics system?
Something along the lines of a paraboloid with the radiating surfaces
at the focus?
There's not a whole lot more to say.
> > > This depends on how patient you are. It could take decades to
> > > get outside the network if you use a highly visible drive to get
> > > outside the net quickly (and thus it's something suspicious to
> > > the enemy). Or if you lob the thing slowly with something like
> > > a planetary mass driver, it could take centuries to get outside
> > > the network.
> > Again, you're assuming stuff without using any numbers.
>
> Just using basic intuition about how long it takes to get around in
> the outer solar system. Even with 300km/s class drives, it takes
> decades to get around.
Really? 20th C Earth did that . . . without even 30 km/s exhausts.
Your numbers don't say what you think they do.
> > But so what?
> > Why shouldn't these sorts of manuveurings take decades?
>
> Well, if you don't mind the enemy knowing exactly where you
> are at all times, because he's suspicious of this rocketship
> zooming out beyond the sensor network at high speed and he
> tracks it with active sensors...then fine. You could very well
> do some "stealthy" maneuver with a heat signature he can't
> detect because you're beyond the passive sensor network.
> But he's tracking you with active sensors anyway, so your
> patience is a wasted effort.
But now you're invoking something else: what are these 'active'
sensors you speak of, how do they work, what is their range and
resolution? Swallowing camels again?
> On the other hand, if you'd rather the enemy not noticed,
> and stealthily lobbed the ship using a planetary mass
> launcher, then it could easily take centuries to get outside
> the outer solar system.
>
> > > This assumes it's even possible to get outside the sensor
> > > net at all. The easiest way to launch the sensor drones is
> > > to just send them out with just enough fuel to accelerate
> > > outward. No deceleration burn when reaching a particular
> > > desired radius from the Sun. Thus, the sensor network just
> > > continuously gets bigger and bigger. Instead of wasting
> > > resources on deceleration burns, you simply periodically
> > > launch more sensor drones to "replace" the ones that get
> > > too far out to be particularly useful.
> > But, uh, if you're doing that, aren't you kinda showing where those
> > sensors are? The rules are most definitely, 'assume sensors can
> > always be perfectly stealthed and never found, while spaceships have
> > to obey the laws of physics.'
>
> Yes, the enemy knows where the sensor drones are. So what?
> It doesn't give the enemy any particular capability to do anything
> about it.
It doesn't? They can't be evaded then, or taken out, or spoofed, or a
combination of the three or something else?
IsaacKuo
> > On Feb 23, 7:56 pm, "dwight.thi...@gmail.com"
> > <dwight.thi...@gmail.com> wrote:
> > > What is this HARD you speak of? Some numbers, please? In fact, an
> > > arrangement that covers 1% of the sky or less should be assumed to be
> > > doable. Assume the mirror is 99.99% reflective.
> > Why would I assume that? The only mirrors we have which are
> > that reflective are only so reflective at extremely narrow
> > bandwidths.
> Uh-huh. You've an economic power, or powers that can support a huge
> space industry. Possibly parts of the solar system off Earth are
> permanently manned if not outright colonized. You've got all sort of
> toys like nuclear thermal rockets, at the least, but at any rate,
> extremely advanced technologies beyond anything we can achieve today.
I generally don't assume the existence of magic technologies,
even if it's entirely plausible that some sort of science which
today would seem magical will be developed.
> Yet mirror technology improves not a whit. Even though there is
> nothing in the laws of physics that forbid such reflectivities, and
> even though such reflectivities have been achieved for certain
> wavelengths.
Maybe there will be clever designs or new science which gives
us neat broadband highly reflective mirrors. Maybe not. I don't
assume the existence of them.
> And you think this is realistic? Or do you think, as I do, that this
> is swallowing camels and straining at gnats? If not, why suppose all
> these other advances, but not advances in optics(and rather
> pedestrian advances at that.)
What other advances do you think I'm assuming? I tend to be
rather conservative in my speculations on future space
technology. The only specific "toy" you mentioned was
nuclear thermal rockets, which I generally do NOT assume
will be used very much, if at all.
I don't like the general economics and technological
developmental path of nuclear thermal rockets, compared
to solar electric. Solar electric is already a mature technology,
and capable of outperforming nuclear thermal. Nuclear
thermal has never even flown once. Solar electric costs
less, it performs better, it doesn't have political problems,
and it's already flown. The basic technology of solar electric
has extensive civilian terrestrial applications, which means
lots of R&D money and effort to make it even better. In
contrast, the sort of nuclear reactors suitable for nuclear
thermal have no civilian applications and practically no
terrestrial military applications. And what's the potential
eventual payoff? A drive system which underperforms
compared to cheaper solar electric. So no, I don't anticipate
nuclear thermal drives.
This is representative of my approach to future technology
speculations--I like to be conservative. If something exists
today, works well, and has great potential for continued
development, then I like it. If not, then I'll need some
convincing.
> > I've done numbers on various "stealth" radiators before, usually
> > coming from the perspective of trying to design one (i.e. I was
> > "pro-stealth"). I don't remember the specifics, but basically I
> > settled on a design with a 60 degree radiation cone. I wanted
> > to design one with a 15 degree radiation cone, but the numbers
> > never came anywhere close to adding up.
> Instead of saying this, why don't you just show the numbers?
Because I don't feel like looking them up or redoing them. You
don't like that? Well, deal with it. I simply don't feel like going
through the effort.
> > If you have a better design in mind, I'm all ears.
> What? A radiator whose output is redirected by an advanced, actively
> cooled optics system?
> Something along the lines of a paraboloid with the radiating surfaces
> at the focus?
> There's not a whole lot more to say.
That's the design which I had worked on before, and rejected
it on the grounds that the amount of active cooling required
exceeded the amount of power available. Well, I didn't
"reject" it, exactly, I just wrestled with the numbers until I
got something which worked about as well as I could make
it (a 60 degree cone).
If you have nothing more to say, then color me completely
and utterly unconvinced.
> > Just using basic intuition about how long it takes to get around in
> > the outer solar system. Even with 300km/s class drives, it takes
> > decades to get around.
> Really? 20th C Earth did that . . . without even 30 km/s exhausts.
> Your numbers don't say what you think they do.
We used gravitational assists and started off from a nice fast
moving Earth, and the only way to get somewhere fast was
a one way flyby. If you want to get around in the outer solar
system, you need to provide all of the delta-v, and you need
4x the delta-v to do a return journey.
Also, consider that Neptune if five times further away from
the Sun than Jupiter. If you want to get from Neptune to
Uranus, it will take decades even with a high performance
drive.
> > > But so what?
> > > Why shouldn't these sorts of manuveurings take decades?
> > Well, if you don't mind the enemy knowing exactly where you
> > are at all times, because he's suspicious of this rocketship
> > zooming out beyond the sensor network at high speed and he
> > tracks it with active sensors...then fine. You could very well
> > do some "stealthy" maneuver with a heat signature he can't
> > detect because you're beyond the passive sensor network.
> > But he's tracking you with active sensors anyway, so your
> > patience is a wasted effort.
> But now you're invoking something else: what are these 'active'
> sensors you speak of, how do they work, what is their range and
> resolution? Swallowing camels again?
I don't care for your tone. Are you interested in convincing
anyone of your argument or are you more interested in just
insulting others?
The relevant active sensors involved would be UV wavelength
lidar. I've had discussions about active sensors on sfconsim-l,
and it's an interesting interplay between beam size and
photon energy. More energetic photons can be focused
more narrowly, but for a given beam power there are fewer
photons so your get less of a return. Also, low energy
photons can be stealthed against, by directing reflections at
off angles. It seems the ideal photon energy would be
somewhere in the UV range, which is low enough to get
a decent return signal but difficult or impossible to stealth
against (Thomson backscatter).
Now, if I feel like it I'll do the math for a specific long range
sensor design.
The basic design is straightforward enough. You'd have a
free electron laser ship in formation with a fresnel lens
drone several thousand km away. The fresnel lens may
have a radius of perhaps 100m or maybe 1km. The lens
both focuses the outgoing laser beam and focuses the
return only a germanium based detector. It only takes
one or two photons for a detection. From here, it's a
matter of math. That's effort, and frankly I don't feel
like doing that effort if it's for the benefit of a rude bastard.
> > Yes, the enemy knows where the sensor drones are. So what?
> > It doesn't give the enemy any particular capability to do anything
> > about it.
> It doesn't? They can't be evaded then, or taken out, or spoofed, or a
> combination of the three or something else?
They're too far away to be "evaded" in any meaningful
way.
You could perhaps shoot missiles at them, but it'll
take years for the missiles to reach them, during which
time the enemy is going to be launching replacement
sensor drones.
You could perhaps shoot long range interplanetary
X-ray lasers at them, which could be a significant
strategy. However, if the enemy also has
interplanetary X-ray lasers, the enemy will be using
them to shoot at your lasers; your lasers are "wasting"
firepower on sensor drones while the enemy is withering
away your own firepower.
Spoofing is a more interesting possibility, but it's
going to be an interesting challenge to try and fool
all of the sensor drones simultaneously in any way
cheaper than just doing the real thing. For example,
suppose you want to fool them into thinking that a
warship is being launched in some direction. You
want some sort of flare which matches the signature
of the drive, and accelerates the same amount as
the drive. If the enemy only had one or two sensor
drones, then maybe you could do something cheap
and clever by pointing lasers at them. But if the
enemy has dozens or hundreds or thousands of
drones, it may be cheapest just to slap a real space
drive onto a cheap rock and fly it just like a "real"
warship.
Isaac Kuo
dwight...@gmail.com
> On Feb 23, 8:42 pm, "dwight.thi...@gmail.com"
>
> <dwight.thi...@gmail.com> wrote:
> > On Feb 23, 8:17 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > > On Feb 23, 7:56 pm, "dwight.thi...@gmail.com"
> > > <dwight.thi...@gmail.com> wrote:
> > > > What is this HARD you speak of? Some numbers, please? In fact, an
> > > > arrangement that covers 1% of the sky or less should be assumed to be
> > > > doable. Assume the mirror is 99.99% reflective.
> > > Why would I assume that? The only mirrors we have which are
> > > that reflective are only so reflective at extremely narrow
> > > bandwidths.
> > Uh-huh. You've an economic power, or powers that can support a huge
> > space industry. Possibly parts of the solar system off Earth are
> > permanently manned if not outright colonized. You've got all sort of
> > toys like nuclear thermal rockets, at the least, but at any rate,
> > extremely advanced technologies beyond anything we can achieve today.
>
> I generally don't assume the existence of magic technologies,
> even if it's entirely plausible that some sort of science which
> today would seem magical will be developed.
Really? What's magical about this technology? What makes it so
difficult? Your whim? Going in the other direction:
http://www.sciencedaily.com/releases/2008/01/080122154610.htm
A material, in the beginning of 2008 mind you, that is better than
99.9% absorbtive.
Don't try to force the issue with lame rhetorical tricks like 'magic',
m'kay?
> > Yet mirror technology improves not a whit. Even though there is
> > nothing in the laws of physics that forbid such reflectivities, and
> > even though such reflectivities have been achieved for certain
> > wavelengths.
>
> Maybe there will be clever designs or new science which gives
> us neat broadband highly reflective mirrors. Maybe not. I don't
> assume the existence of them.
And yet, you _do_ assume clever designs or new science which allows a
massive manned presence in space. As I said, swallowing camels,
straining at gnats.
> > And you think this is realistic? Or do you think, as I do, that this
> > is swallowing camels and straining at gnats? If not, why suppose all
> > these other advances, but not advances in optics(and rather
> > pedestrian advances at that.)
>
> What other advances do you think I'm assuming? I tend to be
> rather conservative in my speculations on future space
> technology. The only specific "toy" you mentioned was
> nuclear thermal rockets, which I generally do NOT assume
> will be used very much, if at all.
You're not? No extremely high exhaust velocities? No scaled-up mass
drivers? No closed-loop life-support? And so on and so forth. No,
you don't get to say it's a minor adaptation of off-the-shelf stuff we
have now.
> I don't like the general economics and technological
> developmental path of nuclear thermal rockets, compared
> to solar electric. Solar electric is already a mature technology,
> and capable of outperforming nuclear thermal. Nuclear
> thermal has never even flown once. Solar electric costs
> less, it performs better, it doesn't have political problems,
> and it's already flown. The basic technology of solar electric
> has extensive civilian terrestrial applications, which means
> lots of R&D money and effort to make it even better. In
> contrast, the sort of nuclear reactors suitable for nuclear
> thermal have no civilian applications and practically no
> terrestrial military applications. And what's the potential
> eventual payoff? A drive system which underperforms
> compared to cheaper solar electric. So no, I don't anticipate
> nuclear thermal drives.
But you are assuming lots of R&D for solar electric; it's just a
matter of throwing money at the problem?
> This is representative of my approach to future technology
> speculations--I like to be conservative. If something exists
> today, works well, and has great potential for continued
> development, then I like it. If not, then I'll need some
> convincing.
You've got it backwards - you need to convince me that it works well,
first. You don't get to decide this by fiat. And the fact of the
matter is, what you're positing _is_ advanced technology, technology
we don't have yet, technology you're assuming is not terribly hard to
develop.
Your priors seem rather arbitrary to me.
> > > I've done numbers on various "stealth" radiators before, usually
> > > coming from the perspective of trying to design one (i.e. I was
> > > "pro-stealth"). I don't remember the specifics, but basically I
> > > settled on a design with a 60 degree radiation cone. I wanted
> > > to design one with a 15 degree radiation cone, but the numbers
> > > never came anywhere close to adding up.
> > Instead of saying this, why don't you just show the numbers?
>
> Because I don't feel like looking them up or redoing them. You
> don't like that? Well, deal with it. I simply don't feel like going
> through the effort.
Pardon me for not just taking your word for it, but I don't; it's
entirely possible you made unjustified assumptions, did some bad
arithmetic, etc.
> > > If you have a better design in mind, I'm all ears.
> > What? A radiator whose output is redirected by an advanced, actively
> > cooled optics system?
> > Something along the lines of a paraboloid with the radiating surfaces
> > at the focus?
> > There's not a whole lot more to say.
>
> That's the design which I had worked on before, and rejected
> it on the grounds that the amount of active cooling required
> exceeded the amount of power available. Well, I didn't
> "reject" it, exactly, I just wrestled with the numbers until I
> got something which worked about as well as I could make
> it (a 60 degree cone).
>
> If you have nothing more to say, then color me completely
> and utterly unconvinced.
Shrug. You're the one telling me that stealth in space is not
possible. It falls upon _you_ to convince _me_ that this is the
case. Not the other way around. Note that at this point, what I've
contributed, basically, is pointing out that John's bit 'wisdom' was
just plain wrong. I am most certainly not trying to convince you that
it is possible. In fact, my position is that we simply don't know
enough to tell yet, because the question turns upon many factors, some
of the economic, which are not amenable to prediction.
> > > Just using basic intuition about how long it takes to get around in
> > > the outer solar system. Even with 300km/s class drives, it takes
> > > decades to get around.
> > Really? 20th C Earth did that . . . without even 30 km/s exhausts.
> > Your numbers don't say what you think they do.
>
> We used gravitational assists and started off from a nice fast
> moving Earth, and the only way to get somewhere fast was
> a one way flyby. If you want to get around in the outer solar
> system, you need to provide all of the delta-v, and you need
> 4x the delta-v to do a return journey.
>
> Also, consider that Neptune if five times further away from
> the Sun than Jupiter. If you want to get from Neptune to
> Uranus, it will take decades even with a high performance
> drive.
I'm curious as to why you would want to get from Neptune to Uranus.
Your assumptions seem to paint a rather inconsistent picture of the
future. With what you've said so far, it doesn't seem likely that
there would be much reason for doing this, and so gravitational
assists from the inner system seem to be a perfectly fine way to get
out there.
> > > > But so what?
> > > > Why shouldn't these sorts of manuveurings take decades?
> > > Well, if you don't mind the enemy knowing exactly where you
> > > are at all times, because he's suspicious of this rocketship
> > > zooming out beyond the sensor network at high speed and he
> > > tracks it with active sensors...then fine. You could very well
> > > do some "stealthy" maneuver with a heat signature he can't
> > > detect because you're beyond the passive sensor network.
> > > But he's tracking you with active sensors anyway, so your
> > > patience is a wasted effort.
> > But now you're invoking something else: what are these 'active'
> > sensors you speak of, how do they work, what is their range and
> > resolution? Swallowing camels again?
>
> I don't care for your tone. Are you interested in convincing
> anyone of your argument or are you more interested in just
> insulting others?
No. And I don't care if you don't care for my tone; in fact, you seem
to think that after you make an assertion, that I have to work to
convince you that it isn't so.
That's not how it works. You've got to justify, defend, persuade,
etc.
Needless to say, I find your presumptions and tone annoying, freighted
as it is with these rather obnoxious presumptions.
For example, all of a sudden you're talking about 'active' sensors,
and I have no idea what they are; the only thing I can think of is
radar. And radar is notoriously power-hungry and has (on the scale of
interplanetary distances) a rather short range. So when you say
something like a highly reflective mirror is 'magic technology', but
that these 'active sensors' are practically off-the-shelf, I think I
am fully justified in pointing out that you're denying one (in my
opinion) rather modest advance, but have no problems waving around
your own favorites which seem (at least to me) way more advanced.
Iow, you just can't drag in a deux ex machina, not explain it, and
think that settles everything.
> The relevant active sensors involved would be UV wavelength
> lidar. I've had discussions about active sensors on sfconsim-l,
> and it's an interesting interplay between beam size and
> photon energy. More energetic photons can be focused
> more narrowly, but for a given beam power there are fewer
> photons so your get less of a return. Also, low energy
> photons can be stealthed against, by directing reflections at
> off angles. It seems the ideal photon energy would be
> somewhere in the UV range, which is low enough to get
> a decent return signal but difficult or impossible to stealth
> against (Thomson backscatter).
>
> Now, if I feel like it I'll do the math for a specific long range
> sensor design.
>
> The basic design is straightforward enough. You'd have a
> free electron laser ship in formation with a fresnel lens
> drone several thousand km away. The fresnel lens may
> have a radius of perhaps 100m or maybe 1km. The lens
> both focuses the outgoing laser beam and focuses the
> return only a germanium based detector. It only takes
> one or two photons for a detection. From here, it's a
> matter of math. That's effort, and frankly I don't feel
> like doing that effort if it's for the benefit of a rude bastard.
Now stop right there!
To call me rude for being skeptical, to call positing a highly-
reflective mirror 'magic technology' . . . and then go on to blithely
describe fresnel lenses with a radius of hundreds of meters (up to
one kilometer!) as not such a big deal is - I'm sorry to say again -
swallowing camels and straining at gnats.
> > > Yes, the enemy knows where the sensor drones are. So what?
> > > It doesn't give the enemy any particular capability to do anything
> > > about it.
> > It doesn't? They can't be evaded then, or taken out, or spoofed, or a
> > combination of the three or something else?
>
> They're too far away to be "evaded" in any meaningful
> way.
>
> You could perhaps shoot missiles at them, but it'll
> take years for the missiles to reach them, during which
> time the enemy is going to be launching replacement
> sensor drones.
>
> You could perhaps shoot long range interplanetary
> X-ray lasers at them, which could be a significant
> strategy. However, if the enemy also has
> interplanetary X-ray lasers, the enemy will be using
> them to shoot at your lasers; your lasers are "wasting"
> firepower on sensor drones while the enemy is withering
> away your own firepower.
Long range interplanetary X-ray lasers. Nope. Just standard tech
here, nothing to see folks, just move along.
Again, pardon me for being skeptical.
I'm not sure you're aware of how wild this really sounds.
Let's take a real-life comparison:
Darn. Can't find it. Maybe you can; it was about the relatively
recent solar sail project that was canceled. Henry Spencer was in on
that one, but my google-fu can't seem to locate it.
If we can't even do halfway decent testing of something as mundane as
solar sails, and you're talking about about fresnel lenses (In Space!)
hundreds of meters across, interplanetary x-ray lasers capable of
shooting down hostile objects . . . and then go on to describe highly
reflective mirrors as 'magic', well, I don't think we live in the same
universe.
And, one last time: I'm not trying to convince you of anything.
You're asserting that stealth in space is 'impossible'; I'm playing
the good skeptic, and have already pointed out some flaws.
Nothing more.
Tim Little
> A material, in the beginning of 2008 mind you, that is better than
> 99.9% absorbtive.
Completely different from a material that is 99.99% reflective. Such
an absorptive material in sunlight will heat up and radiate very
efficiently.
Then again there is no point making a big curved mirror 99.99%
reflective anyway, since then it will reflect sunlight. You're out of
luck either way.
- Tim
IsaacKuo
> > On Feb 23, 8:42 pm, "dwight.thi...@gmail.com"
> > > Uh-huh. You've an economic power, or powers that can support a huge
> > > space industry. Possibly parts of the solar system off Earth are
> > > permanently manned if not outright colonized. You've got all sort of
> > > toys like nuclear thermal rockets, at the least, but at any rate,
> > > extremely advanced technologies beyond anything we can achieve today.
> > I generally don't assume the existence of magic technologies,
> > even if it's entirely plausible that some sort of science which
> > today would seem magical will be developed.
> Really? What's magical about this technology? What makes it so
> difficult? Your whim? Going in the other direction:
I don't know whether it's magical. It's beyond our current
knowledge, which makes it unknown. You claim that
it's easy. I make no such assumption. It might be easy,
or it might be difficult, or it might be impossible.
> http://www.sciencedaily.com/releases/2008/01/080122154610.htm
> A material, in the beginning of 2008 mind you, that is better than
> 99.9% absorbtive.
So what?
> > > Yet mirror technology improves not a whit. Even though there is
> > > nothing in the laws of physics that forbid such reflectivities, and
> > > even though such reflectivities have been achieved for certain
> > > wavelengths.
> > Maybe there will be clever designs or new science which gives
> > us neat broadband highly reflective mirrors. Maybe not. I don't
> > assume the existence of them.
> And yet, you _do_ assume clever designs or new science which allows a
> massive manned presence in space. As I said, swallowing camels,
> straining at gnats.
No, I don't.
> > > And you think this is realistic? Or do you think, as I do, that this
> > > is swallowing camels and straining at gnats? If not, why suppose all
> > > these other advances, but not advances in optics(and rather
> > > pedestrian advances at that.)
> > What other advances do you think I'm assuming? I tend to be
> > rather conservative in my speculations on future space
> > technology. The only specific "toy" you mentioned was
> > nuclear thermal rockets, which I generally do NOT assume
> > will be used very much, if at all.
> You're not?
No, I'm not.
> No extremely high exhaust velocities?
No. Current exhaust velocities of ion drives are more or
less ideal. Increasing the exhaust velocity would decrease
efficiency, reduce acceleration, and increase trip times.
> No scaled-up mass drivers?
No. I don't care for mass driver technology for space
applications. It requires large amounts of initial
investment and R&D, and the payloads are small
and must be robust enough to survive high
accelerations. Traditional rocket drives require
propellant, but the payloads can be larger and
they don't need to survive high accelerations.
> No closed-loop life-support?
We currently live in a closed-loop life-support
biosphere, so that's a proof of concept. I
accept that as a baseline for what sort of
minimum investment might be required for
off-world colonization. That's a truly MASSIVE
investment. On sfconsim-l, I worked out a
conservative baseline design which would
be 18km in diameter and 1.5E12 tons. Just
for the habitat. I accept that much smaller
may be possible, depending on technological
advancement and/or willingness for less
than total self-sufficiency. I do not ASSUME
it, though.
> And so on and so forth. No,
> you don't get to say it's a minor adaptation of off-the-shelf stuff we
> have now.
So, the only choices are between off-the-shelf
stuff we have today, and everything you deem
must be possible. Riiiight.
> > I don't like the general economics and technological
> > developmental path of nuclear thermal rockets, compared
> > to solar electric. Solar electric is already a mature technology,
> > and capable of outperforming nuclear thermal. Nuclear
> > thermal has never even flown once. Solar electric costs
> > less, it performs better, it doesn't have political problems,
> > and it's already flown. The basic technology of solar electric
> > has extensive civilian terrestrial applications, which means
> > lots of R&D money and effort to make it even better. In
> > contrast, the sort of nuclear reactors suitable for nuclear
> > thermal have no civilian applications and practically no
> > terrestrial military applications. And what's the potential
> > eventual payoff? A drive system which underperforms
> > compared to cheaper solar electric. So no, I don't anticipate
> > nuclear thermal drives.
> But you are assuming lots of R&D for solar electric; it's just a
> matter of throwing money at the problem?
At this point, yes. Because the technology is sufficiently
mature, we're at a point where a significant amount of
improvement is indeed just a matter of throwing money
at the problem. There's a limit to how far we can predict
the improvement, of course.
> > This is representative of my approach to future technology
> > speculations--I like to be conservative. If something exists
> > today, works well, and has great potential for continued
> > development, then I like it. If not, then I'll need some
> > convincing.
> You've got it backwards - you need to convince me that it works well,
> first. You don't get to decide this by fiat. And the fact of the
> matter is, what you're positing _is_ advanced technology, technology
> we don't have yet, technology you're assuming is not terribly hard to
> develop.
I don't need to convince you anything. More and more,
it seems clear that you are just going to believe whatever
you want and no one can convince you to budge in
the slightest.
It's crazy that you insist I'm "swallowing camels" by
assuming the existence of lots of crazy advanced
technology. You don't know what I assume. And
even when I explicitly tell you what I assume, you're
still arguing the point. This is ridiculous.
> > > > I've done numbers on various "stealth" radiators before, usually
> > > > coming from the perspective of trying to design one (i.e. I was
> > > > "pro-stealth"). I don't remember the specifics, but basically I
> > > > settled on a design with a 60 degree radiation cone. I wanted
> > > > to design one with a 15 degree radiation cone, but the numbers
> > > > never came anywhere close to adding up.
> > > Instead of saying this, why don't you just show the numbers?
> > Because I don't feel like looking them up or redoing them. You
> > don't like that? Well, deal with it. I simply don't feel like going
> > through the effort.
> Pardon me for not just taking your word for it, but I don't; it's
> entirely possible you made unjustified assumptions, did some bad
> arithmetic, etc.
Sure. But that doesn't change the fact that I don't feel like
redoing the math right now. And you're not providing me
with anything to change my mood.
> > > > If you have a better design in mind, I'm all ears.
> > > What? A radiator whose output is redirected by an advanced, actively
> > > cooled optics system?
> > > Something along the lines of a paraboloid with the radiating surfaces
> > > at the focus?
> > > There's not a whole lot more to say.
> > That's the design which I had worked on before, and rejected
> > it on the grounds that the amount of active cooling required
> > exceeded the amount of power available. Well, I didn't
> > "reject" it, exactly, I just wrestled with the numbers until I
> > got something which worked about as well as I could make
> > it (a 60 degree cone).
> > If you have nothing more to say, then color me completely
> > and utterly unconvinced.
> Shrug. You're the one telling me that stealth in space is not
> possible. It falls upon _you_ to convince _me_ that this is the
> case. Not the other way around. Note that at this point, what I've
> contributed, basically, is pointing out that John's bit 'wisdom' was
> just plain wrong.
What you specifically said was this:
"Sigh. The anti-stealth people were wrong. Period. They
misunderstood a bit of basic physical law."
You're saying that we misunderstood some basic physical
law. When pressed to specify which basic physical law
was misunderstood, you said this:
"The one where Schilling insisted that the radiator had to scale in
inverse proportion to directionality. It doesn't. I gave a proof
(two actually) of why this wasn't so, and consulted with someone in
the physics department who actually knows a little thermodynamics. He
said it was a new one on him. In fact, I had a thread about just this
not too long ago. "
Okay, but it wasn't clear to me what you meant by
"scale in inverse proportion to directionality" nor the
particular relevance with this "basic" physical law
and the difficulties of stealth in space.
> I am most certainly not trying to convince you that
> it is possible. In fact, my position is that we simply don't know
> enough to tell yet, because the question turns upon many factors, some
> of the economic, which are not amenable to prediction.
No, it seems your position is that the anti-stealth people
are wrong. Not that "we simply don't know". That
we're wrong.
> > > > Just using basic intuition about how long it takes to get around in
> > > > the outer solar system. Even with 300km/s class drives, it takes
> > > > decades to get around.
> > > Really? 20th C Earth did that . . . without even 30 km/s exhausts.
> > > Your numbers don't say what you think they do.
> > We used gravitational assists and started off from a nice fast
> > moving Earth, and the only way to get somewhere fast was
> > a one way flyby. If you want to get around in the outer solar
> > system, you need to provide all of the delta-v, and you need
> > 4x the delta-v to do a return journey.
> > Also, consider that Neptune if five times further away from
> > the Sun than Jupiter. If you want to get from Neptune to
> > Uranus, it will take decades even with a high performance
> > drive.
> I'm curious as to why you would want to get from Neptune to Uranus.
Because you are in orbit around Neptune and you'd like to
get to something in orbit around Uranus. Why, specifically?
I don't know, it could be a religious war, or something.
Entire wars can be fought over stupid things.
> Your assumptions seem to paint a rather inconsistent picture of the
> future. With what you've said so far, it doesn't seem likely that
> there would be much reason for doing this, and so gravitational
> assists from the inner system seem to be a perfectly fine way to get
> out there.
Sure, it can be a fine way to get around. It'll still take
decades to get from Neptune to Uranus.
> > > But now you're invoking something else: what are these 'active'
> > > sensors you speak of, how do they work, what is their range and
> > > resolution? Swallowing camels again?
> > I don't care for your tone. Are you interested in convincing
> > anyone of your argument or are you more interested in just
> > insulting others?
> No. And I don't care if you don't care for my tone; in fact, you seem
> to think that after you make an assertion, that I have to work to
> convince you that it isn't so.
"No" is a strange way to answer the question I asked.
> That's not how it works. You've got to justify, defend, persuade,
> etc.
If I were all that interested in convincing you of my
statements, then that's perfectly true. But I'm not that
interested in convincing you of my statements. If I
were, then I'd go through the effort to put up all of
the relevant math. But as I've stated several times
already, I just don't feel like it.
Obviously, if I have a desire to convince you of my
arguments, this desire isn't strong enough to overcome
my laziness to doing all of the relevant math.
But why should I even bother? You haven't shown
the slightest good faith by justifying anything of your
own statements.
> Needless to say, I find your presumptions and tone annoying, freighted
> as it is with these rather obnoxious presumptions.
> For example, all of a sudden you're talking about 'active' sensors,
> and I have no idea what they are; the only thing I can think of is
> radar. And radar is notoriously power-hungry and has (on the scale of
> interplanetary distances) a rather short range. So when you say
> something like a highly reflective mirror is 'magic technology', but
> that these 'active sensors' are practically off-the-shelf, I think I
> am fully justified in pointing out that you're denying one (in my
> opinion) rather modest advance, but have no problems waving around
> your own favorites which seem (at least to me) way more advanced.
You think that something which no one has the faintest
idea how to make and which could easily be utterly
impossible is a "rather modest advance".
You think that radar, an existing technology which we
know can be scaled up to interplanetary scales (we've
essentially used it on interplanetary scales already),
is way more advanced. It may be extremely expensive
and not worth it because there are better alternatives,
but it's something which we could do simply by throwing
money at the problem. (Retrofit a huge ass transmitter
on Arecibo or something like that.)
Okay, you are perfectly free to feel that way. I differ
from that opinion. I'm not going to bother convincing
you otherwise. And I'll tell you right now that you're
wasting your time if you try to convince me otherwise.
BTW, I never claimed any of the technologies I'm
assuming are "off-the-shelf".
> > Now, if I feel like it I'll do the math for a specific long range
> > sensor design.
> > The basic design is straightforward enough. You'd have a
> > free electron laser ship in formation with a fresnel lens
> > drone several thousand km away. The fresnel lens may
> > have a radius of perhaps 100m or maybe 1km. The lens
> > both focuses the outgoing laser beam and focuses the
> > return only a germanium based detector. It only takes
> > one or two photons for a detection. From here, it's a
> > matter of math. That's effort, and frankly I don't feel
> > like doing that effort if it's for the benefit of a rude bastard.
> Now stop right there!
> To call me rude for being skeptical, to call positing a highly-
> reflective mirror 'magic technology' . . . and then go on to blithely
> describe fresnel lenses with a radius of hundreds of meters (up to
> one kilometer!) as not such a big deal is - I'm sorry to say again -
> swallowing camels and straining at gnats.
The difference is, I actually have an idea how to go
and build the thing. Hand me a billion dollars and
I'll get to work. You, in contrast, don't have the
slightest clue how to make a broadband highly
reflective mirror, nor do you even know if it's possible.
As for the UV laser system--I haven't described to
you the specifics and the many really annoying aspects.
I have, however, worked out the big ones to myself.
The design of the free electron laser in particular
has a whole slew of non-obvious technical issues
which only become clear when you hammer down
specific designs and work out various formulas.
The optics side of things is less speculative. There's
current research into the "eyeglass" telescope
configuration, which involves a large diameter
fresnel lens drone spaced far away from the
main sensor array drone. This includes foldable
fresnel lenses which have been tested in the
lab, although not flown yet. It's likely that one of
the next generation of space based telescopes
will use such a configuration, perhaps with the
primary mission of directly imaging exoplanets.
> > > > Yes, the enemy knows where the sensor drones are. So what?
> > > > It doesn't give the enemy any particular capability to do anything
> > > > about it.
> > > It doesn't? They can't be evaded then, or taken out, or spoofed, or a
> > > combination of the three or something else?
> > They're too far away to be "evaded" in any meaningful
> > way.
> > You could perhaps shoot missiles at them, but it'll
> > take years for the missiles to reach them, during which
> > time the enemy is going to be launching replacement
> > sensor drones.
> > You could perhaps shoot long range interplanetary
> > X-ray lasers at them, which could be a significant
> > strategy. However, if the enemy also has
> > interplanetary X-ray lasers, the enemy will be using
> > them to shoot at your lasers; your lasers are "wasting"
> > firepower on sensor drones while the enemy is withering
> > away your own firepower.
> Long range interplanetary X-ray lasers. Nope. Just standard tech
> here, nothing to see folks, just move along.
> Again, pardon me for being skeptical.
Note that I said "perhaps". You're the one claiming
that you could maybe do something to the deep space
sensor drones. I'm being helpful to your cause by
pondering the POSSIBILITY of using extremely long
range interplanetary X-ray weapons lasers.
That said, I actually have an idea of how to go
about making long range X-ray weapons lasers.
The technological challenges are VERY daunting.
There are a lot of aspects to the designs which
I've worked out, but it's easily possible that there
are some important deal-killers I haven't considered.
But at least I have a good idea of a direction to
go with the design.
You haven't even got step one.
> If we can't even do halfway decent testing of something as mundane as
> solar sails, and you're talking about about fresnel lenses (In Space!)
> hundreds of meters across, interplanetary x-ray lasers capable of
> shooting down hostile objects . . . and then go on to describe highly
> reflective mirrors as 'magic', well, I don't think we live in the same
> universe.
Obviously not. You think something which you haven't even
worked out the first detail on is a "rather modest advance".
It could easily be something which is utterly physically
impossible, but it's nevertheless a "rather modest advance".
I have a preference for speculating on technology which
I can at least understand the basics of.
> And, one last time: I'm not trying to convince you of anything.
> You're asserting that stealth in space is 'impossible'; I'm playing
> the good skeptic, and have already pointed out some flaws.
Personally I do indeed think stealth in space is more or
less "impossible". However, my idea of the sorts of
sensors involved are rather different from John
Schilling's. His basic concept involves what I feel is a
rather optimistic assumption of tracking everything all
the time. Maybe the numbers add up, but my gut doesn't
like it. Rather than settle on just one particular sensor
system, I instead pondered the possibilities of different
sensor systems.
In particular, boring old radar is just too good a sensor
system, at least at short range, to discard. Even with
fancy uber-sensitive passive sensors and UV lidar
around, I suspect plain old radar will be used extensively.
And it seems that something which is good at reflecting
sunlight into a narrow cone (to avoid passive detection)
will tend to show up on radar pretty well. That's my
old argument against "stealth in space" from the '90s.
Since then, I've considered different sensor systems
and radar doesn't seem to be as important a component.
Still, radar can be very useful, especially for civilian
purposes.
Isaac Kuo
IsaacKuo
wrote:
> > A material, in the beginning of 2008 mind you, that is better than
> > 99.9% absorbtive.
> Completely different from a material that is 99.99% reflective. Such
> an absorptive material in sunlight will heat up and radiate very
> efficiently.
> Then again there is no point making a big curved mirror 99.99%
> reflective anyway, since then it will reflect sunlight. You're out of
> luck either way.
That's okay since you can still point the dish anywhere in the
hemisphere away from the sun. A more troubling possibility
is if the enemy uses any active sensors. That big curved
mirror will show up very well on active sensors.
Isaac Kuo
Wayne Throop
: Then again there is no point making a big curved mirror 99.99%
: reflective anyway, since then it will reflect sunlight. You're out of
: luck either way.
Oh, come on. The shiny side of the mirror is pointing
away from observers (and presumably, from the sun), and
a mirror doesn't need to be shiny on both sides. Further,
the active cooling allows you to avoid heating up the
absorbing side.
Mind you, an adequately broad-spectrum 99.99 percent reflective
material is getting close to magic tech. If you allow magic,
just have an ultra-dense or extra-dimentional heat sink incorporated
into the thing.
Wayne Throop thr...@sheol.org http://sheol.org/throopw
Tim Little
> That's okay since you can still point the dish anywhere in the
> hemisphere away from the sun.
I mean the *back* of the big dish.
- Tim
dwight...@gmail.com
Sigh. The point is that developing a highly reflective mirror isn't
any less realistic (or more realistic) than some other technologies
being assumed here. As for the rest, I suggest you read up.
IsaacKuo
wrote:
> On 2008-02-24, IsaacKuo <mech...@yahoo.com> wrote:
> > That's okay since you can still point the dish anywhere in the
> > hemisphere away from the sun.
> I mean the *back* of the big dish.
The back of the dish could be obscured by the rest of the ship's
hull. For example, the hull could be shaped like a tetrahedron.
One face has the radiator dish in it. Another face is toward
the Sun. That face reflects sunlight into a narrow cone in an
arbitrary direction vaguely toward the Sun. Due to the
reflectivity of the faces, this hull will show up well on radar
and really well on radar at wavelengths comparable to the
hull diameter. But as for passive visible/IR sensors, the ship
could be pretty stealthy (assuming the existence of the
broadband uber-reflective material).
Isaac Kuo
Charles...@gmail.com
Sure it does. A 8 km/s 10kg self-guiding munition sent to where the
meticulously tracked and plotted spy sat will be in X years will toast
it. If the craft can't maneuver. If they can, then you need to send in
your own hunter-killer sats. If you pre-position your killer sats
there then you can have an even better timeframe. It would come down
to a matter of dollars and cents. How much would it cost to keep up
with this cold war?
Bryan Derksen
> Sigh. The point is that developing a highly reflective mirror isn't
> any less realistic (or more realistic) than some other technologies
> being assumed here. As for the rest, I suggest you read up.
We know how to build nuclear thermal engines, both in principle and in
engineering terms. Prototypes have been tested on the ground, even.
Ditto for mass drivers.
We don't know how to build a mirror that's %99.99 reflective over a
broad spectrum. Not even in principle. How would it work? Handwaving.
There's a real difference here, IMO.
Bryan Derksen
> Sure it does. A 8 km/s 10kg self-guiding munition sent to where the
> meticulously tracked and plotted spy sat will be in X years will toast
> it. If the craft can't maneuver. If they can, then you need to send in
> your own hunter-killer sats. If you pre-position your killer sats
> there then you can have an even better timeframe. It would come down
> to a matter of dollars and cents. How much would it cost to keep up
> with this cold war?
If you're actively shooting down enemy satellites it's not a cold war
any more. You've just started a "hot" war with the explicit condition
that your enemy is going to have an intact sensor net for X years after
the war is declared. If you're depending on stealth to win it this isn't
a good position to be in.
Charles...@gmail.com
Not to be lumped in with Dwight, I find his tone abrasive as well, but
unless my numbers are completely messed up I have a way to deal with
the mirror heating.
I make a couple of assumptions.
1.) I assume that a Helium Turbine design can be pushed to 50%
efficiency. This is not much of a stretch. The linked simulation
software[1] seems to indicate that you can push the GT-MHGTR design to
50.11% with two intercoolers per reheat.
2.) I assume a reflector of 98% efficiency with blackbody radiation.
Polished gold has an emissivity of 0.02, so I assume that would
translate to 98% efficiency.
3.) I assume a 60% efficient[2] Rough Silicon Nano-Wire Multi-Stage
Electrothermal Device[3].
4.) Said Rough Silicon Nano-Wire Multi-Stage Electrothermal Devices
will be on par with roughly 1.5 grams per watt generated. This is
roughly equivalent to modern aircraft engine power to weight ratios,
which since the linked PDF[2] mentions replacing aircraft engines with
ones based on these devices I think that's plausible.
5.) I also assume that it takes electricity equivalent to 110% of the
heat you wish to move to compress helium back to operating
temperatures. (With 10% of that becoming new heat in the system.)
[1] http://www.energyfromthorium.com/javaws/BraytonSim.html
[2] http://www.belarusembassy.org/science/275.pdf
[3] http://web.mac.com/majumdargroup/iWeb/Site/Main_files/NANOTECH%20ALERT%20--%20JANUARY%2018,%202008.html#2
Now, I'm still working the numbers to take into account new issues
that come up, but so far it still looks workable. You end up using a
lot of electricity getting the heat out in that small cone, but you
still have some left over. The problem is that if you want to pump the
heat out it costs you more electricity than heat remaining to pump it
out. So you end up in a "Red Queen's Race" that Isaac mentioned in
another thread. You can never get rid of all the heat because you keep
making more by cycling the heat from the reflectors to the radiator.
However you don't have to get to 0 watts thermal energy in the system.
You only need to get to a few thousand watts thermal energy in the
system and then send the rest to the cold hull radiating at 50 Kelvin.
So you just need to have enough electricity to keep ahead of the heat
until it gets to that level.
The first step is to get as much of the heat turned into electricity
as possible before it enters the radiator system. A 50% efficient
helium turbine combined with a 60% efficient Electrothermal device
leaves only 20% heat in the system. So if we are only taking 5mw heat
into the system that leaves us with 1mw of thermal energy to be rid of
and 4mw of electricity to do it with. This isn't very farfetched,
there is a lot of talk about putting this into play for existing power
plants in the next few years if an efficient way to manufacture these
things can be found, which seems likely. The US DoD is investing into
this; they have plans to field diesel engines with thermoelectric
devices to scavenge heat by 2010 to 2014 IIRC. (I found a US DoE
PowerPoint that discussed this.) There is also another method of
scavenging heat that is also indicated to be 60% efficient if the
silicon version does not pan out, but it is only 60% efficient at much
higher temperatures. (800 Kelvin as opposed to power plant exhaust
temperatures which are roughly 400 Kelvin.)
Next we have a radiator design that can put out the heat in a narrow
cone. (I'll leave the actual physical design to other minds or to the
imagination of the reader.) If the radiator reabsorbs less than 50%
(hopefully much less) of what it emits then scavenging the heat for
electricity with another electrothermal device makes it a little
easier to get rid of what's left in the system by eating some of that
heat and then providing electricity to pump the remainder back through
the system again. Still this is a losing battle, since you still spend
more in electricity to pump and compress it than you regain by
scavenging (assuming my calculations are correct). This is that "Red
Queen's Race" again. Luckily we don't need to get the heat in the
system down to 0 watts per second; we just need to get it down to a
few thousand watts per second, which can be conveniently emitted
through the rest of the hull at 50 Kelvin. So we need to get 1000
kilowatts thermal down to about 3 kilowatts thermal. This seems
feasible to me and did pan out with the original calculations I did.
(I'm working on a new set that is more accurate however. See below.)
This of course ignores the remaining electricity in the system. Does
this electricity turn back into heat by running life support and
shooting weapons? Yes, but you aren't really stealthy if you are
shooting weapons, so you would abandon this clunky emission system and
start radiating through simple radiators and crank up the dial on your
reactor. While you are trying to remain stealthy you just keep the
dial on your reactor at just enough to generate for your life-support
and electronics and to reject the heat that is made in the process of
generating that electricity.
I'm still running numbers on a cycle that can do this. I had made a
program so that I could tinker around with the numbers, but since I
made some possibly bad assumptions in the calculations for various
stages, I could not say that it was within an order of magnitude in
accuracy. I'm still running them, making a new cycle that takes into
account the temperature of the helium at various stages so I know how
many compression stages I need to get the helium back up to operating
temperatures for the radiator system.
My biggest unknown is the calculations for electrical draw and heat
efficiency of compressing helium gas. Simply put, I'm making wild
guesses on how this would work. So my bad assumptions may push the
calculations out of an order of magnitude in accuracy.
Charles...@gmail.com
The Russians were shooting at US spy planes during the cold war and
actually got one too. It ratcheted up the tension but did not provoke
a shooting war. It's part of the cost of doing business, if you are
spying on me then I'm going to soot your spies, if you don't like it
stop spying on me or declare war.
It's not hard to imagine the same situation that provided the
environment for this to occur during the Cold War to apply in a future
battlefield. If you are dealing with nuclear armed warships the
tendency to not want them shooing near your house (or at your house)
is strong reinforcement to suck up your spy sat losses and roll the
dice again with the next wave rather than resorting to all out war.
However it does not need to be X years. At least not X years in a good
position to spy from. If you have hunter-killer sats in the area that
your opponent wants to spy from, or better yet, on the way to the
place that he wants to spy from then you can shoot them down before
they become effective.
It seems that it would become a war of attrition with each side
sending in spy sats, ant-spy sats, anti-anti-spy sats, and so on.
Resulting in not so much good spying for either side.
IsaacKuo
> Not to be lumped in with Dwight, I find his tone abrasive as well, but
> unless my numbers are completely messed up I have a way to deal with
> the mirror heating.
Cool! Like I admitted, I'm too lazy to write up my own
"stealth" designs right now. My current efforts are in
fiddling with free electron laser design.
> I make a couple of assumptions.
> 1.) I assume that a Helium Turbine design can be pushed to 50%
> efficiency. This is not much of a stretch. The linked simulation
> software[1] seems to indicate that you can push the GT-MHGTR design to
> 50.11% with two intercoolers per reheat.
> 2.) I assume a reflector of 98% efficiency with blackbody radiation.
> Polished gold has an emissivity of 0.02, so I assume that would
> translate to 98% efficiency.
Assuming you're radiating in the regime where gold is an efficient
reflector, of course. Roughly speaking, if you want to radiate into
a smaller cone you'll generally want to be radiating at a higher
temperature to compensate.
> 3.) I assume a 60% efficient[2] Rough Silicon Nano-Wire Multi-Stage
> Electrothermal Device[3].
> 4.) Said Rough Silicon Nano-Wire Multi-Stage Electrothermal Devices
> will be on par with roughly 1.5 grams per watt generated. This is
> roughly equivalent to modern aircraft engine power to weight ratios,
> which since the linked PDF[2] mentions replacing aircraft engines with
> ones based on these devices I think that's plausible.
> 5.) I also assume that it takes electricity equivalent to 110% of the
> heat you wish to move to compress helium back to operating
> temperatures. (With 10% of that becoming new heat in the system.)
> Now, I'm still working the numbers to take into account new issues
> that come up, but so far it still looks workable. You end up using a
> lot of electricity getting the heat out in that small cone, but you
> still have some left over. The problem is that if you want to pump the
> heat out it costs you more electricity than heat remaining to pump it
> out. So you end up in a "Red Queen's Race" that Isaac mentioned in
> another thread. You can never get rid of all the heat because you keep
> making more by cycling the heat from the reflectors to the radiator.
> However you don't have to get to 0 watts thermal energy in the system.
> You only need to get to a few thousand watts thermal energy in the
> system and then send the rest to the cold hull radiating at 50 Kelvin.
> So you just need to have enough electricity to keep ahead of the heat
> until it gets to that level.
Ah, here's a critical difference between your concept and
designs I tried to work out in the past. I always assumed a
hull temperature of 3K. In other words, all of the waste heat
had to be dumped into the directional radiator.
My gut feeling is that if a 50K hull temperature is deemed
acceptable, then the easiest method of dealing with waste
heat is to simply make a however big a 50K radiator you
need and forget about the directional radiator.
> The first step is to get as much of the heat turned into electricity
> as possible before it enters the radiator system. A 50% efficient
> helium turbine combined with a 60% efficient Electrothermal device
> leaves only 20% heat in the system.
Where is "the heat" coming from? A fission reactor, perhaps?
In principle, you can make a very efficient reactor that converts
almost all of its heat energy into electricity. Conveniently,
a heat engine is most efficient when the radiator temperature
is a small fraction of the reactor's temperature. For stealth,
this seems like a win-win situation. Your radiator is a lower
temperature, and you have less waste heat to worry about!
Unfortunately, the design constraint isn't just efficiency, it's
also power/weight ratio. The lower the radiator temperature,
the larger the radiator must be for a given power level. For
a 50K hull, you should be able to get decent levels of power
even without worrying about the directional radiator (which
I suspect will be more trouble than it's worth). For a 3K
hull, it doesn't matter how big the hull is--a zillion square
meters multiplied by zero watts per square meter is still
zero watts.
> Next we have a radiator design that can put out the heat in a narrow
> cone. (I'll leave the actual physical design to other minds or to the
> imagination of the reader.) If the radiator reabsorbs less than 50%
> (hopefully much less) of what it emits then scavenging the heat for
> electricity with another electrothermal device makes it a little
> easier to get rid of what's left in the system by eating some of that
> heat and then providing electricity to pump the remainder back through
> the system again.
Assuming that 98% gold reflector, the radiator will reabsorb only
about 2% of what it emits. However, NONE of this heat can be
scavenged for electricity. Assuming you want to keep the reflector
as cool as the rest of the hull, then there's no temperature gradient
downward to work with. The reflector is at 50K, and the rest of
the hull is also at 50K. There's no heat sink to run a heat engine
with.
> This of course ignores the remaining electricity in the system. Does
> this electricity turn back into heat by running life support and
> shooting weapons?
Presumably, the electricity is being used for "something useful",
like operating a (hopefully) stealthy mass driver rocket. This
thread started with the idea of using a mass driver to propel a
stealthy impactor. Unfortunately, this "something useful" might
not be 100% efficient, so it will generate extra waste heat which
must be dealt with somehow.
> My biggest unknown is the calculations for electrical draw and heat
> efficiency of compressing helium gas. Simply put, I'm making wild
> guesses on how this would work. So my bad assumptions may push the
> calculations out of an order of magnitude in accuracy.
Rather than getting bogged down into the details of a particular
heat engine, you can first look at things assuming a generic
efficient heat engine. To a first approximation, the inherent
thermodynamics limitations on a heat engine will be more
significant than the differences between specific heat engine
designs.
Isaac Kuo
IsaacKuo
> On Feb 25, 2:01 pm, Bryan Derksen <bryan.derk...@shaw.ca> wrote:
> > CharlesRCap...@gmail.com wrote:
> > > Sure it does. A 8 km/s 10kg self-guiding munition sent to where the
> > > meticulously tracked and plotted spy sat will be in X years will toast
> > > it. If the craft can't maneuver. If they can, then you need to send in
> > > your own hunter-killer sats. If you pre-position your killer sats
> > > there then you can have an even better timeframe. It would come down
> > > to a matter of dollars and cents. How much would it cost to keep up
> > > with this cold war?
> > If you're actively shooting down enemy satellites it's not a cold war
> > any more. You've just started a "hot" war with the explicit condition
> > that your enemy is going to have an intact sensor net for X years after
> > the war is declared. If you're depending on stealth to win it this isn't
> > a good position to be in.
> The Russians were shooting at US spy planes during the cold war and
> actually got one too. It ratcheted up the tension but did not provoke
> a shooting war. It's part of the cost of doing business, if you are
> spying on me then I'm going to soot your spies, if you don't like it
> stop spying on me or declare war.
This is because the US was doing something that was already
inherently an act of war. Overflying someone's sovereign territory
without their permission is an act of war. The only reason why
it didn't start a war was because the Russians didn't feel like
starting a war over it.
Of course, the US isn't above contriving a trumped up situation
where someone shoots at our stuff to start a war. However, the
US habit is to do this with our stuff in/over international waters.
It'd just be plain embarrasing to try and start a war over the
shooting down of an unauthorized overflight.
With space, the precedent has already been set that outer
space is international airspace. When Sputnik flew, it opened
the door for US spy satellites to orbit over Russia with
impunity. If the Russians can overfly in orbit without asking
any of the hundreds of overflown countries for permission,
so can we.
That said...
> It's not hard to imagine the same situation that provided the
> environment for this to occur during the Cold War to apply in a future
> battlefield.
True enough. The precedent set today could be changed by
future cultural changes and/or specific political developments.
> If you are dealing with nuclear armed warships the
> tendency to not want them shooing near your house (or at your house)
> is strong reinforcement to suck up your spy sat losses and roll the
> dice again with the next wave rather than resorting to all out war.
> However it does not need to be X years. At least not X years in a good
> position to spy from. If you have hunter-killer sats in the area that
> your opponent wants to spy from, or better yet, on the way to the
> place that he wants to spy from then you can shoot them down before
> they become effective.
Unfortunately, hunter-killer drones will need a LOT of delta-v,
compared to a simple sensor drone.
A sensor drone only needs enough delta-v to get boosted out
into interplanetary space. Without braking fuel, this drone will
eventually drift out beyond useful range, but this could take decades.
A decades old sensor drone is going to be obsolete and you'll
want to retire it anyway. So you'll be launching replacement
drones anyway.
A hunter-killer drone needs enough delta-v to get boosted
out, and then braking fuel to get on station, and then more
fuel to intercept a sensor drone. It also needs sensors
of its own to home in on the target. All in all, such a
hunter-killer drone could be much more expensive than the
target it's meant to kill.
(I don't call them "satellites" because they aren't in orbit
around a planetary body.)
> It seems that it would become a war of attrition with each side
> sending in spy sats, ant-spy sats, anti-anti-spy sats, and so on.
> Resulting in not so much good spying for either side.
Or you could get something like the real world history of
spy satellites, where attacking satellites is expensive and
difficult (so far). And attacking them is considered an
act of war because they're considered to be flying in
international airspace.
Isaac Kuo
Damien Valentine
about mirrors that Dwight started) is that any species with A.) the
ability to maneuver asteroids wherever they want, is also a species
with B.) the ability to identify and destroy dangerous asteroids.
Therefore, asteroids wouldn't be an effective weapon against such a
civilization, even if they don't have C.) the ability to detect and
track all asteroids in their home system. And they would have to have
C if they had A and B.
Have I got this right?
John Schilling also mentions "We've seen plenty of proposals for that
sort of thing, all of them assuming the attacker has extensive and
highly sophisticated space
infrastructure and the target is the United States of America of
roughly 2000 AD." Jack, would that work for what you have in mind? A
high-tech culture picking on a low-tech one, and deciding to be
stealthy for political reasons rather than fear of retaliation? Of
course, you'd almost certainly have to scratch the "massive Terran
space fleet that 20 years later non-stealthily nukes the Republick
into near-extinction"...
Luke Campbell
> On Feb 23, 11:16 pm, IsaacKuo <mech...@yahoo.com> wrote:
>
> 1.) I assume that a Helium Turbine design can be pushed to 50%
> efficiency. This is not much of a stretch. The linked simulation
> software[1] seems to indicate that you can push the GT-MHGTR design to
> 50.11% with two intercoolers per reheat.
>
> 2.) I assume a reflector of 98% efficiency with blackbody radiation.
> Polished gold has an emissivity of 0.02, so I assume that would
> translate to 98% efficiency.
>
> 3.) I assume a 60% efficient[2] Rough Silicon Nano-Wire Multi-Stage
> Electrothermal Device[3].
>
> 4.) Said Rough Silicon Nano-Wire Multi-Stage Electrothermal Devices
> will be on par with roughly 1.5 grams per watt generated. This is
> roughly equivalent to modern aircraft engine power to weight ratios,
> which since the linked PDF[2] mentions replacing aircraft engines with
> ones based on these devices I think that's plausible.
>
> 5.) I also assume that it takes electricity equivalent to 110% of the
> heat you wish to move to compress helium back to operating
> temperatures. (With 10% of that becoming new heat in the system.)
I am not sure I understand the basic setup - my guess is you have a
helium turbine as a primary heat engine, then run the thermoelectrics
off the exhaust heat of the turbine? Then you place the
thermoelectric coated heat sink inside of a gold mirror with a window
to let the heat out?
If this is true, there are a few issues to consider. First, the heat
sink for the turbine runs at 300 K. By itself this will radiate half
a kW per square meter, assuming a perfect blackbody. But if the
thermoelectrics are perfect, operating at the very limit of what
thermodynamics allow, they will require a 120 K heat sink to allow 60%
efficiency. At this temperature, you are radiating only 10 W per
square meter from a perfect blackbody - and significantly less if the
thermoelectrics do not have their maximum thermodynamic efficiency (at
100 K, for example, you only get 5 W per square meter, and at 80 K you
get about 2.3 W/m^2).
So how much power do we draw? If we assume the mass driver rocket
takes a MW, then with 120 K radiators we need 100,000 m^2 of radiator
without any mirror at all. That seems rather large, and quite
expensive to coat with the thermoelectrics (barring some breakthrough
in bulk nanofabrication).
Or are you running the radiator at 300 K, and putting the
thermoelectrics on the back-side of the mirror?
Luke
Jack Tingle
That's a reasonable summary. As I said in the original post, it requires
an inattentive target. I don't assume the target is low tech, just a bit
careless.
Just because the target _can_ track all asteroids, doesn't mean it
_does_ track them. Again, for example, the countries surrounding the
Indian Ocean could have tracked tsunamis. They chose not to, for a
variety of reasons.
Cost probably isn't one of them. The entire US NWS has a budget of less
than $900m for 2009. There are probably less than 100 people in their
various research and operations tsunami jobs. Let's be generous and book
them at $200k/year each. That's $20m/year. Major network improvements
over the last few years cost around $20m. The Indian Ocean tsunami in
2004 killed over 200,000 people, and over $7b were donated in aid in its
aftermath. I presume the affected nations must have lost a similar
amount of money, some of which could have been saved with a network
warning. Call it $14b, roughly. That's a 350:1 bet; not bad for
insurance. So rational cost evaluation probably wasn't a big factor.
My bet is that the budget discussions in our inattentive target went
like: "Well, yes Senator, we could check every big asteroid all the
time, but all we need to do is spot check a few dozen a year for
perturbation... We can contract that out to existing observatories...
New hazardous asteroids?... No sir, we haven't found one since the big
push in the middle of the 21st century." Note that they're _not_
suggesting not watching for natural hazards. They're just not thinking
about man-made ones.
As I said before, it would only work once*, and it probably wouldn't
work at all, if you drew attention to it with a big signature of some
kind. A mass driver is the only thing I can think of that would stand a
chance of being stealthy enough, and of moving a small asteroid. And
Shilling is right, if you can't cover or neutralize the EM signature of
the mass driver, that wouldn't work either.
Still, it is a stealthy space attack that at least _sounds_ plausible,
limited though it is. As I say, that should be worth some extra points
for the soon-to-be-blasted First Marshen Republick.
Regards,
Jack Tingle
*The next year's budget discussion would have a lot different tone.
IsaacKuo
> Damien Valentine wrote:
> > So the basic objection to Jack's original post (snipping out the mess
> > about mirrors that Dwight started) is that any species with A.) the
> > ability to maneuver asteroids wherever they want, is also a species
> > with B.) the ability to identify and destroy dangerous asteroids.
> > Therefore, asteroids wouldn't be an effective weapon against such a
> > civilization, even if they don't have C.) the ability to detect and
> > track all asteroids in their home system. And they would have to have
> > C if they had A and B.
> > Have I got this right?
> > John Schilling also mentions "We've seen plenty of proposals for that
> > sort of thing, all of them assuming the attacker has extensive and
> > highly sophisticated space
> > infrastructure and the target is the United States of America of
> > roughly 2000 AD." Jack, would that work for what you have in mind? A
> > high-tech culture picking on a low-tech one, and deciding to be
> > stealthy for political reasons rather than fear of retaliation? Of
> > course, you'd almost certainly have to scratch the "massive Terran
> > space fleet that 20 years later non-stealthily nukes the Republick
> > into near-extinction"...
> That's a reasonable summary. As I said in the original post, it requires
> an inattentive target. I don't assume the target is low tech, just a bit
> careless.
The assumption above is that the target is low tech. "A high-tech
culture picking on a low-tech one"
> Just because the target _can_ track all asteroids, doesn't mean it
> _does_ track them. Again, for example, the countries surrounding the
> Indian Ocean could have tracked tsunamis. They chose not to, for a
> variety of reasons.
Cost is likely the big one.
> Cost probably isn't one of them. The entire US NWS has a budget of less
> than $900m for 2009. There are probably less than 100 people in their
> various research and operations tsunami jobs. Let's be generous and book
> them at $200k/year each. That's $20m/year. Major network improvements
> over the last few years cost around $20m. The Indian Ocean tsunami in
> 2004 killed over 200,000 people, and over $7b were donated in aid in its
> aftermath. I presume the affected nations must have lost a similar
> amount of money, some of which could have been saved with a network
> warning.
That's a rather strange presumption. Even if the aid were
roughly equal to the amount of damage done, the damage
wasn't done to piles of money. The damage was done
to "stuff". Even if seven billion dollars worth of damage
were done, this doesn't imply that the people had seven
billion dollars of cash sitting around.
> Call it $14b, roughly. That's a 350:1 bet; not bad for
> insurance. So rational cost evaluation probably wasn't a big factor.
What's the rational cost evaluation for someone living
paycheck to paycheck vs buying health insurance?
If the money isn't there, the money isn't there.
> My bet is that the budget discussions in our inattentive target went
> like: "Well, yes Senator, we could check every big asteroid all the
> time, but all we need to do is spot check a few dozen a year for
> perturbation... We can contract that out to existing observatories...
> New hazardous asteroids?... No sir, we haven't found one since the big
> push in the middle of the 21st century." Note that they're _not_
> suggesting not watching for natural hazards. They're just not thinking
> about man-made ones.
The more plausible scenario is that the Senator's
decision process is going the other way.
"Well, yes Senator, we NEED to deploy our asteroid
sensor satellites because the existing asteroid
sensor satellites are almost three years old. How
much do we NEED it? About fifteen billions dollars
worth of us needing it, including a hundred FTE's
worth of needing it in your home disctrict. And
including fifteen million dollar's worth of contributions
to your campaign."
> As I said before, it would only work once*, and it probably wouldn't
> work at all, if you drew attention to it with a big signature of some
> kind. A mass driver is the only thing I can think of that would stand a
> chance of being stealthy enough, and of moving a small asteroid. And
> Shilling is right, if you can't cover or neutralize the EM signature of
> the mass driver, that wouldn't work either.
> Still, it is a stealthy space attack that at least _sounds_ plausible,
> limited though it is. As I say, that should be worth some extra points
> for the soon-to-be-blasted First Marshen Republick.
It doesn't really sound plausible to me, because
even in the absence of pork-barrel asteroid sensor
satellites, there are going to be lots of civilian
sensors out there, including ground based
telescopes and transport based sensors.
Isaac Kuo
Tim Little
> So how much power do we draw? If we assume the mass driver rocket
> takes a MW
Better make it much more than a GW. The original scenario was moving
a "giant asteroid" (big enough to devastate the Pacific Rim with
tsunamis) to Earth within ten years. Even with a gigawatt and
minimising energy lost in the "exhaust", that only gets you about a
million tonnes.
> then with 120 K radiators we need 100,000 m^2 of radiator without
> any mirror at all.
Or 100 km^2 for a 1 GW mass driver of perfect efficiency (which is
still too small). At that size you'd need the radiator to be edge-on
to the Sun, else deal with dozens of gigawatts of solar flux.
- Tim
Luke Campbell
wrote:
> On 2008-02-25, Luke Campbell <lwc...@gmail.com> wrote:
>
> > So how much power do we draw? If we assume the mass driver rocket
> > takes a MW
>
> Better make it much more than a GW. The original scenario was moving
> a "giant asteroid" (big enough to devastate the Pacific Rim with
> tsunamis) to Earth within ten years. Even with a gigawatt and
> minimising energy lost in the "exhaust", that only gets you about a
> million tonnes.
Fair enough. 1 GW it is, then.
> > then with 120 K radiators we need 100,000 m^2 of radiator without
> > any mirror at all.
>
> Or 100 km^2 for a 1 GW mass driver of perfect efficiency (which is
> still too small). At that size you'd need the radiator to be edge-on
> to the Sun, else deal with dozens of gigawatts of solar flux.
Of course, that's true no matter what size radiator you choose - if
the incoming solar intensity exceeds the outgoing radiated intensity,
you gain heat rather than losing it. On the other hand, this puts the
radiators along the plane of the ecliptic, which is where the OP
needed them for his scenario.
Luke
Tim Little
> Of course, that's true no matter what size radiator you choose - if
> the incoming solar intensity exceeds the outgoing radiated
> intensity, you gain heat rather than losing it. On the other hand,
> this puts the radiators along the plane of the ecliptic, which is
> where the OP needed them for his scenario.
Yes, under the original assumption that nobody has any sensors outside
the ecliptic. Certainly a gigawatt source will be quite visible if
there does happen to be anyone with IR sensors out there.
- Tim
Luke Campbell
wrote:
Oh, I quite agree. And the cooler you make the radiators, the more
photons they emit for a given power output, and the more visible they
will be.
Luke
Charles...@gmail.com
wrote:
Well, in regards the OP's concept, you have a multi-gigaton asteroid
to work with. How long will it take to heat up X-gigatons from 10K or
whatever the asteroid is at to begin with to a temperature where it
begins to stand out? Personally, I don't hold high hopes for it
succeeding, but you do have a mighty big heat sink to work with,
enough perhaps to hide at least the initial phase of the operation.
Combine that with a radiator hidden in the shadow of the asteroid and
you might be able to do it. Do people look for asteroids in the IR? If
not then you might be able to heat it up to a pretty high temperature
before someone doing something in the IR unrelated to finding
asteroids stumbles over it.
IsaacKuo
> On Feb 25, 8:05 pm, Tim Little <t...@soprano.little-possums.net>
> wrote:
> > Or 100 km^2 for a 1 GW mass driver of perfect efficiency (which is
> > still too small). At that size you'd need the radiator to be edge-on
> > to the Sun, else deal with dozens of gigawatts of solar flux.
> Well, in regards the OP's concept, you have a multi-gigaton asteroid
> to work with. How long will it take to heat up X-gigatons from 10K or
> whatever the asteroid is at to begin with to a temperature where it
> begins to stand out?
The amount of waste heat generated is proportional to the
amount of kinetic energy pumped into the exhaust, which
is proportional to the mass of the asteroid. Thus, the effect
of mass more or less cancels itself out.
However, the amount of kinetic energy pumped into the
exhaust is also dependent on the desired mission delta-v.
The original concept assumed it's okay to wait decades
for the mission to complete, so the required delta-v
might not be so daunting.
> Personally, I don't hold high hopes for it
> succeeding, but you do have a mighty big heat sink to work with,
> enough perhaps to hide at least the initial phase of the operation.
> Combine that with a radiator hidden in the shadow of the asteroid and
> you might be able to do it. Do people look for asteroids in the IR?
Not today, because Earth's atmosphere gets in the way.
Space based IR telescopes currently spend their time
looking at more scientifically juicy targets. It's a gap in
sensor coverage which will be filled in when we expand
significantly into space.
Isaac Kuo
Charles...@gmail.com
> On Feb 25, 1:14 pm, CharlesRCap...@gmail.com wrote:
> > On Feb 23, 11:16 pm, IsaacKuo <mech...@yahoo.com> wrote:
> I am not sure I understand the basic setup - my guess is you have a
> helium turbine as a primary heat engine, then run the thermoelectrics
> off the exhaust heat of the turbine? Then you place the
> thermoelectric coated heat sink inside of a gold mirror with a window
> to let the heat out?
Yes:
Main Cycle:
Stage 0: Reactor - MITEE Nuclear Thermal Rocket
Capacity: 20mw thermal @ 3000K
Efficiency: 75% (adjustable)
Waste: 5mw thermal @ 3000K
Stage 1: Helium Coolant Reservoir at 3000K-1123K
(Cycles helium through reactor to the Brayton Cycle Helium Turbine
and back.)
Input: 5mw thermal at 3000K
Output: 5mw thermal at 1123K (K^4 = 1,590,446,354,641)
Stage 2: Brayton Cycle Helium Turbine
(Device using extensive regeneration and consisting of multiple
compressors and intercoolers feeding a hydrogen turbine.)
Input: 5mw thermal at 1123K
Efficiency: 50.11%
Generation: 2.5055mw electrical
Waste: 2.4945mw thermal at 378.7K
Stage 3: Helium Coolant Reservoir at 378.7K
Input: 2.4945mw thermal at 378.7K
Output: 2.4945mw thermal at 378.7K
Stage 4: Multi-Stage Silicon Electrothermal Nano-Devices (#1)
(Scavenges waste heat from the Brayton Cycle Helium Turbine)
Input: 2.4945mw thermal at 378.7K (K^4 = 20,567,486,479)
Efficiency: 60%
Generation: 1.4967mw electrical
Waste: 0.9978mw thermal at 301.2K (K^4 = 8,226,994,544)
Stage 5: Helium Coolant Reservoir at 301.2K
Input: 0.9978mw thermal at 301.2K
Output: 0.9978mw thermal at 301.2K
I'm considering putting another Multi-Stage Silicon Electrothermal
Nano-Device stage in there and then just having more compression
cycles to further reduce the amount of heat that needs to be moved to
the radiator.
My assumptions about how temperature differential for electrothermal
devices is calculated is probably flawed though.
I assume that the Helium Turbine would be more efficient if the input
temperature was closer to the 3000K that the engine operates at, but I
don't have numbers for that, and so I stick with what I do have a
model for which has an input temperature of 1123K.
In regards to your suggestion to place the thermoelectrics on the
radiator directly:
That's an interesting idea about mounting the thermoelectrics on the
radiator directly. I might use that if the operating temperature
wouldn't be too high to melt them or worse, reduce their efficiency.
What I am thinking is a radiator with a gold (or some unobtanium
alloy) reflector set up to direct the radiation in a small cone. The
actual cone itself does not really matter because I'm assuming that
all of the energy is hitting the reflector and not radiating directly
to space. So the narrowness of the cone is only important for
determining the size of the reflector which I have not gotten to
calculating yet. The reflector will be actively cooled at to about
133K (see below for details) and the coolant pumped to the interior of
the ship for compression back to operating temperatures. (On the way
it passes through some thermoelectric that radiate directly to space.)
The hull at 50K with an emissivity of around 0.99 is roughly 0.35w per
square meter radiated. The reflector at 133K with 0.02 emissivity is
roughly the same. So if we actively cool the reflector to 133K and
pump the coolant through the electrothermal devices radiating directly
to the 3K background (does not need to be on the back side of the
reflector since we would need to actively cool it anyway and can pump
the helium somewhere else) then we might not get 60% efficiency but
it's still something... 130K differential is what kind of efficiency,
50% maybe? I don't know the math, I've been making it up for now.
So then we need to calculate the heat lost from the reflector directly
(0.35w m^2), the heat reflected to space (98%), and the heat absorbed
back into the system from the reflector (2% of the radiator output,
minus 0.35w m^2 that the reflector is emitting directly) and then how
much heat the electrothermal devices take out of it (50%?), which
leaves some amount to be recycled through the system. We also need to
know what temperature the helium is at when it is cycled back from the
electrothemal devices so we know how much we need to compress it to
get it back to operating temps.
Let's see. The energy of 133K is 312900721 (T^4 if I am making a
correct assumption) so if we take 50% out of it we are left with
111.8K (sqrt(sqrt(312900721))/2 or the fourth root of 50% of the
energy we started with.) in the system. So we then need to compress
that back up to whatever temp I run the radiator at.
What I'm doing basically is reducing the amount of heat that needs to
be compressed and radiated as much as possible (and consequently
generating electricity from it) and then taking the remainder and
getting rid of enough that I can pump the rest to the hull at about
50K. I can create internal reservoirs down to 50K or less inside the
ship if that is needed to operate the thermoelectrics. This however
will cost more (on a percentage base) to compress back up to radiator
temps, but since I'll be working with a smaller about of thermal
energy the relative costs can be manageable. (maybe)
Charles...@gmail.com
> On Feb 25, 2:14 pm, CharlesRCap...@gmail.com wrote:
> Assuming you're radiating in the regime where gold is an efficient
> reflector, of course. Roughly speaking, if you want to radiate into
> a smaller cone you'll generally want to be radiating at a higher
> temperature to compensate.
If gold is not an efficient reflector at the radiator temperature,
Chromium as an 0.05 emissivity, not a significant difference in my
design. It will cause me to have to re-juggle the numbers, but I was
doing calculations based on reabsorbing up to 20% of the reflected
energy with reasonable results, so reflecting 95% or 90% or even 80%
can be made to work, but the remaining electricity for other things
gets lower and lower.
> Ah, here's a critical difference between your concept and
> designs I tried to work out in the past. I always assumed a
> hull temperature of 3K. In other words, all of the waste heat
> had to be dumped into the directional radiator.
I'm not going for invisible, just less visible. If you can't be
detected at anything over 10 light seconds, that lends you strategic
and operational stealth. You may not have tactical stealth, but you
don't have that as soon as you start shooting anyway. I'm thinking 50K
is the upper limit, I'd like to get it closer to 25K if I can manage
it. (I think it's possible, but haven't calculated it yet.)
10,000 M^2 at 50K is 3543 watts, at 25K it is 221 watts. It's a big
hurdle, but it's what I'm aiming for.
> My gut feeling is that if a 50K hull temperature is deemed
> acceptable, then the easiest method of dealing with waste
> heat is to simply make a however big a 50K radiator you
> need and forget about the directional radiator.
Well, my design is considering a warship and not necessarily the OP's
asteroid, so some slightly different issues are involved. The problem
with such a big radiator has been covered elsewhere.
Among other things:
1.) The heat addition to the system from sunlight becomes a large
problem.
2.) Your ship becomes more massive and larger target.
3.) From a detection standpoint having a larger cross-section
countermands the benefits you get from having a lower temperature.
(You are cooler but still emitting the same amount of total energy.)
So the combination of the two (you need to have a hull so why not use
it as a radiator) systems will provide the best compromise.
> Where is "the heat" coming from? A fission reactor, perhaps?
> In principle, you can make a very efficient reactor that converts
> almost all of its heat energy into electricity. Conveniently,
> a heat engine is most efficient when the radiator temperature
> is a small fraction of the reactor's temperature. For stealth,
> this seems like a win-win situation. Your radiator is a lower
> temperature, and you have less waste heat to worry about!
I am assuming the use of a MITEE nuclear thermal rocket which can
produce very little waste heat for its output. Further, I am assuming
that the efficiency of that reactor can be adjusted downward so that
more waste heat is radiated into the heat engine system if needed.
> Assuming that 98% gold reflector, the radiator will reabsorb only
> about 2% of what it emits. However, NONE of this heat can be
> scavenged for electricity. Assuming you want to keep the reflector
> as cool as the rest of the hull, then there's no temperature gradient
> downward to work with. The reflector is at 50K, and the rest of
> the hull is also at 50K. There's no heat sink to run a heat engine
> with.
If your hull is 0.99965 emissivity (from Dwight's link or just plain
old 0.98 of normal black carbon) at 50K you are radiating the same
energy that your 0.02 emissivity reflector is emitting at 133K so we
have a useful gradient there. (Particularly if you are radiating to
open space rather than back into a reservoir inside the ship. So would
the business end of those devices radiating into space be warmer than
the hull? It doesn't seem so, or if they are, it is a manageable
difference.)
> > This of course ignores the remaining electricity in the system. Does
> > this electricity turn back into heat by running life support and
> > shooting weapons?
>
> Presumably, the electricity is being used for "something useful",
> like operating a (hopefully) stealthy mass driver rocket. This
> thread started with the idea of using a mass driver to propel a
> stealthy impactor. Unfortunately, this "something useful" might
> not be 100% efficient, so it will generate extra waste heat which
> must be dealt with somehow.
I am assuming that you will only generate enough electricity while in
stealth mode to run your life-support and electronics, so the amount
of additional heat that needs to be dealt with is relatively small. I
will include it into the calculations. (And already did for my
previous set of calculations and it did not seem to push the system
out of balance. There was still a surplus of electricity that could be
reduced to 0 if you kept reducing the initial thermal input from the
plant.)
> > My biggest unknown is the calculations for electrical draw and heat
> > efficiency of compressing helium gas. Simply put, I'm making wild
> > guesses on how this would work. So my bad assumptions may push the
> > calculations out of an order of magnitude in accuracy.
>
> Rather than getting bogged down into the details of a particular
> heat engine, you can first look at things assuming a generic
> efficient heat engine. To a first approximation, the inherent
> thermodynamics limitations on a heat engine will be more
> significant than the differences between specific heat engine
> designs.
I have, assuming generic efficiencies; it works within an order of
magnitude. Now I'm getting bogged down in the details of calculating
the temperature gradient for the different stages of the system and
more accurate calculations on how the different pieces work to give a
more accurate answer on the feaseability of the system.
IsaacKuo
> In regards to your suggestion to place the thermoelectrics on the
> radiator directly:
> That's an interesting idea about mounting the thermoelectrics on the
> radiator directly. I might use that if the operating temperature
> wouldn't be too high to melt them or worse, reduce their efficiency.
Note that your radiator's operating temperature must be lower
than the temperature of the reactor (and much lower in order to
get any sort of efficiency). You're getting bogged down in the
details of specific heat engines and not looking at the big
thermodynamic picture.
> What I am thinking is a radiator with a gold (or some unobtanium
> alloy) reflector set up to direct the radiation in a small cone. The
> actual cone itself does not really matter because I'm assuming that
> all of the energy is hitting the reflector and not radiating directly
> to space. So the narrowness of the cone is only important for
> determining the size of the reflector which I have not gotten to
> calculating yet.
Generally, the reflector is really big. For example,
let's suppose the initial radiator is a 10m diameter circle,
and the desired cone is 2 degrees wide. This means
the distance from the radiator to the mirror needs to be
about 280m. The mirror's radius of curvature to focus at
280m is 560m, resulting in an overall diameter of around
1000m. For this example, the overall diameter needs
to be two orders of magnitude bigger than the initial
radiator (four orders of magnitude greater area).
> The reflector will be actively cooled at to about
> 133K (see below for details) and the coolant pumped to the interior of
> the ship for compression back to operating temperatures. (On the way
> it passes through some thermoelectric that radiate directly to space.)
> The hull at 50K with an emissivity of around 0.99 is roughly 0.35w per
> square meter radiated. The reflector at 133K with 0.02 emissivity is
> roughly the same.
The amount of power is not a good measure of how
easy it is to detect. For temperatures like these,
significantly greater than the 3K background, the
detectability will be roughly proportional to the
number of photons emitted.
> So if we actively cool the reflector to 133K and
> pump the coolant through the electrothermal devices radiating directly
> to the 3K background (does not need to be on the back side of the
> reflector since we would need to actively cool it anyway and can pump
> the helium somewhere else) then we might not get 60% efficiency but
> it's still something... 130K differential is what kind of efficiency,
> 50% maybe? I don't know the math, I've been making it up for now.
At this point, efficiency isn't the problem. Power level
is the problem. You don't just get a 3K heat sink for
free. You need a really really really big radiator to dump
heat into the 3K background, and this radiator will be
radiating away at some temperature above 3K. Unless
you choose a temperature significantly higher than 3K,
the radiator will be HUGE. In any case, it's this radiator
which will be the most visible thing on your ship.
I get the feeling that your essential design approach is
flawed in some sort of fundamental way. You seem to
be looking at sources of heat as potential places to
get energy from, but the real problem that you need to
be tackling is radiating away waste heat. You can't
just turn waste heat into energy, unless you use an
even bigger more visible lower temperature radiator
as a heat sink.
The directional radiator in your design seems to just
be a complex unneeded and unwanted distraction.
Isaac Kuo
IsaacKuo
> On Feb 25, 4:29 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > On Feb 25, 2:14 pm, CharlesRCap...@gmail.com wrote:
> > Assuming you're radiating in the regime where gold is an efficient
> > reflector, of course. Roughly speaking, if you want to radiate into
> > a smaller cone you'll generally want to be radiating at a higher
> > temperature to compensate.
> If gold is not an efficient reflector at the radiator temperature,
> Chromium as an 0.05 emissivity, not a significant difference in my
> design. It will cause me to have to re-juggle the numbers, but I was
> doing calculations based on reabsorbing up to 20% of the reflected
> energy with reasonable results, so reflecting 95% or 90% or even 80%
> can be made to work, but the remaining electricity for other things
> gets lower and lower.
You seem to be making some sort of fundamental mistake,
if reabsorbing 20% of the reflected energy gives reasonable
results. I suspect that this fundamental mistake is thinking
that you can recover some energy from it without emitting
radiation which is even more visible in the process.
> > Ah, here's a critical difference between your concept and
> > designs I tried to work out in the past. I always assumed a
> > hull temperature of 3K. In other words, all of the waste heat
> > had to be dumped into the directional radiator.
> I'm not going for invisible, just less visible.
I get the feeling that you're going for less visible, but you're
actually getting more visible. For a given power level, the
lower temperature radiator is actually more visible, since
it's spitting out lower energy photons (and thus, more photons).
> 10,000 M^2 at 50K is 3543 watts, at 25K it is 221 watts. It's a big
> hurdle, but it's what I'm aiming for.
I think this is where you're losing sight of the big picture.
The big picture is that you've got a certain number of watts
you've got to dump out there. You don't have some fixed
hull size and then you get to choose how much power
to radiate based on the hull temperature. You've got some
amount of power you've got to get rid of, and a lower
radiator temperature just means you've got to make the
radiator much bigger.
> > My gut feeling is that if a 50K hull temperature is deemed
> > acceptable, then the easiest method of dealing with waste
> > heat is to simply make a however big a 50K radiator you
> > need and forget about the directional radiator.
> Well, my design is considering a warship and not necessarily the OP's
> asteroid, so some slightly different issues are involved. The problem
> with such a big radiator has been covered elsewhere.
> Among other things:
> 1.) The heat addition to the system from sunlight becomes a large
> problem.
You can limit this problem with an edge-on radiator and a flat
reflective sunshade. The sunshade reflects sunlight into a
narrow cone at an random angle.
You pretty much don't have a choice about this. Absorbing
sunlight would be a problem even with no on board power
generation at all.
> 2.) Your ship becomes more massive and larger target.
Well, I wouldn't worry about becoming a larger target, but
there are naturally limits to how big and massive something
is before you decide it's stupidly big and massive.
> 3.) From a detection standpoint having a larger cross-section
> countermands the benefits you get from having a lower temperature.
> (You are cooler but still emitting the same amount of total energy.)
You say this as if you had the choice of emitting less energy.
You don't.
> > Where is "the heat" coming from? A fission reactor, perhaps?
> > In principle, you can make a very efficient reactor that converts
> > almost all of its heat energy into electricity. Conveniently,
> > a heat engine is most efficient when the radiator temperature
> > is a small fraction of the reactor's temperature. For stealth,
> > this seems like a win-win situation. Your radiator is a lower
> > temperature, and you have less waste heat to worry about!
> I am assuming the use of a MITEE nuclear thermal rocket which can
> produce very little waste heat for its output. Further, I am assuming
> that the efficiency of that reactor can be adjusted downward so that
> more waste heat is radiated into the heat engine system if needed.
Oh, well then forget about any sort of stealth. The reason a
nuclear thermal rocket "produces little waste heat" is
because it actually produces lots of waste heat but most
of it is conveniently dumped into the exhaust. The rocket
exhaust is going to be rather visible; a Space Shuttle launch
is stealthy in comparison.
> > Presumably, the electricity is being used for "something useful",
> > like operating a (hopefully) stealthy mass driver rocket. This
> > thread started with the idea of using a mass driver to propel a
> > stealthy impactor. Unfortunately, this "something useful" might
> > not be 100% efficient, so it will generate extra waste heat which
> > must be dealt with somehow.
> I am assuming that you will only generate enough electricity while in
> stealth mode to run your life-support and electronics, so the amount
> of additional heat that needs to be dealt with is relatively small.
You're trying to put people on a stealthy rocket ship? Not
good, since people generate plenty of heat. For my "stealth"
attempts, I assume an unmanned drone with virtually no on
board power (other than propulsion). The main source of
heat is from sunlight, even though most of the sunlight is
reflected away with a mirror shade.
Isaac Kuo
dwight...@gmail.com
> On Feb 23, 11:20 pm, "dwight.thi...@gmail.com"
>
> <dwight.thi...@gmail.com> wrote:
> > On Feb 23, 10:16 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > > On Feb 23, 8:42 pm, "dwight.thi...@gmail.com"
> > > > space industry. Possibly parts of the solar system off Earth are
> > > > permanently manned if not outright colonized. You've got all sort of
> > > > toys like nuclear thermal rockets, at the least, but at any rate,
> > > > extremely advanced technologies beyond anything we can achieve today.
> > > I generally don't assume the existence of magic technologies,
> > > even if it's entirely plausible that some sort of science which
> > > today would seem magical will be developed.
> > Really? What's magical about this technology? What makes it so
> > difficult? Your whim? Going in the other direction:
>
> I don't know whether it's magical. It's beyond our current
> knowledge, which makes it unknown. You claim that
> it's easy. I make no such assumption. It might be easy,
> or it might be difficult, or it might be impossible.
>
> >http://www.sciencedaily.com/releases/2008/01/080122154610.htm
> > A material, in the beginning of 2008 mind you, that is better than
> > 99.9% absorbtive.
>
Come on, I know you can make the connection. If I had said in
December of '07 that it was possible to make a material that was
99.99% absorptive, you would have (or could have) made exactly the
same claim as the one you are making now with regards to reflectivity.
I think one needs to distinguish between 'magic' that depends on the
violations of supposedly fundamental law, the 'magic' that is an
engineering nightmare, and the 'magic' that is simply unknown. An
easy example would be Edison trying out materials for the filaments of
light bulbs; by your own criteria, he was engaged in 'magical'
speculation.
> > > > Yet mirror technology improves not a whit. Even though there is
> > > > nothing in the laws of physics that forbid such reflectivities, and
> > > > even though such reflectivities have been achieved for certain
> > > > wavelengths.
> > > Maybe there will be clever designs or new science which gives
> > > us neat broadband highly reflective mirrors. Maybe not. I don't
> > > assume the existence of them.
> > And yet, you _do_ assume clever designs or new science which allows a
> > massive manned presence in space. As I said, swallowing camels,
> > straining at gnats.
>
> No, I don't.
It pains me to say this, but yes you do. You've even given explicit
examples below.
> > > > And you think this is realistic? Or do you think, as I do, that this
> > > > is swallowing camels and straining at gnats? If not, why suppose all
> > > > these other advances, but not advances in optics(and rather
> > > > pedestrian advances at that.)
> > > What other advances do you think I'm assuming? I tend to be
> > > rather conservative in my speculations on future space
> > > technology. The only specific "toy" you mentioned was
> > > nuclear thermal rockets, which I generally do NOT assume
> > > will be used very much, if at all.
> > You're not?
>
> No, I'm not.
>
> > No extremely high exhaust velocities?
>
> No. Current exhaust velocities of ion drives are more or
> less ideal. Increasing the exhaust velocity would decrease
> efficiency, reduce acceleration, and increase trip times.
Those exhaust velocities are 'extremely high', at least as I judge
them. Your second statement simply does not follow, depending as it
does on a multitude of factors. But then again, I am unsure as to
what you think constitutes an ion drive: is it just the thrusters?
Or something more?
I would also note that while ion drives might - might - qualify for
certain unmanned probes, but the fact of the matter is, they're not
even that far yet:
And for manned flights? No, that is way speculative. I might add
that in the context of, say, a Mars mission, the sense is that ion
propulsion will increase payloads, not decrease trip times.
> > No scaled-up mass drivers?
>
> No. I don't care for mass driver technology for space
> applications. It requires large amounts of initial
> investment and R&D, and the payloads are small
> and must be robust enough to survive high
> accelerations. Traditional rocket drives require
> propellant, but the payloads can be larger and
> they don't need to survive high accelerations.
>
> > No closed-loop life-support?
>
> We currently live in a closed-loop life-support
> biosphere, so that's a proof of concept. I
> accept that as a baseline for what sort of
> minimum investment might be required for
> off-world colonization. That's a truly MASSIVE
> investment. On sfconsim-l, I worked out a
> conservative baseline design which would
> be 18km in diameter and 1.5E12 tons. Just
> for the habitat. I accept that much smaller
> may be possible, depending on technological
> advancement and/or willingness for less
> than total self-sufficiency. I do not ASSUME
> it, though.
I CALL FOUL! No way. Closed loop life-support is _very_ advanced
technology. We can't even do it here, on Earth, with even fairly
large mammals, let alone in space, for humans. I know that some
physical-science oriented types tend to be dismissive of these sorts
of issues, but the fact of the matter is, this is an extremely hard
problem. It might, as you say, be eminently solvable, but at this
point we simply don't know. Just as, in fact, we don't know how
difficult it would be to make highly reflective mirrors - that was my
whole point (I'm a bit baffled why you're hung up on this, frankly; I
don't think stealth is dependent upon the development of that
technology. I just picked a number that was easy to work with, namely
0.0001. It could be 0.01, or 0.001, for all I care or know.)
And if you don't have this sort of technology, how do you justify an
extensive manned presence in space? And if you don't have _that_, how
could there be any need for stealth or space wars, ever? You've got
to be consistent.
> > And so on and so forth. No,
> > you don't get to say it's a minor adaptation of off-the-shelf stuff we
> > have now.
>
> So, the only choices are between off-the-shelf
> stuff we have today, and everything you deem
> must be possible. Riiiight.
No, of course not. But you seem to be missing something: the
argument you're attributing to me is actually _your_ argument. I'm
pointing out that you're making assumptions that seem to be arbitrary,
in fact, seem to be driven by the need to assemble a consistent set of
story elements. There is no evidence to suggest the sorts of stories
you wish to tell are in fact feasible.
I don't suppose it would help at this point, but I'm a space-head from
waaay back. I thought that we'd have space stations, a moon base, a
manned mission to Mars by now, with plans on the drawing board for a
Jupiter shot (this was back in the 60's.) I am bitterly disappointed
that none of that has come to pass, and I really, really want there to
be something even so humble as a manned Mars trip before I die.
But such is life. You don't always get the pink pony you wanted -
that really flies.
> > > I don't like the general economics and technological
> > > developmental path of nuclear thermal rockets, compared
> > > to solar electric. Solar electric is already a mature technology,
> > > and capable of outperforming nuclear thermal. Nuclear
> > > thermal has never even flown once. Solar electric costs
> > > less, it performs better, it doesn't have political problems,
> > > and it's already flown. The basic technology of solar electric
> > > has extensive civilian terrestrial applications, which means
> > > lots of R&D money and effort to make it even better. In
> > > contrast, the sort of nuclear reactors suitable for nuclear
> > > thermal have no civilian applications and practically no
> > > terrestrial military applications. And what's the potential
> > > eventual payoff? A drive system which underperforms
> > > compared to cheaper solar electric. So no, I don't anticipate
> > > nuclear thermal drives.
> > But you are assuming lots of R&D for solar electric; it's just a
> > matter of throwing money at the problem?
>
> At this point, yes. Because the technology is sufficiently
> mature, we're at a point where a significant amount of
> improvement is indeed just a matter of throwing money
> at the problem. There's a limit to how far we can predict
> the improvement, of course.
You need to talk with other people who are actually in the field. My
impression is that these are very hard problems, and that the money,
even for your scenario, isn't coming in large gobs any time soon.
That's for manned flight of course.
> > > This is representative of my approach to future technology
> > > speculations--I like to be conservative. If something exists
> > > today, works well, and has great potential for continued
> > > development, then I like it. If not, then I'll need some
> > > convincing.
> > You've got it backwards - you need to convince me that it works well,
> > first. You don't get to decide this by fiat. And the fact of the
> > matter is, what you're positing _is_ advanced technology, technology
> > we don't have yet, technology you're assuming is not terribly hard to
> > develop.
>
> I don't need to convince you anything. More and more,
> it seems clear that you are just going to believe whatever
> you want and no one can convince you to budge in
> the slightest
Um, no. I am very willing to be convinced. I want to be convinced.
But you're not giving me a lot to go on.
.
> It's crazy that you insist I'm "swallowing camels" by
> assuming the existence of lots of crazy advanced
> technology. You don't know what I assume. And
> even when I explicitly tell you what I assume, you're
> still arguing the point. This is ridiculous.
In fact, you've given me numerous examples of what is, indeed,
advanced technology. Maybe you're getting the impression that it's
not because you're comparing yourself to the types who say that Orion
is 'just a matter of assembling bits of present technology'. But
really, you are.
Let me put it this way: if it's not advanced technology, how come
it's not already in use?
> > > > > I've done numbers on various "stealth" radiators before, usually
> > > > > coming from the perspective of trying to design one (i.e. I was
> > > > > "pro-stealth"). I don't remember the specifics, but basically I
> > > > > settled on a design with a 60 degree radiation cone. I wanted
> > > > > to design one with a 15 degree radiation cone, but the numbers
> > > > > never came anywhere close to adding up.
> > > > Instead of saying this, why don't you just show the numbers?
> > > Because I don't feel like looking them up or redoing them. You
> > > don't like that? Well, deal with it. I simply don't feel like going
> > > through the effort.
> > Pardon me for not just taking your word for it, but I don't; it's
> > entirely possible you made unjustified assumptions, did some bad
> > arithmetic, etc.
>
> Sure. But that doesn't change the fact that I don't feel like
> redoing the math right now. And you're not providing me
> with anything to change my mood.
>
>
>
> > > > > If you have a better design in mind, I'm all ears.
> > > > What? A radiator whose output is redirected by an advanced, actively
> > > > cooled optics system?
> > > > Something along the lines of a paraboloid with the radiating surfaces
> > > > at the focus?
> > > > There's not a whole lot more to say.
> > > That's the design which I had worked on before, and rejected
> > > it on the grounds that the amount of active cooling required
> > > exceeded the amount of power available. Well, I didn't
> > > "reject" it, exactly, I just wrestled with the numbers until I
> > > got something which worked about as well as I could make
> > > it (a 60 degree cone).
> > > If you have nothing more to say, then color me completely
> > > and utterly unconvinced.
> > Shrug. You're the one telling me that stealth in space is not
> > possible. It falls upon _you_ to convince _me_ that this is the
> > case. Not the other way around. Note that at this point, what I've
> > contributed, basically, is pointing out that John's bit 'wisdom' was
> > just plain wrong.
>
> What you specifically said was this:
>
> "Sigh. The anti-stealth people were wrong. Period. They
> misunderstood a bit of basic physical law."
>
> You're saying that we misunderstood some basic physical
> law. When pressed to specify which basic physical law
> was misunderstood, you said this:
>
> "The one where Schilling insisted that the radiator had to scale in
> inverse proportion to directionality. It doesn't. I gave a proof
> (two actually) of why this wasn't so, and consulted with someone in
> the physics department who actually knows a little thermodynamics. He
> said it was a new one on him. In fact, I had a thread about just this
> not too long ago. "
>
> Okay, but it wasn't clear to me what you meant by
> "scale in inverse proportion to directionality" nor the
> particular relevance with this "basic" physical law
> and the difficulties of stealth in space.
Right. Perhaps I should have been more clear: the anti-stealth types
who base their misconceptions on this misunderstood 'law'. In fact, I
was thinking of the last several go-rounds:
http://james-nicoll.livejournal.com/1130300.html
http://www.projectrho.com/rocket/rocket3w.html#nostealth
_That_ is what I am replying to: that stealth is impossible because of
a (fictitious) natural law. It may be impossible for other reasons,
maybe, but certainly not that one.
> > I am most certainly not trying to convince you that
> > it is possible. In fact, my position is that we simply don't know
> > enough to tell yet, because the question turns upon many factors, some
> > of the economic, which are not amenable to prediction.
>
> No, it seems your position is that the anti-stealth people
> are wrong. Not that "we simply don't know". That
> we're wrong.
See above. I'm just objecting to using misunderstood natural law as a
springboard for some sort of authoritative conclusion.
Luke Campbell
> I get the feeling that your essential design approach is
> flawed in some sort of fundamental way. You seem to
> be looking at sources of heat as potential places to
> get energy from, but the real problem that you need to
> be tackling is radiating away waste heat.
Let me just rephrase this slightly. If you are running a reactor core
of, say, 1 MW at 3000 K, then every second you are generating 333 J/K
of entropy (approximately, assuming your heat engine is running
reversibly). It is this entropy that you need to somehow get rid of
(preferably without creating much more of it). Any energy you have
left over that you don't use up while dumping the entropy can do
something useful. So looking at this from an entropy perspective, a
black body at temperature T and area A will radiate an entropy
S=4/3 sigma_b T^3 A t
in a time interval t, and where sigma_b is the Stefan-Boltzmann
constant.
Luke
Charles...@gmail.com
> On Feb 26, 9:55 am, CharlesRCap...@gmail.com wrote:
> > On Feb 25, 4:29 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > > On Feb 25, 2:14 pm, CharlesRCap...@gmail.com wrote:
> > > Where is "the heat" coming from? A fission reactor, perhaps?
> > > In principle, you can make a very efficient reactor that converts
> > > almost all of its heat energy into electricity. Conveniently,
> > > a heat engine is most efficient when the radiator temperature
> > > is a small fraction of the reactor's temperature. For stealth,
> > > this seems like a win-win situation. Your radiator is a lower
> > > temperature, and you have less waste heat to worry about!
> > I am assuming the use of a MITEE nuclear thermal rocket which can
> > produce very little waste heat for its output. Further, I am assuming
> > that the efficiency of that reactor can be adjusted downward so that
> > more waste heat is radiated into the heat engine system if needed.
>
> Oh, well then forget about any sort of stealth. The reason a
> nuclear thermal rocket "produces little waste heat" is
> because it actually produces lots of waste heat but most
> of it is conveniently dumped into the exhaust. The rocket
> exhaust is going to be rather visible; a Space Shuttle launch
> is stealthy in comparison.
Indeed? Well it gets back to the question that I had asked in that
other thread which was how detectible is the trail from an Ion drive
or a VASIMR's hydrogen trail? Is it even worth trying to stealth your
craft if your exhaust is going to give you away no matter what? If the
exhaust from an ion drive or VASIMR (or a MITEE) is going to be very
detectible then there is no reason to ever worry about the temperature
of your craft's hull. How different is the exhaust from a MITEE than
from a VASIMR? They both heat and spew out hydrogen. In any case you
can replace stage 0 with just a plain old fission plant if you like
and then I need to make sure I have enough electricity left over to
run a drive of some sort that isn't as 'noisy' as a MITEE. What I'm
working on now is not about the propulsion system, rather it is about
the heat rejection system.
***
To preface this: Obviously something in the math is off. Probably
several things and very badly off. If everyone says it can't be done,
then if I have something that says it can, I have a problem somewhere.
I'm feeling really troubled about the values I came up with to
describe the compression stages, I feel that they are probably more
than an order of magnitude or more off. I'm also unhappy with the
values I got for the Electrothermal devices as well.
***
I'm pretty sure my calculations are flawed in a serious manner, but
I'm not sure how exactly. I am proceeding on the assumption that they
are within an order of magnitude in accuracy. I'm expecting to refine
them as I get a better understanding of the math involved with the
various stages. It's my hope that the end result will be within an
order of magnitude, but who knows if that'll be the case.
Main Cycle:
Stage 0: Reactor - MITEE Nuclear Thermal Rocket
Capacity: 20mw thermal @ 3000K
Efficiency: 75% (adjustable)
Waste: 5mw thermal @ 3000K
Stage 1: Helium Coolant Reservoir + Intercooler at 3000-1123K
Input: 5mw thermal at 3000K
Output: 5mw thermal at 1123K (K^4 = 1,590,446,354,641)
Stage 2: Brayton Cycle Helium Turbine
Input: 5mw thermal at 1123K
Efficiency: 50.11%
Generation: 2.5055mw electrical
Waste: 2.4945mw thermal at 378.7K
Stage 3: Helium Coolant Reservoir at 378.7K
Input: 2.4945mw thermal at 378.7K
Output: 2.4945mw thermal at 378.7K
Stage 4: Multi-Stage Silicon Electrothermal Nano-Devices (#1) ****
Input: 2.4945mw thermal at 378.7K (K^4 = 20,567,486,479)
Efficiency: 60%
Generation: 1.4967mw electrical
Waste: 0.9978mw thermal at 301.2K (K^4 = 8,226,994,544)
Stage 5: Helium Coolant Reservoir at 301.2K
Input: 0.9978mw thermal at 301.2K
Output: 0.9978mw thermal at 301.2K
Stage 6: Multi-Stage Silicon Electrothermal Nano-Devices (#2) ****
Input: 0.9978mw thermal at 301.2K (K^4 = 8,226,994,544)
Efficiency: 60%
Generation: 0.5987mw electrical
Waste: 0.3991mw thermal at 239.5K (K^4 = 3,290,797,817)
Stage 7: Helium Coolant Reservoir at 301.2K
Input: 0.3991mw thermal at 239.5K
Output: 0.3991mw thermal at 239.5K
Stage 8: Multi-Stage Silicon Electrothermal Nano-Devices (#3) ****
Input: 0.3991mw thermal at 239.5K (K^4 = 3,290,797,817)
Efficiency: 60%
Generation: 0.2395mw electrical
Waste: 0.1596mw thermal at 190.5K (K^4 = 1,316,319,127)
Stage 9: Helium Coolant Reservoir at 190.5K
Input: 0.1596mw thermal at 190.5K
Output: 0.1596mw thermal at 190.5K
Stage 10: Multi-Stage Silicon Electrothermal Nano-Devices (#4) ****
Input: 0.1596mw thermal at 190.5K (K^4 = 1,316,319,127)
Efficiency: 60%
Generation: 0.0958mw electrical
Waste: 0.0638mw thermal at 151.5K (K^4 = 526,527,651)
Stage 11: Helium Coolant Reservoir at 151.5K
Input: 0.0638mw thermal at 151.5K
Output: 0.0638mw thermal at 151.5K
Stage 12: Helium Compressor/Intercooler (#1)
Input: 0.0638mw thermal at 151.5K (K^4 = 526,527,651)
Draw: 0.0089mw electrical**
Waste: 0.0009mw thermal***
Output: 0.0647mw thermal at 174.1K* (K^4 = 917,737,696)
Stage 13: Helium Coolant Reservoir at 174.1K
Input: 0.0647mw thermal at 174.1K
Output: 0.0647mw thermal at 174.1K
.
. (Jumping ahead a bit, some 21 cycles of compression are assumed
here.)
.
Stage 55: Helium Compressor/Intercooler (#22)
Input: 0.0855mw thermal at 2800.1K (K^4 = 61,474,381,270,411)
Draw: 0.0120mw electrical**
Waste: 0.0012mw thermal***
Output: 0.0866mw thermal at 3217.3K* (K^4 = 107,143,600,341,952)
Stage 56: Helium Coolant Reservoir at 3217.3K
Input: 0.0866mw thermal at 3217.3K
Output: 0.0866mw thermal at 3217.3K
Total Electricity for Compression: 0.2285mw
Stage 57: Radiator and Reflector
Input: 0.0866mw thermal at 3217.3K
Reflection Efficiency: 98%
Rejected: 0.0849mw thermal
Reabsorbed: 0.0017mw thermal at 3217.3K
Stage 58: Reflector Active Cooling to 50K or less
Input: 0.0017mw thermal at 3217.3K
Output: 0.0017mw thermal at 50K or less
Stage 59: Outer Hull Helium Reservoir at 50K or less
Input: 1700w thermal at 50K
Radiation: 0.3542 watts per m^2 at 0.9994 Emissivity
Hull Area: 4800 m^2 to Eliminate 1700w at 50K
* I'm assuming a factor of 1.743 on the energy for each compression
stage, this is derived from the values listed in the Brayton cycle
simulation software. I feel that I am making bad assumptions on this
though.
** I'm assuming a factor of 0.1401 on the energy (per thermal unit)
needed to compress per stage. Which is also derived from the values
listed in the Brayton cycle simulation software. I feel that I am
making bad assumptions on this though.
*** I'm assuming 10% waste heat on the electricity drawn is added to
the heat moved. This is a wild guess.
**** I assume that if the devices work at 60% efficiency that they
take out 60% of the energy in the heated medium, the energy is
equivalent to T(K)^4 and they leave 40% of that to pass through as
waste. Thus the output temperature is sqrt(sqrt(((T^4)*0.4))) which
could be wildly incorrect and I have the feeling that it is, I think
the output temperature is much lower than that equation represents,
but I don't have another equation to work with.
All of these values could be off by huge amounts.
Also this does not take into account the heat that is returned from
using the electrical generation that's left over. I assume that one
would tune the output of the reactor so that no unused electricity
would be generated.
Tim Little
> The original concept assumed it's okay to wait decades for the
> mission to complete, so the required delta-v might not be so
> daunting.
If it were not an asteroid crossing Earth's orbit to begin with, then
there would be a minimum delta-V requirement - almost certainly enough
for contained waste heat to vaporise it.
If it were a multi-gigatonne asteroid passing near Earth's orbit, it
would definitely be tracked.
Either way, the power requirement for an asteroid of the size being
discussed would be more like terawatts than megawatts. It would be
practically impossible to change its orbit enough to hit Earth at all,
let alone do so stealthily.
- Tim
IsaacKuo
wrote:
> On 2008-02-26, IsaacKuo <mech...@yahoo.com> wrote:
> > The original concept assumed it's okay to wait decades for the
> > mission to complete, so the required delta-v might not be so
> > daunting.
> If it were not an asteroid crossing Earth's orbit to begin with, then
> there would be a minimum delta-V requirement - almost certainly enough
> for contained waste heat to vaporise it.
The original concept didn't specify the target was Earth, but
that's not really the main issue. The main issue is what the
minimum delta-v requirement would be, and it's not significantly
worse for Earth than any other target. Basically, you want
to nudge a far out object, like a Kuiper belt object. All you
need to do is wipe out the object's 3-5 km/s of orbital
speed, and the Sun's gravity will do the rest. Naturally, you
need to time it well, so it will fall onto the target as it passes
by. It'll take decades for the object to fall onto the target,
but that's compatible with the original scenario.
In principle, at least, a mass driver rocket can be pretty
efficient (also specified in the original scenario). With
a mission delta-v of only 5km/s, an average exhaust
velocity of 3-4km/s is fine. This is within the range of
chemical rockets, which is vaguely limited by the
strength of chemical bonds, which will also be vaguely
the limit of how much you can dump heat into a solid
object without vaporizing it. But since the mass driver
can be pretty efficient, you'll only be contending with
a fraction of that energy. I think it should be possible
to dump the excess heat into the asteroid without
vaporizing it.
However, while this is possible in principle I don't think
it's at all practical. Once the asteroid is heated up,
it's going to naturally glow in infrared unless it's very
well insulated. How well insulated? Well enough that
the heat doesn't leak through the insulation even after
decades.
> Either way, the power requirement for an asteroid of the size being
> discussed would be more like terawatts than megawatts. It would be
> practically impossible to change its orbit enough to hit Earth at all,
> let alone do so stealthily.
By starting off with an object with little orbital speed,
the required energy is minimized. The energy required
scales with the square of the mission delta-v. The
power required is the energy required divided by the
time. With a mission time on the order of many decades,
it's okay if the acceleration burn requires a couple
decades.
So, for example, let's suppose the asteroid is a million
ton Kuiper belt iceball. Roughly 3/4 of the mass is
consumed by the mass launcher rocket, leaving a
250,000 ton payload and a mission delta-v of 4km/s.
The energy required is about 2E15 joules. An
acceleration time of two decades is 630 million
seconds, so the power required is about 3 megawatts.
Of course, this is assuming 100% efficiency. With
an 75% efficient mass launcher system, let's call it
around 4 megawatts. So, for this example it's
megawatts, not terawatts.
Depending on the target, total mission time could be
half a century to a century.
And no, I still don't find this method of attack any
more plausible than in my first reply. Even if we
stealth up the asteroid with a neat semi-magic
reflective sunshade and directional radiator, all
it takes is one active sensor scan during all those
decades to blow the surprise.
Isaac Kuo
Tim Little
> The original concept didn't specify the target was Earth, but
> that's not really the main issue.
I think you've forgotten what the original concept was:
"Suppose the First Marshen Republick launches an expedition to visit
various uninhabited asteroids, a-to-b-to-c-to-d. The carefully
selected crew (who all happen to have high security clearances and
tight lips) visit the asteroids and publish their papers and all is
right with the worlds. Ten years later, an unmapped giant asteroid
wipes out the Terran Pacific Rim with giant tsunamis and earthquakes,
and severely inconveniences the rest of the earth with several nuclear
winters."
> Naturally, you need to time it well, so it will fall onto the
> target as it passes by. It'll take decades for the object to fall
> onto the target, but that's compatible with the original scenario.
Not as originally written. The whole thing took place within ten
years, from beginning emplacement to devastation.
> In principle, at least, a mass driver rocket can be pretty efficient
> (also specified in the original scenario). With a mission delta-v
> of only 5km/s, an average exhaust velocity of 3-4km/s is fine.
If your payload is only a quarter of the original asteroid, sure.
> But since the mass driver can be pretty efficient, you'll only be
> contending with a fraction of that energy.
Not just the mass driver: every step of the conversion from potential
energy to kinetic. With near-future technology I'd be dubious of any
claims of more than 30% total efficiency, though I'd accept up to 50%
or so for improvements in the more distant future. I'd be especially
critical of something rigged while plausibly pretending to be mapping.
With a 3-4 km/s exhaust for 5 km/s total delta-V, the minimum energy
required is on the order of 25 MJ/kg of payload. That's easily enough
to vaporise any common asteroid materials even neglecting the fact
that the efficiency will drop as the environmental temperature rises.
> So, for example, let's suppose the asteroid is a million ton Kuiper
> belt iceball. Roughly 3/4 of the mass is consumed by the mass
> launcher rocket, leaving a 250,000 ton payload and a mission delta-v
> of 4km/s.
That's multiple orders of magnitude too small for "wip[ing] out the
Terran Pacific Rim", and too slow to get there in ten years.
> And no, I still don't find this method of attack any more plausible
> than in my first reply.
At least we're agreed in that. We just disagree in some of the
details.
- Tim
Charles...@gmail.com
I found that the numbers I was using lead to a perpetual motion
machine. I looked around a bit and adjusted the compression factor to
1.416 and the draw factor to 0.25. This leads to more satisfactory
numbers (i.e. I believe I'm going in the right direction, which is
less efficiency) but they still amount to little more than wild
guesses.
IsaacKuo
I think you're getting bogged down in the details of the
middle steps, while it may be easier and less mistake-prone
to analyze the thermodynamics of just the front end and
tail end--the reactor heat source and the radiator(s).
Your maximum potential efficiency is 1-Tcold/Thot,
so start with your reactor temperature for Thot and
your desired radiator temperature for Tcold. For example,
you could have Thot be 1400K (melting point of Uranium),
and Tcold be 50K. That leads to a maximum efficiency
of 96.4%.
Along with the desired power level, this gives you and
idea of the amount of waste heat generated. You don't
have to worry your mind about "recovering" any of the
waste heat--you can't! This is the amount of waste
heat you'll have to deal with assuming every last erg
of recoverable heat is recovered absolutely perfectly.
(This is NOT realistic, but it gives an ultimate ceiling
on performance.) The amount of waste heat is
power*(Tcold/Thot)/(1-Tcold/Thot), so if you want, say,
10 megawatts, then the waste heat is 370 kilowatts.
10 megawatts is around the amount of power for a
250,000 ton multi-decade impactor.
Now, we can figure out the surface area required to
radiate that 370 kilowatts. Assuming the radiator is
a perfect blackbody, the irradiance is:
5.67E-8watts/m^2*(Tcold/K)^4, or .35watts.
Thus, 370 kilowatts requires about a million
square meters, or a radiator that's 1km by 1km.
Of course, this radiator is not at all stealth, since
it's radiating over an entire hemisphere. If we had
a perfect reflector, we could in principle use a
100km diameter parabolic mirror to direct the
waste heat into a relatively narrow 2 degree cone.
But we have something less than a perfect
reflector, of course. Let's not worry about the
additional waste heat just yet, and just consider
the mass of the hardware required to cool the
mirror. Assuming most of the vehicle mass is
the mirror and associated hardware, we're
spreading 250,000 tons over a 100km diameter.
This implies a mass budget of only 32 grams
per square meter. Assuming you need at least
two layers sandwiched around near-vacuum coolant,
this only the same order of magnitude as solar
sail material, and this mirror needs to be rigid!
Isaac Kuo
Charles...@gmail.com
> On Feb 28, 8:57 am, CharlesRCap...@gmail.com wrote:
> > On Feb 26, 2:02 pm, CharlesRCap...@gmail.com wrote:
> > > On Feb 26, 11:42 am, IsaacKuo <mech...@yahoo.com> wrote:
> > > > On Feb 26, 9:55 am, CharlesRCap...@gmail.com wrote:
> > > > > On Feb 25, 4:29 pm, IsaacKuo <mech...@yahoo.com> wrote:
> > > > > > On Feb 25, 2:14 pm, CharlesRCap...@gmail.com wrote:
> > > All of these values could be off by huge amounts.
> > I found that the numbers I was using lead to a perpetual motion
> > machine. I looked around a bit and adjusted the compression factor to
> > 1.416 and the draw factor to 0.25. This leads to more satisfactory
> > numbers (i.e. I believe I'm going in the right direction, which is
> > less efficiency) but they still amount to little more than wild
> > guesses.
>
> I think you're getting bogged down in the details of the
> middle steps, while it may be easier and less mistake-prone
> to analyze the thermodynamics of just the front end and
> tail end--the reactor heat source and the radiator(s).
>
> Your maximum potential efficiency is 1-Tcold/Thot,
> so start with your reactor temperature for Thot and
> your desired radiator temperature for Tcold. For example,
> you could have Thot be 1400K (melting point of Uranium),
> and Tcold be 50K. That leads to a maximum efficiency
> of 96.4%.
Thank you very much for that well reasoned and succinctly described
explanation. I am quite grateful for the time that you have spent
crafting it. Despite the simplicity of the mathematics involved it
taught me some very fundamental things about thermodynamics that will
be very helpful in understanding where my design is flawed, and boy it
is flawed. =)
You are absolutely correct, stepping back and looking at the big
picture helps immensely. Even with perfect efficiency (which the
Helium Turbine comes really close to) the maximum efficiency should be
no more than 86.5% from 1123K to ~150K while the design I put forth
has nearly 99% efficiency.
The deal breaker that you indicate (the huge radiator) is where I
would still disagree. The system rejects its waste heat into a ~150K
helium reservoir (mainly because from what I can gather the efficiency
of the electrothermals really starts to bottom out at 150K) so we have
the low point there.
The difference that I'd like to go with is that there is no reason I
can't take the heat that is radiated into the ~150K reservoir and then
compress it though a series of compressor/intercoolers until it is at
a high enough temperature to be rejected from a reasonable sized
radiator. The issue would be if the cost in waste heat and electrical
power would be too high to make that practical.
Your lesson in thermodynamics has given me a lot to think about, and I
thank you again for taking the time to explain this to me. (I'd like
to take some courses in physics beyond Physics 101 that I took 13
years ago, but I'd have to enroll and be accepted into the local state
university to take any. The other local colleges I attend are more
geared towards career type education even if in theory the degrees
they offer are just as good. And when I think about it, where would I
find the time?)
Luke Campbell
> The difference that I'd like to go with is that there is no reason I
> can't take the heat that is radiated into the ~150K reservoir and then
> compress it though a series of compressor/intercoolers until it is at
> a high enough temperature to be rejected from a reasonable sized
> radiator. The issue would be if the cost in waste heat and electrical
> power would be too high to make that practical.
There is a reason. Suppose your reactor core produces 3 MJ of energy
at 3000 K. Because this is produced as heat, you also unavoidably end
up with 1000 J/K of entropy. I am going to simplify things for this
discussion and assume all processes are reversible and heat/entropy
exchanges take place at constant temperature. Now, you can't just
keep that 1000 J/K of entropy around - as your reactor produces more
heat that entropy will pile up and eventually make things bad for
you. So you need to get rid of it. Since entropy is associated with
heat, you get rid of entropy by dumping heat. The amount of entropy S
that is transferred along with an amount of heat Q at temperature T is
S=Q/T or, rearranging, Q=ST (more rigorously dQ = T dS, and you need
to do an integration). So if you expel the entropy at a radiator
temperature of 150 K, you must also lose 1000 J/K * 150 K = 150 kJ of
energy in the process. If, however, you recompress the helium to
radiate it at higher temperatures, say at 500 K, then you also lose
more energy in the process, 1000 J/K * 500 K = 500 kJ. Therefore, the
hotter your radiator, the more energy you must radiate to get rid of
the same amount of entropy. This lowers your efficiency and makes you
more visible. Note that this accounting of entropy along with your
energy is exactly what gives you the thermodynamics-limited
efficiencies that Isaac mentioned.
Luke
Charles...@gmail.com
Thank you for taking the time to write that.
I'm afraid that I didn't understand all of it however. Basically what
you are saying is that thinking of the waste as heat is problematic
and I should rather be thinking of the waste as units of entropy
(Joules per degree Kelvin) 1000 of them in this example.
I also understand that you are saying that regardless of any
efficiencies of compressors or other machinery, that it is more
expensive (in free energy, is that the correct term?) to get rid of
that entropy at a higher temperature. You have to radiate more heat to
get rid of the same amount of entropy.
Okay, assuming I am correct in understanding what you said, then it's
not really a difference in type of problem, but more a difference in
the scope of the problem. It's going to be significantly more
expensive to get rid of the entropy by increasing the temperature.
One question I have is how you derived the 1000 K/J value for entropy
in the example system? What did you have to do to determine that?
Luke Campbell
> On Feb 29, 10:36 am, Luke Campbell <lwc...@gmail.com> wrote:
>
> I'm afraid that I didn't understand all of it however. Basically what
> you are saying is that thinking of the waste as heat is problematic
> and I should rather be thinking of the waste as units of entropy
> (Joules per degree Kelvin) 1000 of them in this example.
It can simplify things.
> I also understand that you are saying that regardless of any
> efficiencies of compressors or other machinery, that it is more
> expensive (in free energy, is that the correct term?) to get rid of
> that entropy at a higher temperature. You have to radiate more heat to
> get rid of the same amount of entropy.
Yes. Exactly.
Free energy has some technical meanings in thermodynamics, referring
to energy-like quantities of systems in thermal or particle
equilibrium with other systems, but I understand what you are talking
about.
> Okay, assuming I am correct in understanding what you said, then it's
> not really a difference in type of problem, but more a difference in
> the scope of the problem. It's going to be significantly more
> expensive to get rid of the entropy by increasing the temperature.
Expensive in terms of energy expenditure, yes. It can end up being
less expensive in terms of mass if you end up with smaller radiators.
> One question I have is how you derived the 1000 K/J value for entropy
> in the example system? What did you have to do to determine that?
I divided the heat produced by the reactor core (3 MJ) by the assumed
temperature of the reactor core and its coolant (3000 K). Things get
more complex if the heat is not exchanged at constant temperature.
Luke
Tim Little
> The difference that I'd like to go with is that there is no reason I
> can't take the heat that is radiated into the ~150K reservoir and then
> compress it though a series of compressor/intercoolers until it is at
> a high enough temperature to be rejected from a reasonable sized
> radiator.
Thermodynamically, this is (at best) equivalent to just using a higher
temperature 'cold' reservoir to begin with, thus losing efficiency.
- Tim
Tim Little
> in the example system? What did you have to do to determine that?
It was actually 1000 J/K, derived from 3 MJ of heat at 3000 K. If you
could get a hotter working temperature you could reduce the amount of
entropy. If you could derive energy directly from million-degree
plasma, you could in theory reduce the entropy to 3 J/K. However,
such hot things tend to have pesky little problems even without trying
to make it 99.9% efficient as well. Just ask the ITER people.
- Tim
Damien Sullivan
>It's been pointed out to you multiple times that your reasoning is
>based upon a factually false premise, something I think you heard but
>did not understand. But given the debacle where you claimed that you
>'weren't convinced' that there was no link between Aluminum and
>Alzheimer's, despite being corrected repeatedly, I'm not guessing
>you're going to publicly change your mind.
http://www.alzheimers.org.uk/site/scripts/documents_info.php?documentID=99
"There is circumstantial evidence linking this metal with Alzheimer's
disease but no causal relationship has yet been proved. As evidence for
other causes continues to grow, a possible link with aluminium seems
increasingly unlikely."
http://www.alzheimer.ca/english/disease/causes-alumi.htm
"Most researchers no longer regard aluminum as a risk factor for
Alzheimer's disease. However, some researchers are still examining
whether some people are at risk because their bodies have difficulties
in handling foods containing the metals copper, iron, and aluminum."
http://www.straightdope.com/classics/a971219.html
"weren't convinced" seems like a perfectly respectable attitude.
-xx- Damien X-)
Damien Sullivan
>I would also note that while ion drives might - might - qualify for
>certain unmanned probes, but the fact of the matter is, they're not
>even that far yet:
>
>http://query.nytimes.com/gst/fullpage.html?res=9C02E2D61E38F935A35753C1A96E958260&sec=&spon=&pagewanted=all
That article is from 1998.
>I CALL FOUL! No way. Closed loop life-support is _very_ advanced
>technology. We can't even do it here, on Earth, with even fairly
>large mammals, let alone in space, for humans. I know that some
How much have we tried?
>And if you don't have this sort of technology, how do you justify an
>extensive manned presence in space? And if you don't have _that_, how
>could there be any need for stealth or space wars, ever? You've got
>to be consistent.
True, and applies to space drives as well, which makes watching you two
argue over nuclear thermal and ion drives somewhat surreal.
>I don't suppose it would help at this point, but I'm a space-head from
>waaay back. I thought that we'd have space stations, a moon base, a
You're not the only one around here.
>manned mission to Mars by now, with plans on the drawing board for a
You're not the only one around here.
>You need to talk with other people who are actually in the field. My
>impression is that these are very hard problems, and that the money,
>even for your scenario, isn't coming in large gobs any time soon.
Not that large gobs are being thrown at space at all.
-xx- Damien X-)