[ExI] Spacecraft (was MM)

[Keith Henson]

On Mon, Jan 3, 2011 at 12:32 PM,
<extropy-chat-request at lists.extropy.org> wrote:

> From: "spike" <spike66 at att.net>

>>>>> That amounts to about 0.002 MOA tracking a rocket through atmosphere.
>>>
>>>> MOA?
>>>
>>> Miinute of arc.
>
>>Ah.  The Hubble, which has been up for 20 years and is based on technology
> at least 10 years before that, has a pointing accuracy of 7 milliarcseconds.
> A milliarcsecond is about 5 x 10^-9 radians, so 7 would be about 35 x 10-9
> rad.  At the end of a 36,000,000 m radius, the error would be ~1.3 m
>
> Ja, there are two different things being discussed here, actually three:
> tracking objects thru atmosphere, tracking a moving object and Hubble
> boresight accuracy.  The Hubble is indeed an impressive control system, but
> it cannot track moving objects very competently.  It is really really good
> at doing what it was designed to do, fix on a dim object and stay right on
> it for long periods.  But it isn't nimble, doesn't need to be for that
> application.  One can steer a battleship with a canoe paddle, if one is
> patient and has a few days to get it done.
>
> For anything that moves or depends on an optical feedback, an arc second of
> accuracy is probably still out of our reach, but there is plenty of useful
> stuff we could do with arc-second class ground based tracking.
>
>  ...
>
>>> Slow tracking gives no room for any perturbations in flight path, right?
>
>>Talk to Spike about the control problem.  But the acceleration is modest,
> around a g, and while the velocity is high, the angular velocity tracking
> the vehicle is low and you get feedback in less than 1/10th of a
> second...Keith
>
> There are ways to do stuff like this using fast steering secondary mirrors,
> adaptable aperture, lostsa cool notions for laser propulsion that were
> actually developed from a weapons program, Airborne Laser:
>
> http://en.wikipedia.org/wiki/Boeing_YAL-1
>
> Booeing was the prime contractor for this, but most of the pointing control
> and accuracy infrastructure was subcontracted to Lockheed Missiles and Space
> Company, in Sunnyvale Taxifornia, and also the Lockheed Advanced
> Technologies Center in Palo Alto, using Lockheed Martin control technology
> and Lockheed Martin control engineers, who incidently worked for Lockheed
> Martin.  Of course we used a Booeing product to carry our control system
> aloft, so those lads deserve credit where credit is due, but the control
> system was pure LMCo, and it is WICKED cool, do let me assure you, clever as
> all hell.
>
> Looks to me like we could adapt the accuracy infrastructure of the ABL to
> fly at about 10 to 12 km altitude and provide second stage ablative boost
> assist, from about 20 km to about 80-ish km altitude.  So first stage mostly
> solid propulsion, second stage ABL ablative boost, third stage throwaway
> H2/LOX?  We would need to get tricky with our optical feedback loops, but I
> think this is doable with current control law.
>
> spike
>
>
>
> From: Alfio Puglisi <alfio.puglisi at gmail.com>

>> Ah.  The Hubble, which has been up for 20 years and is based on
>> technology at least 10 years before that, has a pointing accuracy of 7
>> milliarcseconds.  A milliarcsecond is about 5 x 10^-9 radians, so 7
>> would be about 35 x 10-9 rad.  At the end of a 36,000,000 m radius,
>> the error would be ~1.3 m
>>
>
> You are finally touching points where I can contribute something :-)
>
> Typical pointing accuracy for ground-based telescopes are on the order of 1
> arcseconds, sometimes worse. The Hubble doesn't do much better (some
> references I saw speak of 0.2-0.5 arcsec, which I would regard as very
> good). The number you quote (7 milliarcsec) is for pointing *stability*,
> with feedback from a guide star. Hubble has the advantage of working outside
> the atmosphere, in diffraction-limited mode. A normal telescope inside the
> atmosphere has worse stability (20-30 milliarcsec or more), because the
> guide star feedback is smeared by atmospheric aberrations. A telescope
> equipped with an adaptive optics system (AO) can again reach a stability in
> the single-digit milliarcseconds range.
>
> All these numbers are for fixed, low-speed objects whose trajectory can be
> computed in advance.
>
> Therefore a laser-propulsion device cannot work without an AO-enabled launch
> system, which can keep your laser beam focused on your target and will track
> its motion. This supposes that the payload can give optical feedback about
> its position, but I imagine that the ablation process will make it quite
> bright :-)

Jordin Kare convinced me that CW laser rather than short pulse
ablation lasers are *much* nearer term on a GW scale.  They are also
much more efficient, turning upwards of 90% of the beam energy into
kinetic energy in the exhaust vs 30% for ablation.

And putting the whole Skylon derived stage into LEO solves the sticky
problem of how you get it back to the launch point.

If it is too much trouble for the beam to hit the vehicle, then we
know the ideal location and velocity profile we need to put the
vehicle in orbit.  Maybe we just sweep the beam without feedback and
let the vehicle keep up with it.  Given how fast hydrogen is moving
through the vehicle's plumbing, it should be fairly easy to acquire
the beam and keep it in the right spot.

> Such AO systems are now commonplace on big telescopes, but work in the
> milliwatt regime instead of GW, since they just reflect starlight :-)
> Laser-guided AO systems, where a laser is shoot up in the sky to create a
> "fake" guide star, are starting to work right now, and if I remember
> correctly some of them are planning an adaptive launch mirror. Such lasers
> are on the order of 10 watts. I don't have the foggiest idea of what happens
> when the power is scaled by a factor of 10^9. Having seen the safety systems
> for such lasers, I would imagine that a multi-GW laser will be treated like
> a nuclear-testing ground. At least, I would keep the same distance.

The lasers would be at most multi MW.  It has always been assumed that
they would use adaptive optics.  A ten watt mm beam is 10 MW per
square meter.  The power going up to GEO would be way below that.

Keith