Space & Astronomy/Space and Astronomy (October 1993).iso/mac/TEXT/SPACEDIG/V11_1/V11_176.TXT

Internet Message Format | 1991-07-08 | 19KB

Return-path: <[e-mail]>
X-Andrew-Authenticated-as: 7997;andrew.cmu.edu;Ted Anderson
Received: from beak.andrew.cmu.edu via trymail for +dist+/afs/andrew.cmu.edu/usr11/tm2b/space/[e-mail] (->+dist+/afs/andrew.cmu.edu/usr11/tm2b/space/space.dl) (->ota+space.digests)
          ID </afs/andrew.cmu.edu/usr1/ota/Mailbox/Qa28DSG00VcJINIU4W>;
          Thu, 22 Mar 90 02:50:55 -0500 (EST)
Message-ID: <[e-mail]>
Reply-To: [e-mail]
From: [e-mail]
To: [e-mail]
Date: Thu, 22 Mar 90 02:50:28 -0500 (EST)
Subject: SPACE Digest V11 #176

SPACE Digest                                     Volume 11 : Issue 176

Today's Topics:
         Re: The Amazing technicolour flying coilgun
                  Engineers
              Re: Shuttle escape systems
     Re: Coilgun on a 747 - supplies to orbit at $20/lb?
     HST at the limit: roll deconvolution and other topics (long)
            Re: More space station news...
         Re: SR-71 Record Flight Information
----------------------------------------------------------------------

Date: 22 Mar 90 08:45:28 GMT
From: munnari.oz.au!csc![e-mail]
Subject: Re: The Amazing technicolour flying coilgun

I have been following this discussion about the use of a coil gun on a 747 or
similar aircraft with interest.  However many have disparaged the idea because
of the inability of most (if not all) present aircraft to either fly
high enough or fast enough and be able to absorb the recoil of the gun when
it fired.  I would have thought the obvious solution would have been to design
and build a brand new aircraft, which *uses* the coil gun as a major structural
member.  Hence the aircraft would be designed from the start to absorb the
problems that came with the use of such a launching system.  Additionally this
would allow you to make the aircraft to fit the coil gun, not the other way
around.  While it might not actually be that nice a plane to fly, it would
be a hell of a lot cheaper than a new spacecraft, because it doesn't have to
be rated for exo-atmospheric flight.  I don't imagine it would be all that
pretty :-), in fact I think it would most probably be designed perhaps like
a very long thin "dragonfly" shape with the several hundred metre long coil gun
forming the fuselage.
                     ___
e.g.                |   |
                    |   |
                    |   |
                    |   |                                         ||
              ____________________________________________________||___------|
             |_________________________________________________________|     |
                    |   |                                         ||   ------|
                    |   |                                         ||
                    |   |
                    |___|

    You might have to place the pilot's cockpit in the extreme stern or
asymetically out on a wing to avoid electro-magnetic hazards (the elarged
section on the stern is for a cargo compartment where the payloads to be
launched would be carried and loaded into the rear/breach of the gun).

Brian Ross

------------------------------

Date: 21 Mar 90 20:41:23 GMT
From: amethyst!spock![e-mail]  (Michael Hicks)
Subject: Engineers


    For those of you who work as aero/mech engineers at one of
    the NASA centers such as AMES, KSC, LEWIS,  etc, could you
    answer this following question for me please?

    Approximately how many new engineers fresh out of school does
    your center hire per year?   Do openings come up year round or
    are there certain periods when most of the hiring is done?
    Pros and cons of your work?

    If you can answer any of these questions,  I'd be grateful.


                        Michael Hicks




    [e-mail]

------------------------------

Date: 18 Mar 90 05:17:07 GMT
From: [e-mail]  (Barry Lowell Brumitt)
Subject: Re: Shuttle escape systems

If you are interested in how the parachute aspect of the shuttle escape
system works, I recommend finding an isuue of Parachutist magazine for
March '90.  The lead article is on this subject! Has some interesting
pictures of people testing the system on a Mock-up build into a c-141.

This is published by the USPA (US Parachuting Assoc.) and they can be
reached at1-703-=836-3495. Back issues are $2.

Std. Disclaimer: I am not affiliated financially with the USPA in any
way, though I will be a member shortly.

Barry Brumitt
"Who is John Galt?"
[e-mail]

------------------------------

Date: 17 Mar 90 22:24:59 GMT
From: sam.cs.cmu.edu![e-mail]  (Vincent Cate)
Subject: Re: Coilgun on a 747 - supplies to orbit at $20/lb?


Dmitry Gokhman:
>.... has anyone given much thought about a power source for the sucker?

I looked up "Generator Electric" in the yellow pages and called a couple
places that had the large diesel generators.  A few that I found are:
   150 kw   6,000 lbs    $42,000
   150 kw   8,000 lbs  including 260 gal gas and 1,000 lbs noise shielding)
  1000 kw  20,000 lbs
  1000 kw  22,000 lbs  eats gas at 73 gal/hour, no gas in weight
  1000 kw  32,000 lbs
  1500 kw              in phone book but they were closed

These generators were in no way designed for lightness.  In general
the more valuable one weigh more.  They seem to be using engines that
are about 5 to 20 lbs/hp.  There are engines that are 1 lb/hp (some
even less).  I will make the rash prediction that if designed for lightness
it would be possible to get a 20,000 kw generator that weighed under
50,000 lbs (getting 5x from lighter design and 4x with 2.5x weight increase
due to efficiencies of scale).  To run for a flight it will take something
like 50,000 lbs of gas just for the generator.  It looks like a 747 can carry
around 200,000 lbs of payload (anyone have accurate info on this?).  In this
paper (reference below) they say it takes 1500 MJ to shoot 12.6 kg at
6.5 km/s.  This works out to 416 kw-hours for 12.6 kg.  If we wanted to
do 25 kg projectiles it would take about 1,000 kw-hours each.  With
20,000 kw we could fire about 20 each hour, or 500 kg/hour while flying.
If we fly for 5 hours, thats 2,500 kg per flight.  A plane could make several
flights each day.  Still looks like this could cheaply send lots of supplies
into orbit (if each flight cost $100,000 it would cost about $20/lb to
put stuff into orbit).

Norman C. Kluksdahl:
>Vince Cate
>>If the gun weighs 50,000 lbs the gun will go backwards at 1/1000th the
>>velocity of the 50 lb projectile.  At 1/3 g something moving at
>>15 MPH can stop in a couple seconds and about 20 feet.
>
>Sure.  You want to fly a 747 that has 25 tons shift by 20 feet?
>Does the phrase "death wish" mean anything to you?

COWARD!!!!   :-)   OK, I can do it in about 5 feet such that over
2 seconds it averages being out of place by 2.5 feet.  Pull the gun
back on a spring, let it go forward for 1 second then fire such
that it bounces back into the stretched position of the spring and
latches.  It will be moving forward at about 10 feet/sec when it
fires and backward at about 10 feet/sec right after that.  A 747
is 232 feet long and can have a gross weight up to 871,000 lbs in
flight.  50,000 lbs out of place by an average of 2.5 feet over
only 2 seconds of flight should be no big deal.


Steve Nuchia:
>And a 747 really doesn't fly that high.

True, but it is above most of the air.  I think that at 37,000 feet
there is about 1/4th as much air above you as when you are at sea
level.  Note that the air density has not gone down by 4 because the
temperature is much lower.  I think this will help projectiles alot.

>I do know that the energy storage problem has a solution.  A group
>at Texas A&M (if memory serves) has a railgun power supply "the
>size of a garbage can", using some kind of magnetic/electromechanical
>principle.  The primary generators in the engines or your APUs place
>one limit on the firing rate, but once you charge the secondary
>generator you can pull megawatts out of it for the duration of the
>shot.

WOW!!!!!!!   This sounds really good!!  I would like very much to
hear anything more that anyone has on this!

>A 747 is what, 200 ft long?  Calculate the acceleration required
>to reach orbital velocity from, say, 1200 kts (air speed plus
>rotation, fudged for inclination and wind) in 200 ft.  It won't
>be as kind to your payload as a longer ground based gun would
>be, but you may make it up in reduced heating for some payloads.

True.  It won't be kind at all.  In IEEE Trans on Magnetics
Vol 25 No1 p393 they have 64,000 Gs for 33 meter launch to
6.5 km/s.  This system looks very much like the size of something
you might get on a plane.  The projectile is only 12.6 kg (I
think we could do twice that).

>Interesting idea, but I think you need a high-altitude, hypersonic
>aircraft for your first stage.  A 747 just isn't enough better than
>a ground station to justify the cramped quarters, increased operation
>risk, and weight limits.

A very large, very fast, very high flying plane would be best.
However, I think we might be able to do a good job with just a 747.
You might be right that it is not enough better than ground launch
to justify it but I think that is unclear, so far.


CAN ANYONE ANSWER THE FOLLOWING:

     Not counting energy storage (assuming the device a couple
     paragraphs up) or production, how much does a coilgun that can
     launch a 25 kg object need to weigh?

The answer to this could be the real killer.


Thanks,

  -- Vince

------------------------------

Date: Wed, 21 Mar 90 16:48:04 +0100
From: [e-mail] (Daniel Fischer)
Subject: HST at the limit: roll deconvolution and other topics (long)
Cc: [e-mail]

Thanks to Dan Briggs <14 Mar 90 18:11 GMT> for forcing me to take a deeper
dive into the HST literature. He asks for details about the 'roll deconvolu=
tion' technique that will give the Space Telescope 0.015 arc sec resolution
in the short UV (around 140nm):
> If the optics of the telescope are as stable as we are lead to believe,
> shouldn't the Point Spread Function of the scope be a measurable quantity?...
> Once this is in hand, then deconvolution is downright easy... There must be
> a good reason why HST doesn't plan to do it, "the simple way."
There *is* a good reason, and it can be found e.g. in the paper 'High-resolu=
tion astronomical imaging by roll deconvolution of Space Telescope data' by M.
Mueller & G.Weigelt in the journal ASTRONOMY AND ASTROPHYSICS Vol.175, p.312-
318 (1987):
] Of course, it is also possible to deconvolve only *one* degraded image, but
] in this case the reconstruction has a lower signal-to-noise ratio.
   Here's the explanation why *two* images do a far better job. You have the
true 2D object in the sky O(x,y) and the Point Spread Function (=:PSF) P(x,y)
which you find by looking at any star. The HST delivers, according to the
'incoherent imaging equation', the image I(x,y)=O(x,y)*P(x,y). '*' denotes
convolution, an integral operation and thus very hard to invert. But if you
Fourier-transform the equation ('~' denoting the transformed quantities; I
couldn't tell my terminal to write the ~ above the letters as usual), it
becomes  ~I(u,v)=~O(u,v).~P(u,v); '.' is now a simple multiplication! Now the
inversion ('complex inverse filtering') is straightforward, the Fourier trans=
form of the true object is ~O(u,v)=~I(u,v)/~P(u,v). P and thus also ~P are
presumed to have a long lifetime in the case of the HST, and so ~O and O can
be determined: that's the standard inverse filtering Dan Briggs had in mind.
   BUT: usually ~P will have zeros at many (u,v)-coordinates where you cannot
divide by it, ~O will have gaps, and O will deteriorate. THE SOLUTION: you
take two images of your object but rotate the HST around its optical axis by
>10 degrees in between. Now the zeros of your 2nd PSF (which probably is identi=
cal to the 1st one but rotates with the optical assembly) correspond to other
coordinates of ~I, and you reconstruct this particular coordinate of ~O with
the ~P that doesn't have a zero (or a very small value) there. And if both
~P's are o.k., you take their mean - *that's*all*!
John Roberts <15 Mar 90 14:57 EST> writes in this context:
> I believe the optics are nominally rated at 20 milliarcsec... The CCD pixel
> density, however, limits single photographs to 100mas resolution...
According to the Mueller&Weigelt paper, the rating of the mirror system is 30
to 50mas, and that's why deconvolution is needed to reach the 15mas diffraction
limited resolution in the UV. The 100mas resolution that you read about in
every article about the HST refers to the Wide Field camera (which NASA belie=
ves to yield the most impressively looking pictures); the Planetary Camera
reaches 43mas and thus exploits the optical quality fully by just touching the
PSF (the field of view is smaller, of course). But to deconvolve the PSF, you
need more magnification, and that's why the f/288 mode was later added to the
Faint Object Camera. It sees 6 times sharper still, 'samples' the PSF, but, of
course, its field of view is *very* small (a few arcseconds) - thus each of the
3 cameras has its merits. Computer simulations have shown roll deconvolution to
work with different possible PSF's: convincing samples can be found in the
paper cited above. It works with dim objects (though some artefacts appear if
there are fewer than 100 photons/pixel) and with objects with high dynamic
range like a galaxy. Laboratory simulations are under way, as I hear.
   Dave Ray <14 March 90 1:14 GMT>, Greg Hennessy <4:31 GMT> and Henry Spencer
<17:53 GMT> also raise the question of HST's pointing stability. The HST
OPTICAL TELESCOPE ASSEMBLY HANDBOOK (Version 1.0, May 1989) says on page 35:
] ...some account must be taken of the movement of the optics. The dominant
] jitter is rigid movements of the field of view, caused by movement of the
] spacecraft as a whole. Image distortions caused by vibrations of the primary
] for example will be negligibly small. In fine lock the jitter is expected
] to have an amplitude of 7 mas RMS [that is 0.007 arc sec, not 0.07 or 0.1 as
   some have posted!], for course track, estimates vary but something of order
] 20 mas is expected.
Obviously it is believed that mechanical movements inside the science instru=
ments won't cause significant shaking of the 11.6 metric ton satellite. Does
someone know whether the 7 mas pointing stability could be tested on the
ground? The FINE GUIDANCE SENSOR INSTRUMENT HANDBOOK (Oct.1985) indicates that
in orbit the pointing will *not* be guaranteed by e.g. keeping a star in
crosshairs. On page 9 it says about the Fine Guidance Sensor (FGS):
] In conclusion, the FGS has been designed for guidance purpose (0.007 arcsec
] accurracy), but the primary information used by the Pointing Control System
] is from the Rate Gyros and the FGS information is considered as secondary.
] Competitive astrometry (0.0016 arcsec accuracy) with the FGS has been
] advertised as a complementary function.
By repeatedly measuring the same star they even want to pin down relative
positions to 0.0003 arcsec, measure light deflection by Jupiter's gravity etc.
   Bernd Ebach <15 Mar 90 23:00 GMT> asks about HST's chances to find extra=
solar planets. As was announced at an 'HST PRESS DAY' at ESTEC on 16 Feb.,they
are *slim*: dust on the mirror will cause so much scattered light that
planets close to a star will be washed out even in the best coronographic mode.
Observations of Beta Pictoris (Pictor is a constellation in the deep Southern
sky) will not be made soon: this star is so bright (2nd magnitude, I guess),
that HST's cameras might be damaged if the pointing of the coronograph fails...
   Finally, Greg Moore <17 Mar 90 03:12 GMT> asks:
> ...does anyone know what happened to the second mirror? If it is still
> around, do you think it would be possible to... send it up as a 2nd HST?
This has been proposed indeed: in the journal NATURE Vol.339 of 22 June 1989
p.574 several famous astronomers dream of sending HST's spare mirror right
into geostationary orbit. The only thing they didn't mention was: who will
pay? I haven't heard of this plan ever since...

+- [e-mail] --- Daniel Fischer --- [e-mail] -+
| Max-Planck-Institut f. Radioastronomie, Auf dem Huegel 69, D-5300 Bonn 1,FRG |
+----- Enjoy the Universe - it's the only one you're likely to experience -----+

P.S.: Dan Briggs <14 Mar 90> asks to avoid rivalry between the big interfero=
meters - agreed, but: first you have to *know* all of them and their spec's!
Thanks for his <18 Mar 90>-message clarifying CHARA's work. Optical interfero=
meters are little publicised, and thus mistakes easily happen. That will
surely change when there's one on the dark side of the moon (see NATURE Vol.
344 of 15 March 1990 p.188) ... [ :-) ]

------------------------------

Date: 21 Mar 90 20:43:25 GMT
From: mnetor!utzoo![e-mail]  (Henry Spencer)
Subject: Re: More space station news...

In article <[e-mail]> [e-mail] (Paul Dietz) writes:
>... You still think a NASA space station is a good idea, Henry?

Done right, yes.  They seem to be finding more and more ways to do it
wrong, though.
--
Never recompute what you      |     Henry Spencer at U of Toronto Zoology
can precompute.               | uunet!attcan!utzoo!henry [e-mail]

------------------------------

Date: 21 Mar 90 21:02:40 GMT
From: hp-ses!hpcuhb!hpscdc![e-mail]  (Jack Zeiders)
Subject: Re: SR-71 Record Flight Information

> [e-mail] (Paul Cantrell) writes:

stuff deleted....

>Finally, I personally doubt whether a non-film system such as KH-11 really has
>that kind of resolution. I can easilly believe that a film based system on
>a satellite or an SR-71 might have that resolution. My doubt, however, is not
>based on any expertise in the field, so I wouldn't be too surprised if I'm
>wrong.

more deleted...

    Think about a recon satellite the size of the Hubble space telescope
with adaptive optics feeding hi-resolution sensors. In an adaptive optical
system, thin mirrors are deformed by an array of piezo-electric mechanisms
and monitored by a computer to evaluate the local/average seeing many times
per second to eliminate atmospheric factors by locally bending the optic(s).
This system could feed a wide range of sensors, IR near-far, UV, visible, etc.
A commercially available CCD, Charge Coupled Device, has pixels on the order
of 4 microns, about the same as fine grain black & white film,  what do you
think the gummint has? CCDs can integrate to improve signal to noise ratio.
If cooled this may be many minutes as this rides along in space cooling
should be fairly easy.

    The high relolution radar imaging birds don't care about clouds.

     I am not arguing with the idea there may be a replacement already
on line, just pointing out the capilility of satint may be more than we think.


    I have NO specific knowledge of any military satellite system, just
some basic awareness of some optical systems and readily available hardware.


    Jack Zeiders
    [e-mail]

------------------------------

End of SPACE Digest V11 #176
*******************