MSNBC (JimO) - Hubble debate -- a lot of sound and fury

On 3/31/04 5:38 PM, in article , "Jorge
R. Frank" wrote:

h (Rand Simberg) wrote in
:

On Wed, 31 Mar 2004 21:05:45 +0100, in a place far, far away, "John"
made the phosphor on my monitor glow in
such a way as to indicate that:

The change is 3kmps, which is about a third of the
acceleration that the shuttle needs to get into orbit in the first place.

What's your basis for that number? How much of a plane change were
you assuming?

His number is pretty close to mine (3.05 km/s). I assumed a noncoplanar
Hohmann transfer from HST (310 n.mi., 28.45 deg) to ISS (210 n.mi., 51.6
deg), for a total plane change of 23.15 deg. Delta-V is minimized by
combining most of the plane change (21.76 deg) with the first Hohmann burn.
Of course, both burns must occur at the moment when the RAANs are equal,
otherwise the plane change will be much larger than 23.15 deg.

Someone (Ed Kyle?) figured out a while back that such a delta-V would
require two Centaurs, for a spacecraft HST's size.
Jorge, what do you think about John's idea in an earlier post about using an
ion propulsion system to push Hubble into an ISS compatible orbit? Would
that be a viable option or would the development/ transition fuel costs,
etc. be too prohibitive?

h (Rand Simberg) wrote in
:

I actually ran some numbers on this a few months ago. While you only
get pure plane change at the node, there are several minutes on each
side of it during which it's still pretty efficient to do it (i.e.,
the ratio of plane change to node change still remains above ninety
percent).

Thanks for the details, Rand!


--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.

On 01 Apr 2004 23:54:25 GMT, in a place far, far away, "Jorge R.
Frank" made the phosphor on my monitor glow in
such a way as to indicate that:

(Rand Simberg) wrote in
:

I actually ran some numbers on this a few months ago. While you only
get pure plane change at the node, there are several minutes on each
side of it during which it's still pretty efficient to do it (i.e.,
the ratio of plane change to node change still remains above ninety
percent).

Thanks for the details, Rand!

I mistyped above--I meant the ratio of inclination change to node
change. Your amount of plane change is constant at any point in the
orbit, but it will be a sum of the node change and the inclination
change.

If you really want details, I can send the spreadsheet. It was
actually a pretty interesting problem.

It turns out that if you do a cross-track burn exactly at the top or
bottom, it will always increase inclination--there's no way to
decrease it. In retrospect, this makes sense, the same way that you
can only increase your inclination above your launch latitude,
regardless of whether you launch north or south. But get a little
distance away from that point, and then you start trading off node
change for inclination change (up or down), with minimal inclination
change at the top, and maximum at the nodes.

Chuck wrote in
:

Jorge, what do you think about John's idea in an earlier post about
using an ion propulsion system to push Hubble into an ISS compatible
orbit? Would that be a viable option or would the development/
transition fuel costs, etc. be too prohibitive?

As far as the ion propulsion goes, the basic concepts were proven with DS1,
but that was a smaller beast than HST. To transfer HST to ISS inclination
in a reasonable time (on the order of months rather than years), you'd need
a larger engine than has been successfully developed so far. Some
development cost but no show-stoppers that I can see.

Likewise, using DS1-class ion engines with HST might work but would require
longer transfer times (on the order of years), so you'd need an engine that
can work reliably over a very long period, or multiply-redundant engines.
Again, no showstoppers, just development work and tradeoffs to be made.

In either case, the fuel costs are negligible compared to development
costs. Take all of the above with a grain of salt - I'm a rendezvous guy,
not a low-thrust propulsion guy. I haven't actually run any numbers on how
long these transfers would take.

The biggest development headache may well be automated rendezvous and
capture of HST. All the automated systems developed to date require their
targets to be cooperative to some degree (i.e. equipped with navaids for
the chaser vehicle). The Russian Kurs system requires a rather elaborate
set of RF antennas/avionics on the target vehicle. The systems under
development for ATV and HTV require the target vehicle to have a GPS
receiver and an array of laser retroreflectors.

Trouble is, HST has *none* of those things - it's a completely passive
target, save the (visual) aids on its RMS grapple fixtures and berthing
mechanism. Attaching the navaids to HST becomes a chicken-and-egg problem -
any automated system that could attach the navaids to HST needs the navaids
to be already there in order to get there in the first place.

The *only* vehicle in existence which can rendezvous with and capture such
a completely passive target is the very vehicle which O'Keefe has ground-
ruled out of the solution: the space shuttle. It was designed to be able to
rendezvous with passive targets - hell, it can even rendezvous with debris,
if the situation required it. Its rendezvous radar is capable of skin-
tracking a target with no transponder, and the crew is equipped with
police-style handheld lidars that are likewise capable of skin-tracking an
inert target.

That is not to say that it's impossible to develop an automated system to
rendezvous and capture a non-cooperative, inert, and (by 2007-08) possibly
slowly tumbling target. But the challenge of developing such a system will,
IMO, be greater than that of getting the ion propulsion to work.

--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.

"Jorge R. Frank" wrote:
That is not to say that it's impossible to develop an automated system to
rendezvous and capture a non-cooperative, inert, and (by 2007-08) possibly
slowly tumbling target. But the challenge of developing such a system will,
IMO, be greater than that of getting the ion propulsion to work.

But the simplest and perhaps cheapest solution remains flying Shuttle to
Hubble twice. Once to outyfit it with the hardware that is ready to give it
new set of eyes, and a final one to bring hubble down.

All done with existing technology that was designed exactly just for that, and
only requires that some procedural books be updated to include the examination
and repairs that CAIB has told NASA it had to develop anyways.

And until there is real and realistic functing for some mythical CEV, everyone
should assume shuttle continues to fly beyond 2010. I think 2015 is more realistic.

"Jorge R. Frank" wrote ...
As far as the ion propulsion goes, the basic concepts were proven with
DS1,
but that was a smaller beast than HST. To transfer HST to ISS inclination
in a reasonable time (on the order of months rather than years), you'd
need
a larger engine than has been successfully developed so far. Some
development cost but no show-stoppers that I can see.

The concept as I imagined it was several DS1 class bolted together, or even
an array of lesser thrusters. The idea is based upon an article that gave a
transition time of about 18 months.

The biggest development headache may well be automated rendezvous and
capture of HST. All the automated systems developed to date require their
targets to be cooperative to some degree (i.e. equipped with navaids for
the chaser vehicle). The Russian Kurs system requires a rather elaborate
set of RF antennas/avionics on the target vehicle. The systems under
development for ATV and HTV require the target vehicle to have a GPS
receiver and an array of laser retroreflectors.

It's as much of a problem as you make it. If you can park the hubble within
a few hundred meters from the ISS you can effectively attach a line and real
it in. Even if this is impractical, get it close enough in terms of
deltaV and the shuttle could grab it and then rondavouz with the ISS in
'perfect' safety.

The easiest way to dock the OTV to the ISS would be to give it a false
docking hatch. That way it can be attached to any number of hard-points.

Trouble is, HST has *none* of those things - it's a completely passive
target, save the (visual) aids on its RMS grapple fixtures and berthing
mechanism. Attaching the navaids to HST becomes a chicken-and-egg
problem -
any automated system that could attach the navaids to HST needs the
navaids
to be already there in order to get there in the first place.

Call me simplistic, but what's wrong with putting the OTV near the hubble,
pointing the two at each other manually and then pressing the 'go forward'
button? Add a few dozen 100gram cameras to make sure everything's perfectly
lined up and hen press the 'dock' button.

Besides, the base of the hubble is one big circle with the docking mechanism
in the middle. You can't really get a better bulls eye. By measuring the
apparent size of the circle 'on screen' you can tell how far away it is,
it's closing speed, how far off axis it is and in what direction that off
axis is. Measuring the shape of the circle tells you if the hubble is going
off axis by measuring circle/eclipse

That is not to say that it's impossible to develop an automated system to
rendezvous and capture a non-cooperative, inert, and (by 2007-08) possibly
slowly tumbling target. But the challenge of developing such a system
will,
IMO, be greater than that of getting the ion propulsion to work.

Whatever docking systems required for hubble needs to be developed anyway
for the derbit module. Unless of course NASA wants to harpoon it and 'tow'
it out of orbit. Which would actually kind of work...

John

In sci.space.policy Jorge R. Frank wrote:
Chuck wrote in
:

Jorge, what do you think about John's idea in an earlier post about
using an ion propulsion system to push Hubble into an ISS compatible
orbit? Would that be a viable option or would the development/
transition fuel costs, etc. be too prohibitive?

As far as the ion propulsion goes, the basic concepts were proven with DS1,
but that was a smaller beast than HST. To transfer HST to ISS inclination
in a reasonable time (on the order of months rather than years), you'd need
a larger engine than has been successfully developed so far. Some
development cost but no show-stoppers that I can see.

There is also SMART-1, happily on its way towards Moon. I think the PPS-1350 is
rated to work for 7000 hours of propulsion, so long transfers are ok.


Likewise, using DS1-class ion engines with HST might work but would require
longer transfer times (on the order of years), so you'd need an engine that
can work reliably over a very long period, or multiply-redundant engines.
Again, no showstoppers, just development work and tradeoffs to be made.

In either case, the fuel costs are negligible compared to development
costs. Take all of the above with a grain of salt - I'm a rendezvous guy,
not a low-thrust propulsion guy. I haven't actually run any numbers on how
long these transfers would take.

Just use off the shelf ion engines. Sure, having the PPS-5000 or similar around
that are designed for station keeping 10 ton satellites would be nice, but I
don't think the present options are too meager either. So you should be able to
eliminate teh need to spend large amounts on the propulsion part.

[snip]


That is not to say that it's impossible to develop an automated system to
rendezvous and capture a non-cooperative, inert, and (by 2007-08) possibly
slowly tumbling target. But the challenge of developing such a system will,
IMO, be greater than that of getting the ion propulsion to work.


Surely some of the ASAT technology can be reused in such a mission? Such would
by defaulkt have needed to be able to rendezvous with a non-co-operative target.

--
Sander

+++ Out of cheese error +++

John Doe wrote in :

"Jorge R. Frank" wrote:
That is not to say that it's impossible to develop an automated
system to rendezvous and capture a non-cooperative, inert, and (by
2007-08) possibly slowly tumbling target. But the challenge of
developing such a system will, IMO, be greater than that of getting
the ion propulsion to work.

But the simplest and perhaps cheapest solution remains flying Shuttle
to Hubble twice. Once to outyfit it with the hardware that is ready to
give it new set of eyes, and a final one to bring hubble down.

Yes, agreed. But we're not the ones who get to make that decision.

--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.

Sander Vesik wrote in
:

There is also SMART-1, happily on its way towards Moon. I think the
PPS-1350 is rated to work for 7000 hours of propulsion, so long
transfers are ok.

Thanks, I forgot about those.

Surely some of the ASAT technology can be reused in such a mission?
Such would by defaulkt have needed to be able to rendezvous with a
non-co-operative target.

Not really. "Rendezvous" and "intercept" are two separate problems - with
intercept, the objective is to hit the target hard and destroy it. In
rendezvous, the objective is to capture the target gently, and to do that
requires that the chaser be able to fly around the target to locate its
berthing mechanism, and be able to approach and grapple it gently. An
interceptor doesn't really care about the details of the target and doesn't
have the capability to gently capture it.
--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.

"John" wrote in
news:uzbbc.15$is5.3@newsfe1-win:

"Jorge R. Frank" wrote ...

The biggest development headache may well be automated rendezvous and
capture of HST. All the automated systems developed to date require
their targets to be cooperative to some degree (i.e. equipped with
navaids for the chaser vehicle). The Russian Kurs system requires a
rather elaborate set of RF antennas/avionics on the target vehicle.
The systems under development for ATV and HTV require the target
vehicle to have a GPS receiver and an array of laser retroreflectors.

It's as much of a problem as you make it. If you can park the hubble
within a few hundred meters from the ISS you can effectively attach a
line and real it in. Even if this is impractical,

It is. How do you plan to attach the line, for one? An EVA crewmember can't
go out that far from ISS; SAFER doesn't carry enough gas and is zero-fault-
tolerant. It's really just an emergency rescue device, not like the MMU.

get it close
enough in terms of deltaV and the shuttle could grab it and then
rondavouz with the ISS in 'perfect' safety.

The easiest way to dock the OTV to the ISS would be to give it a false
docking hatch. That way it can be attached to any number of
hard-points.

The ISS end isn't the hard one. It's the initial capture of HST in its
original orbit.

Trouble is, HST has *none* of those things - it's a completely
passive target, save the (visual) aids on its RMS grapple fixtures
and berthing mechanism. Attaching the navaids to HST becomes a
chicken-and-egg
problem -
any automated system that could attach the navaids to HST needs the
navaids
to be already there in order to get there in the first place.

Call me simplistic, but what's wrong with putting the OTV near the
hubble, pointing the two at each other manually and then pressing the
'go forward' button? Add a few dozen 100gram cameras to make sure
everything's perfectly lined up and hen press the 'dock' button.

The OTV won't keep going perfectly forward due to orbital mechanics, which
causes some decidedly non-intuitive effects in LEO. The OTV must have the
capability to constantly correct its lateral alignment as it approaches.

Besides, the base of the hubble is one big circle with the docking
mechanism in the middle. You can't really get a better bulls eye.

Careful with the terminology. It's a berthing target, not a docking target.
The distinction is important. In HST berthing, the telescope has already
been captured by the SRMS, and is moved very slowly over the FSS until it
lines up with the berthing camera. Then the capture latches are closed to
berth the telescope to the FSS. In other words, it was designed to work
with an already-captive target. The tolerances are very tight, tighter than
anything a docking system has achieved to date.

By
measuring the apparent size of the circle 'on screen' you can tell how
far away it is, it's closing speed, how far off axis it is and in what
direction that off axis is. Measuring the shape of the circle tells
you if the hubble is going off axis by measuring circle/eclipse

The Canadians have been working on such a system (SVS) for over a decade,
and they can tell you it's not nearly as simple as you make it out to be.
SVS requires dedicated visual targets on the target vehicle, but still gets
confused whenever lighting conditions change the picture on the camera. HST
would be a particularly difficult target due to its high reflectivity. It
is not an impossible problem to solve, but a lot of smart people have been
working on it for a long time and it still hasn't been solved.

Whatever docking systems required for hubble needs to be developed
anyway for the derbit module. Unless of course NASA wants to harpoon
it and 'tow' it out of orbit. Which would actually kind of work...

Right. For a deorbit system, you don't care about protecting the target.
Better than a harpoon would be a set of spikes on the nose of the OTV. It
is easier to ram than to shoot.


--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.