I've been thinking about the problem of finding extraterrestrial
gold, and I was wondering if someone wiser than I would pick this
idea apart:
satellites from the Earth's surface. Satellites are used in several
industries, including communications, entertainment, military and
science.
If the satellites could be manufactured in orbit, the cost of lift-
off would be eliminated.
The problem that comes to mind immediately is, of course, resources.
With solar power collectors and automated assemblies, the main issues
it seems to me are materials and maintenance labor.
What other problems/advantages does the group see?
I've been thinking about the problem of finding extraterrestrial
gold, and I was wondering if someone wiser than I would pick this
idea apart:
satellites from the Earth's surface. Satellites are used in several
industries, including communications, entertainment, military and
science.
If the satellites could be manufactured in orbit, the cost of lift-
off would be eliminated.
The problem that comes to mind immediately is, of course, resources.
With solar power collectors and automated assemblies, the main issues
it seems to me are materials and maintenance labor.
What other problems/advantages does the group see?
--- In ssi_list@... "invertedzero"
> I've been thinking about the problem of finding extraterrestrial
> gold, and I was wondering if someone wiser than I would pick this
> idea apart:
>
> Orbital Satellite Manufacture. We currently have to launch all
> satellites from the Earth's surface. Satellites are used in
several
> industries, including communications, entertainment, military and
> science.
>
> If the satellites could be manufactured in orbit, the cost of lift-
> off would be eliminated.
>
> The problem that comes to mind immediately is, of course,
resources.
> With solar power collectors and automated assemblies, the main
issues
> it seems to me are materials and maintenance labor.
>
> What other problems/advantages does the group see?
a blank slate with respect to assets in orbit. So if you want to
build a satellite, where are you going to get the solar cells, the
circuit board, the solder, the resistors, the capacitors, the
radiators, the integrated circuits, the memory chips, etc?
If everything except PV cells must come from Earth, then why even
introduce the manufacturing risk and incur the maufacturing
investment for the solar celss.
The launch cost is saved by building everything in orbit but it is a
false economy.
You may sense from my response that the theme is the same as the
orbital economy "dance" we went through a few weeks ago :)
That's my 2 cents worth.
TangoMan
> If the satellites could be manufactured in orbit,
> the cost of lift-
> off would be eliminated.
>
> The problem that comes to mind immediately is, of
> course, resources.
> With solar power collectors and automated
> assemblies, the main issues
> it seems to me are materials and maintenance labor.
>
> What other problems/advantages does the group see?
>
As someone who has built many satellites, I do not
think a satellite could be built without humans.
satellite goes through during construction:
1) Requirements
2) Design
3) Manufacturing
4) Assembly, Integration and Test (AIT)
5) Launch
6) Operations
First off, the requirements and design tend to be
unique and change over time. There is a movement to
attempt to modularize satellite manufacturing so that
you can plug various payloads into a standard bus. At
this point, this is still pretty far off and I am not
at all confident this will happen. So it is going to
be very hard to build a manufacturing facility in
which the requirements and design of a satellite are
going to be consistent enough to build an autonomous
factory to put out satellites.
Now manufacturing and AIT are essentially what you are
trying to move out to space. The real question is can
you keep the recurring costs of these operations done
in space vs on ground to be lower than the cost of
launch. It is this point where I do not think it is
possible due mainly to reasons I mentioned stemming
from r&d phase.
Satellites are just not a mature enough technology at
this point to be built and tested without humans being
there. I think that once we have an established human
colony in space with an established industrial
capability, then yes it makes sense. But a satellite
is just to complicated to be automated by today's
standards. Currently it takes a minimum of 2 years to
build a cookie-cutter satellite (by cookie cutter I
mean one utilizing an already proven design with
little changes). For a more advanced satellite, start
throwing on the years. Even something as small as a
globalstar/iridium satellite took 5 years to develop.
And those were obsolete before they even launched!
Ryan Z
Ryan Z
> Besides, the actual 'work" of assembling satellites
> and the jobs they create is a valuable part of a
> high tech economy.,,, and many times is one of the
> secondary excuses for specding the money as well....
> why do you think that the components for these
> things come from around different political
> precincts...:-)
politics, in the commercial world I think this is less
so. A satellite like Sirius or XM could care less
about politics and just wants to insure that the
satellite gets built to perform the job at hand.
Building and assembling a satellite is damn
complicated and really pushes the limits of our
technology. That is why you see that an aerospace
company employing a large number of phds and masters.
That being said, each government *does* want to insure
that it has people who know how to make these
complicated objects. Thats why almost every nation
attempts to have some in-house satellite manufacturing
program. This is done because national security is
linked to the ability to build and operate space
assets.
Ryan Z
> Building and assembling a satellite is damn
> complicated and really pushes the limits of our
> technology. That is why you see that an aerospace
> company employing a large number of phds and masters.
testing phase and even the design phase has to "prevent failure at
all costs?"
If launch failure and operational failure were not "bankrupting
events" without insurance, do you think the satellites could be
designed and built cheaper, even on an assembly line, so that a
number are launched (because it's cheap) and once in orbit they are
tested for surviving the launch process, then those that made it are
deployed either into operation or parking orbits to be called on
when the lower standards catch up with the satellites after a few
years and it fails.
>
> That being said, each government *does* want to insure
> that it has people who know how to make these
> complicated objects.
Let's just look at communications satellites because each satellite,
I'm sure, is complicated in its own way. Aren't communications
satellites only as complicated as telephone or internet switching
devices or are they is there something more to it? Other than the
receiving and transmitting components.
TangoMan
> Currently it takes a minimum of 2 years to
> build a cookie-cutter satellite (by cookie cutter I
> mean one utilizing an already proven design with
> little changes). For a more advanced satellite, start
> throwing on the years. Even something as small as a
> globalstar/iridium satellite took 5 years to develop.
> And those were obsolete before they even launched!
>
> Ryan Z
satellite and every component has to be radiation hardened, vacuum
chamber tested, and run through a series of simulations because
failure is so expensive once in orbit?
What chews up that time?
TangoMan
totally agreed, after all I did say" the actual 'work" of assembling satellites
> Besides, the actual 'work" of assembling satellites
> and the jobs they create is a valuable part of a
> high tech economy.,,, and many times is one of the
> secondary excuses for specding the money as well....
> why do you think that the components for these
> things come from around different political
> precincts...:-)
politics, in the commercial world I think this is less
so. A satellite like Sirius or XM could care less
about politics and just wants to insure that the
satellite gets built to perform the job at hand.
Building and assembling a satellite is damn
complicated and really pushes the limits of our
technology. That is why you see that an aerospace
company employing a large number of phds and masters.
That being said, each government *does* want to insure
that it has people who know how to make these
complicated objects. Thats why almost every nation
attempts to have some in-house satellite manufacturing
program. This is done because national security is
linked to the ability to build and operate space
assets.
Ryan Z
point is still whether the manufacturing could be done "in-situ" in
Space....at a reduced launch cost.
>
> What is the verdict on that idea?
thing! How do you make one? What supplies are needed? How much
equipment is needed to just make the plain old vanilla circuit board?
Where's that going to come from? How much will it weigh?
My 2 cents = won't happen.
TangoMan
>
> Is this so because failure is such an expensive
> option that the
> testing phase and even the design phase has to
> "prevent failure at
> all costs?"
>
> If launch failure and operational failure were not
> "bankrupting
> events" without insurance, do you think the
> satellites could be
> designed and built cheaper, even on an assembly
> line, so that a
> number are launched (because it's cheap) and once in
> orbit they are
> tested for surviving the launch process, then those
> that made it are
> deployed either into operation or parking orbits to
> be called on
> when the lower standards catch up with the
> satellites after a few
> years and it fails.
>
Part of it is testing against failure, which is why
the AIT phase is generally 8 months. Part of it is
that this is a cutting edge technology and just hasn't
been done yet. A good example of this is many recent
comm sats are looking at moving to Ka-band (form more
traditional C and Ku band) due to the ability that you
can fit more information at this wavelength. You can
also use this wavelength in terms of spot beam
coverage so as to concentrate signals to particular
geographies (very import for direct broadcasting
services).
have to be installed and these can break easily. It
is also very hard (if not impossible) to do
repeatability testing on these. Now in order to work
with this band, completely new equipment had to be
manufactured and new skills had to be learned. I will
not even pretend to know all the complexities of this
other than to say we had to have quite a few phds
working on getting this to work in my company alone.
Thankfully, NASA laid down the ground work for us to
follow with the ACTS satellite in the early 90s. This
demonstration satellite allowed customers in the
commercial sector to have the confidence to order such
a satellite. I bring this up for two reasons: 1) to
show that nasa has a great role it fulfills by proving
specific technologies can work, 2) putting up new
hardware is tough thing, customers shy away from
unproven technologies. Look at what happened with the
Boeing 702 solar array design. They tried to be
innovative but instead produced a spectacular failuer
that destroyed their good name in the satellite
manufacturing world and ultimately costs them
contracts.
>
> Let's just look at communications satellites because
> each satellite,
> I'm sure, is complicated in its own way. Aren't
> communications
> satellites only as complicated as telephone or
> internet switching
> devices or are they is there something more to it?
> Other than the
> receiving and transmitting components.
>
part of the payload of a commsat is that, and that is
the easy part. The hard part is as I mentioned
before, the communication to the ground. You need
some fairly powerful antannaes operating over a very
broad range of frequencies. The higher the frequency,
the higher the power demand. So you have all the very
complicated rf stuff which goes way over my head
compounded with the all the bus related functions
which provide a stable platform for the payload to
rest on. A typical commsat has a minimum of about
9000 parts, not exactly a trivial thing.
Ryan Z
> Does it take so long because you're trying to save
> weight in the
> satellite and every component has to be radiation
> hardened, vacuum
> chamber tested, and run through a series of
> simulations because
> failure is so expensive once in orbit?
>
> What chews up that time?
build new and often time unique equiptment to support
the sateelite. This includes battery simulators,
altitude control system stimulators, solar array sims,
and rf racks. On top of that, you have a rigourous
design process as each piece of equipment must be put
into some kind of cad program and then they have the
joy of putting together all 9000 pieces into a a big
square which needs to fit in the cone of a rocket
(typically 4 or 5 meters in diameter for the bigger
ones).
Prolly the biggest time spent is in the design phase.
That runs anywhere from 1 year to 3 years. The actual
AIT portion is much shorter, often less than a year
for commsats.
Now when you get into imaging satellites, all things
change. An imager is VERY complex, as not only do you
have to deal with assembly extremely precise optics
and making sure they don't get contaminated, you also
have to properly align the instruments and its
temperature control system (some IR imagers need to be
around 70 Kelvin or less to operate!). For these
types, a simple case of a weather satellite takes a
minimum of 5 years to make, though things like GOES
actually fall around 12 years. In general, more
instruments the more work.
Ryan
> But the point is still whether the manufacturing
> could be done "in-situ" in Space....at a reduced
> launch cost.
>
> What is the verdict on that idea?
costs. Assuming you can get the cost of in-situ
resource manufacturing up and a plant put together
which can assamble all the made parts, is that going
to be cheaper or even more efficient than an earth
factory who has to launch? What happens when you need
to upgrade the facility to handle a new technology?
So I would say that the verdict is that until we have
a fairly sustained human presence in space (such as an
island 1 community), we will not be manufacturing
satellites in-situ in space.
Ruam
gold, and I was wondering if someone wiser than I would pick this
idea apart:
satellites from the Earth's surface. Satellites are used in several
industries, including communications, entertainment, military and
science.
If the satellites could be manufactured in orbit, the cost of lift-
off would be eliminated.
The problem that comes to mind immediately is, of course, resources.
With solar power collectors and automated assemblies, the main issues
it seems to me are materials and maintenance labor.
What other problems/advantages does the group see?
> I've been thinking about the problem of finding extraterrestrial
> gold, and I was wondering if someone wiser than I would pick this
> idea apart:
>
> Orbital Satellite Manufacture. We currently have to launch all
> satellites from the Earth's surface. Satellites are used in
several
> industries, including communications, entertainment, military and
> science.
>
> If the satellites could be manufactured in orbit, the cost of lift-
> off would be eliminated.
>
> The problem that comes to mind immediately is, of course,
resources.
> With solar power collectors and automated assemblies, the main
issues
> it seems to me are materials and maintenance labor.
>
> What other problems/advantages does the group see?
a blank slate with respect to assets in orbit. So if you want to
build a satellite, where are you going to get the solar cells, the
circuit board, the solder, the resistors, the capacitors, the
radiators, the integrated circuits, the memory chips, etc?
If everything except PV cells must come from Earth, then why even
introduce the manufacturing risk and incur the maufacturing
investment for the solar celss.
The launch cost is saved by building everything in orbit but it is a
false economy.
You may sense from my response that the theme is the same as the
orbital economy "dance" we went through a few weeks ago :)
That's my 2 cents worth.
TangoMan