Space-Based R&D as a Catalyst for Space Development ( Robber Bar

Impressive response. It must have taken you a lot of work to rummage
through all of our past postings and craft this response. I hope it
didn't take too much of your time, and I must say, that the quality
of your writing voice is much different with this response. It must
be the ability to see the whole picture rather than responding point
by point as the two of us have been doing.

It was a pleasure to read.

> > 1.) You point out that SpaceHab and MULE aren't the same thing
in
> > particular, but only share general features, much like the 747
and
> > the F-16 share similar features. You're counting on me to
> understand
> > the differences between the two planes, and I do, but you make
no
> > effort to detail the differences between SpaceHab and MULE.
>
> That's because I already have numerous times and in numerous
places.
> Here's some of the text from our previous exchange on this subject:
>
> JACK: But people are already doing microgravity research. Granted,
> the market isn't that well developed but there's no denying that a
> market exists. It's a tough case to make, but not impossible.
Surely
> these companies currently paying for microgravity research would
> appreciate a cheaper more effective way in which this research can
be
> conducted?
>
> TANGO: Yes, SpaceHab is such a company and their performace isn't
> burning up the stockmarket.
>
> JACK: No, but their storage racks ride on the ISS - I spoke about
> that problem above. They have to pay NASA to plug one of their
racks
> into the ISS meaning whoever does research in one of those racks
has
> to pay NASA AND SpaceHab, right?
>
> In your followup, you did not address my specific response above.
> Instead you just mentioned the same issue at a different point in
the
> exchange:

I didn't address this question because we both have to edit what we
respond to in order to avoid ever-longer posts, but now that I see
that you were looking for a response, I'm happy to comply.

Yes, they would have to pay SpaceHab and NASA.

>
> JACK: Again, the cost of doing research on the MULE is less for
the
> individual company because that cost is shared among many
companies.
>
> TANGO: But that's how SpaceHab works. Why is MULE cheaper? You're
> just assuming it is, show me why MULE can deliver a service
cheaper
> than SpaceHab, and I'll come on side with you.
>
> JACK: Let's say 10 companies decided to develop and deploy a
> commercial space station. All ten would have access, all ten would
> SHARE the cost of developing, deploying and maintaing that space
> station. Now, let's say you own one of those companies - Tango,
Inc.
> Now, what would cost more, Tango, Inc. to pay to have their own
> individual space station or for them to share the cost with nine
> other companies?

There is a latin term used extensively in economic theory, ceterus
parabus, all things being equal. While a useful concept in trying to
grasp economic principles, it isn't too useful in real life economic
analysis, due diligence, or even policy work, because there are
numerous extenuating circumstances. In fact I can't personally
recall even one instance where I could compare competing schemes
where everything was equal.

So, considering this preamble, when you write that categorically
your MULE can perform a service more efficiently, I believe that
you're ignoring real-life constraints and competitor strengths, and
stressing the academic theory or the wishful thinking of a perfect
scenario inherent with a pet project.

One point for you to consider; if MULE, SpaceHab and ISS were
starting as blueprint ideas and competing equally then your
assumptions about cost allocation would hold true. But as part of
your due diligence you need to investigate your competition, and
with respect to SpaceHab, perhaps they've already amortized their
pod. Now that you enter the market, they may not have to charge for
it any longer. Now you've got an expense that they don't have to
cover. As for the man rating of your competition, I don't understand
why a MULE customer would have to design a module that performs the
research robotically and just receives power, heat dissipation, and
communication from the backpane, couldn't take that same design
expertise and design the same module, with its integrated robotics,
place it into ISS and just receive power, heat dissipation, and
communication from the ISS?

The question for the potential pod renter, is why go with MULE, Inc.
when SpaceHab may do it for less in order to starve your business,
and NASA may love to add a robotic R&D module as an additional
component to the ISS thus buying another constituency to support the
ISS, so they too may give you a cut rate?

> Ahh, but the "body of knowledge" you are referring to now is not
the
> same "body of knowledge" you originally implied. Let me cite you:
>
> TANGO: Just to be clear, I have no doubt that when whole bodies of
> science, technology and engineering develop that completely
> incorporate temperature controls in space, cheap energy, variable
> gravity, and variable vacuum AND when scientists and engineers,
who
> are trained to operate within these new environments, to think
about
> them daily, and know them as intimately as their terrestrial
> associates know gravity, atmospheric presence, atmospheric
pressure,
> then and only then, when thousands of people are working in
concert,
> publishing in journals dedictated to orbital technology, will
unique
> products that cannot be duplicated on Earth be devised. That's a
> hell of a sentence. Basically, when there is an orbital society
with
> a critical mass of scientific talent.
>
> JACK: You do not need to have that scientific talent in orbit to
> evaluate the merits and potential returns from space-based R&D. In
> this day and age, with networks and rapid communication, computers
> and robotics, in-orbit presence is superfluous. This army of
> engineers and scientists who must consider the ramifications of
> gravity on a daily basis can do so from Earth. They don't have to
be
> in space.
>
> TANGO: Seeing the army of talent that is used in product R&D and
even
> in basic engineering of structures leads me to conclude that a
series
> of garage startups are not going to be pumping out revolutionary
> products. I look about and see engineers relying on Physical
Handbook
> type references all the time. No such body of knowledge exists for
> the environment of space and the practice of engineering in space.
> Does it? I'd love to read it if it does.
>
> Your contention that the body of knowledge concerning SPS is
> not "lacking" collapses under the weight of your own words: "No
such
> body of knowledge exists for the . . . practice of engineering in
> space." SPS is a considerable engineering project - and you just
> admitted we lack the body of knowledge to practice such
engineering.
> I would also hold that your assertion that constructing the
> infrastructure "has nothing to do with the 'body of knowledge'" is
> inconsistent with your statement as well.

No, No, No . . . you're taking my concerns about a body of knowledge
with respect to innovation and creation of new science, new
engineering prinicples, and new product development in a unique
environment (SPACE) and applying it to a SPS structure that is not
subject to centripetal forces and use existing techonology, PV cells
or sterling engines, klystron tubes, and can be constructed with
1970s technology and engineering knowledge.

For a concise primer on SPS see:

http://www.permanent.com/p-sps-ps.htm

For information on Japan's plan to launch a SPS testbed system, see:

http://www.space.com/businesstechnology/technology/nasda_solar_sats_0
11029.html

Just a quick look at the NASA Glenn site, reveals:

http://powerweb.grc.nasa.gov/pvsee/facilities/
http://www.grc.nasa.gov/WWW/tmsb/concentrators.html
http://www.grc.nasa.gov/WWW/tmsb/stirling.html

> You further contend that what body of knowledge does exist for SPS
> exceeds that of microgravity research (or more generally, space-
based
> R&D). I dispute that. NASA has a Microgravity Research Program, a
> Microgravity Science Division, Zero Gravity Research Facility, and
> The Marshall Space Flight Center's Space Product Development
Office
> (a commercial space center). NASA has no specific division,
facility
> or office dicated to SPS research that I can discern. The only
thing
> I could find was the Space Solar Power Technology Advanced
Research
> and Technology program.

I'll withdraw my overly broad statement about microgravity research,
because you are correct that there is indeed a lot of research
activity surrounding microgravity, and because almost every activity
in orbit can be classified as a subset of microgravity activity.

What I meant to write, and if you'll look at the context of my
criticism, you see that it would have been more appropriate, is the
body of knowledge concerning commerical research and product
development in orbit is easily surpassed by that of SPS. And that
has direct relevence to MULE, because as you pointed out all of
those NASA and other government centers are dealing with basic
science and it's unlikely that you'll entice a NASA Center to rent a
MULE slot when they have their own infrastructure already in place.
Nevertheless, I don't believe that this comparison adds anything to
our debate. It's like a my dad can beat up your dad argument.

One last point: SPS is a specific application, with very limited and
defined technical requirements and microgravity commerical R&D is
all encompassing, so the quantity of research that SPS has thus far
amassed will one day be surpassed by the multitude of distinct
research avenues pursued under the rubric of commercial R&D.

If you're really interested in SPS research, try to drag Arthur
Smith, Mike Combs, Charles Radley (where are you?) and some others
who frequent this list, into the discusssion. They have a wealth of
information at their disposal, and have provided interesting reading
for me.

> > Look at the potential of mag-lev transit, hybrid cars, floating
> > airports, desalination plants, cybernetic implants, human
genetic
> > engineering, etc.
> >
> > They all have future potential, but I guess because they haven't
> > been implemented yet, then that invalidates their potential.
>
> No, but the fact that they have not been implemented does likely
> invalidate any claim to cost-effectiveness made on their behalf.
>
> > Or one closer to your tastes; MULE has potential but it hasn't
yet
> > been implemented, thus it probably doesn't have potential. You
> don't
> > buy this line of reasoning, do you?
>
> No, for very good reason - a reason you can't cite in other
examples.
> The technologies behind the MULE concept have only recently
matured.
> Computers, robotics, communication - these technologies have now
> ripened to the point where we can maintain an unmanned,
> automated/telecontrolled laboratory in Earth-orbit for the
purposes
> of conducting space-related industrial/commercial applied research
or
> government/university-sponsored basic research. If the technology
> hasn't ripened to this point, then the MULE concept is untenable
> right now.

So, welcome to the club. Your MULE concept now has mature
technology, just like SPS, satellite refueling, and a number of
other concepts and all of these concepts haven't yet been
implemented. On a further note, I'm glad that you've rejected the
logic of the statement from which this paragraph derived.

> JACK: No, of course not. I am saying that microgravity research
into
> materials manufacturing, for example, would likely demonstrate the
> industrial and economic viability of the space environment, thus
> encouraging additional applications.
>
> Later that same email:
>
> TANGO: . . . how does launching a payload to the MULE help lay
down
> infrastructure?
>
> JACK: Again, if it can demonstrate the economic viability of
> exploiting the space environment it might encourage additional
> applications. That doesn't make sense to you?
>

> I have maintained the same premise from the beginning: by
> DEMONSTRATING THE ECONOMIC AND INDUSTRIAL VIABILITY OF SPACE-BASED
> R&D, you encourage increased activity and provide SOME incentive
to
> develop the infrastructure. I have never claimed that the MULE was
> anything more than a potential catalyst for doing "more stuff" in
> space.
>
> Now, the last time I said that (just a few minutes ago, in fact)
you
> said, "Well then I've completely misunderstood your whole
argument."
>
> Previously, however, you had a different response:
>
> JACK: Again, if it can demonstrate the economic viability of
> exploiting the space environment it might encourage additional
> applications. That doesn't make sense to you?
>
> TANGO: Yes it does.
>
> Sorry, I thought we moved past this point - I thought we were in
> agreement. Where's the misunderstanding? If you can explain the
> confusion I will attempt to clarify.
>

I'll try to explain the splitting hair point of difference.

I can agree with your statement below because it is sufficiently
general in allowing that one action MIGHT lead to another.
Possibilities are open. This statement can apply to MULE as well as
to other schemes and I think it's perfectly valid.

"If it can demonstrate the economic viability of exploiting the
space environment it might encourage additional applications."

I have more difficulty in accepting the statement below because you
posit a defined consequence and provide that there is indeed SOME
incentive created. I can invalidate your statement by allowing for
the successful operation of MULE, yet it doesn't produce any
commerically meaningful results, thus it doesn't encourage a
blossoming of orbital research & development. MULE operates
successfully by finding tenants so as to earn a profit and it
operates flawlessly, thus proving its basic viability (thus meeting
your goals) but it doesn't cause the results you specify.

"By DEMONSTRATING THE ECONOMIC AND INDUSTRIAL VIABILITY OF SPACE-
BASED R&D, you encourage increased activity and provide SOME
incentive to develop the infrastructure . . "

In the same light as the above statements, consider our dialogue
below. I'm agreeing that MULE has merit, but so do other schemes.
I'm just not sold that MULE is a unique venture or a driver of
future infrastructure development.

JACK: I agree, but we are a long way from that and surely an
unmanned research laboratory in space would help bridge that gap -
not completely, of course, but would be helpful.

TANGO: Sure it would be helpful, but so would a number of other
ventures.

> My point here, Tango, is that I never uncategorically stated that
the
> MULE was less expensive - quite to the contrary.

But you did, and I don't want to search through all of our postings,
but even in your last posting, to quote:

"Well, that's a case I have to make at some time, but honestly I'd
have to go back and reread the report to which I was referring. As
best I recall, their problem with space-based R&D for the
commercial sector stems in large part from the fact that access to
the microgravity environment using current methodologies is
expensive. With a MULE it is less so."

That's not conditional, it is in fact, quite uncategorical.

> JACK: But people are already doing microgravity research. Granted,
> the market isn't that well developed but there's no denying that a
> market exists. It's a tough case to make, but not impossible.
Surely
> these companies currently paying for microgravity research would
> appreciate a cheaper more effective way in which this research can
be
> conducted?

I'm led to conclude that you believe that MULE is the cheaper, more
effective way that you refer to in the above quote.

> > If you think that building a SPS is a daunting task, consider
> > constructing a chip fabrication plant in orbit without any
> > infrastructure.
>
> Ok, now we're going somewhere. Please respond to this question:
>
> Specifically, what infrastructure is required in space for a chip
fab
> plant to function and mass produce chips in orbit for delivery to
> terrestrial markets?
>
> I have an extensible, scalable infrastructural backplane that
> provides the basics: thermal radiation, power, comm, etc.
Everything
> else the plant needs would have to be provided by the plant owner -
> just like on Earth. I mean, when a terrestrial chip fab plant asks
> the power company for power or the telephone company for
> communications capabilities, they don't get chip fab plant
equipment
> as a bonus for signing up, right? That stuff still has to be paid
for
> by the chip fab plant owner(s).

I'd love to answer your question.

I'll start at the beginning of the process as I foresee it.

You'll need some sort of docking bay into which the sand, or other
raw materials will be delivered.

While the sand is being delivered to the crystal growth section,
your docking bay will have to refit the pod to carry back the
finished product because they'll need extensive protection against
the forces of re-entry, and further, the pod must be cleaned of sand
to the standards of a clean room because grains of sand floating
around microprocessers will probably scratch a number of them. Where
do the trays and other refitting components come from? Are they
launched integrated but kept clean? Are they launched seperately?

I'll assume that a primary unique feature of the chip is the
crystalline structure of the substrate that can only be grown in
zero-g and that you won't just be making a melt, with a resulting
glassy product. This means that you can't just melt your sand, or
other raw materials, and create your boule by cooling the sand.

Thus, you need a crytal growth facility. This process needs a medium
in which the atoms will form into a crystal. You have the unique
advantage of containerless processing and thus avoid contamination
issues, but you still need a flux into which you'll have to create a
thermal gradient to facilitate growth.

Another assumption is that you'll pull the crystal, rather than have
it develop free form. This will allow you to control the form of the
resulting boule and thus ease the task of boule manipulation.

It'll be interesting how you would maintain the demarckation between
the flux and the boule absent gravity. But I'm sure that the R&D
process has solved this problem.

Next, you would take the boule and fabricate, or grind, it into a
uniform shape, to facilitate further automation. So, you'll have a
lot of crystal dust that won't fall to the floor. Thus you'll need a
vacuum intermingled around the grinding equipment, and air filters
to catch the crystal dust. Then you'll need to change the filters
and get rid of the dust.

Next, the boule goes through a bath to remove dust. Now the bath
medium must be cleaned and filtered.

Now the boule goes to the cutting facility. Here diamond saws would
cut uniform wafers. Again you have a dust issue and you have to
monitor the state of the diamond cutting surface, the motors driving
the saws, and the control equipment. I'm not sure how you'll handle
mechanical breakdowns in cutting equipment?

Next, using suction, each wafer will be picked up and stacked, but
you can't rely on gravity to keep the wafers in place, so there is
some other mechanism.

Once you have sufficient wafers to make a batch, then you'll send
them for a polishing in order to remove the imperfection of the
cutting process. Then a more thorough cleaning, this one involving
high-purity deionized water and various low-particulate chemicals.
Then the wafers are moved to a high temperature furnace and exposed
to approximately 1000 C and to ultra-pure oxygen in the oxidation
furnace. Under carefully controlled conditions a silicon dioxide
insulator film of uniform thickness will form on the surface of the
wafer.

Now I'm sure that an orbital chip plant will have a process
different than a silicon chip fab, so you'll have to pick and choose
what features you'll likely encounter. For the above, you'll also
need supplies of water, chemicals, and oxygen supplied, and the
waste captured, and either recycled or disposed of. If you chose to
recycle, the process must be so thorough so as to produce a product
of the same purity. Launch new raw materials or invest heavily in
recycling? Your choice.

Next your wafers move to the photolithography module. Here they will
be masked. On Earth, spin coating is used to apply a layer of light
sensitive film. Absent gravity, how will you get a uniform thickness
of the film applied? Then a microaligner aligns the wafer to a glass
mask through which an intense ultraviolet light is projected through
the mask.

Now the wafer moves to a process where the exposed film is
chemically washed away. Again, purity of chemicals is essential, and
again you have the issue of supply or recycle.

Next the wafer is baked so as to harden the remaining film.

Next it goes to another section and is exposed to a plasma discharge
to etch away the exposed surface of the wafer.

Next the wafer goes back to the bath section, and the baked on film
is removed.

Next it moves through quality control to inspect for defects that
may have occured in the process thus far.

With silicon wafers, atoms with one less or one more atom (boron and
phosphorus, respectively) are introduced in a chamber to the etched
areas so as to alter their electrical conductivity.

Now repeat the oxidizing, masking, etching and doping steps until
you've built up the devices called for in the chip architecture.

Now the backend must be fabricated. Using metal deposition, usually
5-7 layers, alternated with insulators. Again, new chambers, new
supplies, cleaning the deposition chamber between uses, cleaning
supplies, etc.

Next to the passivation section where the final insulating layer is
applied to protect the circuit from contamination.

Next, onto a new etching process where openings are etched to allow
for the top metal layer to be accessed by probes and wire
connections.

Now onto the QA section to test for electrical functionality.
Failure here means rejection, and waste.

Next, the wafer is sent to be sawed into individual chips.

Next to another QA section where the functionality is again tested
and where each chip is checked under a high power microscope to
check for defects. I guess you could do this through teleoperation
as long as you've provided for incredible bandwidth from the MULE to
your earth station.

Once passing inspection, the chips are assembled into a package that
provides contact leads for the chip. A wire bonding machine lays
down wires thinner than human hairs to the leads of the package.

Next a final test of the finished chip.

Lastly, packaging for orbital shipment. You've just completed a 30
plus day process (on Earth.)

(Did you notice how the verbiage got progressivley less in each
subsequent step :)

You'll be in luck if the only unique feature of this new chip is the
wafer. That way you can grow the thing in orbit and return it to
earth for further processing.

More troubling will be if the unique orbital features are at the two
ends of the fab process. Of course, cost minimization will be
analyzed, but you may have to have many Earth-capable process done
in orbit because you can't remove, or don't want to, the wafer in
mid process.

I think a chip fab in orbit is a GARGANTUAN undertaking if it
involves more than just the crystal growth. Many launches required
to deliver mundane materials like sand, bleach, oxygen, water,
filters, etc. Lot's of waste. Need for packaging material, etc.

Let's say your MULE is a billion dollar facility and each module
takes up 50 cubic meters. I can easily imagine, that a billion
dollar Earth-based chip fab would cost 20-50 billion dollars and
with all of those distinct sections in the fab process, the volume
of this plant would be much, much larger than the entire MULE.

I think you're overstating the ability of MULE to expand to
accomodate to be all things to all people. There is a concept called
optimization. For instance, the chip Fab would have internal
plumbing and heat dispersal equipment optimized for its process. In
fact, it may have an exothermic process supporting an endothermic
process, thus no need to dump the waste heat. No need to pipe it
along the backplane of the MULE to the radiators, when you can pipe
it out through the shortest route possible (if there's no backplane,
then you can have radiators right where the heat is created.)

The power supply, heat radiation, communications would be such a
small part of the total cost of a chip fab, that I'm not sure that
MULE savings, if they could be realized, would be incentive enough
to bring about a mating of the Fab and the MULE.

Let's right this discussion. By your own
> admission the MULE has merit - I just haven't been able to make
the
> case for the MULE as a potential catalyst to infrastructure.
Right?
>
> Jack

That's right. As a stand-alone venture it rises or falls on its
merits, but I for the life of me, can't see how it provides a basis
for orbital infrastructure.

Nice work on this post,

TangoMan