future of satellite business (was Re: [amsat-bb] More about the Ariane failure)

[Franklin Antonio <antonio@xxxxxxxxxxxx>]

At 11:46 AM 12/14/2002, Jeff Davis wrote:
>Five years ago there was a surge in satellite demand, driven by an 
>over-enthusiastic expectation for both satellite phones and the dot.com 
>business.

Well, I think the connection to the dot.com bubble is tenuous, except 
perhaps in the case of Teledesic.  The peak in the satellite construction 
business due to mobile satellite telephone system construction or 
anticipated construction a few years ago (Globalstar, Iridium, ICO, ...) 
was real.  But there are larger trends that one must consider to get the 
big picture.

Many of the uses for communication satellites have simply been replaced 
with or substantially reduced by technological change.

Long distance telephony:  Fiber has replaced satellite completely.

Computer networks:  In the 70's some folks were convinced that geo 
satellite communication links would be the backbone of high speed computer 
networks used by large corporations.  May sound strange now, but large 
corporations did use satellite links for computer data routinely at that 
time, and for awhile in the 70s it was a growing application.  If you 
guessed that computer networks would see incredible growth (correct), but 
failed to see the transition to fiber, you might have invested.  Satellite 
Business Systems (a JV of IBM, Aetna, and Comsat, if I remember correctly) 
was formed to supply to that market, and spent about a billion dollars 
building a giant high speed TDMA satellite network, to connect computers at 
major corporations.  It was a complete business failure.  Fiber now 
supplies all the long distance connectivity anyone could want for computer 
networks.

Satellite television:  C and Ku band geo satellites in large number were 
used for backhauls and network distribution to TV stations.  Some home 
users tapped in to the signals as well.  For many years the transmission 
was analog, where one TV signal occupied an entire transponder.  With the 
transition to digital, and the advent of high quality compression, many TV 
signals can occupy the same transponder.  In addition, as these satellites 
have been gradually replaced at end-of-life during the past 10 years, the 
satellites that have replaced them have more powerful transponders, meaning 
they can carry even more per transponder.  A lot of TV signals still go 
thru satellite transponders, possibly as many as ever, but the number of 
transponders required has been greatly reduced.  More and more TV signals 
travel over fiber too, in point-to-point (backhaul etc) applications.

The home satellite TV users of course mostly migrated to special DBS 
satellites designed specifically for broadcast to small dishes at the 
home.  That's a new use for satellites, but there isn't a need for very 
many of this special kind of satellite, so they don't add much to the count 
of new launches required.

Satellite internet delivery: There are a few companies offering a 
satellite-based internet connection to your home.  Unfortunately, these 
businesses use conus-beam Ku band satellites, and these satellites just 
don't have enough capacity to make these businesses profitable.  Future 
multi-beam Ka band satellites may accomplish that trick a few years from 
now.  Meanwhile it has been a mystery why these companies continue.  News 
broke a few days ago that Hughes will cease marketing its service.  Perhaps 
they reached the same conclusion that I did.

Then there's Ka band.  FCC issued a bunch of licenses a few years ago for 
Ka band satellite systems (20/30 GHz).  Many of the proposed systems were 
proposed only to get one of the licenses, and then see what would 
happen.  There were people who were serious, such as Teledesic, but it is 
difficult to discuss Teledesic while keeping a straight face.  The original 
system involved a constellation of over 800 LEO satellites.  This would 
have offered high speed internet access to everyone in the world via 
satellite.  The large number of satellites meant that there would always be 
one nearly overhead, providing a short range (good link budget) and low 
latency.  From a communication point of view, this made great sense.  Only 
problem is it would have required an impossible amount of money.  The 
system was redesigned to user fewer satellites several times and then 
finally abandoned.  They got great PR and generated tremendous hype from 
the fact that two high profile rich guys put in some of the initial money, 
but it wasn't enough buzz to attract other investors.  While the other  Ka 
band proposals were not so bold, most of them will never get built 
either.  Some companies have started construction of Ka band satellites so 
that they can show progress and keep their licenses, but there are very few 
who have plans for those satellites that are capable of making sustainable 
businesses.  Any satellite manufacturer who was planning on much of this 
business for future growth is making new plans.  I'm not saying Ka band 
satellites will not be built.  I'm just saying this will proceed much much 
more slowly than was anticipated.  At Ka band it is possible to build 
satellites with a large number of independent antenna beams.  This 
increases the capacity of the satellite tremendously, presuming that you 
have a large number of users who want different bits 
delivered.  Eventually, this sort of satellite will find a place in 
satellite internet delivery.  That's several years away.  Even when it 
arrives, it will likely not create a demand for a large number of satellites.

Finally, even if all the new systems go away, there's a background business 
replacing geo satellites as they run out of fuel.  As the satellite 
manufacturers stare at their blank order books they keep mumbling to 
themselves that the replacement business will surely come.  I claim even 
that business is now uncertain.  As the C and Ku Geo satellites in orbit 
now run out of fuel over the next 10 years, will all of them be replaced as 
once planned?  Probably not.

Don't get me wrong, there are some great commercial uses for 
satellites.  Sometimes nothing else will do.  However, many of the past 
uses are just going away, and many of the hyped new uses have not 
materialized.  That means a many years of reduced activity in construction 
and launch.

>In 1997 there were some 60 commercial rocket launches, which fell to 28
>last year. This year there will be only 21 commercial satellite launches."

This trend will continue for awhile.

>"But the reality is that no one at present is making money by launching
>satellites.

!!! True for building satellites as well as launching.  I believe that the 
satellite manufacturers order books look even worse than these launch 
counts would lead you to believe.  It means we will see more consolidation 
in the satellite industry during the next 2-3 years.

I know less about the launch industry, but it seems clear that something 
similar must happen there too.  Whether they will downsize, consolidate, or 
some just disappear, I don't know.  Most of the launch operations are run 
by governments who are unlikely to consolidate.  They'll probably just 
downsize.



----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org