🚀 go-pugleaf

In article <BzqBvs.J8H@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes:
>In article <1992Dec23.111923.22269@ke4zv.uucp> gary@ke4zv.UUCP (Gary Coffman) writes:
>>... A ten year life zero defects
>>GEO comsat like K2 is much cheaper than a 1 year life package
>>that costs 20 times less. That's because most of the investment is
>>not in the satellite, it's in the Earth based terminals that use it.
>
>I don't grasp this argument.  It's the same Earth-based terminals either way.
>If you're providing a service, you plan to do so over more than one satellite
>lifetime, either way.  Twenty years of service is cheaper with mass-produced
>short-life satellites, even with your (fairly unfavorable) assumptions.
>
>>... Since the satellite represents a single point failure node...
>
>This is your assumption, not a self-evident fact.  Communications networks
>normally have redundancy to cover predictable single-point failures.
>Even today's gold-plated satellite networks do, despite the expense.

It's the network reconfiguration costs that get you. When NBC had to
reconfigure from K2 to SBS 3 due to a control failure on K2, it cost
NBC $150,000 a *minute* for 4.5 hours until the major ground systems 
were re-aimed. Smaller markets didn't get back on line for over 8.5
hours. That was only 223 ground stations that had to be repositioned.
For many large corporate data networks, the number of terminals to be
re-aimed climbs into the thousands. Most ground stations are manually
pointed and require service personnel to visit the site and re-align
the dish.

Now *scheduled* transfers can be handled in off peak times, but still
require those personnel to visit all those sites. Doing that 10 times
instead of 1 get's expensive fast. Doing it *unscheduled* because of
a failure of the cheapsat, can be really expensive if it only happens
once.

>>... and since for most orbits
>>the satellites aren't retrievable or repairable, and DC won't change
>>that...
>
>Again, your assumption, not a self-evident fact.  Cheap launches change
>almost everything, including the feasibility of retrieval and repair.

I wasn't aware that DC was planned to have a GEO capability, or a large
enough cargo bay to retrieve a major comsat.

Gary
-- 
Gary Coffman KE4ZV          |    You make it,     | gatech!wa4mei!ke4zv!gary
Destructive Testing Systems |    we break it.     | uunet!rsiatl!ke4zv!gary
534 Shannon Way             |    Guaranteed!      | emory!kd4nc!ke4zv!gary      
Lawrenceville, GA 30244     |                     | emory!ke4zv!gary@gatech.edu