Index Home About Blog
Date: 13 Dec 88 06:01:53 GMT
From:!  (Phil Karn)
Subject: Re: Advanced Launch System

>In space it is much different....
>Electrical components that are off the shelf:

As a partial refutation of these points, AMSAT can point to its long series
of successful spacecraft that include many off-the-shelf electronic
components.  A good example is our program to adapt standard commercial
Ni-Cd batteries for space use, prompted mainly by the drying up of our
previous supply of surplus space qualified batteries. A set has been flying
on UoSAT Oscar-11 for almost five years now with complete success.

Ironically, many of our component failures have been of supposedly "military
qualified" parts, even though these account for only a small fraction of all
the parts on our spacecraft. A good example is the the failed transistor in
the switching supply that provided bias to the final amplifier in the L-band
transponder on Oscar-10; it had a JAN-TX rating.

On the other hand, this does not mean that you can build a spacecraft and
expect it to work without a full understanding of the space environment.
The radiation-induced failure of the 4116 DRAMs on Oscar-10 is well known.
Many ordinary materials are completely unsuitable for space use. AMSAT uses
many space-qualified materials (Kapton, silver Mylar, Teflon, etc) in its
spacecraft despite the cost because it's absolutely necessary to avoid
vacuum outgassing; this is one of the things that the launch agencies
require. (When you fly with a primary payload costing hundreds of millions
of dollars, and this primary payload is a remote sensing spacecraft like
Landsat 5 or SPOT 2, they tend to be understandably protective of their
optics).  AMSAT spacecraft go through the same kind of vibration and thermal
vacuum testing that commercially produced spacecraft go through -- shake
table tests that are so loud that hearing protection is mandatory to avoid
hearing damage, and exposure to several days of hard vacuum, with chamber
wall temperatures ranging between -25C to +55C.  Failures and design faults
do sometimes occur, and these are (hopefully) fixed before launch.

The experience of AMSAT and its sister organizations (e.g., University of
Surrey in the UK) has shown that you can build successful spacecraft with
costs far below that charged by commercial vendors, but it takes a lot of
skill, creativity, dedication and a willingness to take risks (not to
mention lots of volunteer labor). Not all of our clever tricks work, but
enough do to make them worthwhile.


Date: 14 Dec 88 22:32:34 GMT
From:!  (Phil Karn)
Subject: Re: Advanced Launch System

>"Isolate everything in an Earthlike environment, then you can use mundane
>hardware to solve the problems of space."...

Interestingly enough, this appears to have been a standard Soviet technique.
Several years ago, as US/Soviet relations started thawing, AMSAT began some
informal contacts with its Soviet amateur satellite counterparts.  As I
heard it, one of the questions we got was "how do you pressurize your
spacecraft?" It took a while to understand why they asked the question,
since we've never done such a thing.

Our spacecraft are designed with lots of opportunities for venting, so
nothing gets blown out during the rapid decompression associated with
launch. It's also important that transmitters reach a hard vacuum condition
quickly in order to prevent corona damage when they are switched on; there
is generally a delay of an hour or so after reaching orbit during which all
transmitters are held off.


Subject: Re: How expensive is space?
From: Henry Spencer <> 
Date: Feb 16 1996

In article <4g0ibf$> (David Anderman) writes:
>>...Please give me the name and author of 
>>the physics text, and the page number where it says anything in space 
>>must be 1000x as expensive as it is on Earth.
>What exactly is the cost factor for space based machines vs. placing
>them on Earth *assuming* that transportation to orbit is free?

Very slightly above 1.0.  Some of the smallsat folks have found that Earth 
electronics, for example, generally runs just fine in space if you do four 

        1. Replace all the electrolytic capacitors (which contain liquid)
        with solid tantalum types.

        2. Replace all adjustable components with suitably-chosen fixed
        ones, so vibration won't alter the values.

        3. Conformal-coat the boards (spray them with plastic) for
        protection and mechanical durability.

        4. Replace all lubricants with ones suitable for high vacuum.

Steps 1 and 4 aren't even necessary if you provide a sealed pressurized
enclosure for the equipment.  (The scan-platform motors on the Voyagers
are in sealed nitrogen-filled housings, for the sake of their lubricants.)

This is for low Earth orbit.  For high orbits and interplanetary
trajectories, you need to think a little bit about radiation resistance,
and choose your semiconductors with an eye on that. 

>As an example,
>what are the factors that make space based communications transponders
>more expensive than terrestrial transponders?

Requirements for absolute minimum mass, absolute maximum efficiency (to
keep the mass of the power system down), and absolute maximum reliability
(to avoid the need for impossibly-expensive repair missions).  There is
nothing in the space environment itself which imposes more than a small
cost penalty.  Even the radiation environment of geostationary orbit
would be at most a minor nuisance if shielding mass was free.

Check out the price of scientific instruments for high-altitude balloons
sometime.  They face an environment nearly as severe as space:  operation
in near-total vacuum, concentrated sunlight, limited power, limited mass,
a need to survive high accelerations (for landing).  They are far cheaper
than the corresponding space equipment, because their launches are cheap
and the penalties for occasional failures are small.
Space will not be opened by always                 |       Henry Spencer
leaving it to another generation.   --Bill Gaubatz |

Subject: Re: design approaches (Launch efficiency)
From: Henry Spencer <>
Date: Mon, 5 May 1997 17:28:56 GMT

In article <5k929h$jnj$>, Doug Hendrix <> wrote:
>>Avionics we could mostly borrow from Pegasus...
>If you mean the INS, then most any space rated RLG box would work. As
>Lockheed discovered on the LMLV, you don't just stick any old box on a
>Space launcher and expect it to work.

At least not if it uses high voltage!  Actually, it's not really enough
for it to be *space* rated, because the problem with the LMLV INS was
during the transition from 1atm to vacuum -- it probably would have worked
fine in vacuum, but in very thin air the HV power supply arced.  The INS
supplier did try to test for this, but didn't do it over a sufficiently
wide range of conditions.

Actually, some of the smallsat people have found that you *can* put most
electronics boxes into space and have them work fine in LEO with a few
simple precautions:  replace electrolytic capacitors (which contain liquid
and will vent), replace trimpots with fixed resistors (trimpots are very
sensitive to vibration and contamination), add a conformal coating (to
keep any bits of floating debris off the wiring), and replace lubricants
on any moving parts.  This is not just theory; they've tried it, it works.

Mind you, if you're trying to fly something like a 200MHz Pentium, then
obviously you need to worry a bit about cooling.  But for less aggressive
electronics in a well-designed spacecraft, that's not a big concern.

>>Uh, why not?  Even assuming that it does have to weigh that much, and that
>>we are stupid enough to use liquid hydrogen, the dry mass is only about
>>400klb... roughly that of a 747.  It's the dry mass that we have to build
>>and maintain.  The gross mass is just fuel.
>All the studies I've seen tie ops costs to Dry weight.  Of course, if
>you get airlines type reliability, that would negate some costs. I
>don't know,  can an airlines turn a 747 around as fast as a 2 engine

Once you deduct issues like the larger number of trolleys for the galley,
probably faster; the 747's engines are much more reliable.  The rule of
thumb is that you can't make money on a 747 unless it spends over half its
life in the air.

>>Incidentally, in 1982 Boeing offered to build the USAF an SSTO -- well, it
>>cheated slightly by using sled launch -- for a firm fixed price of $1.4G...
>Not too long ago, a major aerospace company gave us a $2B price tag
>for a RLV SSTO.  I'd guess that they'd take that back now.

Depends on the assumptions behind that price.  It's not an unreasonable
price for a well-run project done right.  Boeing had, incidentally, studied
their RASV concept in more depth than any other SSTO project before or since;
they weren't taking any risks with that quote.  (Incidentally, that was not
in a press release, but in a formal offer to the USAF, signed by Boeing's
president... and the USAF came very close to taking them up on it.)
Committees do harm merely by existing.             |       Henry Spencer
                           -- Freeman Dyson        |

Index Home About Blog