Re: Japans Moon Ambitions - They're KIDDING, Right ?

From: Will (mclean1382_at_aol.com)
Date: 03/13/05 

Date: 13 Mar 2005 12:20:05 -0800

Rand Simberg wrote:
> On 12 Mar 2005 19:26:20 -0800, in a place far, far away, "Will"
> <mclean1382@aol.com> made the phosphor on my monitor glow in such a
> way as to indicate that:
>
> >> >> >> >Do you think that regular inspections will reduce failures
> >from
> >> >wear
> >> >> >> >and fatigue to a negligable level?
> >> >> >>
> >> >> >> It does for aircraft.
> >> >> >>
> >> >> >
> >> >> >It does for aircraft that don't push performance very hard.
> >> >>
> >> >> Even for those that do. I'm not aware of any failures of the
> >SR-71,
> >> >> or the Concorde, due to undetected wear.
> >> >>
> >> >
> >> >Both are a good deal slower than the maximum speeds that aircraft
> >are
> >> >capable of. And both are maintenance hogs by aircraft standards.
> >>
> >> But not by spacecraft standards. And you avoided my point.
> >
> >I was a somewhat surprised to discover that more information is
> >available on SR-71 failures than I would have expected. 40% of those
> >aircraft were lost in crashes. One was very explicitly blamed on
> >fatigue. Several others were blamed on components failing in service
> >long after flight test were complete. And "failing" is something of
an
> >understatement.
>
> Well, fatigue is something that's hard to inspect for (or at least
was
> back then--I think that we have better ways of detecting it now).
>
> The SR-71 was pushing the envelope much more than the Concorde, and
it
> had different expectations. A space transport will have to be more
> reliable than a military spy plane, by nature of its market. SR-71
> was a warfighting machine, and losing vehicles and pilots
occasionally
> was acceptable. Also, even if a high-rate transport turns out not to
> be more reliable than an expendable, it's still likely to be much
> cheaper to operate. I don't recall any proposals to save money by
> throwing the Blackbird away on every flight....
>
> >> >> >Even commercial aircraft has these error modes: fuel load and
> >flight
> >> >> >plan is calculated for each flight, and a fallible human tells
> >the
> >> >> >vehicle what to do, either from the ground or the cockpit.
> >> >>
> >> >> And yet they had better than a 99% reliability, last time I
> >checked.
> >> >> By at least two orders of magnitude.
> >> >>
> >> >
> >> >After many, many generations of aircraft designs and countless
hours
> >of
> >> >flight experience for each. The first several generations of
> >commercial
> >> >aircraft were a great deal less reliable.
> >>
> >> >> >> But still much more reliable than any existing launcher.
> >
> >That is debatable. The US airmail was losing a pilot about every
> >100,000 miles when it started. On the New York-San Francisco route,
> >that's about a 2% fatality rate per trip.
>
> A NY/SF route consisted of many "trips" (i.e., takeoff/landing) so
> that is a gross overstatement of the percentage. And without looking
> at the statistics, I suspect that most of those were a result of
> weather and primitive navigation equipment, than with reliability of
> the aircraft themselves.
>

If your mission is to get a payload from NY to SF, the measure of
relibility is if it performs the trip intact. I don't after all, get to
count each stage of a launcher as a "trip" to enhance my reported
reliability.

And blaming the weather is overstating the reliability of the early
aircraft. As a system, they weren't robust enough to survive weather
that aircraft fly in routinely today.

> >> >> You've presented no basis, so far, to
> >> >> think that the assumptions that work for aircraft won't work
for
> >> >> spacecraft.
> >> >>
> >> >
> >> >The experience to date. Expendable aircraft are absurd for
> >commercial
> >> >transport, yet expendable spacecraft are profitable.
> >>
> >> Because the limited markets are much different. Not because of
any
> >> intrinsic differences in philosophy.
> >>
> >
> >That's interesting, because there are limited markets for aircraft
as
> >well. Ones where the owner doesn't fly very often. These are
strikingly
> >different from launcher markets with similar flight rates
>
> And yet the owner doesn't throw the vehicle away after each flight
> (well, except for Burt, who occasionally gives his to museums after a
> single flight...)

Yes, that's the part that's different. With an airplane you don't pay
extra for wings (or TPS, if you want to fly fast enough). You need them
even if you *are* willing to expend the vehicle. The only savings from
expendability are possibly dropping or omitting the landing gear.

>
> >> >> Which of them do you think flawed, and why? For instance, why
is
> >> >> having an extensive flight test program a bad thing?
> >> >
> >> >Cost. Given a marginal cost of $100, you will rationally choose
to
> >fly
> >> >a large number of test flights. Increase the cost to $100,000,
and
> >you
> >> >will choose a much smaller number.
> >>
> >> Those are absurd numbers in both cases. Try again.
> >
> >
> >Why absurd? $100 dollars a flight sounds about right for something
with
> >the performance of the first mailplanes.
>
> We're talking about launchers.
>

We're talking about the difference between launchers and aircraft. The
huge difference in operating costs matters.

> >And about $100,000 per flight
> >agrees with what Rutan and SpaceDev have said about the marginal
cost
> >of flying SpaceShipOne
>
> That's because it wasn't designed for low-cost operations.

?

>
> However, a hundred thousand for an orbital flight is in fact very
> cheap, and a hundred of them would only cost ten million.

If Rutan *could* fly ten times as fast for the same price, that would
be cheap.

>For a
> vehicle that's like to cost, at a minimum, several times that, it
> seems quite reasonable now that I think about it.
>

The builder disagreed.

> >> >>Why is a low
> >> >> utilization rate better than a high one?
> >> >
> >> >
> >> >All things being equal, the higher the better. Notice, however,
that
> >> >the most reusable orbital vehicles fly less often than some
> >> >expendables.
> >>
> >> You continue to discuss this as though there is this vast
experience
> >> with reusable orbital vehicles, when in fact there have been
> >> exactly...none.
> >>
> >
> >None that were 100% reusable. But there have been partially reusable
> >orbital vehicles, and fully reusable suborbital vehicles. And quite
a
> >few entirely expendable orbital vehicles that have made a profit for
> >their builders.
>
> I don't know why you think that "making a profit for their builders"
> has any relevance to this discussion. We weren't talking about
> profit--we were talking about reducing launch costs.
>

Which will require profitable vehicles, or they won't fly.
No bucks, no Buck Rogers.

Will McLean