Re: Impractical Spaceflight

From: mark@unicorn.com
Date: Thu Aug 20 1998 - 07:01:57 MDT


Philip Witham [p.j.witham@ieee.org] wrote:
>Agreed, I could see spending hundreds of thousands of dollars on average
>to tow it back between several hundred and several thousand miles, depending
>on how closely you can control the de-orbit burn.

And how much in interest on loans, salaries for people waiting at the
launch site and twiddling their thumbs, etc, etc, while you wait for it
to come back for re-launch? Probably far more than hundreds of thousands
of dollars.

>The dollar has not depreciated by ten to one over that time.

So how much has it depreciated? There's no point whatsoever in throwing
around forty-year-old $50-per-pound figures when we're comparing it to
modern costs.

>The smaller Excalibur is a $20,000,000 launch in recent dollars,
>putting up 100,000Lbs.

And where does that cost figure come from and what does it include? The
shuttle costs only about $100,000,000 to launch, but the total cost of
a shuttle flight when you include KSC etc is nearer a billion. Most of
the cost of most of the SSTO designs I know of comes not from the cost
of a launch (fuel, maintenance etc) but from paying back development
loans, investors, etc. That's why they concentrate so much on simplicity
and fast turnaround.

>And you are right, I ignored vertical landing SSTO's. I would add then,
>that vertical landing is an advantage of having vertical landing
>capability (!), and not of having one stage to orbit.

Name one multi-stage launcher which can land intact. Just one will do. Of
course there aren't any, because they're not designed to, and being big,
tall, unstable designs they're never likely to. They could be designed
to do so, but that would significantly reduce their payload.

>Again, my argument (repeating Truax) is against SSTO, and against
>complexity where simple solutions exist.

Yet you're arguing against a simple single-stage design which lands at
the launch site ready for maintenance, refuelling and relaunch in favor
of a complex multi-stage rocket which has to be towed back from a
landing in the sea, checked for damage, and reassembled.

>If you think so, then design it to land, and develop it sufficiently
>that it stands a small chance of exploding. But this is independent
>of the decision of the number of stages.

But it's never meant to land in its launch configuration, so any changes
you make will cost you payload. The reusable SSTO has to land in its
launch configuration (possibly after dumping fuel), so there's no extra
cost. No simple development of a Saturn-V is ever going to land in its
launch configuration.

>We are talking about a design comparison here. With the same people
>working to the same design goals,

But they're not working to the same design goals. That's the whole point.

> is one high-tech stage with turbopumps, pushed to the utmost in
> structural strength to weight ratio,

Uh, Roton is a LOX/Kerosene rocket; they don't even push it to LOX/LH2.

> more reliable than two pressure fed stages with fewer total engines
>and enormously fewer parts, and large strength margins?

Is the SSME more reliable than almost any other rocket engine regardless
of design? Uh, yes. Yet it's also probably the most complex engine ever
built. Similarly, a modern jet engine is far more reliable than any old,
simple, biplane piston engine. A complex engine can be more reliable than
a simpler design.

And again, it doesn't even have to be. If a Truax monster-rocket engine
in the second stage fails to ignite, your payload probably ends up as
vapor in the stratosphere. If an SSTO engine fails to ignite you just
shut down the rest, climb out of the pilot's seat and grab a coffee
while the engineers fix it.

>Look at the engines alone: (oversimplified) A pump fed motor is a pump
>and a thrust chamber. A pressure fed motor is a thrust chamber. You
>cannot assert that a pump fed motor is therefor more reliable.

Look at a modern jet and an old biplane engine. You can't assert that a
complex jet engine is more reliable than a simple old biplane engine?
Yet it is.

>I'd like to see the statistics on that, do you have a reference handy?

Uh, reading numerous reports on the reasons why launchers failed.

>Clearly ignition failures before launch that resulted in an abort
>(non-catastrophic shutdown) wouldn't count here, so I'm not sure
>that to make of this.

I didn't even mention those. Two-engine Centaur stages, old Ariane
third stage, Energiya test flight, commercial Titan IV, just to name
a few. All lost their payloads because of engine ignition failures,
which is a significant problem with multi-stage designs.

>I don't understand what you said here, a list of what?

A partial list of launchers which have lost their payload because an
engine shut down or failed to ignite. All cases in which an SSTO
would probably have brought the payload back intact, but a traditional
multi-stage design had no choice but to crash and burn.

>I can't remember many in-flight launch failures in rockets that were
>not catastrophic. Have I been missing something? I must not be
>getting something here.

No, you're not. Hopefully my last message will finally have explained it
clearly enough.

>Yes. But with an SSTO, that 1 or 2% comes directly out of the
>performance and structural margins, and forces the ultimate from
>everything.

Yet Roton isn't even using LOX/LH2 engines... nor would Black Horse,
and I don't think Kistler were planning to either.

>With two stages, the payload (or next stage) is about 10-20% of
>the stage liftoff mass.

So? Your first stage increases in mass by 3% and you've lost your payload.
Your Roton mass increases by 2% and you've lost your payload. What's the difference?

>I am not aware of any SSTO design being funded enough to make it to
>operational capability, including the Roton. Yes, I'm making a bet
>on that one that I hope to be wrong about.

Well, the implication of their Web site is that they're building the
first one and it will fly into space in a couple of years. Unfortunately
their site is too damn slow to hunt out many details.

    Mark



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 14:49:29 MST