Index Home About Blog
Newsgroups: sci.space.shuttle
From: henry@spsystems.net (Henry Spencer)
Subject: Re: Scramjets (was: Better deliverables for X-33 program?)
Date: Sun, 26 Mar 2000 00:59:24 GMT

In article <8bemcc$dpo$1@nnrp1.deja.com>,  <ladasky@my-deja.com> wrote:
>> Despite what some snake-oil peddlers will tell you, scramjets are a
>> lousy way to get to orbit.)
>
>Care to elaborate on the reasons why?  I would think that the weight
>savings that come from not having to carry oxidizer would be substantial.

That's the theory.  The practice is not nearly so simple.

For one thing, there's the overwhelming problem that nobody has ever
gotten positive net thrust out of a scramjet in flight.  Ever.  They're
hard to design and hard to test, and so in a fairly fundamental sense they
are still vaporware.  (The snake-oil peddlers make it sound like they are
proven, off-the-shelf hardware, which is emphatically not true.)

For another, they intrinsically have nasty technical problems.  Notably,
they demand good and very rapid mixing of fuel into the air, without
disturbing the airflow much!  That's very close to a contradiction in
terms; it may be possible but it's not easy.

You need something else to get them up to ramjet operating speed, and then
a rocket engine for the final push into orbit.  This adds mass and
complexity, reducing the benefits.

They also have some more general problems shared by other airbreathing
concepts...

They are very heavy by rocket-engine standards.  Even at sea level, air is
three orders of magnitude less dense than liquid oxygen, and is 4/5 inert,
so the machinery needed to handle it is inherently much bigger and bulkier
and heavier.  This is not a good trade; LOX burns off on the way up, but
the engines have to be carried all the way.

The speed range involved in going from the ground to orbit is enormous.
It's rare for an airbreathing engine to operate efficiently over a 3:1
speed range, let alone a 10:1 range.  Coming even close to this tends to
involve extensive variable geometry, which makes them even more
complicated and even heavier.

The heating and drag problems of trying to airbreathe at very high speeds
are monumental.  Also, it inherently takes a lot of energy to get thrust
by pushing on air that is already moving past you very quickly.  The
combination tends to mean that while a scramjet uses less LOX, it uses
a lot *more* LH2... and LH2 is the expensive and bulky part.  LOX may be
heavy, but it is cheap and compact.

Finally, developing high-speed airbreathing engines is very difficult.
Computer models are worthless without test data to back them up, wind
tunnels simply are not up to the job, and development using research
aircraft would be tedious and costly.

In real launch systems, it invariably turns out that if you skip the
scramjets and simply use rockets, the system gets simpler and cheaper.
They may be useful for high-speed cruising flight within the atmosphere,
but they are not a good way to launch things into space.
--
Computer disaster in February?  Oh, you |  Henry Spencer   henry@spsystems.net
must mean the release of Windows 2000.  |      (aka henry@zoo.toronto.edu)


Newsgroups: sci.space.shuttle
From: henry@spsystems.net (Henry Spencer)
Subject: Re: Scramjets (was: Better deliverables for X-33 program?)
Date: Sun, 26 Mar 2000 22:33:16 GMT

In article <8blqch$d2h$2@supernews.com>,
Richard Ireson <richard_ireson@dknet.co.uk> wrote:
>Pardon my ignorance, I know what a ramjet is but what is a scramjet?

It's a semi-acronym for Supersonic Combustion Ramjet.  Ordinary ramjets
peter out at about Mach 6, because they slow the incoming air to subsonic
speed before injecting fuel, and that heats it up; when the incoming air
temperature equals the temperature of a fuel-air flame, no thrust can be
produced by injecting and burning fuel.  That problem can be avoided, in
principle, by injecting and burning fuel in an airstream which remains
supersonic throughout... but that's very hard to do.
--
Computer disaster in February?  Oh, you |  Henry Spencer   henry@spsystems.net
must mean the release of Windows 2000.  |      (aka henry@zoo.toronto.edu)


Index Home About Blog