Index Home About Blog
Date: 8 Oct 1981 00:28:19-PDT
From: decvax!utzoo!henry at Berkeley
Subject: Bussard ramjet speed limit

Since I haven't seen anybody else demolish the myth about the 0.17c
speed limit on Bussard ramjets, here goes.  The article in Asimov's
was not the first recognition of the problem;  contrary to what was
said in the article, this problem has been known for years, but has
not been trumpeted out loud because there is a straightforward fix.
It is reasonably obvious to anyone who does a real analysis of the
Bussard ramjet instead of relying on analogy from "ordinary" jets.
(In the following, all velocities etc. are with respect to the ship.)

The purported speed limit works as follows:  thrust from the expulsion
of fusion-engine exhaust is counterbalanced by intake drag from
fast-moving incoming fuel hitting the ramscoop field, so net momentum
transfer to the gas stream works out to be zero.  With some simplifying
assumptions, the speed limit equals the engine exhaust velocity;  it
is easy to set an upper bound on this based on the reactions involved.

The mistake is to assume that the kinetic energy of the incoming fuel
is necessarily converted to heat or some other useless form.  WRONG!
Suppose instead we decelerate the incoming protons against an electric
field.  The momentum IS transferred, but the energy is stored as
potential energy.  We use that stored energy to further accelerate
the outgoing exhaust by letting the protons fall down the other side
of the same electric-field potential hill.  Other variations are
possible, but the principle remains the same:  use the kinetic energy
rather than wasting it.

There is still some momentum transfer, because the same amount of
kinetic energy does not mean the same amount of momentum at different
velocities.  The exhaust acceleration is less effective at producing
momentum because it is applied to already fast-moving material.  But
this is a lesser effect;  barring losses, there is always a nonzero
net momentum transfer to the gas stream (unless relativity introduces
some subtle complication at extreme velocities;  not my specialty).
Of course there WILL be losses, and integrating this with a ramscoop
may be a lot of fun, but those are problems of technology, not
fundamental physics.

In any event, the ramscoop is the EASY part of a Bussard ramjet, by
current thinking:  the HARD part is getting a decent reaction rate
out of a fusion reaction burning ordinary hydrogen.  Building a
ramscoop is a formidable engineering problem, but a fast proton-proton
reactor involves nasty difficulties of fundamental physics.  It may
be necessary to go to an internal energy source, either ordinary
fusion (the "ram-augmented rocket" scheme) or antimatter.  The latter
is the more interesting:  antimatter-heated rockets have rather (!)
high performance themselves, and adding "free" reaction mass makes
it even better.

To sum up:  the near-c Bussard ramjet is not impossible, it's just
complicated, difficult, and not quite the way it was visualized.

					Henry Spencer

From: (Henry Spencer)
Subject: Re: Space Drive
Date: Mon, 29 Jun 1998 13:37:27 GMT

In article <>,
Trakar <> wrote:
>>You don't need to grab space, just the hydrogen that's in it, fuse it and
>>kick it out the back.  Still using reaction but effective...
>But, the "Bussard Ramjet" style propulsion systems are actually limited to the
>lower fractions of c (60% +/-IMSMC), due to "drag" and "field collection
>efficiencies" (or some such gobbly gook), at least according to the best
>simulations I've seen so far.

If you want seriously relativistic speeds, Bussard-ramjet performance is
quite sensitive to the quality of the engineering that goes into them,
i.e. to the assumptions made in such simulations.  Contrary to occasional
claims made by the ignorant, there is no *fundamental* speed limit on a
Bussard ramjet... but to be effective at relativistic speeds, they have to
avoid even small losses while handling very large amounts of energy.
Considering that we don't currently know how to build even an inefficient
Bussard ramjet, the engineering problems are daunting.

(Nobody who has investigated the physics of the intake in detail has come
up with a satisfactory configuration.  And there is also the non-trivial
problem of building a fusion reactor that can burn ordinary hydrogen at a
useful rate.)

A further small problem is that Bussard ramjets probably are not usable
anywhere near here in any case.  The local interstellar medium is much
thinner than was once thought.  (For that matter, even Bussard's original
paper -- which pre-dates the data that changed the estimates -- expresses
some doubt that the local interstellar medium is thick enough for the
ramjet to work here.)
Being the last man on the Moon is a |  Henry Spencer
very dubious honor. -- Gene Cernan  |      (aka

From: (Henry Spencer)
Subject: Re: future of spacetravel
Date: Sun, 9 Jan 2000 00:02:02 GMT

In article <8585u1$c8r$>,
 <> wrote:
> Has research on the Bussard Ram jet been abandoned then?

Well, it's not like very many people were doing research on it to begin

Bussard himself suspected that the interstellar medium near the Sun might
be too thin to run his ramjet.  And the estimates of its density have been
revised downward a lot since then.

Nobody has yet come up with a workable way of building the intake of a
Bussard Ramjet.  It's very hard.  Also, if you want a pure Bussard ramjet,
you also need a fusion reactor which can burn ordinary hydrogen, which is
also very hard.  You can finesse the reactor problem by taking along
antimatter as an energy source, and just using the interstellar gas as
reaction mass, but the intake problem remains.
The space program reminds me        |  Henry Spencer
of a government agency.  -Jim Baen  |      (aka

From: (George Herbert)
Subject: Hydrogen/Boron Interstellar Ramjet
Date: 18 Jan 2000 01:35:12 -0800

The discussions around related newsgroups flipped a light
on in back of my head.  The key problem for bussard ramjets
has always been trying to deal with the energy lost accellerating
the interstellar medium to ramship speed.  What if you don't?

Specifically, the proton-Boron11 reaction should work reasonably
well using moderate-fractional-C collected interstellar medium.
The reaction products are high energy alpha rays (helium nuclei)
which can be magnetically or electrostatically confined and/or
focused to exhaust to the rear.  They're moving at a goodly fraction
of C, but not so high that it's inefficient.  You have to carry
most of your fuel with you, though, 11/12th of it.  You might
want to do the electromagnetic equivalent of an ejector rocket,
using the alpha rays to accellerate some more interstellar
medium not run through the fusion system, to get a mass leverage
at an Isp penalty (which is probably tolerable).

I need to run the numbers again at a decent time of day,
but it looks sort of interesting so far.

-george william herbert

From: (Henry Spencer)
Subject: Re: future of spacetravel
Date: Thu, 20 Jan 2000 02:27:12 GMT

In article <8630g8$nia$>,
Frank Crary <> wrote:
>>With a sufficiently rapid reaction, e.g. an antimatter-fueled system, it's
>>quite conceivable to use the hydrogen without decelerating it.
>That would (could) cause the same scale problems that come up with

Indeed so.  Not at all easy, but it's not obviously impossible either,
especially in a system where the scooped gas is primarily reaction mass
for energy obtained by other means (e.g. an antimatter reaction).

>>The subtle error is to assume that the energy lost in decelerating the
>>incoming hydrogen cannot be *recovered*...
>...the system you are talking about would need
>perfectly adiabatic compression and expansion. (Well, only
>perfect if you want to recover all the energy, but under
>the circumstances, I think you would need to get very close to that.)

Yes, any losses loom quite large when the speeds get really high -- the
intake and nozzle are handling power which is very large compared to that
of the engine itself.  However, this is a second-order problem, which is
an improvement.

(Also, while energy recovery could in principle be perfect, momentum
recovery can't be -- if the engine adds any velocity to the exhaust at
all, even a perfect energy-recovery nozzle generates less thrust than the
intake drag -- so there *is* a net reduction in engine effectiveness.
Again, though, it's a second-order problem, reducing effectiveness rather
than eliminating it.)

>>...It's quite conceivable to do deceleration
>>reversibly, e.g. by decelerating just the protons against a strong
>>electrostatic field...
>That would take some extremely strong fields... I'm not saying this
>would be theoretically impossible, but it would be quite a bit
>harder than you imply.

Nowhere did I say it was easy!  In fact, I'd expect that a practical
energy-recovery system would have to do something more complicated.
The space program reminds me        |  Henry Spencer
of a government agency.  -Jim Baen  |      (aka

Index Home About Blog