Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!db.toronto.edu!hogg
From: hogg@db.toronto.edu (John Hogg)
Newsgroups: sci.space.shuttle
Subject: Re: SRB solutions
Message-ID: <1989Oct25.100345.2672@jarvis.csri.toronto.edu>
Date: 25 Oct 89 14:03:45 GMT
References: <5474@umd5.umd.edu> <5149@uhccux.uhcc.hawaii.edu> <12973@s.ms.uky.edu> <1989Oct18.172311.22863@utzoo.uucp> <838@cybaswan.UUCP>
Organization: University of Toronto, CSRI
Lines: 37

In article <838@cybaswan.UUCP> iiitsh@cybaswan.UUCP (Steve Hosgood) writes:
>Wouldn't it be possible to design
>the SRB casing to have a venturi nozzle at both ends, but with the top one
>usually plugged by the igniter? If the igniter could be blown off the top
>in case of emergency, you'd be left with an SRB thrusting at both ends. The
>result *ought* to be that you could separate from the boosters.

That sounds roughly equivalent to the SRB destruct charges, except that
the goal of these is simply to reduce thrust to zero by reducing
chamber pressure.  SRB fuel only burns effectively at high pressure.

If this were a valid abort procedure, NASA would have adopted it.  A
small selection of problems includes avoiding debris, the reverse
loading on the attachment struts when the SRB thrust suddenly becomes a
drag in thick atmosphere (remember, normal SRB staging occurs at a
considerable altitude) and the simple aerodynamics of separation in
atmosphere.  I'm sure that there are other (and probably more
significant) difficulties, which somebody else can point out.

>>(For one thing, the SRBs would have to stay under some sort of
>>control for at least a few seconds to avoid problems like having their
>>exhaust hit the external tank, and they rely on the orbiter for control
>>at present.)
>Couldn't they have fins fitted to give them some degree of independant
>steering?

While thrusting, they have independent steering via nozzle gimballing.
This, of course, wouldn't apply after shutdown.  However, the actual
*control* circuitry is in the orbiter; the only signal that the SRBs
receive independently is ``destruct''.  There's no reason in principle
why control packages couldn't be added.  It's just that they'd have to be
man-rated and salt-water-splashdown proof, and would be yet another mass
penalty and source of potential failure.  The same would apply to fins.
Sometimes, safety in increased by leaving things out.
-- 
John Hogg			hogg@csri.utoronto.ca
Department of Computer Science, University of Toronto