Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!bbn!apple!well!tneff
From: tneff@well.UUCP (Tom Neff)
Newsgroups: sci.space.shuttle
Subject: Re: holds in countdown
Summary: planned holds make sense if you know what a countdown IS
Message-ID: <11234@well.UUCP>
Date: 4 Apr 89 22:17:00 GMT
References: <8.UUL1.3#5131@mvac.UUCP> <1989Mar22.175252.1343@utzoo.uucp> <2515@phred.UUCP> <1989Mar28.233955.1843@utzoo.uucp>
Reply-To: tneff@well.UUCP (Tom Neff)
Organization: Whole Earth 'Lectronic Link, Sausalito, CA
Lines: 59

In article <1989Mar28.233955.1843@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
>In article <2515@phred.UUCP> petej@phred.UUCP (Pete Jarvis) writes:
>>>There is no particularly good reason for this; it's just NASA tradition
>>>as far as I know.
>>
>>The buil-in holds are not there because of NASA tradition. They have a purpose
>>in life. Built-in holds are designed into the count at key points in the
>>launch activity sequence where they may need to do some catch-up activity...
>
>You miss the point.  That's an excellent reason for having some slack time.
>The original question, though, was why the countdown clock stops during the
>reserve time.  There's no obvious reason why it couldn't keep counting --
>the lengths of those periods are fixed and predictable if nothing goes wrong.
>It's a bit silly to have the clock stopping and starting when everything is
>going according to plan.

To understand built-in holds you have to understand the countdown.  It
is not like the game clock at Maple Leaf Gardens, ticking away
oblivious to the action taking place down on the ice.  Rather it is an
intricate master *timeline* running down the left side (conceptually
speaking) of a massive set of checklists.  Every one of the myriad
events and actions leading up to a launch has its own specific moment
on the master timeline - its T coordinate, if you will.  This is what
keeps a launch organized, otherwise it would be hopeless because many
steps are interdependent and doing them out of order would be fatal.

Now when the checklists are created the T coordinates are assigned so
they SHOULD correspond to real time if everything goes OK.  In this
ideal scenario you would start the countdown clock at T-24h0m0s or
whatever, and it would unwind in real time right up to ignition.  But
frequently problems arise, major or minor, and a step needs extra time
(real time) to finish.  In this case three things can happen.  The step
could be a super critical one that lots of other things rely on (like
spacecraft power or sequencing equipment), in which case you have no
choice but to take an unplanned hold in the countdown while the problem
is corrected; or it could be something with local scope, with just a
few things relying on it immediately, so you may whip out the pen and
move those things down in hopes the problem will go away before you
have to hold everything; or the problem may be something that isn't a
show stopper just yet but that needs to be handled before the next
major phase of the countdown.  In the last two cases you may need some
slack time.  Thus the planned hold.  Planned holds are like expansion
joints in a bridge, separating major countdown segments.  They let you
take a breath, nail down any remaining loose ends before proceeding
with the next phase -- *without* messing up your intricately
coordinated timeline.  

Since everyone is very busy with events right up the the start of a
planned hold, the launch director will frequently take the hold for a
couple of minutes just to go around the loop and make sure everyone's
in shape for the next phase -- even if nothing's actually wrong.

Sorry for the long windedness of this explanation but it's a frequently
asked question, and I want to make sure the answer is thorough.  If
anyone has corrections or amplifications to this I look forward to
reading them.
-- 
Tom Neff                  tneff@well.UUCP
                       or tneff@dasys1.UUCP