Sender: <marv@lancaironline.net> (Marvin Kaye)
To: lml@lancaironline.net
Date: Mon, 12 Jan 2004 19:01:29 -0500
Message-ID: <redirect-2929584@logan.com>
Reply-To: "Dan Schaefer" <dfs155@earthlink.net>
From: "Dan Schaefer" <dfs155@earthlink.net>
Subject: Fuel Transfer Pumps
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit

Rick Argente asked:

<<Anybody had a pump failure? Number of hours at failure?>>

Rick, if we're talking about the Facet 12 volt units, I've had two
failures - one used for wing fuel transfer and one used for the boost
(back-up for the engine mounted mechanical) pump. Both failed in about the
first 100 hours of operation, though not at the same time. Since that time,
I've had no failures in about another 600 hours. I even went so far as to
install a second, redundant transfer pump in series with the single unit and
of course, haven't had another failure since.

In my past life, I was involved with a lot of qualification testing of
aircraft system components and we had a term for this kind of random
failure. It was called "infant mortality" and referred to such early
breakdowns. Generally, if one plots numbers of failures vs time for just
about any hardware, it invariably starts high (in about the first 100 hours)
and then drops to the inherent failure rate curve. This is not to say that
there won't be an occasional random early failure during large scale useage
(since the plots only depict averages over time for many units) as there
will always be a maverick unit.

One way to help get around this (which I didn't do) is to set up a test
stand which duplicates the useage environment (on - off cycles and run
times) and run it for about 100 hours. Usually, if you have no failures
during this time, the device should last for it's predicted service life.
Which for the Facet pump, is probably quite long indeed (though I have no
hard data to back that up - just a guess). This approach is probably far in
excess of what most of us are willing to take on. The alternative is to let
the first hours of useage occur in service on the airplane, while keeping a
critical eye on such items. If you get past the first 50 - 100 hours you're
probably OK for a long time.

Sorry for the long-winded reply, but you asked!

Dan Schaefer
Early LNC2