Sender: <marv@lancaironline.net> (Marvin Kaye)
To: lml@lancaironline.net
Date: Thu, 16 Sep 2004 16:46:24 -0400
Message-ID: <redirect-413981@logan.com>
Received-SPF: pass
 receiver=logan.com; client-ip=68.168.78.199; envelope-from=dfs155@adelphia.net
From: "Dan Schaefer" <dfs155@adelphia.net>
Subject: Re: transfer pumps-LNC2
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit

Gerard,

I've had two Facet pumps fail in my 235, one as a boost pump installed on
the engine side of the firewall and exposed to the rigors of the high
temperature there, and one as a transfer pump installed under the pilot's
seat. Both failures, according to my A/C log book, were in the first 100
hours and the replacements have been going strong ever since (~ 7 - 800
hours). The third pump (also under the seat) has never failed. This is
reminiscent of what I learned during some of my working days.

In another life, before being swallowed up by the areospace industry, I
designed large electronic power systems (60 and 100KVA variable frequency
motor drives) for industrial use. We quickly learned that final system tests
prior to shipment required full power runs of about 40 hours to ferret out
the inevitable "infant mortality" failures. It appeared that after that
amount of time, the number of component failures dropped drastically to
approximate the calculated Mean Time Before Failure (MTBF) for the
equipment.

Short version, when installing a new part like a Facet pump, you're taking
pot-luck on it's initial reliability as it's probably never been run longer
by the manufacturer than it takes to merely make sure that it actually does
run - probably a few seconds (if that, even).

Bottom line, be prepared for some infant mortality failures early on - they
are almost inevitable.

Dan Schaefer