Sender: <marv@lancaironline.net> (Marvin Kaye)
To: lml
Date: Mon, 13 May 2002 11:10:48 -0400
Message-ID: <redirect-1236302@logan.com>
From: "Gary Casey" <glcasey@adelphia.net>
Subject: engine reliability
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit
Importance: Normal

<<2) Given the history and experience of brands C and L, I still maintain
that it is inexcusable for out-of-the-box new Continental and Lycoming
engines to break crankshafts, shear oil pump drives, and throw
connecting rods.  We should expect (and get) MUCH more.  Certified
engine reliability is NOT what it should be.

3) To expect that a NEWLY designed package will have the same
reliability as a mature one is simply unrealistic.  The prototype will
never be as good as the later, more refined versions.  Early production
units would (should) not be as "good" as later production versions
although the recent Continental and Lycoming experience causes one to
review and question this assumption.>>

Fred says it very well, but I would take some exceptions or maybe just add
some comments.  First, what degree of reliability are we looking for?  I
submit that it should be quite high - like an engine failure per more than
100,000 hours.  And there are only two ways to determine reliability:  By
design and by experience.  By experience, the L/C engines don't fare too
well.  A couple of broken cranks and rod bolts mess up the statistics pretty
badly.  By design they don't fare too well either - a rod bolt really
shouldn't ever fail in a low-rpm engine unless something about it isn't
designed right.  And once a fix is in place the clock starts over - look at
the Lycoming oil pumps - "now it is okay.....oops, no, I meant NOW it's
okay.....well, try this..." and so on.

Would I "expect" a newly designed engine to have equivalent reliability?
That depends on the design as certainly there is no experience to fall back
on.  In this case experience seems to be a poor teacher as there are more
failure modes lurking than there are "test" pilots willing to find them.
The point is that if one wants to do something new it is not necessarily bad
the result necessarily less reliable, but there are many, many design
considerations that must be addressed.  I once asked someone about his auto
engine fuel system that I thought had built-in failure modes - the answer
was "I've flown it for 100 hours with no problems"  as though that were the
proof.  In fact, it was little more than anecdotal information; interesting,
but not of much value.

Gary Casey
ES auto engine project