COZY MKIV Canard Incidence angles

"Marc J. Zeitlin" <marc_zeitlin@alum.mit.edu> Jul 19 09:05AM -0700  

Folks:

I recently brought a COZY MKIV into the shop after a pre-buy inspection.
This plane had not flown much (180 hours or so over 12 years) and hardly at
all in the past 3-5 years (maybe 3-5 hours). It is well built, with very
careful craftsmanship, a nice panel and interior and good contouring. Fast
too, with a cruise prop from Catto. Needs a paint job, but that's only
because the builder made the same mistake I did originally, using the same
water based polyurethane that caused me to have to refinish my plane.

In any case, I flew it up to Tehachapi, stopping at Fox Field in Lancaster
to get a VFR Transponder check done, so it would be legal.

I knew that due to a lack of appropriate ballast, I was flying the plane
slightly aft of the rear CG limit (102.6", as it turns out), so I did not
allow my airspeed to drop below 75 KIAS until I was just above the
runway(s). I noticed, however, that the plane did not seem stable in pitch
- I could not trim it to an airspeed, and it would run away, either faster
or slower, if I let go of the stick or did not have constant pitch inputs.

As we know, aircraft get less stable as the CG moves aft, but the Neutral
Point of a COZY MKIV is nominally at 105" or so, so any CG forward of that
should be at least somewhat stable in pitch (albeit very susceptible to
deep stall at CG's aft of 103" or so). So being unstable at 102.6" is
unusual (or should be).

As some of you may remember, another COZY MKIV that I worked on over the
winter needed a canard incidence angle increase to fix a pitch instability
problem (it would actually reverse trim within the approved CG range - I'd
trim up and it would go down, and vice-versa - that was a fun 11 hour
flight). Increasing the canard incidence angle by 2.5 degrees fixed the
stability issue on that aircraft, and it's flying safely now with a new
owner.

After some minor work on THIS new plane, including a new W&B to verify CG
position (the original W&B was spot-on) and a check of the canard incidence
(it matched the latest template within a couple tenths of a degree), I flew
it again yesterday at a CG of 101" - 1" forward of the rear limit, to
attempt to determine if the plane was stable within the normal CG range. It
was BARELY stable- when trimmed for (what I hoped was) 130 KIAS, it would
oscillate in a phugoid mode between 120 KIAS and 160 KIAS over a period of
about 30 seconds. When trimmed for 80 KIAS, it would oscillate between 65
KIAS and 90-100 KIAS, also over a very long period. When forward of the
rear CG limit (or even AT it), those types of large speed oscillations and
long periods are NOT acceptable (although not necessarily unsafe - there
was no trouble whatsoever controlling the aircraft with hand on stick).

So this plane, too, needs the canard incidence angle increased, and I've
started that work.

Having two aircraft wit this problem - two that had both theoretically been
tested during Phase I, and two that had HAD the canard incidence angle
increased at least once before on each - makes me think a few things.

- One, that even with the new canard incidence angle templates, aircraft
that were built and flown in the early 2000's (one of these first flew in
2001, the other in 2004) had a higher susceptibility to canard incidence
angle issues. I'm not sure what's going on with newer planes.

- Two, that people do not do a good job of stability testing during
their Phase I period (and with all due respect, most people don't do a good
job of testing in general during Phase I - YOU, dear reader, of course are
in the category of people that do wonderful Phase I testing, I'm sure).

[As an aside, a Long-EZ I did a CI on last weekend had been flutter
tested to Vne BY A NON-BUILDER OWNER. I was astounded (and pleased) - this
is rare enough by a builder, much less a subsequent owner. Now, they should
have done a balance on the elevators and ailerons BEFORE doing the flutter
test, but the flutter test is the proof of the pudding - with no flutter up
to Vne (should have gone to Vd, but I'll take what I can get, especially in
an O-235 LE), we KNOW the plane is safe from a flutter standpoint, whatever
the balance.]

- And three, that it's very important to realize that the canard
incidence angle is not the only important angle here - what's critical is
the RELATIONSHIP between the canard incidence angle and the main wing
incidence angle. If the main wing angle is 1-2 degrees too high, then the
canard incidence angle would have to be raised by the same amount, no
matter what the templates show, so that the aerodynamics would be correct
for stability (and deep stall resistance).


It's VERY hard to measure the main wing incidence angles after the fact,
and no-one wants to change main wing angles after painting, as the whole
strake intersection would need to be recontoured. So easier is to change
the canard incidence angle.

In any case, one of the things I was asked to talk about at OSH this year
at the COZY forum is Phase I flight testing and I'll be showing an example
of a test card for a first/second flight. It struck me that, given all the
questions we get regarding how to flight test in Phase I, it might be a
good idea to create a set of detailed test cards for ALL the desired tests,
including stability, etc., for Phase I and publish that for general use
among canardians (obviously, they'd need to be customized per plane, but at
least 80% - 90% of it would apply to everyone).

I'll put this on my list of things to do and publish...

Also, when building a COZY MKIV, it seems clear that it's impossible to
mount your canard at too high an incidence angle - never heard of it
happening - so one should always bias the mounting to the TOP of the
incidence tolerance - have as much incidence as you possibly can when using
the templates.

And lastly, be DAMN sure to get the wing incidence angle correct during the
build - bias it downward, even - and if you need to shim a wing during
Phase I for whatever reason, ALWAYS shim it so the leading edge drops - do
NOT shim wings to raise the leading edge, as this will decrease the
difference in incidence angles between the main wing and the canard. In
other words, if the plane rolls to the right and you want to eliminate that
rolling tendency, drop the incidence of the left wing rather than raising
the incidence of the right wing.

It's always something :-).

--
Marc J. Zeitlin mailto:marc_zeitlin@alum.mit.edu
http://www.cozybuilders.org/
Copyright (c) 2014 http://www.mdzeitlin.com/Marc/

"Marc J. Zeitlin" <marc_zeitlin@alum.mit.edu> Jul 19 09:57AM -0700