X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Tue, 18 Aug 2009 16:32:39 -0400 Message-ID: X-Original-Return-Path: Received: from QMTA10.westchester.pa.mail.comcast.net ([76.96.62.17] verified) by logan.com (CommuniGate Pro SMTP 5.2.16) with ESMTP id 3817213 for lml@lancaironline.net; Tue, 18 Aug 2009 14:56:02 -0400 Received-SPF: pass receiver=logan.com; client-ip=76.96.62.17; envelope-from=mjrav@comcast.net Received: from OMTA02.westchester.pa.mail.comcast.net ([76.96.62.19]) by QMTA10.westchester.pa.mail.comcast.net with comcast id Vnnd1c0050QuhwU5AuvWQD; Tue, 18 Aug 2009 18:55:30 +0000 Received: from mjr ([66.30.29.210]) by OMTA02.westchester.pa.mail.comcast.net with comcast id VuvT1c0034Y010q3NuvVn0; Tue, 18 Aug 2009 18:55:30 +0000 X-Original-Message-ID: <001501ca2035$aa3c6240$6401a8c0@mjr> From: "Mark Ravinski" X-Original-To: "Lancair Mailing List" References: Subject: Re: [LML] Strength vs. stiffness X-Original-Date: Tue, 18 Aug 2009 14:56:59 -0400 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0012_01CA2014.2307A9E0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1983 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1983 This is a multi-part message in MIME format. ------=_NextPart_000_0012_01CA2014.2307A9E0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Gary, I suggest you make your aft fuselage a full cone shape by closing the = vertical tail opening. And I agree to use glass for this work. I (and several others) found this was needed after cracks formed near = the top of the fuselage at the tail. It's a bugger of a place to work. It would be easier with the plane = inverted. I'll send some pics if you need them. Mark Ravinski 360 1470 hrs ----- Original Message -----=20 From: Gary Casey=20 To: lml@lancaironline.net=20 Sent: Tuesday, August 18, 2009 11:24 AM Subject: [LML] Strength vs. stiffness I lost the original question on this, but someone asked about the = wisdom of adding stiffness to the fuselage and whether that would make = it more likely to break. Good question, but the answer is not = straightforward. Generally, stiffness and strength go together, but = they are actually different things: A nylon rope might be strong, but = not stiff while a clay pot is stiff but not strong. Yes, if you make = one element in a complex system stiffer than the others it will absorb a = disproportionate amount of load and might be subject to failure. For = instance, I added unidirectional Kevlar "straps" to my aft fuselage and = they are probably stiffer than the glass fabric under them. They will = absorb more of the force exerted and yes, they might break first or = possible de-bond from the substrate during an overload condition. It = would have been better to design the whole system out of Kevlar, but I = suspect the straps do increase the overall strength and especially = stiffness. Why do we want the fuselage to be stiff, especially in = torsion? Everything that has stiffness and weight has a resonant = frequency - the frequency at which it will tend to vibrate. The tail = will torsionally vibrate at some frequency and this frequency is lower = than it could be because of all the lead weights at the tips. If that = frequency is too low it could be excited by any number of things, one = being aerodynamic loads, possibly resulting in destructive flutter. = Since all three control surfaces are overbalanced it will not likely be = excited by a control surface itself. When I studied the shimmy issue I = looked for things that could have a resonant frequency in the range of = the observed shimmy. Three obvious ones came to mind. There is the = resonance created by the combination of nose strut/mount stiffness and = the gyroscopic precession forces of the wheel tire, a very complex = system. Another is the resonant frequency of the engine in its = isolators. The third is torsional resonant frequency of the tail, which = isn't has high as you might think. To demonstrate, take you fist and = pound on the tip of the horizontal stabilizer - I haven't measured it, = but it appears to be 10 Hz or less. That is in the same range as the = engine. What would happen if the excitation (the nose wheel turning = side to side) was the same frequency as one of the other major resonant = frequencies? It might trigger a catastrophic shimmy event. Why would a = shimmy trigger a torsional vibration of the tail? To start with the = shimmy puts a torsional load into the fuselage and besides inputs a = lateral, or steering load. The CG of the tail is above the neutral axis = in torsion and therefore when the tail goes side to side it generates a = twist. The two forces actually add together, increasing the effect. = The resonant frequency of the nose gear changes with speed, but what = happens if the two other large masses, the tail and the engine have the = same resonant frequency? When the resonant frequency of the nose gear = matches the other two the vibration could get severe. One basic goal of = elastic design is to separate the various resonances so no two can be = excited at the same time. Skip all that - I assume the tail is strong enough (I think), but it = could always be stiffer and I like stiffer. The graceful curve of the = aft fuselage maybe impart slightly less drag and certainly looks good, = but a simple conical shape (Piper Cherokee) is probably more = structurally sound. Gary ------=_NextPart_000_0012_01CA2014.2307A9E0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Gary,
I suggest you make your aft fuselage a = full cone=20 shape by closing the vertical tail opening.
And I agree to use glass for this=20 work.
I (and several others)  found this = was needed=20 after cracks formed near the top of the fuselage at the = tail.
It's a bugger of a place to work.  = It would be=20 easier with the plane inverted.
I'll send some pics if you need = them.
 
Mark Ravinski
360  1470 hrs
----- Original Message -----
From:=20 Gary=20 Casey
Sent: Tuesday, August 18, 2009 = 11:24=20 AM
Subject: [LML] Strength vs.=20 stiffness

I lost the original question on this, but someone asked about the = wisdom=20 of adding stiffness to the fuselage and whether that would make it = more likely=20 to break.  Good question, but the answer is not straightforward.=20  Generally, stiffness and strength go together, but they are = actually=20 different things:  A nylon rope might be strong, but not stiff = while a=20 clay pot is stiff but not strong.  Yes, if you make one element = in a=20 complex system stiffer than the others it will absorb a = disproportionate=20 amount of load and might be subject to failure.  For instance, I = added=20 unidirectional Kevlar "straps" to my aft fuselage and they are = probably=20 stiffer than the glass fabric under them.  They will absorb more = of the=20 force exerted and yes, they might break first or possible de-bond from = the=20 substrate during an overload condition.  It would have been = better to=20 design the whole system out of Kevlar, but I suspect the straps do = increase=20 the overall strength and especially stiffness.  Why do we want = the=20 fuselage to be stiff, especially in torsion?  Everything that has = stiffness and weight has a resonant frequency - the frequency at which = it will=20 tend to vibrate.  The tail will torsionally vibrate at some = frequency and=20 this frequency is lower than it could be because of all the lead = weights at=20 the tips.  If that frequency is too low it could be excited by = any number=20 of things, one being aerodynamic loads, possibly resulting in = destructive=20 flutter.  Since all three control surfaces are overbalanced it = will not=20 likely be excited by a control surface itself.  When I studied = the shimmy=20 issue  I looked for things that could have a resonant frequency = in the=20 range of the observed shimmy.  Three obvious ones came to mind.=20  There is the resonance created by the combination of nose = strut/mount=20 stiffness and the gyroscopic precession forces of the wheel tire, a = very=20 complex system.  Another is the resonant frequency of the engine = in its=20 isolators.  The third is torsional resonant frequency of the = tail, which=20 isn't has high as you might think.  To demonstrate, take you fist = and=20 pound on the tip of the horizontal stabilizer - I haven't measured it, = but it=20 appears to be 10 Hz or less.  That is in the same range as the = engine.=20  What would happen if the excitation (the nose wheel turning side = to=20 side) was the same frequency as one of the other major resonant = frequencies?=20  It might trigger a catastrophic shimmy event.  Why would a = shimmy=20 trigger a torsional vibration of the tail?  To start with the = shimmy puts=20 a torsional load into the fuselage and besides inputs a lateral, or = steering=20 load.  The CG of the tail is above the neutral axis in torsion = and=20 therefore when the tail goes side to side it generates a twist. =  The two=20 forces actually add together, increasing the effect.  The = resonant=20 frequency of the nose gear changes with speed, but what happens if the = two=20 other large masses, the tail and the engine have the same resonant = frequency?=20  When the resonant frequency of the nose gear matches the other = two the=20 vibration could get severe.  One basic goal of elastic design is = to=20 separate the various resonances so no two  can be excited at the = same=20 time.

Skip all that - I assume the tail is strong enough (I think), but = it=20 could always be stiffer and I like stiffer.  The graceful curve = of the=20 aft fuselage maybe impart slightly less drag and certainly looks good, = but a=20 simple conical shape (Piper Cherokee) is probably more structurally=20 sound.
Gary

------=_NextPart_000_0012_01CA2014.2307A9E0--