Conference vmszoo::rc

        Re:                      <<< Note 1587.0 by NYEM1::EVANKO >>>

        I'm not  an  aeronautical  engineer, so the following may contain
        some inaccuracies.
        
        First of all, there are  three  entirely  different problems with
        design:    One  is  aerodynamical  design:    selecting  airfoil,
        moments, areas, weight, incidences, etc.  etc..  The next problem
        is  structural  design:    selecting  the right spar  size,  wood
        weight,  foam  density,  formers,  etc.  etc..  The  third  being
        easthetic (sp?) giving the plane a nice shape.
        
        The impression I have with the two packages that you  mention  is
        that  they  are good at the aesthetic design, but don't give  you
        much help on the aerodynamics and structural side.
        
        As a war-scarred amateur airplane designer I can tell you that it
        is easy  to  design  a  pretty dog that's either over-designed or
        under-designed.
        
        My 2c worth!
        
        Anker

	Regarding -.1	


>        I'm not  an  aeronautical  engineer, so the following may contain
>        some inaccuracies.
        
	 Me either. 

>        First of all, there are  three  entirely  different problems with
>        design:    One  is  aerodynamical  design:    selecting  airfoil,
>        moments, areas, weight, incidences, etc.  etc..  

	 These criteria, would cause one to do alot of head scratch'n.
	 Airfoil selection would be of concern in relation to what it 
	 is exactly you want the aircraft to do. I read the airfoil note 
	 (1113) to see if I could find out what the characteristics of each 
	 airfoil are but, will need to go a little more reading on the outside 
	 to figure this out. I remember reading something awhile ago, where it 
	 described airfoils and what where the benefits/downfalls of each and 
	 suitable applications of where this airfoil could/should be used. 

	 Moments, areas, and weights, are derived from calculations on current 
	 design parameters, right ? You wouldn't know these until you 
	 had a basic planform. Then from these, one could see what 
	 would/wouldn't work and change the planform accordingly.

	 Incidences. Ah, not yet.

>							  The next problem
>        is  structural  design:    selecting  the right spar  size,  wood
>        weight,  foam  density,  formers,  etc.  etc..  

	 OK, let's look at this one.

	 As far a material selection, I've noticed that if you look at a 
	 specific size/type aircraft (56"-66" wingspan), alot of the materials 
	 such as formers, ribs, longerons, sheeting are basically identical 
	 for this size aircraft. Sure there are minor difference, but overall 
	 there are alot of similarities. 

	 Now when you move up in scale, (72"-84" wingspan), again, same size 
	 and type aircraft, material for these aircraft are basically identical 
	 in the same places. I feel that if one wanted to design a 
	 basic airframe, and depending on the size concerned, one could easily 
	 reference the appropriate plans to get an idea of what to use.

>					The  third  being
>        easthetic (sp?) giving the plane a nice shape.

	 What I am looking to do is to take a three view of an aircraft and 
	 use those dimensions as a starting point. I am not looking to 
	 design something new to the world.


>        The impression I have with the two packages that you  mention  is
>        that  they  are good at the aesthetic design, but don't give  you
>        much help on the aerodynamics and structural design.

	 I agree with you here. Aerodynamics being the main concern.


>	 As a war-scarred amateur airplane designer I can tell you that it
>        is easy  to  design  a  pretty dog that's either over-designed or
>        under-designed.
        
>        My 2c worth!
        
>        Anker

	 Don't plan on getting to the point where I'll have night sweats.
	 Just looking for ideas.

    I thought these programs would give you theoretical performance data
    based on design parameters.
    
    In other words, you could design a plane and the program would tell you
    you needed more tail surface, or the plane would scream through the air
    UNTIL you lost the engine, and then it would drop like a brick.

Well, if you're willing to take a look at plans for formers and 
such, then there should be no problem getting normal moments for 
the type of ship you're interested in. Get a catalog and look at 
what is available for the task at hand. Taking several samples 
should give you a ballpark figure for most dimensions. The rest 
falls into asthetics.

SoarTech has the Princeton windtunnel info. The NSP catalog has a 
bit of discussion on each of the popular (read: ones on planes 
they sell) airfoils.

Most of the designers I've talked to are using packages like AutoCAD
to do their plans and such and do a lot of trial and error design 
work.

Designing a plane generally involves mmaking an initially model (or 
modifying an existing one) and making it adjustable in the areas you 
want to try (different wings or engines or tail moments or...) and 
then changing things around until you find something that is close 
to what you're looking for. Then you go into asthetics and 
manufacturability (or you fly the hell out of the prototype because 
the intent was not to build many but just one).

   Regarding Steve's response, ModelCad supposedly does some of these 
calculations and let's you know what the problem is. How detailed ? Don't 
know. But if it lets you know of some type of possible problems, at least 
it's an awareness requiring further investigation.

    In RCM there is an advertisement for "Computer Aircraft Designer/ 
The Electronic Wind Tunnel", which does things like, Aerodynamic Analysis, 
Drafting, Airfoil Editing/Analysis, (etc.). I spoke to them, and as it 
stands now, whatever you design within this application, it will give 
the necessary data that we have been talking about, +/- some small percent. 
Good news is, that when the final version is released, sometime after April 
(No idea as of yet), it will have the ability to import DXF files from CAD 
systems and perform the analysis. You will also be able to perform scans with 
the final version. I guess, once this final version is out, I'm Golden.

   Also, in RCM, one of the Colonel's (Thacker/Johnson, don't remember 
off hand), has done several articles on using one of these Low end CAD 
applications. Will have to go back and see if he says anything about 
figuring it all out.

   Concerning airfoils, I seem to remember Al Casey talking about one 
of his friends (Parsons ??) using one type of airfoil for his designs 
since he was familiar with it's characteristics and performance within 
the envelope. Which leads me to believe that other designers (Platt, 
Ziroli) would also stick to the same general airfoil for their designs 
for "TYPE" aircraft. You thinks ?????

   I am currently taking an AutoCad course, and that's what made me think 
about this. You can design something bare bones, and then modify each piece 
to your liking, put it altogether, and from there do the calculations to see 
if it will work. 

    Most CAD programs are designed to help you make drawings and not to
    design things for you.  They do help you do a better job but you still
    need to have the creativity to tell the program what to draw.  The end
    result is a high quality set of plans if you give the CAD program high
    quality input.
    
    Charlie
    

    RE CAD TOOLS.... for drafting.
    
    In office here we use Unigraphics [UG] (the main tool used in Corp
    Engineering) which is fairly high tech and can be used interfaced into
    solid models etc. from 3D wire frames... and then the skys the limit. 
    But to get there for model aircraft you would have to put in a very high
    investment. I don't anyone dedicated enough.
    
    I have used UG as a staight forward drawing package for my 2M design
    "Browser". The real benifits that I found...
    	o  its very easy to stretch, move, mirror, scale, copy (several
    	   times) rotate, Etc., Etc, and so once you have a design you can
    	   quickly manipulate a different variant.
    
    	o  also I decided to formulate the body design around the radio
    	   components. Here the features for murging and smoothing lines and
    	   curves provided a classy SP.
    
    WE also use AutoCad here but more as an architechural tool.
    
    If anyone wants any specific info let me know.
    
    Trev
    
    

We use UG (10.2 at the moment) for testing the graphics boards we're 
doing 3D support for.

    Here is a real world design problem that I will face this weekend.  I
    have cut a set of pink foam cores for a 7-10 cell electric sailplane.
    The cores will be sheeted with obechi.  I am looking for the lightest
    wing that has adequate strength for the intended task(fast flyer with
    high G turns).  The wing is 65" long, has an aspect ratio of about
    12-to-1, and utilizes a 9% airfoil. All up weight will be 40-55 oz.
    depending on construction, motor and #cells used.
    
    The question is - "To spar or not to spar?"  In other words does this
    wing need a spar, if so how many, of what construction(spruce, carbon,
    etc.).  Will the foam be sufficient to act as a shear web, or should
    a full depthe shear web be used?  If so what thickness, what material,
    etc?  Maybe laminating a thin carbon skin to the obechi during wet
    layup would eliminate the need for a spar and greatly simplify
    construction.
    
    This is not a totally rhetorical question, as I would love to be able
    to enter this information into a computer and have the magic answer
    pop out maybe as follows: - 1) 2# Blue foam wing with .04" obechi
                                   and .004 carbon skin
                                2) 1.4# pink foam with 1/4" spruce spar
                                   to 1/2 span, bottom surface only.
    
    Unfortunately I do not think things will ever be this easy.  The 
    program would certainly be large and complex(more than $100).
    
    British F3B champ Nic Wright who holds a degree in mechanical
    engineering has developed software aimed at designing the
    "optimal" F3B model.  Given Nic's extensive empirical experience,
    I believe his program would be quite good, but in the end 
    empirical testing will always override straight calculations.
    And Nic ended the interview by saying that his many successes and
    failures have given him a "gut feel" for what will work and what
    won't.
    
    So in the end I think computers are great for crunching the numbers
    (stress calculations and such) but cannot be substituted for 
    experience with flying machines. 

Smooth turns would be xG's while you need something that will survive 
panic pull-ups of 2xG's 8^)

If all your high G loading is going to be positive, consider a piece of 
.007 CF laminate as a vertical spar at the high point with a 1" piece of 
CF on the bottom skin over this. That combined with the obechi should be 
pretty bullet proof with little weight penalty over the CF on the sheeting 
alone. I'd just slit the panel and insert it from the bottom. It doesn't 
have to be full depth.

The foam provides little strength contribution so blue versus pink versus 
white is noise level. Extruded foam is better since it bonds the skins 
better (more continuous contact) than beaded foams. The strength of this 
configuration is in the stiffness of the obechi spanwise when held in a 
curved form. The curve of the airfoil with contribute to the non-flexing 
of the wing and the top skin will be in compression while the bottom wing 
is in tension. The typical failure in this mode is delamination of the 
top skin followed by compression of the rest of the structure. The vertical 
CF will take some of this load while the bottom sheet CF will prevent the 
vertical CF from punching through the bottom skin and also help with the 
tension on the bottom skin. CF on the top sheeting will help partially 
with the compression loading but continuous bonding is required to prevent 
delamination failures. Another strength improver is light weight fiberglass 
(.6oz or .75oz) under the TOP sheeting which improves the strength of the 
sheeting in the curved shape (which is why the top skin is better than the 
bottom one for this feature)

All this info is based on experience, reading, and some failure testing I 
did about 2 years ago. In almost all my test cases, it was the delamination 
of the top skin that caused the failure. Right up to where the top skin 
pulled foam away with it.

Oh yeah... sand the CF laminate to put some "tooth" on it for the glue to 
grab. This seems to come through with one side very smooth from the layup 
process.

As a simple exercise, take a sheet of paper and hold it curved in an 
inverted U and then press down on it between two books or other supports. 
You'll be impressed with the difference over a flat sheet.

BTW: if you loading will be both positive and negative (fully aerobatic) 
then just turn the wing over and apply the same logic to both sides 
uniformly.

    re: -1
    
    Jim,
    
       Thanks for the info., I particularly like the idea of the light
    fiberglass laid up on the top surface.  I think I will gamble with
    this wing and lay up 1.5 oz. fiberglass on the top surface with
    a strip of .007 CF to half span on the bottom.  If I break this
    wing, I will report along with the move that resulted in the
    destruction.
    
    FWIW - USA F5B member Bob Sliff told me that testing of the SE-XX
    proceeds as follows:
    
   - Mold wing with desired combination of Rohacell, carbon, fiberglass.
    
   - Place plane in vertical high speed dive  
    
   - Pull out with full up elevator
    
   - If wing breaks, it's back to the drawing board
    
   *Sliff claims the current SE-XX cannot be broken by any imposed flight
    load!

Well, we all know that it's not IF it breaks but WHEN 8^)

please note any delamination that takes place and if there were any voids 
in the location. Dry layup is something you have to watch out for when 
trying for super lightness.

high speed dives will also sort out flutter tendancies.

With the 1.5oz glass you should be able to eliminate the use of the 
trailing edge glass. I have about 10' of 12" wide 2.9oz CF that I'm 
considering using this way. Haven't decided whether to use it with glass 
or obechi.

    I have a wing design question:
    
    -If you could only place a single spar in a wing, on either the upper
    or lower surface, which would you choose?  The plane will sustain
    high positive G loads.  Reasoning behind the choice appreciated.
    
                           
                                                    Thanks,
    
                                                    Jim 

>    -If you could only place a single spar in a wing, on either the upper
>    or lower surface, which would you choose?  The plane will sustain
>    high positive G loads.  Reasoning behind the choice appreciated.

	I choose upper spar.  It will do against pressing force. 
	You may want some fiber on lower surface which will withstand for
	stretch.  This combination will cope with high G.

	Katsuya

    The reason I asked the question was I have seen conflicting examples
    from commercial products as follows:
    
    1) My Weston Waco 550 came with a single spruce spar on the upper 
       surface.
    
    2) "We recommend that you build the Sunfly's wing as a one piece 
       structure with a length of carbon fiber on the underside." -
       quoted from Hobby Lobby Catalog, referencing the Aeronaut
       Sunfly electric sailplane.
    
    I am trying to get a better understanding of the forces that affect
    a wing(tension,compression,shear).  Given the variety of construction
    variations, as illustrated above, I am not sure if I have a correct
    understanding of the forces.
    
    By my thinking, a wing undergoing a high positive G-load(such as
    a dive with steep pullout) experiences tension on the bottom surface
    and compression on the top surface, the difference in direction of
    these two forces representing the shear.
    
    If this scenario is true, I would agree with Katsuya's recommendation
    to use an upper surface spar.  This obviously assumes that the wing
    is more likely to fail in compression than tension.
    
    Is this correct?
    
    
    Thanks,
    
    Jim

I agree with .14. A spar on the top surface will help with compression. The 
problem is that foam itself fails under tension so you still need a tension 
spar on the bottom surface. I agree with the combination CF/kevlar on the 
bottom and a wood (basswood/spruce/balsa) on the top.

Your examples 1 and 2 are really comparing apples and oranges. Each 
manufacturer has chosen the best material for their spar location choice.

I'm lazy so I use CF in both locations to avoid cutting the wood spar slot.

    re: -1
    
    Jim,
    
       Do you just lay CF tape(.007") between the foam core and the
    wood sheeting?  Or do you route in a thin trough to lay it in.
    
    Thanks,
    
    Jim

I typically use cloth, not laminate so I just lay it between the skins 
and foam and press. I don't notice any ridge. For the vertical I talked 
about earlier, I DO use laminate. The cloth I have is 4.5oz/ft and only 
.004" thick. Since the obechi is .024" thick, a quick sanding leaves no 
trace.

    On my pattern wings I used the .007 CF and I sanded a groove for it.  I
    glued a piece of 400 paper .5 inch wide onto a hardwood block.  I ran
    the block along a straight edge and it cut a perfect groove.  The depth
    was the thickness of the paper.  I then sheeted with 1/16 balse using
    Epoxy.
    
    Charlie
    

    I guess that we are in agreement that a wing under a positive
    G-load places the upper surface under compression and the lower
    surface under tension, with the difference between these two
    forces being the shear force.
    
    In built up construction, typically end grain balsa is glued between
    the upper and lower spar to keep the spars from separating(hence,
    buckling the wing) under load.
    
    I guess that it would be safe to assume that an upper spar glued to
    a foam wing core need only pull away from the foam to cause a failure.
    It seems that if the upper and lower spar are not tied together with
    a shear web of substantial strength, then the spars with foam in
    between acting as the shear web are not doing that much.
    
    Does this make sense?