thickness percentage for composite

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
I looked in the archives for this but don't seem to find much on this topic.

The original Tipsy Nipper wings were made of typical wood and fabric construction and the main spar was a massive laminated heavy built up type identical the the Corby Starlet and th "One Design".

For my version I would like to use composite construction. I have a complete set of plans for a Rutan LongEz and the wing construction he used is something I am considering for my hershey bar wings. The question I have is thickness %.

Looking at lift and drag charts I really aught to figure out the thickness I should be shooting for before delving into airfoil selection. Using some quick math it seems on composite wings 15-18% thick can really make out for a heck of a strong wing. Using Rutans solid wing construction is there any rule of thumb as to a ballpark minimum thickness.

The wing size is yet to be decided but it seems the span will be about 18-20' long with about a 54" cord. Estimated gross weight will be around 700 lbs. Cruise I figure would be about 120-130mph. I'm not looking for someone to figure it out for me because that's half the fun but I seem to be having trouble finding information as to what would make for a real strong wing using this method. Based on charts, 15-18% thick on some airfoils seems to offer just a slight increase in drag.

I plan on building a wing then testing until total failure. Hopefully on a youtube video.
Mike
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
With the thicker wing, you can get your strength with less weight, and if you need it, you get volume for things like fuel storage or retractable gear. Your bird really does not need the volume for either, so it really comes down to a tiny increment of drag vs a lower empty weight. Many folks, including Harry Riblett, will tell you there really is no point in going thinner than 15%, and up to 18% comes at very little drag penalty.

The real issue with optimization comes down to figuring out what your performance needs are, and then finding the lightest way to get that. Here are a couple things to recognize. The weight of the spar (for a given gross weight and g rating) will go with the first power of the wing span and with the inverse second power of thickness (not percent thickness). The airplane's top speed down low will go with the inverse of wing loading, while its max altitude and climb rate will go with the inverse of span loading. Hmmm. Most of the rest of wing weight will go with the first power of wing area...

If I was building a fiberglass wing for a 90 -120 knot airplane, I would build it with a 15% (minimum) thick wing, little taper in the inner half, a 30% taper in the outer half, and with an aspect ratio between 6 and 9. Yeah, its planform would look like a C172 wing... That will keep the weight down, the climb effectiveness up, drag down and good roll rates.

If I was building a fast cross country airplane for spanning this continent, the wing would be about the same, except that it would be around 17-18% thick, and aspect ratio of 10.

Now if it was to be metal, I would likely build with no taper. Why? Multiplicity of rib tools will make you crazy. In fiberglass, there is no penalty, only benefits to tapering, and it is no more difficult to build.

And why are you thinking about making a really strong wing. You want to build the whole airplane to be sturdy at whatever g-s you decide it to be. Building a part that will carry more g's than anything else in the bird just makes it heavy.

Billski
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
Hi Billski, how's it going?

Here are two photos of the original Tipsy Nipper wing. You can see that it's a one piece wing joined by the front spar. The wing is mounted on the top of the steel tube fuselage by 1 bolt on the left and right side. on the rear of the wing mounting tabs also allow the wing to be joined by means of a bracket.

As far as strength I want it to be plenty strong because I don't yet fully understand the limits of composite construction. Another reason is the way the wing is mounted to the fuselage. Somehow I have to get two bolts through the spar.

From what I understand you think 15-18% thick is a good number to shoot for. I have also heard this as well. I am trying to read and understand the book "theory of wing sections" and although a bit complicated, from what I understand the drag penalty of many of the common airfoils at 15-18% really isn't that much like ya say.

I do think I want to use a simple rectangular wing.
 

Attachments

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
And I am trying to tell you that the only good reason to build a rectangular wing is ease of construction, and that it does not apply to building a Rutan style wing. It does apply to airplanes with wood or metal ribs. If you just want a rectangular wing, knock yourself out...

One of things that you want to avoid is tapering the % thickness. This leads to thin tips which tend to misbehave. But if instead, you keep a constant % thickness while you taper the foil toward the tips, the world has found that the foil will still stall cleanly and behave well.

If I were building an airplane with an absolute minimum of power, meaning it has to fly at very low airspeeds, low wing loading is essential to flight. The Wrights figured this one out and so did Dr Macready (sp?).

Once you are into more normal airplanes, it all becomes a trade of landing speed vs top speed vs altitude capability and keeping the weight down. So, if you want to mimic the plan form of the Tipsy Nipper in fiberglass (instead of building it tapered), it will have more roll inertia, the ailerons will require bigger control forces than they could have, it's max speed in level flight will not be as high, and it will weigh more than it could have if you would have just built a tapered wing. I am even talking the same span and area, just making the root a little larger, and the tip a little smaller.

To help you make sense of all this optimization, you not only need wing theory, but you also need some structural stuff. For a Rutan wing, it is not hard but you will need to do some beam theory and shear flow theory.

Basically, you need to calculate the needed spar cap and shear web thicknesses at a number of places along the wing based upon the build up of shear and bending as you go inboard. The wing skin will be three UNI cloth at 50% cloth/50% epoxy everywhere, the foam is 2 lb/ft^3. When you do this with the wing rectangular and again with it tapered, you will find that your shear web is about the same total weight, but the caps (and most of the weight) get lighter quickly as the root section gets bigger. And the moment of inertia drops rapidly too. Will the ailerons be smaller? Yeah, but so will the wing tips, and the limits on roll rate come down to the helix you can draw with the foil and deflected aileron, so smaller tips don't need as large an aileron...

Then you can set up the math on an Excel spreadsheet. Use a design factor of safety of 2.0 for composites, and the real world of composite design ends up with the spar caps being almost double what basic theory says you will need, while the webs need to be about 1.5 times the raw value.

As to attaching the wing to the fuselage in this way, all of the shear loads from wing torsion will have to be transferred into the main spar and maybe into a drag spar. At the attach points, you have choices. If the bolt holes run fore-aft, you can just build in some laminated wood cores in the fiberglass webs, thicken the web locally, bolt sandwich plates through the web, and then machine some flanged bushings for the bolts. Like everything else, this will require some analysis. If the bolts can be oriented up-down, you will layup some pretty substantial lift tabs.

Oh, if you decide that you would like a little more span to give you some more altitude capability or further reduce induced drag, tapering the wing helps you avoid adding too much more weight due to the increased span. Hmmm. And your wing chord goes down somewhat if you keep the same wing area, so your trim drag drops too, further helping with speed and altitude...

Ah well, be careful. Optimization can be a tempting exercise in itself...

Billski
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
Billski, Thanks for the interesting post! I'm going to re-read it a dozen more times.:grin: Just what I need.......more to think about.:gig: I do understand what your saying and it actually is starting to make sense to me.:ban:

Back to the books. Mike
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
While it's true the Rutan method doesn't reward straight wings, there are other composite construction techniques that do. For instance, if you mold your skins in sections a la Strojnik. And with the aspect ratio so low, I don't think the penalty of a straight wing on handling will be anything like that on a long skinny wing. BTW, the three view of the Tipsy that I saw shows a moderate taper.


Since the Tipsy has a low aspect ratio, and therefore a large chord, if you use a 16 percent thick airfoil and use the Rutan method, you're going to have a very large fraction of the weight of the aircraft in blue foam. Maybe 80 lbs of it! I think if you're going to use the Rutan method, and you went to a thin airfoil, you'd save a lot more on foam weight than you'd spend on spar cap weight. I'd venture to say that compared to the skin, foam, ailerons, etc. that the weight of the caps would be almost insignificant. I think a hollow method would probably be much lighter, whether that was composite or some other method. You could make a high aspect ratio wing, but then it wouldn't be anything like a Tipsy. I suppose composites would be good if you wanted to make the wing really clean, but there's no point in that unless you redesign the fuselage extensively as well.

It's my guess that an aluminum spar and a wood or metal wing would be easiest and somewhat lighter. The aluminum spar would probably be easy to bolt to. Maybe later I could figure out what an aluminum I beam might weigh. Of course, you might want to taper it somehow, but I'm guessing that it would be pretty light in the first place and wouldn't be costing you all that much weight. Another nice feature is that if you didn't put too many nasty holes in it, you could be really sure how strong it was even if you didn't test it to destruction.

I also think if you're not trying to go super low drag and laminar flow on everything, that the original airfoil (wasn't it something like a 43016?) would be a good choice and you wouldn't easily find something better. At speed, the fuselage is probably where the drag is anyway.
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
lr27, what you said here really made me think, or re-think anyway. Ya know...... my Taylorcraft has a real interesting wing. Wood spars and alum. ribs. The wing as a whole is very light weight. A composite was my first thought because of fuel storage issues. Tipsy owners complain about only a 7 gal tank kept in front of the spar crossover. 7 gallons of fuel doesn't work well for me and if I had a tapered wing with fuel storage composite wings seemed to make sense. You are correct about the Tipsy having a tapered wing. I need to study my options a little further on what the lightest weight wing would be while serving my needs and being easiest to build. Aluminum is something I have not really thought of yet but may well be a good option. What I know for sure is although Tipsy owners love the airplane for the aerobatics I would like it to be a good all around airplane first, aerobatics second. Thanks for the input, more for me to think about. Mike
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
Wow, thanks Seb, what a cool find. More to read:grin:. I'm going to spend some time looking and learning what I can from that. Much appreciated. Mike
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
If you don't already read Peter Garrison's stuff, you might find some items of interest below:
Flying Magazine
Melmoth 2
Peter has designed, built, and extensively flown two aircraft. One was metal and the other is composite. Keep in mind that he does tend to get pretty fancy, and ends up putting a LOT of work into his aircraft. On the other hand, his stuff seems to work. But he might not be a guide to designing as simply as possible.

He's also good at presenting reasonably accurate explanations of aerodynamics that are fairly simple.
---------------------
The Taylorcraft wing is light partly because of the struts, which greatly reduce the required strength.
---------------------
I'm not saying all composite wings are bad, but if the insides are all 2 lb. foam, then a Rutan style wide chord wing like this one will be heavy. I'm sure there are lighter ways.

Many years ago, Sport Aviation ran a multi part article exploring weight of composites vs. metal (and maybe other materials, I can't remember). Turns out the lightest was aluminum with stringers. Of course that's with good design practices, it's not an automatic result. And the differences were not enormous. Plus I'm guessing the results might be a little different depending on who the designer was. For example, Voyager was composite, and had an enormously high gross weight to empty weight ratio of 4.3, even though it also had a pretty high aspect ratio.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
I just did a first estimate calculation. I found the Tipsy Nipper's data, and the plan view shows a modest taper.

With a constant chord wing, at 15% thick, the wing will be 7.38" thick, and the foam will be 72ft^2*(7.38/12ft)*2lb/ft^3*0.62 = 55 pounds of foam. Go up to an 18% thick wing, and the foam grows 20% (11 pounds), go down to 12% and it loses 20% (11 pounds)... So yes, foam weighs something. The decision on thickness hinges on how much you take out of the spar vs the foam weight.

I did a quick check on it too, and the spar only changes about 3 pounds when you go from 12% to 15% in this case. So, for the Nipper, it pays you to go thinner with a foam wing...

At some point it does pays you to build hollow fiberglass instead of solid foam. But where is that break point? We have already talked this one to death elsewhere. Hollow wings, besides requiring tooling and vacuum bagging to make, need the same skin on the outside no matter what, and add a skin in the inside of the core foam, ribs and flanges and a drag spar. So we take out 55 pounds of blue foam, but we put in two BID plus 3/8" of 3 pound foam on both the top and the bottom, and the epoxy to attach the glass.

14oz/yard*2 plies*2 skins*2.0(for epoxy)/9 ft^2/yard = 0.72 lb/ft^2
3.0 lb/ft^3*3/8"/12"*2 skins = 0.18 lb/ft^2
So you would be adding 0.90 pounds per square foot on 72 sqare feet or about 65 pounds. So making it hollow has already added 10 pounds, and we don't have any adhesive holding the spar to the skin, or any ribs. It looks like hollow will be heavier. A fiberglass airplane wing has to be pretty big to justify being hollow over being solid cored. Space Shio One and the White Knight had hollow wings, but solid core control surfaces. Other reasons to go hollow are places to put fuel, landing gear, control bays, etc. Not part of a Nipper...

One other factor in what thickness and foil to build - It looks like the spar goes through the fuselage near the panel, and is forward of the thickest part even of 23012 foil. You spar depth will go down as you shove the spar forward, raising the spar weight.

Enough for now.

Bill
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
Hi Bill, you have made some interesting points.

The data I have on the Tipsy shows similar findings as it concerns taper. The leading edge is about 2 deg. and the trailing edge is close to 5 deg. Last night I spent a couple of hours researching composite wings and there weights. I was shocked to discover how much weight variation there is depending on wing length, cord, and thickness. Your right when you say foam does weigh something.:shock: I somehow came up with math that said the wings weighed lighter than what you came up with, I'm going to go over it again tonight and see where I made the mistake in adding it all up.

There must be some point where a designer searches out the best means of construction that lends itself well to a particular design, be it aluminum, composite or wood. Based on your math and mine aluminum might be a lighter method. ARRR! More research. Thanks for the time you spent, Mike
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
Fiberglass allows better aerodynamics and can generally make for a faster airplane because of that.

Aluminum will generally allow a lighter airplane than fiberglass.

The big reason to build in a medium is because you like working in it and can communicate with it. When you like it, it is fun. If you don't like it, the build becomes a job. Bad thing to do to your hobby...

Have fun with your design!

Billski
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I didn't get the same answers as Bill. Figures I saw said about 20 foot span, 80 square feet. As a total guess, I'm saying that maybe the foil has 75 percent as much area as a box drawn around it. So 80/20 is 4 foot chord time 4 times 0.16 times 0.75 is 1.92 times 20 is 38.4 times 2lbs/ft^3 is 76.8 lbs of foam. Of course some of that foam is replaced by something heavier.

I don't think you necessarily need two layers of 14 oz glass on either side of foam. I think those 4 layers wouldn't necessarily need any foam. Strojniks single thickness skins were something like a millimeter. (.039"). I think that's probably like 3 layers or less. Still, it probably would make sense to make the rear part of the wing fabric covered.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
Aerocrafter: Span is 19.67ft, area is 80.70ft^2, so chord for a rectangular wing is 4.1 feet.

For foam, I made the assumption that 2 ft is in the fuselage, so 17.67 ft * 4.1 ft is 72 ft^2 of wing filled with foam. Front half of the wing section looks like an ellipse. Area of an ellipse is d1*d2*PI/4, divide that by two to get an approximation of the front half. Back half looks like a triangle, d1*d2/2, divide that by two. Add them up and 0.62*t*c is a reasonable approximation for section area of foam on an unknown airfoil. Can you get closer? Yeah, tell me the foil and we can work it up too, but this is close enough for trade off study.

When I have done the computations (admittedly for much faster airplanes) for stress in skin panels with aero loads on them, plain layup skins (no cores) end up needing small panel sizes between supported edges, which translates to frequent ribs and longerons, which generally means more weight than a foam cored skin does. And when you remember that the aero loads trying to take the wing skins away from each other are carried by the ribs, then you have to design the inside skin to move the load from each skin to the rib.

To optimize a composite wing, I would start with four BID/ one UNI outer skin with no cores and figure the rib and spar count, then do a 1/2 inch core with 3 UNI outer and 2 BID liner and do the ribs, and do a solid foam core covered in 3 UNI. Compare weights. After that, you can do the in betweens (if they looks like they might be favorable).

It could be that the Tipsy Nipper and the Alex Strojnik's birds are slow enough that the uncored skin and its necessary ribs and spars is the light way to build this wing. But I am skeptical. Here is why:

Deflections go with stiffness and panel dimension squared - Skin (no cores) has stiffness that goes with thickness cubed and E. For BID at 0.040, 0.040^3*2.3Mpsi is 147. For aluminum, 0.025^3*10.5Mpsi is 164. For the same deflections as aluminum, your panels will need to be slightly smaller than they would be with aluminum. Have you ever looked at an aluminum wing in flight and seen how much the skins buldge between ribs and longerons? Maybe you are not worried about skin deflection spoiling laminar flow...

Stresses go with thickness squared times E and panel size squared. Glass is 0.040^2*2.3Mpsi = 3680, and aluminum is 0.025^2*10.5Mpsi = 6562. Your panels will need to be about 3/4 the size of if the wing was aluminum...

I really don't like the idea of designing to use a bunch of ribs. Skin deflection will spoil laminar flow, the build gets difficult because of all the ribs, they end up heavy, and the skin ends up with substantial bumpiness and/or a lot of micro. Many birds out there use 3 UNI on the outside of the core and, if it is hollow, 2 BID inside. It is about the mininum. Use a reasonable core, and it only needs ribs to close the ends and maybe a couple in between... And it usually is lighter too. But each of us has to do our own optimization...

Now Handprop, if you want to build a rectangular wing hollow, I have a simple technique for making the wing skin molds that only took me about 10 hours per mold, and you only need two. But you can skip that nasty mold work by building solid core, and replicate exactly the mild taper of the original with a simple fun build.

Billski
 

handprop

Well-Known Member
Joined
Apr 17, 2009
Messages
156
Location
wisconsin
Hi Billski

I read your post today a few times and I am still absorbing the info. I appreciate you doing some of these calculations, it's kind of nice because when I see the solution you come up with it allows me to work backwards from you answer and figure out how you came up with it. At least I have a starting point for my own calculations and reference.

Most of what you said makes sense to me the more I read it. It's a little tough for me to fully understand composite construction based on internet information I can only read about, so I got real lucky and was able to contact a fella building a Cozy composite about 60 miles from my home and tomorrow night he is having me over for a few hours to help him lay-up composite wings. I can't wait because once I can hold parts and work with it a little things start to make a little more sense. Beyond that I once again ordered more books on composites. Composites really seem like a fun means of building.

You mentioned you have a trick for making wing skin molds. Would you happen to have any pictures you could post, it sounds real interesting? Mike
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
Between Strojnik's speed record airplane, using an engine installation that looks like it has really lousy propellor efficiency, and the reported performance of the S2 sailplane, I think we can be pretty sure that his building technique results in laminar flow. As far as wing skins with no cores, what about 4 layers uni, perpendicular to each other, vacuum bagged, or maybe even infused? You ought to be able to get quite a bit stiffer that way. Strojniks ribs aren't all that heavy, either. Just foam.

Also, Schreder's sailplane wings were aluminum over foam ribs. They seem to have a pretty good record on performance, too.

I suspect that the foam cored skins could be thinner and still work. If you made them thinner, you could make the foam denser, and then the skin could be lighter. They'd still be a lot stiffer than .025" aluminum, and they wouldn't dent under finger pressure.

Not quite sure why a lot of micro would be necessary if you made the skin panels in a mold and were careful with the ribs. The accounts of building Rutan style sure seem to include a lot of micro, though. However I gather that the better the workmanship, the less micro is required.

However, there's really no point in going laminar on a Tipsy unless the fuselage is completely done over too, and the landing gear is properly faired, and lots of other aerodynamic optimization is done.


Aerocrafter: Span is 19.67ft, area is 80.70ft^2, so chord for a rectangular wing is 4.1 feet.

snip
It could be that the Tipsy Nipper and the Alex Strojnik's birds are slow enough that the uncored skin and its necessary ribs and spars is the light way to build this wing. But I am skeptical. Here is why:

Deflections go with stiffness and panel dimension squared - Skin (no cores) has stiffness that goes with thickness cubed and E. For BID at 0.040, 0.040^3*2.3Mpsi is 147. For aluminum, 0.025^3*10.5Mpsi is 164. For the same deflections as aluminum, your panels will need to be slightly smaller than they would be with aluminum. Have you ever looked at an aluminum wing in flight and seen how much the skins buldge between ribs and longerons? Maybe you are not worried about skin deflection spoiling laminar flow...

snip
I really don't like the idea of designing to use a bunch of ribs. Skin deflection will spoil laminar flow, the build gets difficult because of all the ribs, they end up heavy, and the skin ends up with substantial bumpiness and/or a lot of micro. Many birds out there use 3 UNI on the outside of the core and, if it is hollow, 2 BID inside. It is about the mininum. Use a reasonable core, and it only needs ribs to close the ends and maybe a couple in between... And it usually is lighter too. But each of us has to do our own optimization...

snip

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
Several points on glass.

I am postulating here, but it appears that we really need 3 plies for general sturdiness and to avoid perforations by air show morons and flying critter strikes. How to distribute that? Well Burt tells us 1 UNI the long way and 2 UNI at +/-45 is about right. Other people do similar things... This gives us some bending and torsional strength. I took that hint.

When I decided to vacuum bag, I looked further into cloths under vacuum bagging. I found that when you vacuum bag woven cloth it is lighter than in open wet layups, but not as light as you might expect. The interstitial spaces between woven yarns gets filled up with resin. But when I vacuum bagged BIAX and TRIAX, I got still lower resin fractions with laminates that are fully epoxy loaded. You do have to design with them in mind because TRIAX has roughly half of its fibers at 0 degrees, half at +/- 45 degrees...

Micro on solid foam cored wings is supposed to be for smoothing not building. The right way to do it is shown here - Finishing a composite airplane and the canard builders all talk about the Prime Directive (put it on once, take it off once). What remains on the airplane is pretty darn thin. You also do this with cored panels and uncored panels. With cored skins, it works pretty much the same as with solid cores - they are stiff and as accurate as your mold, and I sure hope your mold was accurate. With uncored skins, your ribs will try to print through the skin, and sanding forces will distort the skin between ribs too. You are likely to stop sanding with more micro still on the wing than with cored skins or solid cores, and you are likely to have more waviness when done too.

Some folks build with uncored skins by laying them up on glass, then bending them over the ribs. Well, there is the phenomenon that occurs when you do this with plates - just look close at a factory built airplane or most plywood skinned wings - the edges will try to be higher than the middle, and you will get inward sags between ribs. You will have to fill that sag if you want the surface straight and accurate. It comes down to aesthetics if you are not trying for laminar flow. Anyway, at four pounds to the gallon for dry micro, you are around 8 pounds for a 0.020" thick fill on the Tipsy Nipper wings. Maybe double that for a bumpier structure. Not huge. The builder has to figure out what is important to themselves...

I know that Strojnik's airplanes did amazing things and he probably made laminar flow over a lot of the bird. No arguments there. His airplanes flew on close to min power. Min power airplanes are first about low wing loadings, then about low drag at those low wing loadings. When your wing loadings are low and your q is low, the panel deflections may be small enough to maintain laminar flow. The laminar target for waviness I recall is 0.004"/2". My concern is that with soft skins (uncored skins are soft) in an airplane with higher wing loadings, higher q, and higher Re, you will drive deflection waviness to trip liaminar flow. In the Tipsy Nipper, I agree that it is probably is not important.

My conclusions are still that his lightest fiberglass option is probably to build on hotwired cores. Even to me, a glass wing seems incongurous to the Tipsy Nipper. If he wants and/or needs his lightest option for a wing, he should look at wood, aluminum and fabric.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,688
Location
Saline Michigan
Cheap simple female molds and vacuum bagging parts.

There are more threads on this topic, just use the search function here and on Canard Community.com. Sorry, my pics have not been scanned yet into digital form yet.

The molds go a few inches beyond the part spanwise. It is several sections of blue billet foam, hotwired and bonded together, then lined with roof flashing or plastic laminate, sealed with micro and a sealed tape is put on over the perimeter.

I set up my root and tip foil cordinates in Excel, did linear interpolations to get the foil shapes beyond where the wing ran (the mold has to go several inches beyond the part) and also came up with coordinates for intermediate places along the foil (I can accurately hot wire up to about a four feet). The coordinates for the foil were padded by the thickness of the mold liner and its attaching epoxy.

I made templates for the several cuts, trimmed the "skin" and squared up my flotation billets, and hotwire the mold segments. Then with mixing sticks hot glued to the mold surfaces to help with alignment, I got everything aligned. A long bow with a string works great as a straightline for checking your alignments. Bond them together with micro, then spoon feed the joints to seal them up. Once you have a monolith of foam, you can put a tape around the perimeter for sealling your bag film. Then I cut my liner and bond it in using epoxy and vacuum bag techniques.

I used 36" wide coated roof flashing for my liner, and had to rought up the coating with sandpaper to make it stick to the foam. Other folks have used plastic laminate (Formica). The best reference I have on vacuum bagging is from Gougeon/ West System.

My wing skins are laid up in this order:
Peel ply;
Perforated ply;
Pell ply;
Triax cloth;
3/8" PVC cores;
Biax cloth;
Peel ply;
Perforated ply;
Batting;
Bag film.

The first peel ply allows excess epoxy to find its way out from between the foam and the mold. I punch my foam with a 1" pattern of small holes (tiny brad filed square and sharp) to also help get excess epoxy out of the layer. Peel ply on the glass helps to prepare for finishing and bonding. Peforated ply allows you to get excess resin out of the part without allowing batting to draw epoxy and leave dry areas.

Compared to just hotwiring solid cores and building per Rutan et al, it is complicated and not as light. So all of my tail and control surfaces are solid foam core.

Enjoy your build!

Billski
 
2
Top