What do you think of this wing building method?

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
I'm pretty close to buying the plans for a plane that is becoming popular in Europe but seems to be virtually unknown elsewhere. The plane is a Gaz'aile 2 http://gazaile2.free.fr/index.php .The plane itself is wooden but the wing is composite. I'm interested to know what you think of this wing building method. http://gazaile2.free.fr/englishInformations.pdf
It is fairly high aspect, untapered and fairly heavily loaded at 18lb/sqf.
The spar is made up of a foam shear web with timber and pultruded carbon caps. I can remember reading on here that using carbon and timber together like this is a no no. There are around fifty of these aircraft flying with some over 1000 hours so it obviously works.
The wing is a laminar flow section with very closely spaced PVC foam ribs, then covered in plywood. Looks like a quick, simple and reasonably cheap way to build a wing.

Mark
 

cluttonfred

Well-Known Member
HBA Supporter
Joined
Feb 13, 2010
Messages
7,003
Location
World traveler
I seem to remember a recent thread mentioning another aircraft by one of original Gaz'aile 2 team members, who split off to do his own thing. Anyone?
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
I'm pretty close to buying the plans for a plane that is becoming popular in Europe but seems to be virtually unknown elsewhere. The plane is a Gaz'aile 2 http://gazaile2.free.fr/index.php .The plane itself is wooden but the wing is composite. I'm interested to know what you think of this wing building method. http://gazaile2.free.fr/englishInformations.pdf
It is fairly high aspect, untapered and fairly heavily loaded at 18lb/sqf.
The spar is made up of a foam shear web with timber and pultruded carbon caps. I can remember reading on here that using carbon and timber together like this is a no no. There are around fifty of these aircraft flying with some over 1000 hours so it obviously works.
The wing is a laminar flow section with very closely spaced PVC foam ribs, then covered in plywood. Looks like a quick, simple and reasonably cheap way to build a wing.

Mark
I like it. Pretty, basic construction, looks like decent performance and room. The wing method on the docs I was able to pull up would be better described as wood with some carbon reinforcement.

It is not that wood and carbon can not work well together, as they can, but that the combination is usually heavier than other ways. The reason is simple. By the time you have enough carbon fiber in the part to carry much load, the wood is carrying almost nothing, and so is just a rather heavy core. In most cases, the wood could be changed out for foam except where the hard points are. It is a redesign to do this with all attendant effort and risks. In a wing with low loading, foam ribs and ply skins, a wooden shear web and graphite cloth (or Graphlite rod) caps might indeed simplify build and be suitably light.

If one were not averse to hot wired cores and vacuum bagged skins, one might be able to build a lighter wing, control surfaces, and tailplanes on hot wired cores, Graphlite caps, carbon cloth webs and skin vacuum bagged on... Oh forget I said that, stick to the plans, they are proven, there are other builders in the community, and you will be flying a lot sooner. Sorry, I could not help myself - when I look at some designs, I think how much simpler they would be to build another way...

Billski
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
I seem to remember a recent thread mentioning another aircraft by one of original Gaz'aile 2 team members, who split off to do his own thing. Anyone?
Yes, that would be the Chouchen. From what I can gather one of the first builders wanted to change a lot of the original design so there was a parting of ways. He did things his way , ended up with a heavier airframe, a non laminar flow wing and a Jabiru engine. He is generously offering the plans for free.
Serge Pennec, the designer, wanted to produce a plane that was easily affordable by using a diesel car engine. The engine is heavy so the airframe had to be light to fit into an MTOW of 450kg.
He succeeded and proved it by flying from Britany to Corsica, the furthest straight line in France, a distance of 1400klm (870 MLS). He used 51 ltrs (13.5 US gals) on the 7.5 hr non stop flight. That's 3.5ltr/100klm (78mpg). There are over 40 flying with many different engines.
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
I like it. Pretty, basic construction, looks like decent performance and room. The wing method on the docs I was able to pull up would be better described as wood with some carbon reinforcement.

It is not that wood and carbon can not work well together, as they can, but that the combination is usually heavier than other ways. The reason is simple. By the time you have enough carbon fiber in the part to carry much load, the wood is carrying almost nothing, and so is just a rather heavy core. In most cases, the wood could be changed out for foam except where the hard points are. It is a redesign to do this with all attendant effort and risks. In a wing with low loading, foam ribs and ply skins, a wooden shear web and graphite cloth (or Graphlite rod) caps might indeed simplify build and be suitably light.

If one were not averse to hot wired cores and vacuum bagged skins, one might be able to build a lighter wing, control surfaces, and tailplanes on hot wired cores, Graphlite caps, carbon cloth webs and skin vacuum bagged on... Oh forget I said that, stick to the plans, they are proven, there are other builders in the community, and you will be flying a lot sooner. Sorry, I could not help myself - when I look at some designs, I think how much simpler they would be to build another way...

Billski
Yes, I like it as well. I too started redesigning it in composites but soon realised Serge Pennec is no fool and has already designed a very light plane that can be safely built by inexperienced amateurs. It has a huge support base with excellent plans and over 4000 construction photographs. All the metalwork can be bought at a very reasonable price, including the parts to build a well proven belt drive PRSU. I'm glad Google knows French.

The spar is interesting in that the carbon tapers off to zero with the timber carrying the loads from there on. The wings are easily removable.

I looked at this plane a while back but ignored it as too light at 450kg MTOW; in Australia we can have 600kg if registering as a LSA. Serge tells me it is certified in Enland for 500kg and there is a 600kg MTOW example flying too, with beefed up spar. https://www.forum-ulm-ela-lsa.net/viewtopic.php?f=464&t=5667. I would use a 90hp Diesel.

Mark
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
While on the subject of wing spars I have noticed that carbon spar caps can get very thick. In mast making we would never have more than 2mm thickness of carbon uni without an off axis layer. I usually use light double bias glass. I have seen many airplane spars with very thick layups of Graphlite or hand layed uni. Isn't shear a concern?

Mark
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
Can not give a good answer now. Gota crew coming to flip over my fuselage. For now, see if there are any tendencies for airframe problems, particularly in the spars. If the design has a good history, trust that. Also, look up edge delamination of caps in Jones. More serious response later...

Bill
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
While on the subject of wing spars I have noticed that carbon spar caps can get very thick. In mast making we would never have more than 2mm thickness of carbon uni without an off axis layer. I usually use light double bias glass. I have seen many airplane spars with very thick layups of Graphlite or hand layed uni. Isn't shear a concern?

Mark
OK, I promised to get back to you. First I will talk about why you see what you see in airplane spars, and then give my standard advice about making sure the design you decide to build is OK, then build to the plans.

Wing spars for airplanes with cantilever wings tend to have a lot of bending moment relative to their shear load and are the item with the single biggest loads in the airplane. As such, they tend to be a major focus for the weight conscious designer. Since they are internal to the wing, their aerodynamic shape is non-critical, and we are relatively unconcerned about how much the wing deflects under load as long as the tips do not drag on the ground or hit each other in flight. The biggest stiffness issue with an airplane wing is wing flapping flutter modes, and that is also spar bending resistance. Hardpoints are only common at the root, loading tends to be nicely spread out everywhere else, so the design can be highly optimized around carrying bending and shear, with no compression and little torsion imposed.

Conventional sailboat spars are a complete aerodynamic shape, with integrated sail mountings, and are usually braced column-beams with many points of hardware attachment, large loads imposed at standing rigging and for controls, sail hoisting/tensioning, and so on. The result is there is need for much of the mast to be capable of being hardpoints, must be able to carry large compression, considerable torsion and bending. Not having done big analysis, I will suggest that the bending moment to shear ratio is smaller in conventional sailboat spars than in cantilever airplane wing spars, and that the sailboat spar carries much more torsion and local loads from various connections.

Get into wing mast type rigs and they look like a cantilever wing on an airplane...

How do the differences affect design?

The airplane wing carries big bending, more modest shear, and carries hardly any of the torsion in most designs - the skin of the airfoil serves as a shear flow torsion box with far greater GJ than the spar, so the spar does not really need to be designed for GJ so much as it has to tolerate the deflection that the skin gives it. In most composite wings the skin torsional stiffness is so high that the spar sees little torsion and some designers hardly pay attention to it. When one analyzes using {F} = {e}[ABBD] for your sections, solving for {e}, you get all strains automatically, and torsion is usually very small, so those guys neglecting it in the spar are probably OK anyway. Decompose to get lamina strains and failure criteria checks, and everything is covered anyway...

With bending being huge and thickness being relatively modest, the spar caps must be large indeed. Even with the skins also contributing some EI to the wing, the caps do tend to get substantial. Min design g is 3.8, and many of us recognize the high speed capability of very clean composite airplanes by designing to higher g's with 6 being common. Typical FOS is 2.0 in composite airplanes, so now you ar designing for 12 g's so you can fly to 6 g's. Integrate 12 g's worth of lift over the semi-span and that is a bunch of bending. To keep My/I small enough means pretty big I, and that is all cap size. On my 2150# airplane with a 29' span, that is something over 670,000 in-lbs of moment. Now at the same time, the shear web at the wing root is much smaller, around 14,000lb, and requires only a relatively light shear web by normal sizing methods.

Now this is not to say that the shear web does not have to attach to the caps, but with little torsion in an airplane spar, you do not need fibers on the bias out near the outer surface of the spar the way you might in a sailboat spar that does see substantial torsion loading. In spars where Graphlite rods make up the caps, you do some interleaving of shear web material into the layers of rods. But in more conventional UNI tape based caps, the tendency is to wrap a couple web plies around the outside, and the rest in a C shape between the caps with a wide faying surface to bond the caps and webs together. You should check that your interlaminar shear loads are in bounds when designing.

When you look at the shear web immediately adjacent to the caps, the shear web not only deforms in shear, but it extends or contracts with the caps, so you have to include that strain in the webs too. Doing the analysis with matrix algebra and doing failure criteria checks, you find that with standard cap and web sizing, the web will fail at much lower load that you had expected. I can tell you that beefing up the web is the heavy way to get to strength. Nope, you mostly beef the spar caps while also bumping the number of plies in the web. You can even optimize the spar for weight by finding successful schedules at differing numbers of plies in the caps and converging upon the lowest weight solution. Then move to the next station along the spar, and repeat out the semi-span. You get to strength at lowest weight with some pretty substantial caps, while the shear webs look pretty flimsy. Remember that the skins end up carrying some load too, and spar does not carry substantial torsion.

You end up with some pretty substantial looking caps, almost flimsy looking shear webs, except around hard points, where the webs get much more substantial. Now you have thick edges on the caps, and some have shown a tendency to delaminate... Jones covers this pretty thoroughly in his book. He has analysis and solutions too. Thinning the edges, notching the edges, and others. Interesting thing about this is that at big FOS and then bumping the number of cap plies further to make the shear web survive at min total weight, the edges of the caps tend to be pretty safe in little airplanes.

So, we are back to looking at a spar that is not built like you are used to in a sailboat. Yep. Different loading schemes, little in the way of hard points, little torsion in the spar, and overbuilt caps to make overall strength at min weight. But it looks wrong to a person with your experience? Get over it. The beauty of building an established design is you can look at their history of being flown and how people have found the airplane to handle and behave. If it has a good history and they fly well, stall reasonably and are fun to build, why would you mess with it? Build it, test it, fly it, be happy. On the other hand, if a design has been found to be difficult to build, handles poorly, or has a history of structural issues (you know, little things like breaking up in flight or growing cracks that require new wings, that sort of thing) why go near the design? Some build communities have found that a design is fine except for... they have a known problem. If the community also has a known fix that has been shown to work and be durable, you still have a solution. But to fix things that ain't broke - please do not do that.

I would look over the type's history and see what the build/fly community thinks of their airplanes, and decide what to do based on that. And please stick to the plans with the exception of known fixes to known problems. You will build quicker, have more fun doing so, and be flying much sooner with a better performing airplane than if you start messing with the design details... Too many overweight poor performing airplanes because the builder use thicker ply here, substituted the next larger fasteners and bearings there, and so on. Ugh!

Billski
 
Last edited:

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
Thanks Bilski, very thorough answer and now I understand. I am no composite engineer (the maths does my head in) but I do have a good understanding in relation to boats because I've done a fair bit of it and I have a good relationship with a composite engineer who, like you, is generous enough to explain it to me without going into the maths. So I know lots of rules of thumb one of which was the off axis layers. When I noticed a lot of wing spars not using it (Marske etc) I was intrigued. So thank you for explaining.

I wasn't about to redesign the spar and have already abandoned the folley of designing a composite plane myself, it would take way too long. I'm happy get on with a proven design.

BTW, You may be interested to know, the carbon masts I build these days are all free standing of tapered round section. I hand wet the laminates flat on the bench, roll them around a mandrel which is covered in a tubular bag, place them in a female mould, bag around the mould then vacuum the laminate onto the inside of the mould. It would be a good way to make very light control rods. For bigger free standing masts (or one offs) where I don't want to make a mould I create a precured bias laminate which is the outside layer and the final shape (I cap and seal the ends. I then do the flat layup around the mandrel as per the last method only this time I inflate the bag to about 20 pounds. There's a bit more to it than that but that's the basics. I've made fee standing masts up to 18m using that method.


Mark
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
I wasn't about to redesign the spar and have already abandoned the folley of designing a composite plane myself, it would take way too long. I'm happy get on with a proven design.
Smart man!

BTW, You may be interested to know, the carbon masts I build these days are all free standing of tapered round section. I hand wet the laminates flat on the bench, roll them around a mandrel which is covered in a tubular bag, place them in a female mould, bag around the mould then vacuum the laminate onto the inside of the mould. It would be a good way to make very light control rods. For bigger free standing masts (or one offs) where I don't want to make a mould I create a precured bias laminate which is the outside layer and the final shape (I cap and seal the ends. I then do the flat layup around the mandrel as per the last method only this time I inflate the bag to about 20 pounds. There's a bit more to it than that but that's the basics. I've made fee standing masts up to 18m using that method.

Mark
Cool stuff. If I ever visit your continent, I should come to the shop. So your free standing masts are tapered, carry big loads in any direction, must have increasing section as you go towards the root, must have the structure able to handle the accumulated shear and bending pretty much the same in all directions. That would have to drive a mast shape that is tall truncated pyramid and is structurally the same in all directions.

One thing I did not mention about airplane wings. While they do not carry all loads in the same direction, they are pretty close to that. At Vd on the envelope diagram, we are at the highest airspeed and make max g's at relatively low angle of attack. At Va, we are at the lowest airspeed where we can make max g. At Vd we are usually around 3-5 degrees, at Va around 16-18 degrees. For analysis, we usually model the wing flat, zero incidence and that is pretty darned close to Vd condition, and is the condition where the drag spar and skins give the least contribution to bending stiffness and shear handling too, so is the most conservative condition to design for. Now try to pull max g near Va, the AOA goes up to stall to get max g, and the lift is angled forward 15 or more degrees. If you bother to transform the wing laminate around the 25%c point and compute things, the skin and drag spar contribute more bending and shear handling capability. Usually, this means the wing is stronger at the same lift at high AOA than at low AOA, so if you are good at zero AOA, you are better at higher AOA.

I suspect that your free standing masts are part span and support a pivoting wing or sail system. I also suspect that these carry the loads at the top of the mast and near the base with bearings from the wing, with the mast going only part way up the sail or wing. Correct, or am I all wet? The differences are real and for very good reason...

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
Oh, the method you use could be very effective for making control push-pull rods. How do you remove your parts from the molds? Hydraulic pull machine? Hydraulic pressure in the mold? Do you have a min wall that will work?

Most folks just use an appropriate diameter steel or aluminum tube of 0.035 wall. Buckling resistance drives diameter, and EI is the figure of merit.

I have made tubes using composite braid on disposable mandrels (I like fluorescent lamp tubes, I have written about how I do this). With braid, you can stretch them down to mandrel size and get more favorable fiber angles than with flat wrapping square weave cloth at +/- 45 degrees. I am using them in glass to avoid interference with buried antennas, and they are not a weight save over metal tubes. They could be lighter than metal in graphite fiber, but then I am back to messing up the antenna. This all means that straight tubes are slightly oversized at the middle and grossly oversized everywhere else...

Autoreply advocates for wet layup of tapered graphite fiber tubes over fabricated paper or veneer cones (IIRC, he has written about this on HBA too), then wet tab them together in the middle. If we were to go with dedicated tooling, there are vendors who do straight and tapered mandrels that are removed with big hydraulic machines, but they require a flange and enough wall thickness for mandrel extraction. While the flange can certainly be removed later, I wonder if the wall thickness can be kept small enough and still get the part off the mandrel. Autoreply is usually building only a couple of his pushrods in each design, he is making truly thin walled tapered tubes, and he is definitely one of the smarter people in the room. I would be willing to bet that his wet layup over paper cones may be close to if not at optimal that can be built. Perhaps you can move the state of the art!

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
Answers like that are part of the reason that Billski is making slow progress on his own aircraft!
LOL. Aw c'mon, I don't do that sort of post every day. Did a vacuum bag layup of two parts yesterday, have a secondary layup on one after I finish the current round of chores.

Things that do slow me down are the recovery from this retina surgery (still screwed up vision), harvest and canning time for tomatoes and peppers, chores around the house, bike riding, shooting, flying the storebought airplane, helping friends with their airplanes, being sociable with our friends, and doing design work on details of the coming parts on the homebuilt. In a couple weeks there will be a week with family and friends in Minnesota, then family and friends in Michigan over Thanksgiving. There are worse things than being busy. Back to the nominal chores...

Billski
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,372
Location
North Carolina
if you want a something done, give the job to a busy person!
Pushrods. I haven't tried this yet, maybe someone else has. Take a hollow mandrel, do your layup and bag it. Then hydraulically inflate the tube up to, I dunno, 2/3 yield and pull a vacuum on the layup. When you let the pressure out of the tube, it'll shrink away from the cured laminate. Should make mandrel removal easier?
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
10,475
Location
97FL, Florida, USA
Has anyone used carbon fiber sleeves over a fluorescent light tube to make push/pull rods?


BJC
 

cluttonfred

Well-Known Member
HBA Supporter
Joined
Feb 13, 2010
Messages
7,003
Location
World traveler
Back to the Gaz'aile 2, is anyone offering the PSA diesel or gasoline conversions ready to fly? With a right prop and enough wing area, even the little diesel could be interesting with just a 26.5 liter/7 US gallon tank giving three hours in the air plus an hour in reserve.

DIESEL
PSA Citroën AX/Peugeot 106
1400 cc
55 hp @ 5000 rpm
86 Nm/63 ft-lb max torque @ 2500 rpm
Toothed belt redrive 1.88:1 ratio
Fuel: diesel, JetA1, heating oil, bio fuels
Consumption: 6.3 liters/1.7 US gallons per hour

GASOLINE
PSA Peugeot 205, 306/Citroën ZX, C2, Xantia, Saxo, BX
1400 cc
80 hp @ 5600 rpm
118 N.m/87 ft-lb max torque @ 3800 rpm
Toothed belt redrive 1.88:1 ratio
Fuel: unleaded gasoline
Consumption: 13.3 liters/3.5 US gallons per hour
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,300
Location
Saline Michigan
Has anyone used carbon fiber sleeves over a fluorescent light tube to make push/pull rods?


BJC
I have used glass and Kevlar braided tubes over flourescent tubes, and it works great. I believe I have posts on about the details. It should work every bit as well in graphite fiber, and they are available in a bunch of sizes and weights form the same supplier. The reason they work well is because the fibers are nominally at 45 degrees, and they can be pulled down or bulged up in diameter. I have tightened the sleeves with a succession of zip-ties and cyanoacrylate and then vacuum bagged or wrapped with electrician's tape. I have done graphite cloth repairs on complex shaped windsurfer parts with cloth, epoxy, and electrician's tape to debulk and make the cloth conform to the part underneath. They all work.

Billski
 
  • Like
Reactions: BJC

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
110
Location
Bellingen, NSW. Australia.
Oh, the method you use could be very effective for making control push-pull rods. How do you remove your parts from the molds? Hydraulic pull machine? Hydraulic pressure in the mold? Do you have a min wall that will work?
Billski
My mast moulds are split horizontally so it isn't an issue. The moulds need to have a series of holes drilled to let the air and excess epoxy out to the bleeder. There is no minimum layup. I often use pvc tubing as a mould. It depends on the end use but I usually split them with a 1mm cutter blade. This does end up with a slightly deformed circular section but often that isn't an issue. For shorter tubes, I cut down the length of the tube at half the wall thickness. When cured this allows me to run a knife through the cut allowing me to peel the pvc tube away. Of course this is only good for shorter tubes as the mandrel and layup has to be inserted from the end rather than laid in a half tube. It is always better to vacuum to the inside of a tube rather than around the out side, the fibres don't get kinked and it leaves a nice surface finish.

One advantage of using a split tube is the ability of making bent tubes. I have already posted on here a big carbon eagle I made using this method. http://www.homebuiltairplanes.com/forums/attachment.php?attachmentid=64740&d=1502885942. Probalbly not a lot of use for us here but handy sometimes.

You can also use pressure. I had an order from P&O the shipping company for 80 carbon masts for telemetry on top of container yard forklifts in Seattle. They were 5m long 80mm dia. and I had a month to get them out. I made two fibreglass moulds with bolted flanges. The mandrel had a flexible silicon heating element down the middle so we would inflate the bladder and heat. We were making six masts a day out of two moulds.

Filament wound tubes and masts are beautiful things but expensive and only available in the mandrel sizes they have. Making your own tubes is much cheaper and very easy.

Mark
 
2
Group Builder
Top