wing loading, span, flying speed, landing speed, charts.
- Thread starter OwenN
- Start date
Design materials change for vtol/stol design:
All ribs, spars, keel stringers to be replaced by 1/16 ply, 3 ply.
Looking at the 1/5 scale drawings, unsupported skin spans are around 200 mm (6 inches) in the longest dimension.
The deepest wing section is 50mm (2 ins) this raises the case of significant shear between ribs, spars, and skin.
Also I am now using ply spars with doubled center sections, making 3 layers in the doubled sections, so it makes
sense for all spars to be full depth 1/16 ply, and if spars, why not ribs? - every 200mm, anyway.
There are advantages in unsupported span with ply in the fuselage sides, as well.
The fuselage rib profiles (the circular frames with the big hole in the center) don't need to be
as wide at the sides. - There are significant straight sections in the fuselage walls.
To counter the expected shear loadings, all doublers and load spreaders will now be 1/16 ply, and spars will have 1/16 x 10mm (3/8 ins)
capping strips. (epoxy bonded to the skin).
With the continuous span, and epoxy coating, the 1/32 balsa skin should be sufficiently taut and resistant to bending and puncture.
the maximum loading is 3gs, or 150 oz/sq ft - what is this over a sq inch? = about 1.1 oz/sq ins- not enough to cause
significant deflection or skin stress. in a 6 inch sq area - that is 36 sq ins, or 2.5 lb evenly distributed over the surface.
this is similar to 1 lb spread over the center 4-inch-sq (16 sq ins) and like a thumb applying 3.5 oz to the very center.
This would provide minor deflection if the 6 x 6 area was only supported at the edges, but nowhere near a rupture load.
I can do some calcs, assuming that most of the load is supported in the grain direction .
This is similar to a built-in beam-not exactly, as that is based on bending being resisted by the built-in ends.
Most of the load here is resisted by the skin tension, and very little by the skin bending modulus.
No doubt there is a worked example somewhere on Google that I can look at.
The epoxy coating should be enough to prevent rupture across the grain.
It has similar properties to polyester monofilm.
It should be flexible enough in a thin coating to resist cracking.
A possible failure mode is buckling under compression across the span. This may need additional lengthwise stringers,
1/16 sq balsa?? Additionally, rib spacing may be reduced to 3 inches using alternate 1/16 in ribs.
Do I need to do this? How could I calculate any improvement?
Assembly of the ply structure can be with standard CA glue, thick,. (Superglue).
Since this stuff is very hard and brittle, and not very strong, all joints
will be given a coating and fillet of 15 minute epoxy glue, to add some toughness.
This gluing arrangement reduces the need to hold the joints in alignment while the glue sets.
I may also use some balsa cement, as this is thick and sticky, and helps hold material in place.
Any other ideas on sticky glues that assist with assembly?
Several questions now arise:
1) Bass ply is cheapest, at $ 100 nz per sq mt. Birch is $ 191 nz per sq mt.
Is there any reason why Bass may not be suitable?
2) should I bore lots of holes in the ply?
My estimate is spars have useful sections of 10mm x 1.6mm (3/8 x 1/16 ins).
and rib sections can be 5mm x 1.6mm.
Holes save 35 % of weight with spars, and 56% of weight with wing ribs.
Is it worth while cutting all these holes? I estimate there will be 1/2 sq mt of ply, at 0.3 kg total. (Bass)
Total weight estimate of the airframe is 500 g, so this is a substantial fraction. I will recheck after I have done all the patterns.
What do you think?
I don't want to notch any ribs or frames if I can get away with it. (This is about using 1/16 sq stringers).
This just means frames need deeper sections, and makes a lot of work.
Similarly with lightening holes; I will need to buy a set of wood hole saws in a range of sizes.
I do not think cutting each hole with a fretsaw is practical.
I notice basswood averages 450 kg/m2 density, and birch 600 kg/m2 density,
and similar proportions probably apply to strength.
However, if I can remove 35-55 % of the area, strength can't be that critical.
Possibly moderate strength, hardness, stiffness are the key features?
1/16 balsa gets a bit squidgy across the grain in sections over 3/4 of an inch deep.
Balsa 1/16 spars won't take much compressive side loading in the middle of a span between ribs.
The epoxy bond to the skin supports it a bit.
Use of ply spars will distribute aerodynamic pressure a bit better.
Looking at the properties of Basswood vs balsa, there are some interesting points.
1) balsa is stronger in tension along the grain, but weaker for all other characteristics (metrics).
2) balsa is 29% the density of basswood.(lighter, 1/3 the weight.)
Bending yield-bass = 40.7 MPa (maximum compressive load in the outer fiber)(bending stress lingo
)
Axial strength Balsa = 73 MPa (tension)
Crushing across the grain-balsa = 1 MPa
Basswood crushing across the grain = 2.4 MPa
Behaviour in shear is probably critical to this design- that is what the doublers and load spreaders are for.
I will look those up and do a few calcs.
Bending yield-balsa-not given - but compressive strength along the grain is about 12 MPa, which gives some idea of its bending behaviour.
It is likely to fail by local crushing when bent.
These characteristics make balsa suitable for end-grain composite with fiberglass.
The benefits of Basswood vs Birch are that it is lighter, while providing adequate strength and stiffness in 3-ply.
Any ideas on the above points?
Anybody keen to try and read this lengthy epistle?
Comments are welcome.
All ribs, spars, keel stringers to be replaced by 1/16 ply, 3 ply.
Looking at the 1/5 scale drawings, unsupported skin spans are around 200 mm (6 inches) in the longest dimension.
The deepest wing section is 50mm (2 ins) this raises the case of significant shear between ribs, spars, and skin.
Also I am now using ply spars with doubled center sections, making 3 layers in the doubled sections, so it makes
sense for all spars to be full depth 1/16 ply, and if spars, why not ribs? - every 200mm, anyway.
There are advantages in unsupported span with ply in the fuselage sides, as well.
The fuselage rib profiles (the circular frames with the big hole in the center) don't need to be
as wide at the sides. - There are significant straight sections in the fuselage walls.
To counter the expected shear loadings, all doublers and load spreaders will now be 1/16 ply, and spars will have 1/16 x 10mm (3/8 ins)
capping strips. (epoxy bonded to the skin).
With the continuous span, and epoxy coating, the 1/32 balsa skin should be sufficiently taut and resistant to bending and puncture.
the maximum loading is 3gs, or 150 oz/sq ft - what is this over a sq inch? = about 1.1 oz/sq ins- not enough to cause
significant deflection or skin stress. in a 6 inch sq area - that is 36 sq ins, or 2.5 lb evenly distributed over the surface.
this is similar to 1 lb spread over the center 4-inch-sq (16 sq ins) and like a thumb applying 3.5 oz to the very center.
This would provide minor deflection if the 6 x 6 area was only supported at the edges, but nowhere near a rupture load.
I can do some calcs, assuming that most of the load is supported in the grain direction .
This is similar to a built-in beam-not exactly, as that is based on bending being resisted by the built-in ends.
Most of the load here is resisted by the skin tension, and very little by the skin bending modulus.
No doubt there is a worked example somewhere on Google that I can look at.
The epoxy coating should be enough to prevent rupture across the grain.
It has similar properties to polyester monofilm.
It should be flexible enough in a thin coating to resist cracking.
A possible failure mode is buckling under compression across the span. This may need additional lengthwise stringers,
1/16 sq balsa?? Additionally, rib spacing may be reduced to 3 inches using alternate 1/16 in ribs.
Do I need to do this? How could I calculate any improvement?
Assembly of the ply structure can be with standard CA glue, thick,. (Superglue).
Since this stuff is very hard and brittle, and not very strong, all joints
will be given a coating and fillet of 15 minute epoxy glue, to add some toughness.
This gluing arrangement reduces the need to hold the joints in alignment while the glue sets.
I may also use some balsa cement, as this is thick and sticky, and helps hold material in place.
Any other ideas on sticky glues that assist with assembly?
Several questions now arise:
1) Bass ply is cheapest, at $ 100 nz per sq mt. Birch is $ 191 nz per sq mt.
Is there any reason why Bass may not be suitable?
2) should I bore lots of holes in the ply?
My estimate is spars have useful sections of 10mm x 1.6mm (3/8 x 1/16 ins).
and rib sections can be 5mm x 1.6mm.
Holes save 35 % of weight with spars, and 56% of weight with wing ribs.
Is it worth while cutting all these holes? I estimate there will be 1/2 sq mt of ply, at 0.3 kg total. (Bass)
Total weight estimate of the airframe is 500 g, so this is a substantial fraction. I will recheck after I have done all the patterns.
What do you think?
I don't want to notch any ribs or frames if I can get away with it. (This is about using 1/16 sq stringers).
This just means frames need deeper sections, and makes a lot of work.
Similarly with lightening holes; I will need to buy a set of wood hole saws in a range of sizes.
I do not think cutting each hole with a fretsaw is practical.
I notice basswood averages 450 kg/m2 density, and birch 600 kg/m2 density,
and similar proportions probably apply to strength.
However, if I can remove 35-55 % of the area, strength can't be that critical.
Possibly moderate strength, hardness, stiffness are the key features?
1/16 balsa gets a bit squidgy across the grain in sections over 3/4 of an inch deep.
Balsa 1/16 spars won't take much compressive side loading in the middle of a span between ribs.
The epoxy bond to the skin supports it a bit.
Use of ply spars will distribute aerodynamic pressure a bit better.
Looking at the properties of Basswood vs balsa, there are some interesting points.
1) balsa is stronger in tension along the grain, but weaker for all other characteristics (metrics).
2) balsa is 29% the density of basswood.(lighter, 1/3 the weight.)
Bending yield-bass = 40.7 MPa (maximum compressive load in the outer fiber)(bending stress lingo
)Axial strength Balsa = 73 MPa (tension)
Crushing across the grain-balsa = 1 MPa
Basswood crushing across the grain = 2.4 MPa
Behaviour in shear is probably critical to this design- that is what the doublers and load spreaders are for.
I will look those up and do a few calcs.
Bending yield-balsa-not given - but compressive strength along the grain is about 12 MPa, which gives some idea of its bending behaviour.
It is likely to fail by local crushing when bent.
These characteristics make balsa suitable for end-grain composite with fiberglass.
The benefits of Basswood vs Birch are that it is lighter, while providing adequate strength and stiffness in 3-ply.
Any ideas on the above points?
Anybody keen to try and read this lengthy epistle?
Comments are welcome.
Last edited:
quorneng
Master member
OwenN
You say you want VTOL/STOL yet you are considering using ply instead of balsa? and adding epoxy to all the joints. For any sort of vertical flight weight. or rather the lack of it, is a vital factor.
Ply is heavier than balsa unless the ply is very thin and thin materials do not work well in compression. In addition thin materials have thin joints placing more emphasis on the strength of the glue in the joint. Epoxy is even heavier than ply.
Remember the requirement is not to build the strongest possible structure but one that gives sufficient strength for the required duty at the lightest weight. Not easy to do and requires a good understanding of the likely forces that will occur in each element of the structure.
It is worth noting that most commonly used model plane materials when 'built up' into a lightweight structure are most likely fail where elements are under compression or at glue joints when under tension.
This suggest using materials that are light enough to be used with a sufficient thickness that it can handle the likely compression loads and to try to arrange that glue joints are in compression or shear (which it is good at) rather than tension.
I can only suggest you study known successful designs doing the same job as you as you are after and try to understand why it was done the way it was.
There is no substitute for experience.
I hope this helps.
You say you want VTOL/STOL yet you are considering using ply instead of balsa? and adding epoxy to all the joints. For any sort of vertical flight weight. or rather the lack of it, is a vital factor.
Ply is heavier than balsa unless the ply is very thin and thin materials do not work well in compression. In addition thin materials have thin joints placing more emphasis on the strength of the glue in the joint. Epoxy is even heavier than ply.
Remember the requirement is not to build the strongest possible structure but one that gives sufficient strength for the required duty at the lightest weight. Not easy to do and requires a good understanding of the likely forces that will occur in each element of the structure.
It is worth noting that most commonly used model plane materials when 'built up' into a lightweight structure are most likely fail where elements are under compression or at glue joints when under tension.
This suggest using materials that are light enough to be used with a sufficient thickness that it can handle the likely compression loads and to try to arrange that glue joints are in compression or shear (which it is good at) rather than tension.
I can only suggest you study known successful designs doing the same job as you as you are after and try to understand why it was done the way it was.
There is no substitute for experience.
I hope this helps.
It is difficult to decipher what the designers were doing when looking at an existing plan.
I don't intend to rely on any simple butt joins for tension. The ply membrane-type spars are sufficient to take any bending loading.
My concern is distributing shear into the skin at a level that balsa will stand. The actual tensile strength of the balsa is irrelevant now.
I was going to use the skin as the t-top of the spars, and have a stressed skin, but the requirement to join to ply has superseded that.
1/16 spars and 1/32 skin is a bit suspect under compression, anyway. There needs to be a substantial compression member.
(* newly discovered important fact for me!*)
The ply spar membranes came about from having to pick loads up from the center spars-I might as well continue them right across the span.
My new idea for a skin-supporting sub-structure is to have sub-ribs of 1/16 x 1/8 ply, which are self-supporting.
The spanwise stringers can also be 1/8 x 1/8 balsa, just butted up to the ribs. They don't need to carry spanwise loading, and rely on epoxy bonding to the skin. This way, I avoid having to notch ribs, eliminate the center parts of alternate ribs, and don't need to put cap-strips on the skin support grid. Also-3/8 is a bit massive for cap-strips. 1/4 will do.
Thinking about that-there is no need for any full ribs. they can all be sort of flying buttresses!
The full depth spar webs provide enough vertical support.
I wanted a permanent hollow in the inner front part of the wing, anyway.
Wires can be pulled through as needed.
The flexi-control cables for the rudders will need to be more built in.-I can add them before the top skin goes on, and after the center span joint is made.
The skin has to be carefully matched to the ribs and spars, so I get a good epoxy glued joint. The epoxy is thick and gelled, so it builds up a good glue bead.
The bead also has inherent strength as a gap-filler.
Note that compressive loads are taken at the skin-spar and spar-cap junction, which is effectively an I-beam.
The spec at the moment has 4 times the lift thrust to the projected weight. I can afford to go to 2.5 and still get good vertical lift
effect.
This allows overall weight to go from 1 kg to 1.6 kgs.
If equipment is projected to weigh 700 g, that leaves me
900 g for the airframe. At present, I am aiming at 500g, and will do some mass surveys as I generate more patterns, and get a better
idea of what components go into the airframe.
I will also do a better survey of hardware weights, to make sure my targets are OK.
The equipment survey only covers heavy stuff like motors, battery, and actuators at present.
I need to add-FPV bits, receiver, escs, flight controller, flexi push-pull cables, any turnbuckle assemblies,
various clamps, straps, rubber padding, small fittings, props, motor mounting plates (aluminium) screws, nuts,
elevator horns , clevis joints, pins, cabling, etc.
Re: epoxy-I think all joints between ply pieces should be epoxy. I think it is also better for getting a reliable joint than balsa cement,
and CA glue is really the pits! - only suitable for tacking parts together.
Epoxy coating the outside is no worse that using sanding filler, dope, and then monofilm.
Paint is sort of doubling up on the monofilm, but most aerobatic "pattern ships" use glossy painted surfaces, and generally
they have quite low wing loadings in the 15 to 25 oz/sq ft range.
I don't intend to rely on any simple butt joins for tension. The ply membrane-type spars are sufficient to take any bending loading.
My concern is distributing shear into the skin at a level that balsa will stand. The actual tensile strength of the balsa is irrelevant now.
I was going to use the skin as the t-top of the spars, and have a stressed skin, but the requirement to join to ply has superseded that.
1/16 spars and 1/32 skin is a bit suspect under compression, anyway. There needs to be a substantial compression member.
(* newly discovered important fact for me!*)
The ply spar membranes came about from having to pick loads up from the center spars-I might as well continue them right across the span.
My new idea for a skin-supporting sub-structure is to have sub-ribs of 1/16 x 1/8 ply, which are self-supporting.
The spanwise stringers can also be 1/8 x 1/8 balsa, just butted up to the ribs. They don't need to carry spanwise loading, and rely on epoxy bonding to the skin. This way, I avoid having to notch ribs, eliminate the center parts of alternate ribs, and don't need to put cap-strips on the skin support grid. Also-3/8 is a bit massive for cap-strips. 1/4 will do.
Thinking about that-there is no need for any full ribs. they can all be sort of flying buttresses!
The full depth spar webs provide enough vertical support.
I wanted a permanent hollow in the inner front part of the wing, anyway.
Wires can be pulled through as needed.
The flexi-control cables for the rudders will need to be more built in.-I can add them before the top skin goes on, and after the center span joint is made.
The skin has to be carefully matched to the ribs and spars, so I get a good epoxy glued joint. The epoxy is thick and gelled, so it builds up a good glue bead.
The bead also has inherent strength as a gap-filler.
Note that compressive loads are taken at the skin-spar and spar-cap junction, which is effectively an I-beam.
The spec at the moment has 4 times the lift thrust to the projected weight. I can afford to go to 2.5 and still get good vertical lift
effect.
This allows overall weight to go from 1 kg to 1.6 kgs.
If equipment is projected to weigh 700 g, that leaves me
900 g for the airframe. At present, I am aiming at 500g, and will do some mass surveys as I generate more patterns, and get a better
idea of what components go into the airframe.
I will also do a better survey of hardware weights, to make sure my targets are OK.
The equipment survey only covers heavy stuff like motors, battery, and actuators at present.
I need to add-FPV bits, receiver, escs, flight controller, flexi push-pull cables, any turnbuckle assemblies,
various clamps, straps, rubber padding, small fittings, props, motor mounting plates (aluminium) screws, nuts,
elevator horns , clevis joints, pins, cabling, etc.
Re: epoxy-I think all joints between ply pieces should be epoxy. I think it is also better for getting a reliable joint than balsa cement,
and CA glue is really the pits! - only suitable for tacking parts together.
Epoxy coating the outside is no worse that using sanding filler, dope, and then monofilm.
Paint is sort of doubling up on the monofilm, but most aerobatic "pattern ships" use glossy painted surfaces, and generally
they have quite low wing loadings in the 15 to 25 oz/sq ft range.
Tench745
Master member
It seems to me that you could answer a lot of your own questions and save a lot of headache by reading through the Airfield Models website. The "Designing Radio Control Model Aircraft" series in particular http://www.airfieldmodels.com/infor...f_model_aircraft/rc_aircraft_design/index.htm.
Actually, I recommend starting with How to Design and Build Lightweight Model Airplanes and then go back to the above series.
Actually, I recommend starting with How to Design and Build Lightweight Model Airplanes and then go back to the above series.
Thank you. I shall read that. I am also reading this one, but so far it is more about dynamics-I am up to chapter 8.
file:///C:/Users/nieuw/Desktop/RC_Model_Aircraft_Design_ALennon_partA.pdf
The aircraft really should not be built too light. Construction methods suitable for Pattern Ships would be more appropriate.
Some of the plans I've looked at, my comments would be "Really? What were they smoking when they drew that?"
Why do they put tubes and pins in wings? Do they really need to remove the wings? I suppose a 100 incher is a bit difficult to fit in your average car???
file:///C:/Users/nieuw/Desktop/RC_Model_Aircraft_Design_ALennon_partA.pdf
The aircraft really should not be built too light. Construction methods suitable for Pattern Ships would be more appropriate.
Some of the plans I've looked at, my comments would be "Really? What were they smoking when they drew that?"
Why do they put tubes and pins in wings? Do they really need to remove the wings? I suppose a 100 incher is a bit difficult to fit in your average car???
Quinnyperks
Legendary member
Airplanes are not easy to fit in cars. 100 inch espescially
I have a wagon with fold-down rear seats, so a 44 incher should fit in there OK. I will measure.
1) Whenever I see the detail with the steel rods into brass tubes, I tell myself "that is wrong!".
It should be hollow tubes at least half the section depth, and no more than 4x the diameter long.
Tubes should have their own spar set each, boxed in. The spar should be capable of supporting the wing loading, or about half for two spar sets.
How are the wings latched in place?
Through-pins with end clips give minimum supporting structure.
Toggle clamps need a lot of support,
and don't allow the tubes to transfer the bending load.
The outer tube socket needs to be well integrated into the spar-maybe 4 square sections tapering together, well "plated" in?? 4x the diameter is enough to transfer loads to the spar, keep socket edge loadings down to a reasonable level, and to help reduce any free play.
You definitely don't want a separate main spar, full depth, and spreading out in an exponential curve at the wing root!
This makes 4 spar square wooden sections altogether.
You could use less, but that complicates the tube-socket "boxing-in".
2) I have identified a redundancy in my wing proposal-if 1/16 ply cap strips go into corners, any corner gussets can be quite small.
Cap strips can transition to cross-grain balsa at the more curved sections, and on the outer wing panels.
Also, if the area is well away from holes or other stress raisers (like spar ends), the cross-grain balsa can be used as cap-strips.
3) My proposed front web-spar gets closer to the leading edge towards the end, so needs to be curved along its length.
I will need to make multiple wing section templates to get these spars set up correctly.
1) Whenever I see the detail with the steel rods into brass tubes, I tell myself "that is wrong!".
It should be hollow tubes at least half the section depth, and no more than 4x the diameter long.
Tubes should have their own spar set each, boxed in. The spar should be capable of supporting the wing loading, or about half for two spar sets.
How are the wings latched in place?
Through-pins with end clips give minimum supporting structure.
Toggle clamps need a lot of support,
and don't allow the tubes to transfer the bending load.
The outer tube socket needs to be well integrated into the spar-maybe 4 square sections tapering together, well "plated" in?? 4x the diameter is enough to transfer loads to the spar, keep socket edge loadings down to a reasonable level, and to help reduce any free play.
You definitely don't want a separate main spar, full depth, and spreading out in an exponential curve at the wing root!
This makes 4 spar square wooden sections altogether.
You could use less, but that complicates the tube-socket "boxing-in".
2) I have identified a redundancy in my wing proposal-if 1/16 ply cap strips go into corners, any corner gussets can be quite small.
Cap strips can transition to cross-grain balsa at the more curved sections, and on the outer wing panels.
Also, if the area is well away from holes or other stress raisers (like spar ends), the cross-grain balsa can be used as cap-strips.
3) My proposed front web-spar gets closer to the leading edge towards the end, so needs to be curved along its length.
I will need to make multiple wing section templates to get these spars set up correctly.
Still working on the 1/5 scale drawing. That is a little too small to follow for things like engine mounts, but I will add the
cap-strip , gusset system, and scan it- I can try out my jpg compositor!
It is around A2 size-550 x 450, or 21 x 18 ins.
Watch this space!
Actual full size detail will have to wait until after Christmas. I will push 1/5 scale as far as it will go-down to about
5 mm resolution. I show 1.5 mm detail as being 5mm thick visually.
I still have lots of detail to add at 1/5.
Re: hardware mass check-up to 762.5 grams and counting-now on to bits and pieces. (1.7 lbs).
cap-strip , gusset system, and scan it- I can try out my jpg compositor!
It is around A2 size-550 x 450, or 21 x 18 ins.
Watch this space!
Actual full size detail will have to wait until after Christmas. I will push 1/5 scale as far as it will go-down to about
5 mm resolution. I show 1.5 mm detail as being 5mm thick visually.
I still have lots of detail to add at 1/5.
Re: hardware mass check-up to 762.5 grams and counting-now on to bits and pieces. (1.7 lbs).
Well, I have read the RC-model aircraft book by Lennon, and the suggested articles on light weight building.
1) I now know that 1/64 ply is worth a look for gussets and membranes,
and I can make my own ply,
and that shop-brought ply is usually warped, and it is good to make my own.
Cross-ply 1/32 balsa , epoxy, to make biaxial 1/16- which is better than using shop-bought ply.
Add weights to hold it flat while curing.
It is lighter, and can be used in place of ply, if I make allowances for some lower-level properties in certain areas.
Suck it and see-punish the assembly a bit and get a feel for what it will stand up to!
2) Ribs weigh next to nothing, so use a few!
3) I still like my idea of 1/16 ply membrane spars, using cap strips and the skin as the compression member.
The use of generous beads of epoxy and several layers should increase the compressive load-bearing strength at the "I" top
up to the 12+ MPa level.
4)You don't need to slavishly use spruce spars. A shaped box ply spar will work as well-just do the calcs first!
This is particularly appropriate if you want the wings to be removable, with pin-sockets tied into the spars.
You can even have a "Y" shaped spar or a bent spar if it is a box, and the sides, top, bottom are lapped so all joints don't occur on one plane.
This is good for making internal wing clearance for retractable gear, flap, and aileron actuators.
The main wing tube joint can even be at the 30% point from the leading edge, take all the spar load, and the second pin can just be an
anti-rotate stub.
Side joints can be butt-joined.
Top and side are lapped only at the edge.
You can add face-laps for top and bottom surfaces, if they have joins.
It is not necessary for the sides.
5) I also still like the use of "flying buttresses" as ribs, letting the web-spars provide vertical strength.
If they are cross-plied balsa, and 3mm deep x 1.5mm wide, they can handle a 6 inch span without collapse if weight is applied to the
center of the span. I will try that. they may need doubling to 1/8 inch wide as well.
6) I now realise that balsa joins quite well with CA glue, but use the thick stuff, otherwise it runs everywhere!
The balsa will fail by crushing in bending before the glue will, and it penetrates balsa well.
I still think epoxy is a better choice for ply , as it is tougher, and ply is not as prone to collapse in bending.
7) If I make my own 1/16 ply, it will be lighter, and cutting lots of holes is not necessary.
8) Extra stringers are not required if ribs are spaced about 1.5 inches.
9) I know how to detail the hinge area on the control surfaces with full width wedges, to emulate a rounded elevator rear,
with center cutout to clear the hinges and hinge holding laminations, and opposite wedges on the "plane" side to fill in the gap.
Cutting the long wedges and getting them a regular shape will be a challenge-you really need a micro-bandsaw!
They are tapered to suit the trailing edge taper, in many cases.
10) There are things to learn about spiral instability and vertical surface design, and types of flaps. - see Mr Lennon's book (PDF)-check Google.
There are also some more handy tips on RC model dynamics.
1) I now know that 1/64 ply is worth a look for gussets and membranes,
and I can make my own ply,
and that shop-brought ply is usually warped, and it is good to make my own.
Cross-ply 1/32 balsa , epoxy, to make biaxial 1/16- which is better than using shop-bought ply.
Add weights to hold it flat while curing.
It is lighter, and can be used in place of ply, if I make allowances for some lower-level properties in certain areas.
Suck it and see-punish the assembly a bit and get a feel for what it will stand up to!
2) Ribs weigh next to nothing, so use a few!
3) I still like my idea of 1/16 ply membrane spars, using cap strips and the skin as the compression member.
The use of generous beads of epoxy and several layers should increase the compressive load-bearing strength at the "I" top
up to the 12+ MPa level.
4)You don't need to slavishly use spruce spars. A shaped box ply spar will work as well-just do the calcs first!
This is particularly appropriate if you want the wings to be removable, with pin-sockets tied into the spars.
You can even have a "Y" shaped spar or a bent spar if it is a box, and the sides, top, bottom are lapped so all joints don't occur on one plane.
This is good for making internal wing clearance for retractable gear, flap, and aileron actuators.
The main wing tube joint can even be at the 30% point from the leading edge, take all the spar load, and the second pin can just be an
anti-rotate stub.
Side joints can be butt-joined.
Top and side are lapped only at the edge.
You can add face-laps for top and bottom surfaces, if they have joins.
It is not necessary for the sides.
5) I also still like the use of "flying buttresses" as ribs, letting the web-spars provide vertical strength.
If they are cross-plied balsa, and 3mm deep x 1.5mm wide, they can handle a 6 inch span without collapse if weight is applied to the
center of the span. I will try that. they may need doubling to 1/8 inch wide as well.
6) I now realise that balsa joins quite well with CA glue, but use the thick stuff, otherwise it runs everywhere!
The balsa will fail by crushing in bending before the glue will, and it penetrates balsa well.
I still think epoxy is a better choice for ply , as it is tougher, and ply is not as prone to collapse in bending.
7) If I make my own 1/16 ply, it will be lighter, and cutting lots of holes is not necessary.
8) Extra stringers are not required if ribs are spaced about 1.5 inches.
9) I know how to detail the hinge area on the control surfaces with full width wedges, to emulate a rounded elevator rear,
with center cutout to clear the hinges and hinge holding laminations, and opposite wedges on the "plane" side to fill in the gap.
Cutting the long wedges and getting them a regular shape will be a challenge-you really need a micro-bandsaw!
They are tapered to suit the trailing edge taper, in many cases.
10) There are things to learn about spiral instability and vertical surface design, and types of flaps. - see Mr Lennon's book (PDF)-check Google.
There are also some more handy tips on RC model dynamics.
Comparison of the strength properties of birch, spruce, birch ply.
birch = 8170 psi compressive, 16600 psi tensile, density 510-770 kg/cu mt, depending on source, species
spruce = 5610 psi compressive, 10,200 psi tensile, 430 - 780 kg/cu mt, depending on species.
metric: birch = 56 MPa, 114 MPa (50%)
Birch ply - design stress tensile = 47.6 MPa, actual tests show 56 to 100 MPa.
For design use, use birch ply at around 20 MPa (compressive, beams)
Birch wood - use about 30 MPa compressive, along the grain.
Spruce wood-use about 20 MPa compressive, along the grain.
This shows that birch 3-ply is competitive with spruce wood for spars.
It has an advantage in that spars can be bent or curved in plan view, and can also be easily merged together.
You still need to do the bending stress calculation for a box-beam to ensure spar strength is in line with expected
aero loading- full flip-stop at top speed, (20-30 Gs) pull out from a full dive with normal trim (10 Gs), or low impact manoeuvres of 2-3 Gs.
(loops, rolls, banked turns).
Stress formulas, and section moment calculations are available in wikipedia, or elsewhere with a Google search.
It is easier in si units, otherwise you need to find the correct multiplier factors between mass, weight, strength, length units.
birch = 8170 psi compressive, 16600 psi tensile, density 510-770 kg/cu mt, depending on source, species
spruce = 5610 psi compressive, 10,200 psi tensile, 430 - 780 kg/cu mt, depending on species.
metric: birch = 56 MPa, 114 MPa (50%)
Birch ply - design stress tensile = 47.6 MPa, actual tests show 56 to 100 MPa.
For design use, use birch ply at around 20 MPa (compressive, beams)
Birch wood - use about 30 MPa compressive, along the grain.
Spruce wood-use about 20 MPa compressive, along the grain.
This shows that birch 3-ply is competitive with spruce wood for spars.
It has an advantage in that spars can be bent or curved in plan view, and can also be easily merged together.
You still need to do the bending stress calculation for a box-beam to ensure spar strength is in line with expected
aero loading- full flip-stop at top speed, (20-30 Gs) pull out from a full dive with normal trim (10 Gs), or low impact manoeuvres of 2-3 Gs.
(loops, rolls, banked turns).
Stress formulas, and section moment calculations are available in wikipedia, or elsewhere with a Google search.
It is easier in si units, otherwise you need to find the correct multiplier factors between mass, weight, strength, length units.
I have an interesting idea on how to fabricate the large-tube wing joiner for a 100 inch plane.
1) The tube has to line up with the wing dihedral.
2) We are getting into drain-pipe sizes here: maybe 2 inch tube?
Possibly there may be a hardware-related tube about that size. Material doesn't matter.
Brass would be good, to hold the retaining pin, but a brass cross-tube could be bonded in.
3) The tube should be glassed into the center spar structure on the fuselage.
glass fiber composite needs to be applied inside the tube.
4) The socket end can be made from glass fiber composite as well, integrating into the wing
ply box spar. You need a fairly thick layer of wax on the tube - maybe 5 thou??
to allow for composite shrinkage. Look up allowances for shrinkage for glass fiber /resin composite.
Otherwise, the joint won't come apart. Not too much, or the joint will be sloppy.
Any extra slop could be taken out by painting dope on the tube.
6) Then bore for, and bond in the brass tube liners.
7) This tube pin/socket is now complete!
8) Use an elastic gasket at the wing join, to prevent jacking damage to the facing surfaces.
This may also take up any retaining pin slop.
9) Lubricate the tube joint with wax to stop it welding together!
1) The tube has to line up with the wing dihedral.
2) We are getting into drain-pipe sizes here: maybe 2 inch tube?
Possibly there may be a hardware-related tube about that size. Material doesn't matter.
Brass would be good, to hold the retaining pin, but a brass cross-tube could be bonded in.
3) The tube should be glassed into the center spar structure on the fuselage.
glass fiber composite needs to be applied inside the tube.
4) The socket end can be made from glass fiber composite as well, integrating into the wing
ply box spar. You need a fairly thick layer of wax on the tube - maybe 5 thou??
to allow for composite shrinkage. Look up allowances for shrinkage for glass fiber /resin composite.
Otherwise, the joint won't come apart. Not too much, or the joint will be sloppy.
Any extra slop could be taken out by painting dope on the tube.
6) Then bore for, and bond in the brass tube liners.
7) This tube pin/socket is now complete!
8) Use an elastic gasket at the wing join, to prevent jacking damage to the facing surfaces.
This may also take up any retaining pin slop.
9) Lubricate the tube joint with wax to stop it welding together!
More good ideas! I like the blade into socket idea. - line with carbon fiber tape?
This makes dihedral easy!
https://www.rcuniverse.com/forum/sc...gn-3d-cad-174/6944473-wing-tube-dihedral.html
The tongue in slot should really be in similar proportions to the tube-maybe 2/3 width to 1 height?
This has similar bearing capability to a tube, and allows enough cross-section to act as a spar.
Laminate from layers so the dihedral angle change can be drawn on the face of the ply.
Make from ply, paint with epoxy, lubricate with wax. A simple through-pin with a spring clip would do to retain it, too.
The socket should be made from face-on ply, laminated, in a channel section. A simple edge-glued box is not strong enough,
as you are applying tension across the grain.
This can then have the box-spar sheets lapped over it to maximise the glued area.
Close fitting: The tongue should have a smooth finish and have a slight taper.
Then the epoxy-fill method can be used to get a close fit.
Wax everything the epoxy can touch, if it squeezes out!
This makes dihedral easy!
https://www.rcuniverse.com/forum/sc...gn-3d-cad-174/6944473-wing-tube-dihedral.html
The tongue in slot should really be in similar proportions to the tube-maybe 2/3 width to 1 height?
This has similar bearing capability to a tube, and allows enough cross-section to act as a spar.
Laminate from layers so the dihedral angle change can be drawn on the face of the ply.
Make from ply, paint with epoxy, lubricate with wax. A simple through-pin with a spring clip would do to retain it, too.
The socket should be made from face-on ply, laminated, in a channel section. A simple edge-glued box is not strong enough,
as you are applying tension across the grain.
This can then have the box-spar sheets lapped over it to maximise the glued area.
Close fitting: The tongue should have a smooth finish and have a slight taper.
Then the epoxy-fill method can be used to get a close fit.
Wax everything the epoxy can touch, if it squeezes out!
Last edited:
I just measured my car (wagon) with the back seat down. The proposed size of 42 inch span by 69 inch length just fits!
the extreme length is due to the jet-fighter proportions. I already chopped 2 inches out of the middle, leaving just enough room
for the propellers. looking at the front view, the 9 inch propellers look quite small relative to the span.
Hopefully big enough to take advantage of "blow-over" lift, otherwise I shall have to add blower tubes at right angles with 3 inch props,
which adds more weight.
I am also experimenting with "Pull-over" lift on the front wing, with large elevators.
The theory is that it will cause a low pressure area above the front wing, and the static air pressure below will push it up.
I have abandoned tail-sitter ground stance, and want it to pivot up a bit before doing a normal forward launch, using ground effect
and blow-over lift to keep it off the ground until it reaches flying speed. This is effectively VTOL, but at an angle.
Landing will also be at an angle, but also be a "blow-over" type landing, at very high angle of attack, with very short roll and low forward speed.
I am hoping to get past 30 degrees angle of attack, and below 5 mph ground speed. This is more a STOL type landing, or tail-sitter with a little forward motion.
the extreme length is due to the jet-fighter proportions. I already chopped 2 inches out of the middle, leaving just enough room
for the propellers. looking at the front view, the 9 inch propellers look quite small relative to the span.
Hopefully big enough to take advantage of "blow-over" lift, otherwise I shall have to add blower tubes at right angles with 3 inch props,
which adds more weight.
I am also experimenting with "Pull-over" lift on the front wing, with large elevators.
The theory is that it will cause a low pressure area above the front wing, and the static air pressure below will push it up.
I have abandoned tail-sitter ground stance, and want it to pivot up a bit before doing a normal forward launch, using ground effect
and blow-over lift to keep it off the ground until it reaches flying speed. This is effectively VTOL, but at an angle.
Landing will also be at an angle, but also be a "blow-over" type landing, at very high angle of attack, with very short roll and low forward speed.
I am hoping to get past 30 degrees angle of attack, and below 5 mph ground speed. This is more a STOL type landing, or tail-sitter with a little forward motion.
I am now looking at fitting some ducted fans: 1x 3 inch behind the cockpit, and 2x 4inch behind the main wing spar. This is well behind the MAC, which will also be the COG. The main objective is to add a bit more front lift, and roll stabilise the aircraft, and help control pitch.
I propose splitting the gyro pitch/roll control between the elevators, elevons, and the 3 motors.
I want the nose to lift, but stabilise between 35 and 45 degrees, until a reasonable altitude is obtained.
This makes optimum use of the main prop flow.
Once the aircraft is at a reasonable height, the flight path can be flattened a bit to improve acceleration.
The lift fans are to shut off completely once flying speed is reached,-about 40 mph.
There are several questions:
1) should I try to get the rear fans closer to the COG?
2) do the rear fans need to be 4 inch? Would 3 inch do the job?
I expect flow from the main props to be able to fully lift the wing. Most lift will occur at the wing AC, however.
The MAC is further forward.
I am also trying "pull-over" lift on the front wing, which may be insufficient.
3) should I add extra micro-actuators and folding slats over the holes for the fans? Air passing through the wing holes,
particularly, will reduce wing lift.
the extra motors are only about 16 g each, plus props and tube structures.
Adding slats and actuators may double this weight.
4) do I need to add a pitot tube and module, and loop the airspeed back into the control system?
Will Ardupilot cope with this, or does it need extra computing/software to be added?
I propose splitting the gyro pitch/roll control between the elevators, elevons, and the 3 motors.
I want the nose to lift, but stabilise between 35 and 45 degrees, until a reasonable altitude is obtained.
This makes optimum use of the main prop flow.
Once the aircraft is at a reasonable height, the flight path can be flattened a bit to improve acceleration.
The lift fans are to shut off completely once flying speed is reached,-about 40 mph.
There are several questions:
1) should I try to get the rear fans closer to the COG?
2) do the rear fans need to be 4 inch? Would 3 inch do the job?
I expect flow from the main props to be able to fully lift the wing. Most lift will occur at the wing AC, however.
The MAC is further forward.
I am also trying "pull-over" lift on the front wing, which may be insufficient.
3) should I add extra micro-actuators and folding slats over the holes for the fans? Air passing through the wing holes,
particularly, will reduce wing lift.
the extra motors are only about 16 g each, plus props and tube structures.
Adding slats and actuators may double this weight.
4) do I need to add a pitot tube and module, and loop the airspeed back into the control system?
Will Ardupilot cope with this, or does it need extra computing/software to be added?
1) flow through the vertical tubes shouldn't be a problem, as the lower tube edges are sharp, and the normal airflow won't make it round the corners.
2) I don't need a pitot tube, just a ground based on/off swich for the motors. The flight profile can be limited in the flight controller-initial pitch-up, flattening out at about 2-3 meters altitude.
2) I don't need a pitot tube, just a ground based on/off swich for the motors. The flight profile can be limited in the flight controller-initial pitch-up, flattening out at about 2-3 meters altitude.
I have advanced a bit now, I am looking at tilt-motors, and only one front thruster, limited to 15 amps-I really blew my amps budget!
The wing has changed to suit. I think I posted that in a different thread-the one about 5 motors and one battery??
I am reverting to belly launch and landing, possibly vtol/stol??
This is similar to the belly launch, modified tailsitter launch profile.
I am looking at ducted props under the main wing, plus a third rear wing or tail plane, which mainly carries the elevons.
I have moved the front elevators closer to the front of the ducts, to encourage pull-over lift.
Several problems exist for this design:
1) The planned stall-out of the main wing will lose most rear lift, reverting to high angle of attack, under-wing lift only.
This may cause a rather discontinuous flight experience, and need assistance from gyro pitch and roll stabilising.
Normally this is considered an unrecoverable stall with a Canard, but with the extra thrust, high under-wing flow, extra tailplane,
and pull-over front lift, it should be manageable in this case.
2) The Thrust line is offset from the COG, so thrust and flight surfaces will oppose each other in a tailsitter descent.
3) The thrust line is lower than the profile center of area, so it will be prone to pitch-up on acceleration.
What do you think?
My poor 1:5 scale drawing is due for another major whiting-out!
This is similar to the belly launch, modified tailsitter launch profile.
I am looking at ducted props under the main wing, plus a third rear wing or tail plane, which mainly carries the elevons.
I have moved the front elevators closer to the front of the ducts, to encourage pull-over lift.
Several problems exist for this design:
1) The planned stall-out of the main wing will lose most rear lift, reverting to high angle of attack, under-wing lift only.
This may cause a rather discontinuous flight experience, and need assistance from gyro pitch and roll stabilising.
Normally this is considered an unrecoverable stall with a Canard, but with the extra thrust, high under-wing flow, extra tailplane,
and pull-over front lift, it should be manageable in this case.
2) The Thrust line is offset from the COG, so thrust and flight surfaces will oppose each other in a tailsitter descent.
3) The thrust line is lower than the profile center of area, so it will be prone to pitch-up on acceleration.
What do you think?
My poor 1:5 scale drawing is due for another major whiting-out!
