OK....airfoil nerds I need your help!

[David L.]

OK...so I think I want to design my own plane from scratch for aerial photography. I want to haul heavier video or still cameras in the 1-2 pound range. I thought for about a millisecond using my slowstick for this but the resulting mental picture was not very appealing;)

So here I am thinking about all the qualities and characteristics I want this airframe to have. I want to be able turn off the motor
and be able to slowly glide/float with plenty of stability. On the other hand, when I turn the motor on WOT, I want to be able to really get around without to much problem...no need to go vertical obviously but it should be able to climb with authority as well as fight winds fairly well in the 15mph range...preferably 20mph. I would also like the wing to have a very efficient/high lift airfoil. I am leaning towards a custom cut foam wing for no good reason other than building simplicity and hopefully strength.
I can't decide if I want a straight wing or a wing with dihedral. Also, what are the handling differences between a biwing and a single? Knowledgeable opinion here would be appreciated.

Yesterday, out into cyberspace I go searching for the perfect airfoil or at least some guidelines for this airframe. The more I read the more I realize that airfoil design can only be understood by those that work for NASA. Some of this stuff is fascinating and at the same time mind boggling! I ran across an airfoil that is used for human powered flight, it is called the Wortman FX63-137. It seems like this would be a good candidate for my application but really I have no clue.

I finally think I have the answer and download Compufoil...seems like the right thing to do. WOW! I never knew that an airfoil was so complicated. Off I go fiddling with this program...and getting absolutely nowhere! I realize I am out of my league here. I call to all who really dig this sort of thing.

First off, how in the world do you design your own airfoil? Is it a matter of artistic preference? Do you just draw up something that looks pretty in compufoil and then slap it on your plane? How do you know what the lift, drag, etc. characteristics are without building your own wind tunnel? I know there must be a way but seriously, this stuff seems way...way over my head. I do want to learn how to understand all this stuff so I can be a real nerd but it seems like it would require at least a semester in aeronautical engineering!

Here is what I really want to know: Foam wing? Dihedral or straight? Wortman FX63-137 airfoil? Tips for using Compufoil?

I know I'm asking allot here but any tid bits I can get would be greatly appreciated! Thank you. David

[Sparky Paul]

If you are at Square Zero, you cannot design a good airfoil on your own. Don't try that.
Just about every way possible to get from the leading edge to the trailing edge has already been tried, and those efforts that work, kinda, are in the public domain, in the files that are attached to most airfoil programs.
There's several ways to choose an airfoil.
First, look at what works on something that does what you want. Use that.
Second, ask what someone recommends.
Third, pick the cute one. :)
For your purpose, the airfoil can't be too high a lift.. that means drag, and your power is limited. ( Presuming electric).
You know you need penetration.
So the shape you want will be in the utility glider style... semi-symmetrical for a low pitching moment, making trim changes slight,
and thinnish, say no more than 12%.
I don't know if Compufoil can do this, but Profili can look for similar airfoils...
Beginning with the Clark-Y as a base, it finds:
Clark X
Clark Y 11.7%
Clark Y2
Clark K
Clark W
E193-12
Goe 693
Goe 796
Clark V
Goe 285
NACA 3412
Davis
SB 96 12.7/3.0
SBC3
Goe 623
Goe 436
Goe 438
SB 96 11.6/3.0
Eppler 67
NACA M15
..
For simplicity, I'd use the old standard, Clark-Y.
.
For the wing... Dihedral. a few degrees with ailerons, much more without ailerons.
Foam is heavier than air.
Balsa/spruce MonoKote structure.
A Kadet Seniorita wing kit will do the jobl
No need to reinvent a wheel, which might be wobbly. :)

[Ollie]

You don't want to design your own airfoil because there are thousands to choose from and a few will fit your application better than any you are likely to design yourself.

To select an appropriate airfoil you have to have an idea of the speeds an sizes (chords) that you want so that you can find the range of reynolds numbers and from that the range of profile drag coefficients aswociated with the range of coefficients of lift the airfoil is to perform over.

In level flight under power the thrust is equal to the total drag. If the drag is a little high it just means you need a little more power to fly level and a lot more to climb briskly, which is easy to come by.

During the glide drag reduction is a high priority because any increase in drag degrades the glide angle and increases the sinking speed. At the best glide angle the induced drag is exactly half the total drag and at the lowest sinking speed the induced drag is more than half the total drag. That makes induced drag reduction the number one priority in the list of possible drag reduction opportunities. The induced drag is proportional to the square of the coefficient of lift and inversely proportional to the aspect ratio. For a high performance in the glide mode a high aspect ratio wing is a very high priority. If you want to take pictures of big chunks of landscape then you need to fly at a fairly high altitude and a big wing span is an advantage for visibility.

On the other hand, for aspect ratios above about 8 or 10 the wing structure becomes more complicated than just wood over foam to have adequate bending strength.

Now, to go further we need some specific objectives for the minimum stalling speed, the maximum rate of climb and any limitations you want to put on the size of powerplant and the size of the model for storage and transportation.

The airfoil selection comes later.

[Ollie]

Something like a 1/4 scale Fournier RF4 motor glider would be about the right size and configuration. See:
http://www.mrmodels.fsnet.co.uk/p5kits.htm

This would require dead stick landings after shutting down the engine. If you want to be able to restart the engine after glide mode then the size should be enough bigger to carry the extra weight of the restart equipment.

[David L.]

Thanks for your replys. I would like the stall speed of this model to be around 10-13 mph....I have no idea if that is realistic or not. As far as climb rate...I don't know what to say other than what Motocalc says "this is a good climb rate for a basic trainer". The climb rate of a trainer or a little better would be plenty. Wind penetration is important to me.

As far as the plane itself, I am envisioning a stick or tube type fuselage (kind of like the slowstick) I'm not really going for scale or pretty looks but rather more of a work horse type of plane...no need to look like a real world full scale aircraft with decals and a detailed paint job. Certainly, I won't mind if it looks "cool" but that is not necessary. (All I really need is my name on the side with an "If found call:......:D )

I would prefer an airframe that could be easily modified, cameras strapped on, motors easily tinkered with and wings easily detached for transport. A very simple wing mount would be a plus...rubber bands would be acceptable. I may have to design some aerodynamic housings for cameras in order to keep drag down. I will only settle for electric power...the outrunner sort preferably (though not critical...I just love the quite power).
Flight times would need to be at least 25 minutes or more. LiPoly's would be my battery of choice.

I still like the idea of foam wings...more because of durability than anything.

Ollie, does this give you something better to go on? Thanks! David L.

[Sail 'n Soar]

Thanks for your replys. I would like the stall speed of this model to be around 10-13 mph....I have no idea if that is realistic or not. As far as climb rate...I don't know what to say other than what Motocalc says "this is a good climb rate for a basic trainer". The climb rate of a trainer or a little better would be plenty. Wind penetration is important to me.

Your desired stall speed along, with the series of foils Sparky recommends, translates to a wing loading on the order of 4 to 7 ounces per square foot. A light wing loading like that on a conventionally tailed planform usually doesn't correlate with good wind penetration.

With 1-2 pounds of camera payload you are most likely dealing with at least a 2 - 3 pound aircraft weight less payload, or an all up flying weight in the 3-5 pounds range minimum. At 7 ounces/squar foot (~13 mph stall speed) you will require somewhere between ~ 1,000 and 1600 square inches of wing area.

Wing span for ~ 7 aspect ratio wing will be between 7 and 9 feet. If you follow Ollie's suggestion and go the RF-4 route (AR~11), then you will be dealing with an 8+ to 11+ foot span wing. This will not be a small model, which the 2 - 3 pounds of motor, airframe, battery, and radio figure seem quite optomistic!

Recalculating the above for either the 10 mph stall or a heavier airframe will just make your notional model bigger yet. Unless this is the size plane you had in mind you may want to rethink your design goals.

[Ollie]

You want good wind penetration, slow stalling speed, simple comnstruction, a foam cored and wood skinned wing, electric power, 20 minute flight time, and a payload of 2 pounds. The bigger the plane, the more expensive, harder to transport and store. The bigger the plane, the easier to come close to your wish list. Big planes are expensive in both airframe and propulsion. So, give us a maximum allowable wing span and a budget cap.

You can't have it all so its time to prioritize your wish list. This will allow us to trade off the least desirable wishes in favor of the most desirable.

[Sparky Paul]

This one performed well. S3021 airfoil.
The "fuselage" is a 3" dia. mailing tube, totally adaptable.
Probably wouldn't care to have 2 pounds of camera in it though.
1 pound OK.

[Sparky Paul]

This version has more room inside, using the same wing and power. But it doesn't have the performance of the pod-boom combination.

[David L.]

Thanks guys for the responses! This is fascinating. I certainly realize that this plane would be large. I have no problem with a 7-8 foot wing span...just as long as there is some method of separating it in the middle for transport. Also, what are the possibilities of stacking the wing in a Biplane configuration? What I am willing to sacrifice? I can sacrifice payload carrying capability down to about 20 oz's rather than 2 pounds. (These days there are lots of quality lightweight photo options...I just want some tolerance to play with...i.e. more battery weight for longer flight times.) As far as my budget...hmmm lets see what we can do for $500-$800. David L.

[David L.]

Sparky, Those planes look pretty neat, especially the top one.
The balsa, mylar combo seems scary though. I guess I don't like the thought of hours of rebuilding a balsa wing if I crash. Would a foam wing have any advantages over a balsa construction. I am willing to be convinced to use balsa mylar if necessary. David L.

[Sparky Paul]

Based on volume alone, a 2 lb/ft^3 foam wing of that size would weigh 1/2 what the balsa/film version does.. but adding the required spars and covering would eliminate that advantage.
It's really a personal choice.. the wing illustrated broke too severely to be repaired.. one side is fine the other is land-fill. I look at it with the idea of a new right half, but I dunno..
I prefer working with wood over foam, since I can't cut foam.
I'm thinking of the referenced Seniorita kit wing, with 48" wood to replace the 32" spars etc in the kit. Getting lazy in my older age. :(

[Ollie]

Here are some of the trade offs. Weight saved in the airframe can be applied to a better rate of climb, a lower stalling speed, a smaller and less costly propulsion system and a better glide.
Pultruded carbon fiber spar caps have about 20 times the strength to weight ratio of wood. Such spar caps will allow thinner airfoils for better wind penetration and higher aspect ratios. Higher aspect ratios will improve rate of climb and decrease power requirements as well as improving glide ratios and reduce sinking speeds.

A pay load with low frontal area and mounting inside a streamlined fuselage will greatly improve wind penetration and reduce power system requirements.

Saving weight in the airframe generally makes construction more difficult and complex.

[David L.]

Ok, tell me what you think of this set up...this is just fumbling with the numbers in Motocalc. Obviously, Motocalc can only give me a very rough idea of how this thing might fly...or might not fly.

This is what I have come up with.

90 inch wingspan. 18" at the fuselage and 10" at the wing tips. Foam core with ability to be separated in the center.

Glider style airfoil....just guessing.

Approx 1260" of wing area.

Twin carbon fiber tubing fuselage. (I'm thinking a twin fuselage might help with the mechanics of joining the wing in the middle as well as allow for more options in mounting cameras.)

Pusher prop in back between double fuselage. (leaves front wide open for camera view)

Uberall Nippy black 1812/100 motor. (Does it have the same power running in reverse for a pusher prop configuration?)

I would like the total airframe with motor and batteries to weigh in at about 48 oz without the camera. Is that weight reasonable?
I have no idea what all the building materials will weigh.

With camera...between 55 and 60 oz total all up flying weight.
This should give me a wing loading of around 7 oz per sq ft.
and an approx. stall speed of maybe 13mph.

Opinions? Thanks for putting up with my lack of aerodynamic savy. Sincerely, David L.

[Sparky Paul]

A 10 cell 3000 mah NiMh pack weighs 22 oz.
Don't have any experience with Li-pos..
A 9oz/ft^2 loading will perform well.
Use a very high range ESC. These burn up easily when run close to their current limit.

[Ollie]

Assuming one pound per cubic foot foam and an airfoil about 10% thick, The cores will weigh about 10 ounces. Assuming 1/16 inch thick medium balsa sheeting, the sheeting weight will be about 16 ounces. Allowing for adhesives, film covering and, wing joiner structure, the wing will weigh about 2.5 pounds. That only leaves about a half pound for motor, batteries, speed controler, receiver, servos, fuselage pods, landing gear, prop, tail booms and tail.

I think you should go for a stalling speed in the neighborhood of 22 FPS or 15 MPH. That should be achievable with an S3021 airfoil and an all up weight including camera of about 80 ounces. Try that on Motocalc and see what the motor, speed controller and battery weights will be for 20 minute running time at half power.

You can make better use of the wing sheeting by going to a 96 inch span and decreasing the root and tip chords by an inch. This will increase the aspect ratio to 7.3 and make the plane more efficient too. You could also save about four ounces of wing skin weight by sheeting the wing with light contest balsa for all but a 6 inch wide diamond of medium balsa on the thickest part of the top of the wing. This won't decrease the bending load that the wing will carry by very much.

[Ollie]

Here are some more things you can do refine the wing design. With a span of 96 inches, the tip chord should be increased to 11 inches and the root chord decreased to 15 inches. The reduction in taper (increase in taper ratio) will greatly improve the tip stall margin and allow the wing to achieve a coefficient of lift , with the S3021 airfoil, of 1.1. This will allow the wing loading to be increased to 10 ounces per square foot with a stall speed of 15 MPH. Of course,the gross weight should be kept as low as possible but you may find it difficult to do so and the 10 ounce per square foot wing loading is more or less an allowable upper limit for a gross weight of 5.5 pounds. I think you should use 5.5 pounds when determining the power requirements.

The wing bending strength will allow 10G maneuvers without folding the wings.

[David L.]

I think I could live with a span of 96 inches. As far as wing covering....are there other lighter weight options for covering than balsa? The balsa covering just seems heavy but I guess its not that bad. I've heard of people using thin fiberglass and polycrylic....probably not any lighter. I suppose that the covering is not only for pretty looks but also for strength? I was hoping that I would need to do nothing more than insert some carbon fiber reinforcements, coat with polycrylic or something of the sort, then a final mylar covering.

As far as stall speed...I suppose 15mph isn't to bad. A faster plane might give me more stability I suppose. I definitely want this plane to be slow flying though.

Geepers! 5.5 pounds seems heavy! Is that all up flying weight?
These things get heavy fast as they get bigger. I was thinking that since my slowstick has an all up weight of 20 oz. and can carry a 8oz load fairly well, then a 50 oz model should be able to handle at least 20oz fairly well. I can see that not all things are completely proportional that way. I am willing to live with 5.5 pounds but I certainly want to take every reasonable and cost effective measure to keep weight down.

Can you explane what "aspect ratio" means?

Have you seen the "winglets" on the tips of some wings?
what are those for and would they be usefull for me if I decided to get rid of the dihedral in order to simplify the joint in the middle?

David L.

[Sparky Paul]

This has flown well at 122 oz.. 2 sets of 10x3000 NiMhs.
Uses a Speed 700, direct, 10x6 prop.
I've found that providing for controllability and durability add more weight and size relative to the package weight when extended performance is desired.
A full-fuselage such as this though is restrictive in "look-angles" for the camera(s).. the forward/aft pod style is more adaptable.
The internal volume of this wing is 1100 cubic inches.. the other wing is 550 in^3.
.
"Aspect ratio" is the ratio of span to chord..
For a non-tapered wing, it's the span divided by any chord..
Tapered wings, such as 100" above, root chord is 15", tip chord is 12", so the average chord is 13-1/2".
100/13.5= 7.4

[Sparky Paul]

This one is a spritely performer... made to the SAE 2003 rules, it came overbuilt and heavy.
But could take a 2 pound or 5 pound cargo as high as you'd like.
OS 61 FX motor.
But the field of view is limited to sideways and down.
The plate the main wheels attaches to is actually a small tray, which is removeable to get access to the cargo bay, which is about 5"x6"x12".
Advantage over electric is the weight. No large batteries to take away from the lifting capacity.
Disadvantage is the inability to stop and restart the motor in flight.
.
A link to the planes I flew last year.. lots of camera planes..
http://www.angelfire.com/indie/aerostuff/Multipix-01.htm

[winmodels]

Hi David l.

Going through your initial "specification" my initial thoughts lead to a plane on the lines of a modern photo recce RCV.
If electric drive and the envisiged endurance is taken for granted the payload (camera) will not be more than approx. 10% of the total weight. As outlined in the early answeres this will automatically lead to a large model. If you also want low Vmin and good penetration the wing need to have high lift flaps.
On the other hands your budget looks about right (without RC gear) for a DIY project.

Regards

Klaus

[Ollie]

Here is a straw man design for comment:
Wing:
Span 96 "
Root chord 15"
Tip chord 11"
Airfoil S3021
Wing wt. 2pounds
Skinned with 1/16 balsa on top, 1/32 balsa on bottom and 1 #per Cu. Ft. styrene foam core.

Tail:
Moment arm 36 inches.
Inverted V-tail area 280 sq. inches
Foam core covered with 1/32 contest balsa.
Airfoil S8020
Wt 6 ounces

Tail booms:
Tapered carbon fiber tubes
Wt. 3 ounces.

Camera:
Wt. 16 ounces

Propulsion;
Motor AXI282010, 6 ounces
ESC JES 30-3P, 1 ounce
Battery LiPoly 6, 3270 mah cells (3S2P) Wt. 14 ounces

That leaves 11 ounces for landing gear, fuselage pods, radio and prop (10-6) to a gross weight of 5.5 pounds and a wing loading of 10 ounces per square foot. If you keep it aerodynamicall clean it will make good headway under full power in a 20 MPH wind.With an airspeed of 35MPH, 15MPH ground speed ina 20 MPH head wind, the wing will be poerating at a coefficient of lift of 0.2. The induced drag coefficient will be 0.0017, the profile drag coefficient will be 0.009 and the parasitic drag coefficient will be around 0.0004 for a total drag coefficient of about 0.0211. That means tha the thrust required for level flight at an airspeed of 35 MPH would be less than 10 ounces. No problem! We probably could get away with half the battery and save about 7 ounces. since flight would be at half throttle or so except during climb.

[winmodels]

So to be realistic, we are actually talking of 200+oz AUW that's about 5.5 to 6 kg and a span of 120" ~ 3m.
At an AR of 10 this gives you 0.9m^2 and a wing loading of 6.3kg/m^2 or 63gr/dm^2 ~ 20oz/sq".
For a decent flying speed -say 45mph - this requires ca' 2kW of power and a take off speed of 15mph seems possible.
The attached picture is only food for thought.

Have a go!! There are enough people here to help.

Klaus

[winmodels]

Hi Ollie

we should better coordinate our endevours :-), but as one can see there are always several approaches to a technical exercise.

Regards

Klaus

[Sparky Paul]

Something like this... :D

[David L.]

Awesome! You guys have been a real help. Though I can't understand half of the terminology...."parasitic drag coefficient"...it sounds like what I need to do now is start thinking about how this thing is going to be put together.
It sounds like the Axi will be an awesome motor for this beast.
Hobby Lobby has it for $96.50...a pretty cheap but powerfull brushless option I think.

If it is able to penetrate the wind as you estimate, I will be more than happy with that.

I talked to the guy at Flying Foam today and it sounds like it may be easier to build a detatchable foam wing this size without any dihedral. I will probably go with some sort of "plug in" type of arrangement for joining the halves.

David L.

[Ollie]

Yeah, Paul. You do nice work.

The wing with its thin soft skin on the bottom won't take much abuse. We need about 7 to 10 degrees of dihedral to provide ground clearance or landing gear. The landing gear is mandatory unless you can find a suitable folding prop and the motor can be wired to run backwards. The landing gear has to be long enough so that the plane can rotate on take off without the prop hitting the ground.

[Sail 'n Soar]

Originally posted by Ollie
Here are some more things you can do refine the wing design. With a span of 96 inches, the tip chord should be increased to 11 inches and the root chord decreased to 15 inches. The reduction in taper (increase in taper ratio) will greatly improve the tip stall margin and allow the wing to achieve a coefficient of lift , with the S3021 airfoil, of 1.1. This will allow the wing loading to be increased to 10 ounces per square foot with a stall speed of 15 MPH. Of course,the gross weight should be kept as low as possible but you may find it difficult to do so and the 10 ounce per square foot wing loading is more or less an allowable upper limit for a gross weight of 5.5 pounds. I think you should use 5.5 pounds when determining the power requirements.



All mostly true. Checking out the S3021 polars at RN to 300,000 on http://www.nasg.com/afdb/show-polar-e.phtml I come up with a CLmax of ~1.15. Then going to http://aero.stanford.edu/WingCalc.html, the 1.15 max section CL translates to an average CL of ~ 1.05 for ollie's wing geometry. The impact will be either having to design to about 8 oz less gross weight, or accept a stall speed of about 4% higher.

No big deal.

[Ollie]

For a video system see:
http://www.wirelessvideocameras.com/public/specs.htm#HLF11
This system would only weigh about 8 ounces if you could run it from the propulsion battery. That would be another hqalf pound saved. You might have tolengthen the nose a little to balance the model with this light weight camera. The fuselage pod wouldn't have to be much larger in diameter than the AXI motor.The main limitation is the 1100 foot range of the telemetry.

[David L.]

Thats cool Sparky! Thats almost as I envisioned it. I'm not sure about the inverted V-tail though. How would that effect flying manuverability as opposed to a more traditional tail configuration?

Also, the pusher prop is not something I know enough about. I've heard that misaligned thrust on a pusher can be a real handful. I really want a pusher configuration though so I will make it work.

Now what about stability? What would be the most stable configuration to put this altogether. Sparky's CAD drawing looks like it would be pretty stable but of course I really have no experience with this sort of airframe.

David L.

[Ollie]

The tail size, together with the tail moment arm length will result in an adequate horizontal tail volume coefficient of 0.41. This will provide plenty of pitch stability with the proper CG location of about 30 to 35% or the chord. The vertical tail volume coefficient will be somewhere around 0.03. This will provide more than adequate yaw stability. With a flat wing there will be no spiral stability. With a dihedral angle of 7 degrees per side there will be adequate roll and spiral stability. With a dihedral angle of 10 degrees per side there will be ample roll and spiral stability. With an included angle of about 105 degrees between the tail panels the handling will be very nice.

With enough dihedral the pilot work load will be fairly light. In other words, it will be a pussycat to fly but, not very exciting or aerobatic. It will be stable enough that it won't require the pilot's constant attention or a very high skill level to guide it. Because of its size, its reaction to gusts or turbulence will be more sedate than a smaller model.
If the AXI will run properly in reverse, it won't require pusher props even though it is a pusher configuration. I think it is just a matter of reversing two of the three leads between the ESC and the motor.

Several ounces of wing joiner structure can be saved by making the wing one piece, joining the halves together at the proper dihedral angle with several staggered layers of light fiberglass cloth in laminating epoxy.

Sail and Soar,
Thanks for correcting my error on wing maximum lift coefficient. Please check me out on the tail volume coefficients.

[Sparky Paul]

Originally posted by David L.
Thats cool Sparky! ...

Also, the pusher prop is not something I know enough about. I've heard that misaligned thrust on a pusher can be a real handful. I really want a pusher configuration though so I will make it work.

...

David L.
.
Few people have designed and flown pushers.
There's a lot of myths around them.
A high mounted motor can push the nose up or down if it's angled wrong, but it must be really wrong..
What is true is the motor being aft of the c.g. makes for a lot of nose weight, and/or a longer nose to put the weight.
The booms add to this problem.
Using CF pushrods will alleviate the effect of the length of the rods.
This one flew well...( the wing was destroyed on another plane)
Span was 99"...
As Ollie mentions, these things need not be aerobatic. The less active the better. Even consider the addition of a wing-leveler like the FMA Co-pilot. The less workload on the pilot the easier it is to position the plane properly

[David L.]

So how many channels of control would this plane require?
V-tail mixing? Ailerons or flaperons? 5 channels...pitch, roll and yaw, throttle plus a free channel for shutter release? David L.

[David L.]

Several ounces of wing joiner structure can be saved by making the wing one piece, joining the halves together at the proper dihedral angle with several staggered layers of light fiberglass cloth in laminating epoxy.

What if the joining section in the middle of the wing were straight with the dihedral shaped into the last 2 feet or so on each side?

I definitely want the plane to be easy to fly...very stable and easy to guide around. Hopefully with a very smooth and vibration free flying style. I want to avoid as best I can the many herky jerky videos that are so common with RC aircraft videos.
As a matter of fact, smooth silky broadcast quality would be great...probably aided by some in camera optical vibration reduction.

By the way Ollie and Sparky, your knowledge of this stuff is very impressive. I wish I could understand all the terminology. Thank you for your willingness to share so freely. I am like a sponge, I love this stuff. I will glean what I can. David L.

[Sparky Paul]

Originally posted by David L.
So how many channels of control would this plane require?
V-tail mixing? Ailerons or flaperons? 5 channels...pitch, roll and yaw, throttle plus a free channel for shutter release? David L.
.
The planes I've posted are all 4-channel, rudder, elevator, motor, camera.
No aileron surface.
But...
I mix the right stick (aileron) to the left stick (rudder). This eases the takeoff brain-fade. It's called the 1-4 Mix.
I've flown left stick steering on the ground, right stick steers in the air, so often that I have difficulty taking an aileronless plane off with rudder on the left stick.
The servo is plugged into the rudder channel on the receiver; the mix lets the right stick move that servo as well as the left.
MOF< in the pusher, there's a seperate servo for nose-wheel steering, instead of a rudder per se.
One servo in each wing for the seperate "horizontals", which use the V-tail mix, not Ail-Elev, on that plane and the other large one. The 1-4 mix runs the turning feature of the vee-tail, and channel 2 works the elevator function.
I've used channel 4, the rudder, on planes I hand launch which have no rudder anyway,
and channel 5, which is usually a 2 or 3 position switch for the camera trigger.
The 6-channel radios will do the job, as I believe they all provide V-tail mix, and 1-4.

[Ollie]

David,
If you want to understand more, I recommend Model Aircraft Aerodynamics by Martin Simons. The explainations are in plain english. Most of the formulas are contained in the Appendices where you can dig them out if you need to do a calculation. The text is well illustrated.

A wing joiner system has to transfer the bending loads from one side to the other. If the foam is used to transfer the bending load between the joiner and the wing skins, the joiner tubes and joiner bar have to be long and of large diameter to keep from crushing the foam. There are several ounces of structural weight involved in the joiner system. If the joiner is short, plywood ribs have to be set into the foam cores to transfer the bending loads from the joiner bar to the wing skins. This still requires several ounces of structural weight and also complicates the construction. If you can transport and store a one piece wing, it is by far the simplest and best way to go.

I think I can save you about 4 ounces and about $60 in battery cost by examining the power requirement more closely.

In level flight at a constant velocity with the plane trimmed for no load on the elevator, the thrust will be equal and opposite to drag and lift will be equal and opposite to weight. Referring back to post number 22, at an air speed of 35MPH (51.3 FPS) the lift coefficient of the wing will be 0.2 and the coefficient of total drag will be about 0.0211. The ratio of lift to drag is equal to the ratio of the lift and drag coefficients. Therefore L/D= .2/0.0211=9.48 Since the weight is assumed to be no more than 5.5 pounds the drag force is no more than 5.5/9.48=0.58 pounds. In one second the drag force acts through a distance of 51.3 feet. Power is feet times pounds divided by seconds. Therefore the power to move the model at 35 MPH is 0.58 times 53.1= 30.8 foot pounds per second. There are 550 foot pounds per second in a horsepower. Therefore, it takes 30.8/550=0.056 horsepower. The propeller probably delivers that horsepower at an efficiency of about 75%. The shaft horsepower required to drive the propeller will be about 0.056/.75=0.074 horsepower. The motor will probably not be very efficient at this power level so I'm assuming an efficiency of 50% (it's probably closer to 60%). The power input to the motor will be less than 0.148 horsepower. There are 746 watts in a horsepower so the input to the motor will be less than 110 watts. The battery voltage is 11.1 volts so the current will be about 10 amperes. See:
https://www.fmadirect.com/site/fma.htm?body=Products&cat=28
Using the calculator to size the battery yields six, 2000 miliampere hour cells in a 3S2P configuration and a weight of 10 ounces with connectors and wiring.

With the weight savings from the one piece wing and a smaller battery, the weight is down to under five pounds and the cruising duration is up to over 20 minutes after a brief climb out at full power. Also, at full throttle in level flight, the top speed is over 45 MPH but the power duration at full throttle is onlu about 8.5 minutes. The stalling speed is back down to 15MPH. This level of performance is conditional on a smooth and accurate airfoil contour as well as keeping the outside of the aircraft as free as possible of drag producing stuff like fat fuselages, rubber bands and pegs, antenna wires, switches, fat wheels, unstreamlined landing gear struts, control linkages, etc.

[Ollie]

I have reviewed the design recommendations and have two comments:

1. The Jeti speed controller has to be upgraded to 40 amps per Paul's recommendation. The 30 amp rating I originally suggested is cutting it too fine.

2. The full throttle current with a 10-6 prop will be around 29 amperes. The reason for switching from the 3 ampere hour low rate cells to 2 ampere hour high rate cells isn't just about money and weight. The 3 ampere hour pack had a maximum discharge rate of only 24 amperes. The 3S2P configuration of 2 ampere hour cells has a maximum discharge rate of 32 amperes. Take care not to prop the motor for full throttle over 32 amperes.

[Sparky Paul]

Adding landing gear, and moving the motor to the front... and fancying up the wing. :)

[Ollie]

I've been thinking about refining the wing design to achieve a two piece wing that weighs no more than two pounds with joiner.

The only way to reduce the foam core weight without reducing the wing area is to thin the airfoil. The AG25 could serve as the root airfoil, the AG26 at the middle of the semispan and the AG27 at the tip. See:
http://www.charlesriverrc.org/articles/drela-airfoilshop/markdrela-ag-ht-airfoils.htm
This family of airfoils goes from 7.58% thick at the root to 6.11 % thick at the tip compared to 10.6% thickness for the S3021 airrfoil. The new foam cores will weigh only 64% of their predecessors weight.The wing skins will be 1/32 inch thick medium balsa for torsional stiffness. The bending loads will be carried by carbon fiber tube spars. See:
https://www.cstsales.com/Carbon/carbon-tubes.htm
The cores will be done in four 24 inch span panels. The root panels will be equipped with 3/4 inch ID carbon tubes and the tips cores will be equipped with 3/8 inch diameter carbon tubes. The carbon tubes of the tip panels will extend two inches beyond the big end of the core to engage the tube in the root panel. A thin balsa plug will be inserted 2 inches into the 3/4 inch diameter tube at the end that engages the tip panel to act as an epoxy dam. The end of the protruding 3/8 inch tube will be capped with a thin balsa disc.
The cavity in the 3/4 inch tube will be filledabout 3/4 full with a mixture of micro balloons and epoxy. then the tip panel tube will be inserted into the cavity, the panels butted and carefully aligned to each other, any excess epoxy wiped off and the panels taped together while the epoxy cures. This will permanently join the root and tip panels. After the epoxy has cured, the joined panels will beskinned with 1/32 balsa sheet using polyethylene wood glue spread as thinly as possible on the balsa and the cores misted with water. The core beds protected with waxed paper and the cores and skins pressed together under a hundred pounds or more of weights while the polyurathane cures. The skinned wings will have the root ends sanded to a 6.5 degree angle to accomodate that much dihedral in each panel. Thin sheet balsa dams will be inserted three inches into the carbon tube at the root end. A 3/8 inch ID carbon tube 6 inches long will be capped at each end to keep epoxy out. The cavities in each carbon tube spar at the root will be filled 3/4 full with a mixture of micro balloons and epoxy. The capped 3/8 inch carbon tube will be inserted into both cavities of epoxy and the wing halves brought together and propped up to dihedral angles of 6.5 degrees under each wing half.The wing halves will be taped together in alignment while the epoxy cures. when the epoxy has fully cured, the wings will be cut apart with a razor saw, cutting through the 3/8 ID carbon tube. The wing roots again block sanded to the correct dihedral angle and 1/16 inch thick birch ply root ribs added with 3/8 holes to accomodate the joiner bar A 3/8 inch diameter carbon joiner bar 6 inchs long will plug the wing halves together. An indexing pin and tube will be added to the root ribs to align the wing halves. the wings will then be covered with Ultracoatlite or Oracover Lite. If the wings are to be plugged into the fuselage sides then the joiner rod has to be longer to accomodate the fusealge width.

[David L.]

Wow! Sounds really cool Ollie! Have you inspired yourself to make one of these? I think I am understanding having the wing made in four 24" sections.

Sparky, I really like your drawings...that really helps me visualize things. That must be a really cool program.

David L.

[Ollie]

$400 or $500 including motor, ESC, two sets of batteries and special charger are a high price to pay for a little altitude so that I can start soaring. I'm afraid I am adicted to unpowered flight. Besides I have three unfinished projects in the work shop that I need to finish first.

[winmodels]

Hi Guys

Very interseting suggestions!!

I went a little bit over the top with the power on my previous posting, but I still think that ca' ten to twelve pounds OAW is probably what I would aim for.
Have a look at

http://www.aerosonde.com/aircraft

To underline my thinking I have attached a calculation spreadcheet, which I have started to develop ca' 12 years ago and which has been used ever since by many people with great success.
(didn't work yet - I still have to find out how to convert an Excel file to a format accepted here)
There is also the longstanding rule to have at least 50W/lb installed power, better 100W/lb, which ten years ago was quite difficult, but really no problem nowadays. Nevertheless for duration -NiMH or Lipoly- one should have enough capacity, so 70oz for the battery doesn't seem to be too much.

I don't think that the choice of such a thin airfoil section as the AG 24 or AG25 is a very wise one, unless you have a mile of tarmac for take off. If you build it stiff enough to avoid flutter it will probably be heavier than a thicker wing. The foam core itself can always be hollowed out, so the weight of the core is not really an issue.

However the first suggested Wortmann FX63-137 has in my view too much camber and therefore a large moment coefficient, which requires a large tail volume and hence high trim drag.
There are lots of other section available f.e. the FX 60-126 which better fill the bill. I think it is important to have at leat 12% thickness at the root chord, which should not create any problems at the expected Re nos. of more than 200,000.

Look at the wing of that aerosonde UAV, looks pretty thick to me!

A last point: I would fit an EDF to this plane. This leaves your hands intact if you decide to hand start and it never breaks a prop. :D :D

I'm going to design and build one!!!!

Klaus

[MikeSt]

[QUOTE]Originally posted by winmodels
[B]Hi Guys

Very interseting suggestions!!

http://www.aerosonde.com/aircraft

(didn't work yet - I still have to find out how to convert an Excel file to a format accepted here)

Klaus

Use WinZip http://www.winzip.com/ to create a compressed .zip file contaiing the Excel file, then upload the .zip file.

Mike

[Sparky Paul]

Originally posted by winmodels
...
A last point: I would fit an EDF to this plane. This leaves your hands intact if you decide to hand start and it never breaks a prop. :D :D

I'm going to design and build one!!!!

Klaus
.
I have flown a few Speed 400 powered photoplanes, but wonder if there's an EDF with enough grunt...

[Ollie]

The first Aerosonde to cross the Atlantic used a sailplane wing from RNR. The wing had an aspect ratio of somewhere between 13 and 14. it used an SD7037 airfoil of 9.2 percent thickness. The proposed design with an aspect ratio of 7.3 and an AG24 root airfoil of 8.4% thickness will be easier to build with a foam core than the hollow molded wing of the aerosonde. The AG 24 wing's spar won't have to be as massive because of the much lower aspect ratio. Also, there will be absolutely no problen taking off with the AG24 airfoil with a wing loading of under 10 ounces per square foot and the recommended power plant. It will almost leap into the air at full throttle.

The aerosond was larger, had a much higher stalling speed and was more heavily loaded because of the heavy fuel load at takeoff and because of the extensive telemetry and navigations systems and their associated power supplies. Its mission was quite different and consequently its design was was different in accordance with its functions.

[winmodels]

Mike

thanks a lot, I should have noticed

Lets try this then

Klaus

[MZanders]

winmodels, that's a nice sheet! Automaticly generated? What software do you use?

[Ollie]

Klaus,
That's a much better example. The aspect ratio is higher (which I prefer). The airfoil is similar to the AG series but thicker and higher camber, allowing a higher wing loading and less difficulty in achieving the target weight.

The only draw back I see is that the client, David, has opted for a smaller wing span than 3 meters.

BTW, the spreadsheet in your example can be a great labor saving tool for the design process and guide the aspiring designer in the right direction.

[winmodels]

MZANDERS:

what do you mean -outomatically generated- I wrote it :D

Ollie:

the customer doesn't always know what he wants:-)
what's a few feet when it comes to visibility?
One should also bear in mind that a too lightly loaded plane makes an unstable platform.
I think that's what I had in mind from the outset
now let me get on with the EDF version

all the best

Klaus

[MZanders]

Originally posted by winmodels
MZANDERS:

what do you mean -outomatically generated- I wrote it :D


Neatly done! As Ollie says, it saves quite some time..
It really looks like an export of some RC plane design program.
Crisp & clean. Thanks!
I'm thinking of a similar design and will have much use for this spreadsheet.

[winmodels]

Hi MZanders

here is a different version, where you can better see which cells can be changed according to your design.
It is also quite important to use reasonably correct values of Cwp for different profiles. Profili2 from Stephano is the best I have seen so far for this task. There is also a much better version of my program available which calculates rate of climb and static stability. It still needs translation into english, but if your german is good enough..;) ;)

On my way to Holland (~once a month for business) I pass Leuwen on the E40. Is that where you live?

Regards

Klaus

[Ollie]

Klaus,

Would you be so kind as to run your spread sheet with the following input?

Span 2438 mm
Root chord 381 mm
Tip chord 280 mm
Length fuselage pod 914 mm
Height fuselage pod 80 mm
Width fuselage pod 80 mm
Tail area 30 square decimeters
Tail span 914 mm
Tail average chord 197 mm

Wing 900 gm.
Fuselage pod 225 gm.
Tail and booms 200 gm.
Under Carridge 160 gm.
Motor ESC and Prop. 226 gm.
Battery 312 gm.
Radio 113 gm.
Camera 454 gm.

Please use the FX126-10 airfoil since the polars are more readily available than the polars for the AG series airfoils.

This would be a kindness to me to ensure that I have not made a blunder in my calculations.

[David L.]

Wow! Its nice to see that this basic design seems to be popular in many different functions. I can see I still have alot to learn. I just ordered the book that Ollie recommended on aerodynamics. I am excited...but this won't be a project I will complete over night. I will probably give myself about a year for construction because I want to do things right. I am in no rush. In the meantime, I will continue to use my Slowstick for AP. Please keep the data coming! I am listening! Thanks. David L.

[MZanders]

Originally posted by David L.
I just ordered the book that Ollie recommended on aerodynamics. I am excited...
You won't be dissapointed!! It's a great book!

Originally posted by Winmodels
Hi MZanders

here is a different version, where you can better see which cells can be changed according to your design.
It is also quite important to use reasonably correct values of Cwp for different profiles. Profili2 from Stephano is the best I have seen so far for this task. There is also a much better version of my program available which calculates rate of climb and static stability. It still needs translation into english, but if your german is good enough..

On my way to Holland (~once a month for business) I pass Leuwen on the E40. Is that where you live?

Regards

Klaus
Thanks for clarifying the spreadsheet! I'm thinking of a similar setup, only 1m50 and speed 400 or so. Brushless/Lipoly perhaps later. I'm sure it's going to be usefull.

And yes, that's where I live. Our flying field is actually only 200m from the E40, near the exit "Haasrode". Drop by if you see a plane flying.. :-)

Cheers!
Maarten

[Purdue Aero Man]

If you're going electric, do everything to can to reduce weight in your aircraft. Look at taper ratio's of about 0.3 as a way to try to approximate the elliptic wing loading, and a decent amount of washout can be a great help in both reducing tip stalls and unloading the tips. Ollie's suggestion of using the tail volume coefficient as a way to determine the proper hoeizontal tail size is a decent suggestion, though a more highly accurate approach would be to use the static-margin calculation approach. It should be in the aerodynamics book you are getting, but may not be there because it is more of a design/control issue. If you need that, I can dig it up for you.

As for the aerial photography, I am working on a 14.4 ft aircraft designed to haul 20 punds of gear into the air for surveying/reconnaisance. Power is form an OS 1.7 cu in. fuel injected engine, with an endurance in the 30-45 minute range. The payload includes a 5 megapixel digicam along with a full-sized laptop. If anybody has any interest in it, I can post some pics when the actual building starts. As for your project, good luck with it.

[Ollie]

Purdue Aero Man,

Even with 5 degrees of wash out and a taper ratio of 0.3, a wing with an aspect ratio of 7.4 will still stall at the tip first and its lift distribution at small angles of attack will be negative at the tips.

The aspect ratio 7.4 wing with a taper ratio of 0.73 and untwisted will have as good or better lift distribution at all angles of attack and a superior stall characteristic. It's Oswald efficiency factor will be 0.9918, less than one percent worse than a purely elliptical lift distribution so far as induced drag is concerned. Tip stall margin is even more important in models than in full scale because there is less stall warning in models.

Model Aircraft Aerodynamics by Martin Simons does cover the subject of static margin and even gives a method of determining the neutral point with a clever little homemade analog computer.

For many airfoils at low reynolds numbers, the moment coefficient versus angle of attack, over the range of useful angles of attack, has two regions of negative slope and a region of positive slope between them. This means that the neutral point isn't stationary as it is at much larger reynolds numbers. See:
http://www.charlesriverrc.org/articles/drela-airfoilshop/markdrela-ag-ht-airfoils.htm
Also, Summary of Low-Speed Airfoil Data, Vol. 3 by Selig, etal. Because of this complication, the most practical method of establishing pitch stability is by flight testing, starting with excess stability with a forward CG and gradually moving the CG aft while noting the stability until acceptable stability is obtained over the useful range of angles of attack of the model. In a camera plane, long period oscillation after a disturbance will be better than short period oscillation so, a small "static" margin would be desirable.

Defending one's designs publicly can be fun or not depending on how good your design decisions are. The real test is how well they meet their objectives in flight.

[Sparky Paul]

Ollie ran your numbers..
how come your battery is so teeny? :)
The other is the baseline Luftsonde output

[winmodels]

Hello Ollie

here we go.
As you can see we are talking of two very different planes!
As could be expected, your design uses less than half the power o mine, but mine may be a more stable photo platform and probably can stay in the air (under power) much longer.

When you use the spread sheet, have the profile polars at hand, because you still have to imput the Cwp values by yourself.
And please don't use theoretical (xfoil) polars, have at least one properly measured source for comparison.

Have fun and best regards

Klaus

[winmodels]

Sparky Paul

I see you must have changed some numbers on the spreadsheet but didn't correct the Cwp values accordingly. I would think that the results then are not really correct anymore.
The task would be helped if the hand in-putted values were minimised, meaning that the spread sheet (or program) user doesn't have to input data, which can be derived from other sources. I am working on that now, but until then the user must change ALL values in the pink cells. The problem with the "other sources" is actally that real (measured) values have to be digitalized - and thats a long process if you do it by hand. Bear in mind that the whole thing started quite some time ago (1982) to give me a tool to analyse electrically driven models with the aim to find out how they "tick".
Today other programs are available to synthesize parameters for the design of EPA (Electrically Powered Aeroplanes).

Have fun and best regards

Klaus
Winchester Models

[Ollie]

Paul,
I planned for two of the following 3-cell packs from FMA:

Kokam High Discharge LiPo Electric Pack
2000 mAh, 3 cell series (11.1V), heatshrunk with Deans Ultra connector, high discharge (8C / 90%)___
Size: 100mmH x 63.5mmW x 19mmT
Weight: 156 grams
Ratings: 8C/ 90%
Outputs: 11.1V Nominal, 2000 mAh

Klaus,
I have the wind tunnel measurements of the FX126-10 which were done at Princeton about 1988 by Selig, Donnovan and Fraser. I will use then for running your spread sheet.


_

[winmodels]

Hi Ollie

looking forwards to the results.

[winmodels]

BTW

the output of the B.M. is only as good as your imputs. :-)

[Ollie]

Span 2438
root chord 381
tip chord 280
mean chord (geom) 914
mean chord (dyn) 338.176
Wing area 2.228
Aspect ratio 2.67
Loading wing only 11.73
Loading wing and horiz.stab. 10.85
Length of fuselage (pod) 914
Height of fuselage 75
Width of fuselage 75
Length of boom(s) 914
Par. Drag area 0.006
Stab. Span (hor.) 914
Mean stab chord 197
Stab. hor. proj. area 0.180
Stab aspect ratio hor.ptoj. 4.64

Wing 907 g
Fuselage 200 g
Stab + boom(s) 241 g
UC 160 g
Motor + gear+
prop. 226 g
Battery 312 g
RC-gear 113 g
Camera 454 g

Total weight 2.613 kg

Flight- Velocity - Lift
velocity pressure Coefficient RE-No
m/sec mph q cA RE-No corrected
0 0 0 0
4.5 9.945 12.35 0.93 106525
5 11.05 15.25 0.75 118362
6.5 14.365 25.77 0.45 153870
8 17.68 39.04 0.29 189379
9.5 20.995 55.05 0.21 224887
11 24.31 73.81 0.16 260396
12.5 27.625 95.31 0.12 295904



Cwp Cwi
0 0
0.017 0.10349
0.015 0.0679
0.012 0.02377
0.011 0.01036
0.01 0.00521
0.0108 0.0029
0.0108 0.00174

Glide angle Drag

Cw total Ca/Cw ges N
0 0 0
0.132 7.04 7.6
0.091 8.3 5.86
0.039 11.39 3.52
0.023 12.64 3.49
0.017 12.6 2.66
0.015 10.53 3.04
0.014 8.94 3.47

Power
required
W (Cwp * 1000)
0 0
89.95 17
65.1 15
50.8 12
61.99 11
56.13 10
74.37 10.8
96.52

Sorry for the poor format. It results from incompatibility of my Macintosh program format.

I think the low powers (watts) calculated are probably because the numbers do not include the inefficiency of the prop and motor. I think the powers calculated must be for level flight as well.

[Sail 'n Soar]

If you're going electric, do everything to can to reduce weight in your aircraft. Look at taper ratio's of about 0.3 as a way to try to approximate the elliptic wing loading, and a decent amount of washout can be a great help in both reducing tip stalls and unloading the tips.

Even with 5 degrees of wash out and a taper ratio of 0.3, a wing with an aspect ratio of 7.4 will still stall at the tip first and its lift distribution at small angles of attack will be negative at the tips.

From my perspective, except for very low aspect ratio delta wings, about the only advantage the .3 or higher taper ratio offers is a lighter wing structure. As Ollie was indicating, In order to get good stall characteristics you need to have really large wash out. Yet to get reduced induced drag through a more elliptical lift distribution you need wash in. In either case, you get very non-optimum lift distributions at off design angles of attack. The one way you can get both reasonable lift distributions and stall characteristics is to go to stransition to significantly higher lift/higher camber wing sections as you go from the root to the tip. Of course, that requires a different airfoil at every wing station, which presents additional CAD design and construction challenges.

Experiment on the following Stanford site and draw your own conclusions. http://aero.stanford.edu/WingCalc.html

[Sail 'n Soar]

Running the numbers again, the .3 taper ratio wing with no washout is very close to an elliptical lift distribution in terms of induced drag. However, all of the other negatives still stand.

[winmodels]

Sail 'n Soar

straight wing taper ratios of less than 0.6 for model aeroplanes (and may be even full size, except high subsonic transports) are absolute nonsence. Every sensible textbook and proven wing calculation program will tell you. There are, as allways, exeptions to the rule, but I ask myself in our "little" world "why bother".
What I can't understand is, that these things come up again and again, despite the fact that they have been dealt with sufficiently in the past. Can't newbees reed the available literature and not try to reinvent what is already there? There are still enough unknowns where they can grit their teeth on!

Klaus

[Ollie]

Camcorders are already available that weigh only 5-1/2 ounces. Within the year credit card sized camcorders using solid state image storage media will very likely become available. Image stabilization software may become available as a post processing item. The technology is changing faster than you can design and build a plane to take advantage of it. LiPo batteries are already revolutionizing electric powered flight. There are the prospects of much higher energy density power sources in the form of miniature fuel cells on the horizon. Within two or three years the planes we have considered in this thread will seem like dinasaurs.

The things that have lasting value in all this are the scientific and engineering principles involved. Learning has more lasting value than stuff.

[Sail 'n Soar]

Originally posted by winmodels
Sail 'n Soar

straight wing taper ratios of less than 0.6 for model aeroplanes (and may be even full size, except high subsonic transports) are absolute nonsence. Every sensible textbook and proven wing calculation program will tell you. There are, as allways, exeptions to the rule, but I ask myself in our "little" world "why bother".
What I can't understand is, that these things come up again and again, despite the fact that they have been dealt with sufficiently in the past. Can't newbees reed the available literature and not try to reinvent what is already there? There are still enough unknowns where they can grit their teeth on!

Klaus

Klaus,

To whom are you replying? You are preaching to the croud! My post was to provide a counter opinion to Purdue Aero Man's recommendation to use the .3 taper ratio. Purdue's answer wasn't wrong and was directly addressing Ollie's comment about minimizing weight. The problem is that it was incomplete in that it did not address the significant issues of why you would not want to go with an extreme taper, hence my first post. Unfortunately I made a false assertion in my first post, wash in is not required to have a very close to ideal lift distribution on the .3 taper ratio wing. The second short response was to correct that error.

You ask, why do these things come up again and again? It falls right in with the other repeating discussions of collision theory, etc., etc. New members keep joining e-zone, some of which have the same old questions and others have with the same old answers.

[Purdue Aero Man]

Ollie,

Having negative lift out at the tips is sometimes a desirable characteristic. I remember sitting in a lecture given by James Raisbeck of Raisbeck Engineering, listening to the new mod kit they are developing fo the older Lear 35's. One of the features of the mod was a horizontal winglet, the "Bat-wing" is what they called it, which was a cranked section from outboard the aileron to the back of the tip tank. The purpose of the winglet was to control the tip-vortex, as well as actually negatively loading the tips. Of course we all know that minimum induced drag is not a result of an elliptic wing loading :D . For drag reduction in incompressible flow, M < 0.3 , the best results are USUALLY obtained with a tapered wing with a ratio of 0.3, but having a straight trailing edge. And remember that wing design is a series of compromises, with the ACTUAL optimal loading being somewhere between a linear distribution and an elliptical distribution. Bending moments are very important, especially as AR goes up.

[Ollie]

Purdue Aero Man,

The light weight spar structure of this wing already provides for accelerations of 10 G's which is more than adequate. The need to taper the wing to reduce its weight and roll and yaw moment of inertia in this aircraft is of extremely low priority.

Ludwig Prandtl, long before either of us was born, proved that, for an unswept (straight lifting line) wing, an elliptical lift distribution results in the lowest induced drag possible. Induced drag minimization is a high priority in this design because of the low speed requirement imposed by the client. A higher aspect ratio would be more efficient but the client has imposed a span limitation too.

From my perspective, the value of this design exercise is to show how conflicting objectives can be resolved through compromises to best realize the mission of the aircraft. Your recommendation for a highly tapered wing would severly compromise the low speed handling qualities of this aircraft without delivering justifiable benefits.

If this design were an aerobatic aircraft, with a high priority for snappy snap rolls, a highly tapered wing would be justified. If this were a full scale aircraft with stall warning, more wing taper might be justified. If this aircraft were pushing the limits of spar material strength and stiffness to weight ratio, a more highly tapered wing might be justified. The mission of this aircraft does not include justification for any highly tapered wing, period!

Sail and Soar has given you a link to the Stanford Aero Engineering Department's on-line wing design program. To better inform your opinion, you should use it as a homework assignment. If you choose to take the assignment it is to design a single taper wing with tip stall margin and with an aspect ratio of 7.4, using an airfoil with maximum lift coefficient of 1.1, to have minimum induced drag at at wing lift coefficients between 0.2 and 1.0. If you can improve on my wing design's induced drag at any lift coefficient by two percent or more, while maintaining some tip stall margin, I'll buy you a beer.

[Purdue Aero Man]

Originally posted by Ollie


Ludwig Prandtl, long before either of us was born, proved that, for an unswept (straight lifting line) wing, an elliptical lift distribution results in the lowest induced drag possible. Induced drag minimization is a high priority in this design because of the low speed requirement imposed by the client.


The above statement is in fact true. However, it does not cover the entire realm of wing planforms, and is perhaps, with no intent to offend, the most simplistic view of induced drag reduction. In order to address the entire issue of induced drag, however, one must look at the direct result of (and some people will argue the cause of) lift, which is downwash and upwash. A concept not introduced until graduate level aerodynamics is the concept of the Trefftz Plane, which is viewed from behind the wing in the wake of the aircraft. At an AoA of 0 degrees, with elliptical loading, both a swept and unswept wing will produce uniform downwash in the Trefftz Plane, as in the case of the unswept wing above that you hav alluded to. However, if the angle of attack of those wings is increased (or decreased), the straight wing will continue to produce uniform downwash, but the swept wing will not. At this point the swept wing is no longer producing the minimum induced drag, and a non-elliptic wing loading will be needed to get induced drag to the minimum.

[/QUOTE]
From my perspective, the value of this design exercise is to show how conflicting objectives can be resolved through compromises to best realize the mission of the aircraft. Your recommendation for a highly tapered wing would severly compromise the low speed handling qualities of this aircraft without delivering justifiable benefits.
[/QUOTE]

For this please see my previous posting that a lift distribution is somewhere between linear and elliptical. To minimize induced drag, you must find the best relationship between the wing spar size and weight (primarily to resist bending) and the AR and the lift distribution. Changing the lift distribution may increase induced drag until you recaculate that the drag actually decreased because you were pulling less wing spar aloft. Compromises between all of these different areas and requirements are very necessary. What may suit one pilot for a particular pilot may not necessarily suit another pilot for the absolutely same mission. For the aerial camera platform I am working on, an straight-untapered wing was chosen, because of simplicity and the fact that we can hide the drag behing the thrist of an engine. As for the Trans-Atlantic, Point-to-Point Aircraft I am also working on, minimum induced drag is one of the driving factors in the design, as 1 or 2 counts of drag over those ranges makes HUGE differences in A/C size and weight.

[Ollie]

I hope you can carry enough fuel to make it.

At a wing loading of 10 ounces per square foot and an aspect ratio of 7.4, the weight of a well designed wing spar to carry 10G inertial loads is one of the smallest items in the weight budget.

Have you even bothered to look at:
http://aero.stanford.edu/WingCalc.html
? Criticise that.

[Sail 'n Soar]

And remember that wing design is a series of compromises, with the ACTUAL optimal loading being somewhere between a linear distribution and an elliptical distribution. Bending moments are very important, especially as AR goes up.[/

Purdue,

First time I heard that was aa a sophomore about 40 years ago in my aircraft structues course. But the way it was presented at the time is the ACTUAL LOADIING was between a linear distribution based on the actual wing planform and an elliptical distribution. In the pre PC days the prof was providing an easy algorithm for determining approximate wing loading for structural design. In my aero class we derived the relationship between ellipticical lift distribution and minimum induced drag. When designing a wing which will have a good compromise between induced drag, weight and stall characteristics you are far better off using a more accurate lift distribution calculated either from writing own program or using those already out there on the web such as the previously mentioned Stanford site.

Again, check out the Stanford site specfically looking at where the max CL occurs and the value of e as functions of wing CL. Something about a picture is worth a thousand words!

Sail

[David L.]

O.k. folks, it seems as though there are differing opinions as to what would be best for an AP plane with the characteristics I am after. Like someone said..."sometimes the client doesn't know what he wants" I am willing to go along with that. I "think" I know what I want but with my limited experience, it is very likely that I may actually want something else and not realize it.

How about scraping the original requirements and start from scratch. What would be your ultimate AP plane...list the characteristics it would have in plain english...so I can understand.
Thanks! David L.

[Ollie]

David,
There is no absolutely objective and correct solution to the value free problem as you have restated it. Ultimately, you will have to set limits and priorites. Lacking those limits and priorities there will be little agreement on what the design should be. Having said that, I am willing to apply my own personal values and priorities to the design assignment. In other words what I propose will be more for myself than for you, David. It will be right for me and, perhaps, no one else.

I will begin by picking the payload which will be the most compact, light weight camcorder I can find in the hope that its technology won't be too dated by the time the model might be built.

I'll post again when I have a candidate camcorder payload.

[winmodels]

Klaus,

To whom are you replying? You are preaching to the crowd!

Sail ‘n Soar

Well - actually - I don’t think I intended to reply to your posting, rather more like continuing it. I didn’t intend to “preach to the crowd” either. If it came across like that I can only offer you my apologies.

For the actual theme (or at least the taper ratio part) of the thread it seems that I’m not so wrong.
When you compare the span-wise lift distribution and lift coefficient distribution you will find that in the design under consideration at a taper ratio of <0.5 and now wash out the wing tips are too highly loaded at low speed, so there is a very good efficiency but a danger of tip stall. If however you use a taper ratio of ~0.6 and 2 deg. wash out the efficiency is of the same order with a high tip stall margin. Using a taper ratio of 0.3 one will have to employ a high degree of wash out for the slow speed case. At high speed this will cause a lower efficiency.
The problem with the Stanford applet is that it shows both of the above mentioned distributions but leaves the observer to come to his own conclusions and interpretations. If we really want to perform an accurate analysis we also have to account for varying Re numbers as outlined in the book by Schlichting/Truckenbrodt. It would be nice to have a PC program for such computations – I haven’t seen one yet.
Here is an interesting site, where the Trefftz plane is explained.

http://www.desktopaero.com/appliedaero/appliedaero.html
(which also includes the Stanford applet)

Klaus

[Ollie]

Starting from scratch, I would select a camcorder like the Panasonic DSnap Digital SD Camcorder that only weighs five ounces and will record 20 minutes of MPEG2 DVD quality video-compression images at 30 FPS on a 512 MB SD card. Its dimensions are 4 X 5 x 1 inches. It has built in image stabilization. In addition to the in-camera stabilization I would add the FMA Copilot that stabilizes the plane in pitch and roll. The FMA unit only weighs one ounce.

Rather than design a plane from scratch I would adapt Mark Drela's Aegea 2-meter full house design for electric power and the camcorder. He has designed the Aegea for better performance and strength to weight ratio than I ever could. You can study his remarkable design in the files section of :
http://groups.yahoo.com/group/Allegro-Lite/?yguid=108420033
I know of no one that does more thorough aerodynamic and structural design than Dr. Drela.

The only thing to change would be the fuselage pod to accomodate an AXI 280820 outrunner brushless motor (2.75 ounces) a JETI 30-3 ESC (1 oz.), a 10-6 propeller, a 2S2P 2000 mah LiPo battery pack (7.6 ounces) and a mount and fairing (1 ounce) for the camera (5 ounces) over the wing. The gross wing loading with camera and mount would be about 11 ounces per square foot and the wing loading without camera would be about 9 ounces per square foot. With camera aboard the power duration at full throttle would be about 12 minutes. And the power duration at half throttle would be about 25 minutes.

[AndyKunz]

I would want video downlink and the ability to point still and video camera (independently). It would be nice if they could lock on the image at the center of the FOV while the aircraft orbited, for instance.

Andy

[Ollie]

Andy,

There is a license free video down link system shown at:
http://www.wirelessvideocameras.com/public/specs.htm#HLF11

It is more compact than the camcorder but only weighs a few ounces more than the camcorder if it can be run from a LiPo pack. The propulsion system might be redesigned to use three LiPo cells in series so that the video system could run off of the propullsion battery voltage. Its main limitation is the 1100 foot telemetry range.

[Sparky Paul]

There should be two videos, with downlink.
One looks forward for the pilot, the other mounted in a pan-tilt for the observer.
Downlink so the recording capacity isn't limited by onboard capacity.
Pusher, to free the front end for the widest field of view when panning.
The imager in the Pencam has a 640x480 resolution, which would be quite acceptable for piloting the airplane.. the signal downlinked instead of written to a card.

[Ollie]

Paul,
It looks worthy of a government contract!

What do you think the on-board piloting system equipment and payload would weigh?

[Sparky Paul]

Ollie, I have no feel for the weights.. Maynard did a marvelous feat with an all-up weight of 11 pounds..
I've seen some of the AeroEnvironment stuff and it has basically no weight at all! :)
Tossing out the "corder" part of the camcorder, and the image processing in a digital camera reduces the operational stuff to a couple or three imagers, which weigh nothing.
The down-link hardware is also incredibly small.
The FMA Co-pilot is also not significanly large or heavy, so the original request for 1 to 2 pound load could include all the fancy technical stuff, leaving a lot of airplane lifting capacity for batteries, which also are dropping weight almost daily!
There should be serious consideration to a GPS aided flight , since the plane will easily be capable of exceeding visual range.
If the camera focal length is typical, then the plane will need to fly a long ways just to get images of different areas.
The downlink and onboard control of a zoom feature would be handy here, when an area of interest needs a higher resolution image.
.
It IS starting to sound like a gummint proposal ! :)
Researching another subject I ran onto a UAV about the size this one would be, with an AUW of 90# !!!! payload of 45#, and flight endurance of 30 minutes.
.
If this is typical, we're way light on everything!

[Sail 'n Soar]

Researching another subject I ran onto a UAV about the size this one would be, with an AUW of 90# !!!! payload of 45#, and flight endurance of 30 minutes.

If this is typical, we're way light on everything!


The 90# for a 7-foot platform is indicative of a much higher flight speed - and a lot larger surveillance coverage requirement. I've also seen prototypes of that size for future manned Navy surveillance platforms. I believe the weights you are discussing in these posts are more on target for David's project, especially considering his initial stall speed goals.

[Sparky Paul]

Sail, I think David was somewhat overwhelmed with technical requirements.
I don't know, for instance, what the design stall speeds are for any of my airplanes.
As any of my originals evolve from existing planforms and weights which I know have worked, as long as I don't depart too far from that I can have confidence the new one will fly, if not the same, at least in a similar manner.
The photo plane I prefer to use is a Gentle Lady.. A good 2m glider, which took to electric power (more weight) nicely and now carries the camera (more weight).
It doesn't float like a GL without those mods, but I don't try to get in to real areas. :)
As the GL in glider form is commonly ballasted for performance when it's needed, I had no qualms about adding "ballast" in the form of the e-stuff and camera.
.
So for David's plane something on the order of a high-end TD or Electric will work.
Modified for the motor and camera, which for me means a pusher motor with the camera up front.
The pusher motor works well with a twin-boom, which permits a wide latitude in motor/prop combinations. And the twin boom can use a conventional vertical, but the monkey-motion Johnson bars need to wiggle the rudder are more of a pain than a simple inverted vee, which also provides good ground clearance both for takeoff and lack of snagging something when landing.
.
Doodled some more on it this morning... looks like this..

[Sail 'n Soar]

Sail, I think David was somewhat overwhelmed with technical requirements. I don't know, for instance, what the design stall speeds are for any of my airplanes. As any of my originals evolve from existing planforms and weights which I know have worked, as long as I don't depart too far from that I can have confidence the new one will fly, if not the same, at least in a similar manner.

Sparky,

I translated David's stall speed comments to mean a model like wing loading. You may not know your stall speeds, but I bet few if any of your's are over 24 oz/sq ft. With my reflexes and experience level, I'm more of a ~1#/sq ft modeler. I'd guess the 90# craft you showed had a wing loading in the 5 - 10 #/sq ft range, a pretty large departure from what most modelers would fly with confidence.

Sail

[Sparky Paul]

With the basic requirement of flying A# for B minutes, and (understood if not stated) a reasonably easy to fly airplane, it's not too difficult to develop the shape and size..
A plane will weigh C oz. to carry the payload for that time..
A decent wingloading is 8-12 oz/ft^2... so the plane's basic size is known.
D sq.in. for C oz.

The aero guy then takes those parameters and does the math stuff to find a suitable wing profile, which allows for a reasonable landing speed, and a good cruise.
The wing size dictates the tail volume stuff.
The aero guy works this out.
The customer can take those specs and construct the plane, with assurance it will perform and never have to know he might need a Cl to design to.
It's nice to know, but it doesn't dictate anything. It falls out of the parameters used to meet the criteria for this particular purpose.
The aero guy picks the profile(s), washout(s), taper(s)... tail volumes to the best of his ability and says "This will work". But it's also the nature of the beast that some other parameters will do the same job when combined properly.
That's debateable. (QED) :)

[Salto]

David L,

Have you started making anything yet? After such a build up, we sure would like to see the end result.

Graham.

[David L.]

Hi Salto! No official constuction has begun as of yet. I just got back from a 3 week photo project in Albania. Now thats a place that could use some sort of diversion...a hobby if you will. I was thinking what a hit it would be to bring my slowstick over there. Would be the talk of the town for decades probably.
As for the airplane, I am wondering if I want less wind penetration and more lift. That is...I love the ability of my slowstick to fly out of tight spots...very little runway. Can I have both? Would an undercambered wing be the way to go for something like this? I will probably have a foam wing cut shortly as well as two carbon fiber booms ordered. This will be a lengthy project. I am in no hurry and I want to do it right. When things actually start coming together I am sure I will have lots of questions and pics to show.
Thanks for your interest. I can't wait to actually get started. David L.

[Ollie]

Having plenty of thrust available, together with the S3021 or AG series airfoils mentioned, and a wing loading around 10 to 12 ounces per square foot will result in a plane that almost leaps off the ground and also has very good wind penetration at full throttle. At half throttle or less it will easily maintain altitude and cruise for a long time. Too much camber will hinder wind penetration.

Prototype development often requires some back tracking to correct an unanticipated problem that arises during construction or flight testing so include that in your expectations. There are guys here that can help you over those sorts of problems. Build the airframe first, keeping to the weight budget. Weigh everything before assembly and make sure it fits the weight budget. Buy the propulsion systen after the airframe is almost complete and the weight budget is assured. Don't buy the camera until the flight testing is complete. That way you can take advantage of the best technology as it emerges.