New threads and interesting conversations directly in your inbox. Sign up now and get a daily summary of the latest forum activities!
Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Starman, Aug 27, 2009.
This is the design I'm working on:
Clearly your design resembles our latest high performance military fighters. To make such a design perform with such a shape, you may need an equally similar set of control systems and powerplant. What I'm saying is that you are wanting one thing, but trying to get with it with something else. For instance, it's a pusher; that will require a rather tall landing gear for ground clearance at rotation at the end of your take off roll. The handicap of an oversized landing gear will tend to drive other design decisions that will rob your plane of it's potential. I could go on and on. High performance general aviation designs have evolved and have been optimised to quite a degree such that planes of similar purpose, also tend to resemble one another in many ways. All of the various compromises that go into any design have been worked out, thus the similarity you see. Only new technology will allow breakthroughs into newer, different looking designs.
You want to use your vertical surfaces for roll control. Your verticals have much less moment arm than what is available at your wing tips where ailerons have always been located. There are thousands of reasons why planes today look and function the way they do. People like yourself have actually built their unconventional design, but they don't seem to "takeoff" in popularity because they don't actually perform in a way that people want.
I appreciate that you are testing out your ideas and soliciting criticism, and that's good, very good. I would suggest that you travel much further down the learning path before cutting your first piece of metal, or even making another drawing. This forum is a great place for you to be. I would read it all, and read every book you can get your hands on. Study how planes are made by looking at those that are being built, restored or scrapped. Go to Oshkosh next year. Join your local EAA chapter. They will probably want to steer you into building a kit plane, and that's fine too. But if you really want to design, you have far more learning ahead of you.
I appologize if I am to-the-point, but life is too short to be heading into cul de sacs. I've done some of that myself.
Thanks Gary. I am a member of EAA, have a low member number, and am going to be going to some local chapter meetings. I already know almost everyone will be against this so I'm not showing them, just the technical adviser and the inspector, perhaps not the technical adviser as he will be in unknown territory and may well act accordingly.
The horizontal tails will be for roll control, not the vertical ones.
I realize the landing gear will be tall but it's the main compromise I 'loose' on while gaining several others. I know why most of the fast experimental designs look like each other but there are plenty of designs that look different.
My primary criteria for making this is that it have outstanding visibility in all directions (pusher), that it look very good, that it be very adaptable, and that it be easy for me to build. This design has all that.
I wasn't 'trying' to copy the latest military fighters. It's just that it's the only shape that fits my requirements and I'm building this, I don't care what anybody says! Much better to go through a blow by blow analysis as this goes on. Just because it looks like a fighter doesn't mean it will be too heavy, and it will have enough power. I'm aware of a lot of compromises so please go ahead and state your other concerns. I'm a designer, and I don't need to design something like everyone else already has. I'm even designing my own airfoils because I learned something about airfoils from a wise old airplane builder with tons of high level expereince.
By the way, the steel tube subframe that I was talking with you about in the steel tube strength thread is the one drawn in blue. That's the first one I need to tackle.
Your planform is very short coupled which can make this plane difficult to control.
Pushers have a way of pushing their pilot below the sod in crashes that might have been survivable in an engine-forward configuration.
I'm concerned that your horizontal tail surface is will stall easily, because it's so close to the wing.
What can I say, I'm a nut about safety.
A high wing design does provide excellent visibility of the earth.
I think we all go through this phase of designing an airplane. I'm chuckling because yours so closely resembles a set of sketches I did long ago myself (see attached). That was 1992, and still in the 'canard era', but it was a neat little ship.
As you move forward in developing your design, you're going to find that the choices you've made here will provide you with some challenges, not the least of which are the landing gear issues described earlier.
And I personally think that the level of adaptability you're trying to express in one airframe is a larger design issue than you realize. You're essentially trying to have one airframe be both an ultralight (or at least very light aircraft) and a high-performance airplane. That's a simply huge challenge, and more, perhaps, than you grasp at this stage. The design requirements for the two kinds of aircraft are very different at many levels.
Not saying you can't build an airplane along these lines, but just be prepared for what lies ahead. You have a lot of studying and learning to do, despite what you've accomplished in the past. Designing an airplane is not a trivial endeavor, and one such as this, doubly so.
Well that's something juicy to think about, so I'll think about that.
True, but still, there are plenty of pushers and people live with it; and mine will be waaay stronger than anything else you've seen. My aluminum tub cockpit will be like that of a race car, and being boat shaped in front it will ride up and surf the dirt in most cases if the angle isn't too steep. I may not use a pointy front but round it a little. When I was discussing the aluminum tub cockpit in the other thread someone mentioned crumple zones. Mine will have crumple zones, but front engine craft don't have crumple zones because engines don't crumple.
If you check you will see that the horizontal tails are very large compared to the wings so they should not stall. If it looks like using the tails only for roll control could cause one of them to stall then I can add spoilers or ailerons to the main wings or use a differential control for the elevator roll function. The large horizontal tails could make the short coupled problem more interesting though.
OK, I thought some about the short coupled problem. Flying wings and deltas are much more short coupled and seem to function all right. I can test and refine that kind of thing when I get to the RC model stage. First I'm building the cockpit tub and while I'm doing that I'll make a simple little two dimensional flat glider to check initial CG and CL.
Later while I'm building the steel sub frame and adding the hard points to it I'll make a larger control line model to refine the CG CL situation and settle on my initial wing planforms.
Later while I'm adding the small stuff in the cockpit and hooking up the engine and gear and stuff I'll build change my model to an RC to check out the three dimensional stability ... then I'll make the wings.
Concerning the fighter look alike idea. Think of it more as being something lie a single place VeriEasy but with the canard removed and a horizontal tail added, straight wings, and a drive shaft extension, or like doing those same things to Topaz's design.
I've been going through this design phase for well over twenty years, round and round in circles, all around the universe, including ones a lot like your drawing shows, the merry go round just stopped! I used to prefer canards but now I don't like their stalling characteristics compared to conventional.
At first the gear will be fixed but figuring out where to store it could be interesting. There is a lot of real estate in that F15 style rear section to put stuff but I would prefer gear that folds forward so I can extend it just by gravity if need be.
I'm aiming the design requirements at high subsonic with the short wings and the earlier big fabric one will just have to suffer with some excess weight, the fabric wings are just temporary anyway, so I can fly it before spending much time making fine wings.
That's just the kind of challenge that's required to get me interested.
Thanks for the feedback, keep it coming.
Forward engine compartments can be designed to crumple by making the cockpit cage stronger than the engine mount. The bigger issue is that in a pusher, you have 250 pound chunk of hot, throbbing steel barreling into your back in a sudden stop scenario...I'd rather have that thing in front of me paving the way through who knows what you may be augering into.
Short coupled designs like supersonic stealth fighters are only controlable with the assistance of sophisticated sensors and computer controls.. The taller your landing gears are, the wider they must be, which requires a stronger wing structure, all of which adds weight, all of which reduces performance. High subsonic speed is going to require a very powerful engine, which is going to take a larger airframe to support. Do some research to discover what sort of airplanes have been able to accomplish such a feat. That will give you some idea as to how this design of yours will grow from how it appears now. To achieve such high speeds will require a great deal of optimizing...that is to say that you will have to give up many other things to get there. The sacrifices you have to make to achieve one goal may make it impossible to achieve others. In another thread you indicated you wanted to build this plane from thick 6061 plate stock. There is no way you'll be able to achieve high subsonic speeds from a machine made of such a heavy and ineffecient material.
The size of your tail in no way will determine what it's stall characteristics are. If it gets blanked out by the wing, you may certainly find yourself in an unrecoverable stall/spin.
Due primarily to the large difference in the reynolds numbers of a scale RC airplane and the full size airplane, there is no similarity whatsoever between the performance charateristics of the two ships. That's why they say that a bubble bee cannot fly...that is if it were scaled up.
The knowledge required to design a plane is an exponential leap beyond most earthbound vehicles. A high performance plane requires even more, not to mention the $$$.
I'm very familiar with that feeling, but ... oh well.
Engine mounts can provide crumple, that makes sense, but usually there isn't much room for crumpling, and the gas tank is sandwitched between the red hot engine and the human, so in a crash the tank ruptures and splashes fuel on the engine just in time for the pilot to land on top of it =)
That's because their CL is ahead of their CG, they are made to be unstable
I'm going to attach the gear to the steel subframe, which will be attached to the strongest places in the aircraft fuselage and not to the wing structure. Besides the gear won't be that tall, more like a VeriEasy (I have a feeling I keep spelling that wrong). I'm going to see if I can use one of their landing gear bows for the non retractable version.
I just like the idea of being able to go high subsonic in a dive, partly for that safety margin (and for the adaptability aspect). I don't expect to have enough power to go over 250 for several years at least, and if I go with very high power I can easily replace what is required to do that, which is why it's going to be adaptable. Also, I'm using 3/16" and 5/32" 5086 for the fuselage tub, it will be a little heavy but not enough to be a problem.
(remind self to get parachute for self and for plane) That thought had occurred to me, but what about flying wings and tailless deltas? Having the elevators close to the trailing edge is kind of like having a control surface right at the trailing edge. Anyway, if blanking could be a problem then I thought of a V tail. I think an upward v tail could make blanking worse so it would have to be a downward V tail. What I've done for now is put the horizontal tail lower then the main wing by a little bit, I didn't show it that well in my sketches. Maybe still lower would help the blanking scenario.
I'm sure there is some similarity. I don't care for making models anyway ... but maybe a large 1/4 or 1/3 scale model would be better to get a more real life feeling. Besides, whats a feller to do if you don't have a model. X plane? What?
$$$ This thing will be relatively cheap in the first several iterations.
So maybe downward pointing horizontal tails would be a good idea?
Thank you for the advice.
Ermmmm, no. Their neutral point is ahead of the CG. An entirely different animal than the center of lift. (Or the Coefficient of Lift, which is the normal thing people mean when they write "CL".)
Based upon your sketches, the gear required will be substantially taller than that of a VariEze. Remember that the airplane needs to rotate (raise the nose) for takeoff and landing, and the landing gear needs to provide adequate prop clearance (several inches) under that condition with at least one tire flat. Striking a prop blade becomes very expensive - you need to tear down the engine and have the crankshaft inspected for cracks. (That's not just a visual process and, for certified engines, requires the services of an A&P.)
Just out of curiosity, do you have your pilot's license? How much time do you have as PIC of an aircraft?
Actually, it's an entirely different thing. What GES is talking about is the horizontal tail becoming buried in the turbulent wake of the stalled wing, which will effectively stall the tail as well - which means your control power suddenly goes away. In your case, given the fact that you're using "tailerons", your control power goes away on both the pitch and roll axes at the same time.
It will, but consider that the extremely short-coupled nature of your design may make it a bit of a handful to fly, depending upon a number of factors. Once you get into your stability and control calculations, you'll see what we mean.
Put a big accent on "relatively". Nothing is "cheap" in flying.
Aerodynamically, yes, to a point. Now you have to think about the structural and control-system implications.
On the general topic of being close coupled:
Tail size - The tail does several things. It counters pitching moments, stabilizes the bird in the axis of interest, and gives you the ability to change attitude. The pitching moments come from airfoil pitching moments and differences between the neutral point and the CG. Neutral point will be influenced by the projected area of the fuselage forward and aft as well as the airfoils. Stability comes from having the CG forward of the neutral point. Mathematically, the sum forces in any direction must be zero, the sum of rotation moments
must be zero, and the first differential of summed rotation moments with respect to angle of attack must be negative.
You need sufficient tail power, often described as tail volume. It is area of the tail times the moment arm from CG to 1/4c point of the tail, and has units of length cubes, thus, volume. This is usually normalized by using a tail volume coefficient.
Vht = Sht*Lht/(Sw*MAC) - Pazmany and Thurston talk about it in their books. Pazmany says that 0.30 appears to be absolute minimum, with .45 being typical and 0.70 being used on ships with big moments, big flaps, etc. Thurston recommemends a min of 0.55.
Vvt is done two ways. Pazmany lists it as Vvt = Svt*Lvt/(Sw*b) and hints at 0.033 being a minimum, and is a check that you have enough vertical tail against adverse yaw and to yaw the airplane. Thurston uses Vvt = Svt*Lvt/(Sw*MAC) and recommends a min of 0.30, which is a check of the same type as is used in the horizontal tail.
What these do is give you enough tail power to give stability and controllability, but they are monkey see, monkey do engineering. So you also have to check CG versus Neutral Point in both pitch and yaw over your range of CG and under worst cases. Pusher prop windmilling will be destabilizing two ways - pusher props making negative power are destabilizing by themselves, and it churns the airflow going into the tails. Then there is being able to bring the airplane into landing configuration (forward cg limit, prop windmilling, flaps out, long gear legs out) to stall while in ground effect.
Once you have enough tail arm times tail area, you also need adequate damping in each axis. Now damping is tougher to get a handle on, but we do know this - it goes with tail area times arm squared. Hmmm. That means that the tails may have to grow some more area to make the thing want to settle down after a gust, bump, or control input...
And then there is the issue of tail blanking at high AOA.
All other things being equal, we tend to slide the tails further aft when we can. They get smaller and the lift that must be generated to counter them also gets smaller, which makes for a more efficient bird. In this type of bird, they can actually serve to prevent prop strikes too.
Starman, nice looking design. The resemblence to the F-15 is obvious.
Is this the design you were talking about that you said will have a lifting tail? Stability becomes a little tricky under that condition.
You might consider researching the Rans S-11. If I remember correctly, it looked very similar in planform. The designer (Randy Schlitter) was seriously injured during the testing, and I think it may have been attributed to the interactions of the blended wing-body. You may be able to get some useful information that you can apply (or avoid) in your design.
Best of luck,
NTSB Identification: CHI97LA302 .
The docket is stored in the Docket Management System (DMS). Please contact Records Management Division
Accident occurred Sunday, September 28, 1997 in HAYS, KS
Probable Cause Approval Date: 6/26/1998
Aircraft: Rans, Inc. S-11 PURSUIT, registration: N4299Y
Injuries: 1 Serious.The pilot, who worked as an aviation mechanic for the aircraft manufacturer, installed vortex generators on the wings and fuselage of a Rans S-11. He then flew the S-11, while a chase plane followed. The S-11 pilot slowed the airplane to let the chase plane catch up. The pilot of the chase plane reported that the S-11's altitude was about 1,500 feet AGL, but the airspeed was slow. He observed the airplane doing slight Dutch rolls and '...just mushing along.' The S-11's nose pitched down and started a 1/4 spin to the left. The aircraft started to pull up from the nose down spin attitude, and then made a shallow arc before impacting the ground. The pilot of the S-11 was not a test pilot. He had a total of 3 hours in the S-11. The S-11 pilot reported the flight was a personal flight. The company did not have a policy for test flying airplanes. The engine and flight controls exhibited continuity.
The National Transportation Safety Board determines the probable cause(s) of this accident as follows:
failure of the pilot to maintain adequate airspeed, which resulted in an inadvertent stall and collision with the terrain. The pilot's lack of experience in the make and model of airplane was a related factor. Full narrative available
The lack of affordable jet engines has not totally deterred homebuilt designers from jet like designs. A recent example was the RANS S-11. This was not a replica of a specific aircraft but was more of a generic space age jet in appearance (after the F-117 anything appears possible), but propeller-driven. The prototype was destroyed in a crash following an engine failure, however it is reported that flying qualities (difficulty in achieving a power-off flare) contributed to the severity of the damage. Perhaps fly-by-wire shapes should have fly-by-wire systems.
>Stability comes from having the netral point forward of the CG.<
Isn't this a typo?
Thanks Lucrum. I thought it was Randy Schlitter flying the plane, but it sounds like it may have been someone else.
Just found the report. Both of the S-11's crashed. NTSB reports indicated that both crashes resulted in serious injuries to the pilots. The first was with Randy Schlitter in 1991. It said:
THE ACCIDENT AIRPLANE LANDED HARD AFTER AN ENGINE FAILURE AT ALTITUDE. THE AIRPLANE CAUGHT FIRE AFTER IMPACT. ENGINE DISASSEMBLY REVEALED A SEIZED PISTON IN NUMBER 1 CYLINDER.
By what I remember, his injuries were quite severe.
The one Lucrum mentioned happened in 1997.
Thanks, I'll do that. I recall seeing pictures of the Rans S-11 before, and it screams pusher. Really the only reason for having a shape like that is because you have a pusher, making that shape into a tractor means they were mostly concerned with style but it being a tractor just looks like really bad taste to me. It would look sooo much better as a pusher.
I appreciate your sharing it though because I hadn't checked out the reason for the accidents. It looks like insufficient elevator authority and the windmilling prop destroying some lift off of the center section were the main problems.
I'll work on the elevator and tail authority, thanks for the equations Billski.
As far as windmilling props is concerned, I'm going to be using a Mazda rotary because they never quit. If you can start it it will run untill you turn off the ignition - unless you run out of gas. Still I will see about getting the prop a little farther away from the center section trailing edge.
I was using CL as center of lift, and I use cl as coefficient of lift, but I'm getting the message that you don't say center of lift on this forum. Don't worry, I'll get the jargon down before too long, then I'll sound like and expert
Yes, I know about rotation and flat tires Well, it's a lot like a VariEze in the back, differences are that the prop is about two and a half feet behind the engine (needing longer gear), but the prop diameter will be smaller (meaning shorter gear). I also might angle the thrust line upwards to get more clearance but you are right, the VariEze gear probably is to short. If somebody gives me some for free I might add extensions to the gear holder just to get it rolling. Actually I was first thinking of getting some used Cessna flat steel spring gear, which will probably cost a lot less than a VariEze gear, but I don't really like springy gear. When I make my own gear, which I may well do from the start, I'll be copying the F18; it will be a very long stroke trailing link with weak springs and a great new invention called big shock absorbers. It will not bounce, and I'm going to rent a crane and drop it from pretty far up in the air to test it. There's going to be a built in hard point at the CG on top of the fuselage for lifting it and for checking CG.
This is referring to having the horizontal tails with negative dihedral.
That type of thing isn't a problem for me with this design. I'm using a singe steel tube on each side to go back to the tails and I'll make a steel fitting, which can be made at any angle and then turned to any angle on the tube, and just weld it on. It will be another heavy duty hard point which also holds the vertical tails. I prefer one big strong hinge rather than a lot of little teeny wimpy ones. However, if I decide not to have an all moving surface and instead use a hinged surface then the tail attachment point will just be a hard point and I'll have to make some little hinges. I work with big strong steel stuff all the time, it's my nature and so it's the path I'm taking. Figuring out push rods, torque tubes, and bell crank geometry will be fun and easy, and I'll make it adjustable in the beginning.
Concerning tail authority, if need be, it will be easy to just make the subframe tubes longer and move the tails farther back.
Replying to Jan's comment: Aahhhhhh, my dyslexia is showing! Fixed it. Yeah, CG has to be forward of Neutral Point...
The Varieze has the advantage of being a canard, which means that the gear can be placed quite far aft to protect against propstrike and the lever arm of the canard can still easily lift the nose for rotation. In a standard configuration that uses the rear elevator to push the tail down requires that the main gear be placed farther forward, leaving the door open for a propstrike. Still, the dimensions and angles can be designed to keep the prop clear of the ground in all but the most extreme rotation angles. Kick-up of debris and small rocks into the prop is another matter, though.
Separate names with a comma.