I think we have all heard the phrase " if it looks good...it will fly good ( or well ) "

True up to a point I suppose. The Tacoma Narrows Bridge looked good too, but I guess they chose the wrong NACA airfoil on that one.

I sympathize a little with the manufacturers because builders are opting for crazy engine combinations, fat wheels, bulky landing gear, bubble doors, and more, I suppose that's what STCs are for on certified aircraft, but in EAB it's a free for all. There are also design tradeoffs, Longer wings with more area are fine for STOL, just you are going to float on landings, pick up more parasitic drag, and crosswind landings will help put hair on your chest. Other shorter wing designs will have you landing shorter and bouncing like a pogo stick. Pick your poison.

I suppose practically speaking, the only way for a user to check all this out is to go to OSHKOSH and do a demo ride. I want a smooth aircraft that I can take my hands off and when trimmed right will fly true. If I add a little throttle, I don't want it to yaw 180 degrees and pitch up vertically by 90 degrees. Decent rudder authority is always welcome, as is minimizing noise, vibration, and harshness (NVH).

It seems some planes just "get it right", whether out of luck, good design & engineering, or both. The SuperCub, Cessna 180/185, and Fieseler Storch STOL come to mind. Others are clunky, and will always be clunky regardless of engine, prop, rigging, or other mods. A pig with lipstick is still a pig.

I'm just hoping that in 2014 there aren't too many pigs, kit form or otherwise, on the loose.

Many successful aircraft designs have simply used recycled, proven, and low-risk design elements. The actual performance is often a stone's throw from the initial back-of-the-napkin estimates for low speed (<160mph or so) flight.

As for simulation, the price tag for good 3d CFD code is often above 15k, and becomes somewhat important as one looks at higher airspeeds (compressible flow). My code is just a few years old, and would cost around $20k to update. But good CFD is not nearly as important as good engineering inputs to feed into it, a fact born out by years of watching really smart engineers with little actual experience develop great looking, colorful, time-intensive, and detailed models that generate absolute garbage out as a result. A few hours with a spreadsheet, some good assumptions, some small models, and some great experience will often yield better results. CFD is generally a verification and optimization tool, not a fundamental design tool.

As for the CAD side of things, it has evolved into the best way to design and develop parts that go together well, output machining files, and examine subtle structural issues. Structural analysis for light weighting, for example, takes a lot of time to do by hand for many kinds of geometries compared to using FEA software. The same rules about garbage in/garbage out still apply, and strong basic engineering skills are still where the money is at for making expensive FEA code work well. A person still needs how to draw a free body diagram, know their way around a Mohr circle, and interpret an S-N curve. Fatigue, stiffness, strength, modal analysis where appropriate, thermal design, etc., are all some of the many disciplines that get rolled up into a successful aero design.

As for testing, very few small GA companies really find a need or benefit from tunnel testing from anything I've witnessed. For many small GA planes, the general design guidelines, reasonably good fundamental engineering, good fab techniques, and comprehensive flight testing make a successful program.

Denali
"flies right", short of some extreme bad characteristics, is a very subjective thing. I learned to fly in a Citabria, and for 9 years after flew an old straight back/straight tail Cessna 150C. So for me anything that flies similar to those aircraft "flies right" to me. Even later model 150's seem to be sluggish and have heavy controls to me, and Cherokees have horribly heavy unresponsive controls IMO.

Plus, part of my primary training included spins. Spins in an aircraft with good recovery characteristics are a kick! So any aircraft that doesn't recover quickly from a spin, I consider to have poor flight characteristics. Citabrias recover from spins in 1/4 turn, my 150C recovered almost instantly. I've never flown one, but people who've done it claim a Super Cub requires 1 to 1-1/2 turns to recover from a spin. So in my book that's a poorly designed aircraft. Obviously a lot of people disagree with me on that one.

So my point is, what's your experience? What planes have you flown that you like their handling? Which ones didn't you like?

Lesuther said:;

....... CFD is generally a verification and optimization tool, not a fundamental design tool. .......

Fabulous post. It nailed down a lot of what I was thinking. I bel ievemany of the "fundamental designs" are what they are...kind of like the common basic bob sled designs in the Olympics... "form follows function".

I don't consider myself knowledgeable about aircraft designs.

So, in looking at some of the popular designs for today's 2 and 4 place EABs in kit or scratch build plans-only form, are they fairly well optimized ? Or are they "okay" but could use a lot of tweaking? Are some manufacturer's models far better tweaked and optimized than others?

I get the feeling that many of the Scratch build guys continue to evolve their project based on what they read and hear from other builders. They are the hard core rock star dudes of the builder community ! Other quick-build guys who then go out and get 200 % Builder Assist help probably just go with what they have regardless of whether continuing refinements become available or not

Denali wrote:So, in looking at some of the popular designs for today's 2 and 4 place EABs in kit or scratch build plans-only form, are they fairly well optimized ? Or are they "okay" but could use a lot of tweaking? Are some manufacturer's models far better tweaked and optimized than others?

Everything is a compromise. Ease of construction is important for the success of a kit, so simple curves, Hershey bar wings, and square lines dominate most designs. Nobody seems to care about winglets, or elliptical wings, or a lot of variable geometry, or hard to make fairings that could make the difference between cruising at 140 mph and 155 mph, or climb a bit faster, or shave 15 pounds of weight off, because they would be less likely to get built if the complexity of the fab starts shooting up, the weight might eat into useful load, or they exhibit less familiar or desirable characteristics during slow flight. 80% of optimization can be done fairly quickly with 20% of the effort, and the remaining 20% of potential optimization can require 80% of the time (and almost always a lot more).

That being said, there is a lot of room for tweaking any design. I saw a fellow who did nothing more than build a set of replacement wings, add 18" to the span of his RV, taper the wings a bit, and beef up the tail and spar, and gain 12 mph in cruise speed. It cost him a long year and a half of steady effort to design and build those wings and make the other structural and speed mods. But he enjoyed the process.

I agree with Denali about many older airplanes that just fly right. And some don't. I ruptured a disk in the middle of my spray career and found that the CallAir A9-A or B model didn't kill your back and legs like a Pawnee does. It has ribs in all the control surfaces. Working those flat ailerons, rudder, and elevator in a Pawnee in a rough wind will cause your knees to start knocking halfway through the day. Some of you guys with money should rebuild one of the old four place Calls. The Calls in Afton, Wyoming knew how to build a really nice flying airplane.

If anyone is interested, there is a free MIT online course starting in a couple of days:

https://www.edx.org/course/mitx/mitx-16-110x-flight-vehicle-aerodynamics-871
http://www.youtube.com/watch?v=YE54SRkkQ1U

From their website:

"About this Course

This course covers the physics, concepts, theories, and models underlying the discipline of aerodynamics. A general theme is the technique of velocity field representation and modeling via source and vorticity fields, and via their sheet, filament, or point-singularity idealizations.

The intent is to instill an intuitive feel for aerodynamic flowfield behavior, and to provide the basis of aerodynamic force analysis, drag decomposition, flow interference estimation, and many other important applications. A few computational methods are covered, primarily to give additional insight into flow behavior, and to identify the primary aerodynamic forces on maneuvering aircraft. A short overview of flight dynamics is also presented."

This class integrates some ODE's, vector calc, and some other math concepts. There are many other resources available that do not if people prefer...on the MIT website as well.

I'll pass. I just got my pre-course homework for PAX River. I think we'll cover this stuff in that course.

I'm excited but it's going to be a long year.

As mentioned, there are some excellent resources out there. My friend Dave Lednicer (Analytical Methods) is one of the best, And I believe his services are affordable for smaller (non-Prime) companies. Believe it or not, big names like John Roncz are available for smaller companies... he did the new spin-proof wing for the Icon flying boat. I am certain there are others.

Don't forget that Rutan did a lot of early aero testing with his Mojave Wind Tunnel - a pickup truck or station wagon with the test article mounted on a stand, driving around the dry lakes in the desert, with tufts and oil films and smoke pots.

As far as the good airplanes staying in the same attitude with changes in power... all you need is a few washers the size of your engine mount bolts, and an old free flight model airplane builder

And if you ever doubt the validity of the old free flighters, even in today's hi-tech aerospace environment, just look at the de-thermalizer on the world's first successful private spaceship !