Lately I've been entertaining myself checking out Free Flight Gas designs. Mostly because I like their appearance: graceful polyhedral wings, long thin tailbooms with big tailplanes, and nasty little pressure-feed engines. FAI F1C is an interesting technical accomplishment, especially by some talented Russians who seem to be hybrid engineers/jewelers, but I prefer lower-tech elliptical planforms. I think Billy Hunter's "Satellite" design belongs in an art museum.

OK, check out the aerodynamics. The "Free" in FF has to mean a big margin of inherent stabilty. After all, they fire these things off into hat-sucker thermals out in the California Kitty Litter deserts.

The time-honored FF gas formula seems to be a short nose moment with the engine tucked right up under the leading edge, high aspect wing with a thin (sometimes undercambered) airfoil section, long tailboom with a big lifting stab, and the CG back at 80% of the wing chord or more...

Huh?

CG at 80+%???

I've (mis)-trimmed enough R/C planes to know what happens if the CG gets back anywhere close to where the CP might be.

Could the FF's stab and tail moment be creating enough lift to move the combined wing/tail neutral point behind that rearward CG? Even with airspeed changes from power phase to minimum-sink glide speed? From upwind/downwide circles?

I tried backing the CG up beyond 60% on a flight simulator glider, and pushing the stick forward to control the ballooning. I got wild 180+ degree rotations around the pitch axis. Negative stability.

I can see why a FF designer would want a rearward CG to work. If a contest glider has to carry around the weight and drag of a tailplane, they'd like to get some positive lift out of it (although F3B/Thermal Duration R/C gliders don't try to.) Might even cheat a maximum-wing-area class rule if they don't count tail area.

I watched some AMA A/B Gas Free Flight's fly, and pitch oscillations following that vertical climb always decayed nicely into a stable floating glide. Random distrubances during the glide caused some porposing, but generally it damped out pretty well.

How do they do it?

I am sure the mattamatiks types will launch in here with all sorts of complex equations, but it works because these things are single speed airplanes, and that tails are loaded so that the lift centre is someways behind the main wing trailing edge. That way, even with an 80% CG, the balance point is still forward enough of the lift to keep the thing pitch stable. This works even if the tail section is symmetric, by the way. You can do this too, I like these things as well, and I have a Satellite 1000, with 3 channels and a 20 sec. engine run. Although I now have a multi speed model, (elevator control) the CG only had to move to 75% to make it stable in pitch at all speeds the model is capable of doing. On a nice bouncy day 45 min flights are regularly accomplished, you can even do 'DT'd' type descents, full rudder and full up elevator results in a nice spirally/stalled recovery from all sorts of heights without using up much airspace. Because of the vertical climb, short engine run, and vertical descent, I fly mine in a local park, muffled engine of course. I reckon more people should be doing it.
Evan, WB #12.

It's not just the size of the tail that matters. It's way back there. The force from that area however far back it is is called "tail moment".

Very few aeronautic things are simple enough that a sound byte covers it. Trying to describe stuff in single sentences usually doesn't work either, no matter how long the sentence.

Not only is that tail way back, but you also mention those long, narrow wings. The airfoil chord also has a major role in how much tail moment is needed to control pitch. Or more exactly a narrow wing chord needs less tail moment than a wide one. Wing chord and tail moment, two things that matter and should be considered.

This stuff isn't sound byte simple.

BTW, trying to think of a CG location as just somewhere on the wing chord leaves out 3 of the 4 major players that work toward pitch stability. And those are just the major players.

There have been a couple of articles back in the 50's about free flight stability. I did some figuring comparing tail volume with CG position and came to the conclusion that the difference between a super stable Ramrod and a marginal Civy Boy amounted to a range of about 10% of MAC for the CG. Stable at the anterior position and marginal at the rear position. One can play with tail volume numbers and move this 10% range about anywhere you want, even a couple of inches behind the wing.

I went out one day with a free flight and a roll of solder and made flights with the CG moved around. With the CG forward the airplane would try to loop. I gradually moved the CG back. As it moved back the airplane took a shallower and shallower angle of climb. I actually got it to make a shallow dive on the last try. Fortunately it was a short engine run and I did not lose the airplane.

PIMMNZ, you built a R/C assist Satellite??? Very cool. I'd love to build one like this. Can you give more details? Where did you locate the R/C gear? What servos? Anything special about the control surfaces, linkage, etc? What engine?

Not really the forum for this discussion Balsacutter, but in short;
1, Satellite 1000, series 75 plan.
2, TT .40 with r/c carb and muffler on radial mount shifted 1/2" forward of the plan firewall position.
3, 2oz tank behind engine mount, inside fairing behind the mount, this is now a sheeted 'tube' around the fuselage box instead of the block shown on the plan.
4, Three JR mini digital servos in line inside fuselage behind tank, the rear two operating the rudder and elevator, forward one operating throttle/shutoff. All through standard nyrod type pushrods.
5, Fin moved back far enough to allow hinge line rudder to be aft of tailplane trailing edge. (easy linkage)
6, Elevators around 10% of tailplane on T/E either side of fuselage (Big tail - 45% of wing!) Tailplane section modified with a bit of 'Phillips entry'.
7, Fin and tailplane permanently fixed to fuselage, tailplane at -2deg to wing.
8, L/E of wing sheeted back to mainspar, top and bottom, centre dihedral braces doubled in length and thickness. (Cos of the elevators!)
9, Battery and JR610 RX in pylon under the wing, servo leads led through from fuselage to pylon.
10, Cover with transparent film, looks real cool!
Evan, WB #12.

Oh yes, the fuselage under the tank and servos is removeable for access.
Evan.

The aft CG is also the "up" elevator . The plane is designed to fly in circles; if it flies straight a head the nose goes up and it stalls, but in a turn the pitch up is enough to keep it from falling out of the turn. Making it a regular rc plane requires you to move the CG forward so it can stop the pitch up stall routine. The decalage is a speed control for the glide; if set right it finds a happy medium to fly at; if you move the CG forward without zeroing the decalage you tuned the stab into up elevator so it wants to loop.

Balsacutter, picture worth a thousand words. The other two are a Koster 'Cream' variant and a HTL 'Tequila' from a '60's Aeromodeller Annual. Both scaled to over 2 metre span for a simple 'climb/glide' contest we try to fly. All 3 chan, balsa/ply/film.

As Jim Thomerson found pitch stability on a free flight is about getting enough but not too much. You want enough to let it recover from a stall or tubulence induced dive but not too much. Too much results in a model that wants to loop under power just as he found. I know it sounds counter to what most RC'ers think but it's been proven successfully by many FF'ers over many years. In fact I've seen far more planes crash during initial powered flights from having the CG too far FORWARD than I have from having it too far back. This applies only to FF models where it's not possible for the pilot to add down trim when things turn nasty.