Stab design from a technical perspective (many questions)

[SoarNeck]

Hi all,

I was curious if I could get some help with how to optimize the tailfeather design for my new F3B sailplane. I am aware of the general guidelines concerning appropriate tail volumes and how to calculate them, but since I'm trying to optimize everything else on the model, I want to get away from rules of thumb as much as possible. Some might call me masochistic, but what the heck. These are semi-random questions, but all are useful for me to organize my thoughts.

Stab design:

(T/F?): In level flight and when the model is balanced properly, the stab is really just providing enough lift to balance the pitching moment produced by the wing when the lifting force being produced is enough to balance the force of gravity. A nose heavy condition provides an extra force that the stab must offset (trim drag...boo).

In climbing or diving flight, the lift vector no longer directly opposes gravity, so the lift requirement for the wing changes. The wing has to adopt a different angle of attack (with a correspondingly different Cm).

Calculating the lift required for a turn would require overcoming inertia, which might be a complicated calculation for a model that doesn't exist yet.

I can calculate the pitching moment produced by the wing from the formula M = Cm x (rho/2) x (v^2) x l x S. Then, knowing the tailboom length, I can calculate what the force required would be at that distance.

Am I correct in remembering that Cm is referenced to a point at 1/4 chord?

Can I approximate the center of pressure for the stab to be at 1/3 chord to know where to resolve the required force to offset the pitching moment?

I assume that stability calculations would be relevant to stab design to minimize oscillations, but could someone briefly explain what they might be?

I would guess that the choice of a stab airfoil would be one that could develop the required forces at reasonably small angles of attack. Is that correct? Knowing what the forces (required lift) are would be of help there!

Cambered elevator:

Even if it's not worth trying to calculate the required size of the stab, would it be worth chasing a stab airfoil that develops less drag than the traditional symmetrical section when operating in "level flight" conditions (majority of speed & distance)? I would still need to know the lift required to balance the model for level flight, but some drag reduction might be possible. Turning conditions might be harder to design for.

I can't recall much of the debate regarding cambered tail sections, but what was the final verdict?

My thoughts:

Symmetrical sections produce equal lift in both + alpha and - alpha conditions.

However, given that it is more of a requirement to pitch the nose of a model upwards (climb/turn), wouldn't it be beneficial to chose a stab airfoil that might be biased towards "lifting the tail downwards"? That would point to a cambered section mounted inverted, in my mind. You would then be able to compromise the airfoil design so that you would produce less drag for positive angles of attack on a polar ("lifting the tail down"), with some sacrifice of extra drag if you had to push the nose down. Comments?

Maybe I will just use V(h) ~=0.55 and be done with it! :)

Cheers,
Adam

[Sparky Paul]

Why change the point of reference on the horizontal to 1/3rd from the 25% point which all profile information uses?
Isn't the ideal situation in cruise one where the stab isn't lifting at all?
Aren't climb/dive situations of transient duration, and seldom exist long enough to be of any practical consideration in the design?
T/D planes have evolved to a point where any improvement is more likely subjective than objective, and it's how the pilot guides the plane with the least disturbance to the lowest drag trim situation that results in the longest flight... and being able to see air helps. :)
I'd prefer a flying tail to an elevator as they have less drag than a flapped surface with a deflected flap.

[Ollie]

Viurtually every design decision involves conflicting objectives that have to be balanced against each other at some design point that favors all the model's objectives. Just thinking about, quantifying and prioritizing what you want the model to do is the place to start because it provides quantitative standards for making the tradeoff decisions.

On the matter of tail area, here are some things to think about. For a given tail volume, a long tail moment with small tail area gives a favorable reduction in parasitic drag and increased damping of pitch and yaw oscillations. Parasitic drag isn't a big part of the drag budget at thermalling speeds but the lower the parasitic drag, the lower the wing loading can be for a given penetration ability and lower wing loading allows less angle of bank in a thermal turn than a high wing loading. Therefore, parasitic drag reduction affects thermalling ability both directly (small effect) and indirectly (large effect). The cost is possibly increased nose weight to balance, increased moment of inertia in pitch and yaw and, increased required stiffness in the tail boom. The penalties can be minimized by a carbon fiber tail boom with the fibers running lengthwise and only enough diagonal glass cloth to give a little hoop strength. The strength to weight ratio of the tail should also be maximized and matched to the tail load in a hard zoom launch. Weight can be saved by using the lightest control linkage system that will do the job. Studying the designs of Dr. Drela will pay off in helping you to see what favorable trades can be made.

I agree with Paul that many existing designs will be very hard to beat and that pilot skill trumps performance differences almost every time.

For most of us the joy of engaging in the design process is the justification because like the typical contest flier, the best we can reasonably hope for is to get closer to the best.

[Sparky Paul]

The three tasks in F3B require a one-size-fits-all airplane... good L/D, good high speed and manuverability.
For the duration and distance tasks, normal practices for best gliding speeds are well known, but the turn-arounds in distance and speed ask for something a bit more than merely slippery at a single L/D.
A camber changing wing is almost dictated by the tasks.
Getting that good is probably more important than the small amount of improvement possble at the tail.
I've read where some guys in the speed task don't turn as much as split-S to go to the opposite direction. This can do strange things at the end of a long tail boom. :)

[Ollie]

Good point, Paul. The tail boom has to be designed for stiffness. the down load on the tail can be very high in high speed turns and in hard zooms. If the aeroelasticity is too high, up elevator control is lost. That's another good reason for an all moving stab that is aerodynamically balanced. It can tolerate a little more tail boom bend if the throw is great enough.

You could save the plane while loosing the contest by doing an outside maneuver instead of plunging uncontrolled into the ground if aeroelasticity caused lack of response to up elevator and you had enough altitude.

Running some rough calculations, at a speed of 125 FPS a stab with an area of half a square foot and a maximum lift coefficient of 0.7 could produce a maximum download of about 26 pounds!

[SoarNeck]

Hi guys,

I've built and flown high performance sailplanes before, so I'm definitely well aware of the structural requirements. I'm just trying to get entirely away from "rule of thumb" design as much as I can, except as a final check.

I finally finished the additions to my spreadsheet that would allow for 21 panel local cl/ local lift/ local cd/ local drag/ local cm/ local moment analysis of a wing, so I'm quite prepared to deal with the wing. I know the aero and pilot-load requirements for f3B in detail, and already have about 12 hours into a wing design.

Since I spent so much time on the wing, it didn't seem right to just copy the tail volume of other similar sailplanes.

There's plenty of room to improve on existing designs, which are largely compromised for production (or just based on older airfoils/ideas etc). I've also quickly found how little most mfgrs really know!

[shoe]

Hey Adam,

Are you planning on posting the most recent version of your spreadsheet on your website? I'd like to check it out. Thanks

[SoarNeck]

Not yet, but I plan to later. Send me a PM with your email address and I'll send you a copy.

Cheers,
Adam

[BMatthews]

So many questions...:eek:

:D

Like you I went through this a few years back when I was designing my own glider. Other factors came into the picture and the model never was finished but that's another story. But I did wrestle with a lot of the same questions as you are now.

If you choose an airfoil that has a low Cmo you can get away with sizing the tail and boom so that you're closer to the smaller end of the acceptable range of Tvh. In fact if it was symetrical you can get away with a tail that's well under the minimum accepted value since symetrical airfoils do not have a changing Cmo but you probably don't want that. What you need to do is spreadsheet up a LOT of designs and add in the Cmo of the airfoils and the Tvh for each along with some sort of empiracal value of how the model flew. This info must also take into account the Cmo of the airfoil with the various flap positions involved. It won't do you any good to minimize the stab size based on a total flap range of -3 to +3 only to have the model turn into a monster when you drop the flaps to 90 for landing or 20 for a good launch.

When I was deciding on the airfoil for the tail I was thinking of minimal drag in the common thermal turn. As such I chose a very low cambered airfoil that showed it's minimum drag at a low but still positive Cl and just decided to live with a slight drag when cruising at CL=0 or very close to it so that I could get the minimum drag when in a positive but shallow thermal turn. I was hoping that the reduction in drag at the tail would partially offset the increase in wing drag as the Cl, and thus the Cd, rose while in the turn. I still think this is a viable option but it would require further study to learn what the tail's Cl range is between level flight with the tail load close to Cl=0 and whatever it would be in, say, a 20 to 25 degree thermal bank. From there I would chose an airfoil that offered me the same Cd at thes two points knowing that the minimum drag would be while circling at about 10 degrees of bank. But while that's fine for thermal duration you're talking F3B where SPEED is king. In this case I'd stick to symetrical and learn which option gets the most lift with the least drag at the max Cl required for sharp turns, a symetrical full flying stab or a conventional hinged elevator and fixed stab. It's not as clear a choice as you would think I suspect.

If you can determine how much extra G load is required during a given bank turn you can determine the lift required (as in it's the same or the model would not keep on flying... :D ) From that it's an easy task to determine the Cl for the wing in order to generate that much lift. Most of the guys here would use a formula but I cheat. I go to FoilSim and enter in my model's sizes and then up the weight to reflect the turn loading and input a flying speed. FoilSim politley returns the Cl required to stay in the air. Yeah, so I'm lazy... It's close enough to set up some simple range of numbers and it's as close to reality as you need just to get into the ball park.

Now the big question is "Is it all worth the trouble?" Probably not since all this trouble probably amounts to either hitting the perfect flight path or missing it by as little as a yard or two but it makes for some fun brainstorming.

[Ollie]

Since the wing has around ten times more drag than the horizontal tail and you have put 12 hours into the wing design, then you can rationalize putting around 1.2 hours into the horizontal tail design. ;)

Oops, times up. ;) ;) ;)

Pick a red line airspeed. (A worst case maximum airspeed would be the terminal velocity in a very steep dive. The angle of dive that gives the maximum terminal velocity is the angle that results in the wing drag minimum and the corresponding coefficient of lift.) Calculate the down load that must be provided by the stab at that red line speed based on a CG location. Choose the stab airfoil and using its maximum lift coefficient, calculate the minimum stab area required to maintain control at that speed. Choose a stab airfoil that does not have a jog in the coefficient of lift versus alpha plot near zero. Take that minimum stab area and the tail moment arm length and calculate the pitch damping factor. Then make a judgement call about the desirability of increasing the damping factor to reduce low amplitude porpoising initiated by gusts or pilot input.

The greater the pitch damping factor the closer you can fly the plane to the wing's maximum lift coefficent without stalling. In other words, the bigger the pitch damping factor, the less stall margin a given pilot can maintain without stalling in gusty conditions. if you can fly with neutral static stability, then there will be no low amplitude porpoising to deal with and damping factor is no longer a consideration. So, stab design comes down to a matter of matching piloting skill and allowable pilot work load.