HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Jun 10, 2012, 06:42 PM
12th Pursuit Squadron
TheAeronut's Avatar
Garland, Tx.
Joined Mar 2002
1,111 Posts
Discussion
Control authority of V-tail vs. Cruciform

An exchange has started in a thread in the Foamies (Scratchbuilt) section as to whether a Properly Designed V-tail will have the same control authority as a conventional Cruciform tail. I would love to see some input and/or discussion by some of the world class brains on this forum. Hopefully I will learn something. I am not too bothered whether this theory is proved or disproved - I am more interested in getting to the facts.

Thanks in advance for your input.

J.P.
TheAeronut is offline Find More Posts by TheAeronut
Last edited by TheAeronut; Jun 10, 2012 at 06:51 PM. Reason: correct spelling... :)
Reply With Quote
Sign up now
to remove ads between posts
Old Jun 10, 2012, 07:10 PM
Registered User
United States, FL, Indialantic
Joined Mar 2012
184 Posts
Looking forward to some good conversation. I've been following the OSG thread and am very interested in this topic.
Calla969 is online now Find More Posts by Calla969
Reply With Quote
Old Jun 10, 2012, 10:12 PM
Don't take your guns to town
vespa's Avatar
Thousand Oaks, CA
Joined Mar 2004
2,607 Posts
In theory, of course a V has the same control authority, though the V has more pure pitch and pure yaw authority than a X and less combined pitch+yaw.

In practice, of course not. V's are always undersized since people are not smart enough to read the directions and make the V the right size. Furthermore, V's typically have hinged surfaces at ~25% chord whereas X's often have full flying stabs and 50% rudder hinges.
vespa is offline Find More Posts by vespa
Reply With Quote
Old Jun 10, 2012, 11:24 PM
12th Pursuit Squadron
TheAeronut's Avatar
Garland, Tx.
Joined Mar 2002
1,111 Posts
I accept that a v-tail that is not properly designed is not the equivalent of a well designed cruciform tail. You also make the valid point that few people know how to design a v-tail thus most are poorly, even badly designed - if you could even call most 'designed'. Heck, I am a proponent of the v-tail and I don't consider myself to be a 'good' designer of v-tails.

Most of what I know I learned from Don Stackhouse. Ask Joe and Don is a wonderful source of information for the more technically minded model avation enthusiast. Thank you Joe and Don.

All that said, I would prefer that this thread focus on v-tails done right. Feel free to use not done properly as an example, but please try to tell us how to do it properly as well. Better yet, provide comparisons between cruciform and v-tails.

J.P.
TheAeronut is offline Find More Posts by TheAeronut
Reply With Quote
Old Jun 11, 2012, 06:33 AM
Herk
HerkS's Avatar
Virginia USA
Joined Jun 2007
1,647 Posts
An accurate answer to your question depends on the purpose of the aircraft.

While a V-tail is quite suitable for a large sailplane, (and many other types) it is not likely to be an effective choice for a 3D acrobatic design. Most (but not all) designers of competition DLG hand launch sailplanes would choose a cruciform tail in preference to a V-tail.

As always the intended mission-purpose of a model (or of a full size aircraft) is the first step in deciding on the many choices and compromises.
HerkS is offline Find More Posts by HerkS
Reply With Quote
Old Jun 11, 2012, 09:23 AM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,076 Posts
The points that Herk brings up are valid, but apply to specific situations. If you have a design whose mission profile has an extremely high priority on one particular parameter (such as yaw damping for DLG's, or the ability to keep control axes as isolated as possible from each other for aerobatics, because current aerobatic fashion focuses on a rectilinear coordinate system), then the advantage goes to whatever system does the best job of focusing on that particular issue.

However, if the design requires a balance of parameters, then these constraints do not apply, or at least not in the same manner. At that point other considerations come into play, such as efficiency over the entire mission envelope (not just during one action, such as a large rudder input), and operational concerns such as structural weight, loads during landing, or clearance for tail-mounted engines, or for staying out of the way of supply trucks and baggage carts.

The big thing to keep in mind in all of this is balance. Yes, a DLG needs lots of yaw damping during the initial stages of launch. However, it also has other requirements during the rest of the flight, which overall constitute a far greater portion of the flight than the fraction of a second it spends trying to damp out yaw disturbances immediately after release. Does that mean that the cruciform tail that does so well during the initial moment after release is still the best choice overall? Maybe, maybe not. It all depends on the details.

In my experience, one of the most common failings in airplane design comes from getting "tunnel visioned" on one point of view about one specific parameter or mission segment, in isolation from the other parameters. You have to look at the entire mission profile. You also have to be very careful about any "simplifying assumptions" in your analysis, particularly the subtle hidden ones you didn't realize you were making. "Engineering sacred cows" often hide in these.

As far as V vs conventional, properly designed they can both have the same control authority. Period. As far as them always being designed wrong, I suggest you buy a 2M Chrysalis plus a fuselage+tail kit, and see how they compare.

As far as how to do it, you need to start by making the total area the same. That's the single biggest factor. After that, for tail dihedral there is the question of constant control authority (linear relationship between tail angle and rudder vs. elevator authority) vs constant stability (a squared relationship between the two). The tail dihedral determined by the two approaches does differ by typically a small amount.

However, there is also (for models in particular) the effects of the V-tail having better span loading and/or Reynolds numbers for its panels than the smaller panels of an equivalent T or conventional tail, and also that whole issue of "destructive interference", which in my experience seems to be more of an interference with the tailcone, since on pod-and-boom designs it seems to be less of an issue. Lots of hairs to split on these issues, and each design will be affected differently by how these concerns balance out in its particular case. However, with proper fine-tuning, they can all be brought into balance. The final choice of tail dihedral typically will be based on a consensus of these various concerns, not just the number calculated from one method or the other.

As far as control surface sizing, that is not a V-tail issue. I have built all-flying V-tails (in particular a tube-launched UAV that had to have a folding tail), and they work fine. The details of control surface chord depend a lot on Reynolds number. As the Reynolds number goes down, the hinge location needs to move forwards, regardless of whether we're talking about a rudder or elevator on a tail, or a flaperon on a wing.

In general, I am not a fan of all-flying control surfaces. An all-flying tail needs to be larger than an equivalent 2-element surface to have the same control authority. The all-flying surface changes only its angle of attack when deflected, while a 2 (or 3) element surface changes both its angle of attack and its camber, which generally allows it to make more control force (lift) than an all-flying surface of the same size. One exception can occur at really low Reynolds numbers, where the amount of allowable camber increase can become very limited. Low-Re airfoils need to have very modest amounts of thickness and camber compared to airfoils for higher Re's. If the Re is low enough, you could be better off without any camber change at all, which would argue in favor of an all-flying control surface.

It's also generally easier to get good stiffness and strength, with less structural weight, from a 2-element surface than an all-flying surface, where the primary structure is also typically a moving part, and the load transfer typically occurs in a very small space, with consequently poor structural "leverage".

However, as always, there can be exceptions. What matters in the end, as always, is how well the designer finds the balance between the various mission requirements, and the various design parameters.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jun 11, 2012, 12:01 PM
12th Pursuit Squadron
TheAeronut's Avatar
Garland, Tx.
Joined Mar 2002
1,111 Posts
And we have a couple of winners... Thank you gentlemen.

Now for a specific detail. One specific claim was that a v-tail could not have as much control authority when the command was full elevator and full rudder. I say BS, as although only one side of the v-tail is deflected the force vector is more nearly perpendicular to the loaded surface while the same input on a cruciform tail produces a total force vector more nearly 45 deg. off of the surfaces. I think that the net result should be much the same.

I suppose that I must yield on the implied argument from the original thread that cruciform is better for aerobatic aircrarft on the basis that it is easier to set up and fine tune for specific yaw and pitch response. Still I think that total control power (sorry if I use this term in a manner inconsistent with engineering dialect) would be the same for any combination of inputs. Should you be sufficiently determined to do the set up and fine tuning, a v-tail would be a valid and workable design option for almost any aerobatic aircraft as well as any other craft that you would use a tail on.

J.P.
TheAeronut is offline Find More Posts by TheAeronut
Reply With Quote
Old Jun 11, 2012, 12:37 PM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,076 Posts
Blanket statements such as "a v-tail could not have as much control authority ..." are INHERENTLY false. Such statements imply that there is no size and setup of V-tail that could possibly have the same control authority as a conventional tail, which is of course ludicrous.

Yes, a V-tail with the same PROJECTED areas (same area in the top view as the conventional stab+elevator, and same in the side view as the fin+rudder) has the problem you describe. It can make the same rudder force, OR the same elevator force, but not BOTH at the same time. This is the "projected area method", which results in a badly undersized tail. The problem is not that it's a V-tail, it's that the tail is too small.

A properly designed V-tail with the same total area can indeed make the same rudder force, elevator force, or combined rudder+elevator force, as the equivalent conventional tail. Yes, there are a total of four control combinations where only one half of the tail is deflected, because in those directions that one half is perpendicular to the required net force, and only that half of the tail is needed to equal the effects of the combined rudder + elevator forces of the equivalent conventional tail.

There are four corresponding conditions for the conventional tail (up or down elevator only, left or right rudder only) where only one surface of the conventional tail needs to be deflected to equal the combined efforts of both V-tail panels working together.

In both cases, we are looking at the specific conditions where one particular tail type is strongest, while the other is operating at the point where the factors are against it.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jun 11, 2012, 01:10 PM
Don't take your guns to town
vespa's Avatar
Thousand Oaks, CA
Joined Mar 2004
2,607 Posts
Hmm... after a little more thought I believe I had it backwards. All else being equal (hinge locations, tail volumes, etc.) A V should have higher peak control authority in the combined yaw+pitch direction whereas the X would be stronger in pure commands. Here's my reasoning:
- There's a maximum Cl (AOA/camber) for any surface.
- There's a loss of efficiency (Cl reduction) with conflicing controls (e.g. elevator on V, or ele+rudd on a X)
- Therefore a V should have a lower max CL in conflicting directions (pure pitch or yaw) and X should be lower with combined pitch+yaw.
- Of course a V has higher Re, AR, and thus Clmax so that shifts things a little, but likely not enough to overcome the losses of the conflicting controls.

At any rate, V's are not great for aerobatics, DLG, DS, or any other aircraft which may see extreme sideslip angles since large yaw excursions cause one half of the V to blanket the other and produce a pitch response. And though a V typically has structural advantages, X-tails have the potential to be even more structurally efficient whilst also mitigating some of the fuselage boundary layer problems (e.g. Supra).
vespa is offline Find More Posts by vespa
Reply With Quote
Old Jun 11, 2012, 02:54 PM
An itch?. Scratch build.
eflightray's Avatar
South Wales U.K.
Joined Mar 2003
13,399 Posts
My only 'V' tail, and it works surprisingly well.
eflightray is offline Find More Posts by eflightray
Reply With Quote
Old Jun 11, 2012, 06:16 PM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,076 Posts
Quote:
Originally Posted by vespa View Post
...At any rate, V's are not great for aerobatics, DLG, DS, or any other aircraft which may see extreme sideslip angles since large yaw excursions cause one half of the V to blanket the other and produce a pitch response...
That takes pretty extreme angles of attack (close to 90) for that to be significant. About the only common cases where it's significant are spins (where a V-tail tends to force the nose down and break the stall) or a ground loop, where the tail tends to lift itself clear of the ground, pivoting around the nose and holding the tail clear of ground obstructions. We used to call that the "Monarch Pirouette"
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jun 11, 2012, 09:42 PM
12th Pursuit Squadron
TheAeronut's Avatar
Garland, Tx.
Joined Mar 2002
1,111 Posts
That's a Pretty bird eflightray, I hope it flies as good as it looks.

Don, thanks for that last post. As simple as it was, it did add to my knowledge of v-tail characteristics. Simple things like this allow me to better use them in future builds.

I don't remember my monarch doing a 'monarch pirouette'. Of course, now that we are talking about it, it makes me think that I need to dust it off and fly it again. Great little plane. As my friend said, that thing will go up on a mouse fart. Lots of fun to catch a bump at 15-20 feet and work it up to several hundred. I did that under ten feet one time. What Fun!

J.P.
TheAeronut is offline Find More Posts by TheAeronut
Reply With Quote
Old Jun 11, 2012, 10:01 PM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,076 Posts
I watched Joe Hahn do that once with one of our prototypes from less than three feet. Caught what looked like zero sink right over second base at the ball diamond down the street from his house, where we did a lot of our testing. He cranked it over into a 50 degree bank, wing tip just inches off the ground, started making three foot diameter circles with the inside wingtip, gaining a couple inches with each circle. Turns like that require some special techniques, like using twice as much "up" elevator as what would stall the plane in level flight, just to compensate for the curvature of the airflow due to the turn (there are some articles in "Ask Joe and Don" on our website that discuss this). After a couple dozen turns he was at about ten feet, and could flatten out the turn a little. A couple minutes later he was at about 300 feet, and rolled out to do some rolls and other aerobatics before looking for another thermal.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jun 12, 2012, 07:48 AM
Registered User
Blacksburg, VA 24060 USA
Joined Feb 2000
3,204 Posts
Being an old bear with very little brain left, I don't think anyone has mentioned adverse yaw of V-tail airplanes without ailerons. Comments?

Jim R.
JRuggiero is offline Find More Posts by JRuggiero
Reply With Quote
Old Jun 12, 2012, 09:30 AM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,076 Posts
Adverse yaw is a property of wings with ailerons, not tails. The down-going aileron has more drag than the up-going one, causing the plane to yaw in the opposite direction from the roll input. An exception is wings with a bell-shaped lift distribution ("BSLD", such as some of the Horten flying wings) with ailerons at the tips that do not go too far inboard, which can have zero adverse yaw, or even proverse yaw.

As far as yawing of planes without ailerons, that is also a property of wings, not tails. A plane with no ailerons uses yaw, coupling with the wing's dihedral and/or sweep, to cause roll. If you don't yaw the plane first, it won't roll. If you don't have dihedral effect (from dihedral, sweep, or interactions between the wing and fuselage), it won't roll, even if yawed.

The amount of yaw needed to get a given roll rate depends on the amount of dihedral effect present. More dihedral effect, less yaw needed to achieve a given roll rate.

Be careful about using sweep alone for dihedral effect. The amount of dihedral effect from sweep varies in proportion to the lift coefficient. At high speeds and correspondingly low lift coefficients, the effect tapers off, till at zero lift it goes away completely.

There is a balance between fin effect and wing dihedral. Too much fin and you get spiral instability (the plane wants to steepen its bank angle all by itself until it winds itself into a "graveyard spiral"). Too much dihedral and you get dutch roll. If too much of your dihedral effect comes from sweep, you can get a plane that has spiral instability at high speeds and dutch roll at low speeds. It can be tricky trying to land a plane that thinks it's a falling leaf all the way down final approach.

Jim, you might be thinking of adverse roll. Yes, the rolling effect of an upright V-tail is opposite the direction of yaw during a rudder input. However, the span of most V-tails is so small compared to the span of the wing that this effect is negligible. The biggest problem is that some folks who don't understand V-tails mistakenly think they're supposed to control roll directly, instead of controlling yaw like any other tail, and set them up as if they were ailerons, then are surprised when their controls are backwards. A V-tail is a tail, just like any other tail, and tails control yaw, not roll. There is also an adverse rolling effect from an upright fin+rudder on a conventional or T-tail, which may or may not be significant depending on how tall the rudder is. Reportedly this was one of the reasons for canting the fins inward on the SR-71, to direct the lift vectors for the rudders to go through the C/G, and eliminate the problem.

Jim, one more thing you might be thinking of: in a thermal turn, an airplane needs more lift coefficient on the "inside" wing tip than on the "outside" wing tip, to compensate for the difference in airspeeds between the two. On an aileron airplane this can be done with a small amount of aileron deflection. If the plane has no ailerons, it needs to be yawed to the outside of the turn to make the extra lift on the inside wing tip. This has nothing to do with the tail type, it's because the inside wing tip is going around the turn at a smaller radius from the center, and therefore has a lower airspeed.

In a thermal turn, the airflow past the plane is curved relative to the plane. If the airflow at the wing root is perfectly aligned with the fuselage, at the tail its blowing upwards, and in towards the center of the turn. This tends to pitch the nose down (which is why in a really tight turn you may need gobs of up elevator, more than what's required to stall the plane in level flight, just to compensate for this effect), and it tends to yaw the plane towards the outside of the turn. If the tail design, tail moment arm, wing dihedral, etc., are just right, the amount of yaw could be precisely what's needed to compensate for the difference in airspeeds between the wing tips. The plane will go around the turn all by itself like it was on rails, without any inputs from the pilot (other than the added "up" elevator). Tricky, and it generally works within only a limited range of turn radii and bank angles, but it can be done.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Last edited by Don Stackhouse; Jun 12, 2012 at 09:50 AM.
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Split v-tail suitable for yaw control? VTOLicious Modeling Science 27 Apr 06, 2012 02:05 PM
Discussion MKS Nuke 3: any good for V tail controls? Masterpiece Radios 0 Oct 21, 2011 09:35 AM
Discussion v-tail vs inverted v-tail tymbrewolf Foamies (Scratchbuilt) 17 Jan 07, 2011 04:42 PM
Discussion t tail vs v tail vs conventional SPARTMAN Sailplane Talk 8 Nov 09, 2006 12:29 PM
SS V Tail vs Inverted V Tail CRASHED AGAIN Parkflyers 9 Oct 29, 2004 10:24 PM