I have also built several ships with a swept forward TE and used a single horn with absolutely no ill effects.  I also know that two World Champions and another gentleman with a second in the World Championships who do the same.  Billy Werwage (Ares, Juno), Bob Hunt ('80 Genesis and Crossfire) and Al Rabe (Bearcats and Mustangs).  All use a single horn with apparently no ill effects.  It is so insignificant that there is completely free controls.

I am presently working on Pat Johnson's P-47 (the big one) and it has dihedral with only one horn.  Doesn't seem like it should work but it does.

BIG Bear
RNMM/AMM

Its completely free if and only if you allow the horn tails to slide sideways in a slot. Anything else springs the horn. Billy and Bob Hunt can confirm that. How much effect it has compared to other issues is an entirely different matter.

    Brett

I use a square to make the tang on the horn 90 degrees to the trailing edge. Its close enough to be undetectable. So if you don't feel it its not there so it must be perfect. This has always been a problem with engineers. That's one reason Burt Rutan was able to go into space from a junk yard. He didn't need a over engineered $5000.00 door handle he just used car parts to do the same job, that were close enough.

We're pulling away from your nifty build thread, so I'm going to try not to comment too much more on this.  

You can go overboard in both directions on this.  You can equate "good enough last time" with "perfect", then go on to build a failure, or you can equate "nothing is ever perfect" with "don't risk anything" and stand shivering on the brink of a big pool filled with money and never take the plunge.

Yes, when Burt Rutan is personally supervising the build you can use old car parts to build quality airplanes.  But this doesn't work as well when you have a whole bunch of people working together on a project.  Burt can do it because he's brilliant: he can look at a part and sum it up immediately.  Burt is also not building airplanes in quantity, for great masses of idiot pilots.  Lesser engineers, building for just anyone, have to analyze the bejesus out of any part, then insist that it be made exactly the same every time, with controls -- when you spread that expense over 100 airplanes, you end up with a $5000 door handle.  I've seen the "it's perfect, just use it" attitude cause numerous projects fail, and I've seen as many products with exceedingly lame features leave a wake of irritated and disappointed customers behind, and I've even watched from afar as a client company collapse in the middle of a really fun project because "version control is too expensive".  And, we're all seeing Boeing jet liners with batteries that burst into flame -- clearly they didn't spend enough money there.

On the other hand, I've seen projects that I felt had strong potential never get launched because there was a Nervous Nellie at the helm, I've seen time and money spent flogging problems that were not there and could have been solved with a hypothetical door handle from a junk yard, and I've had a client company collapse because the owner was unable to triage a bunch of problems in a new product, solve only the worst show-stoppers, and ship before he hit bankruptcy.  You don't hear about those failures as often because they're not nearly as showy -- but it does happen, and if you keep your eyes open you'll see it happen.

If the surface flops down both when you hold the plane upright and inverted, then the amount of spring is negligible and you don't need to do any more work on the hinging and controls -- but the spring back is still there.

If the surface flops down both when you hold the plane upright and inverted, then the amount of spring is negligible and you don't need to do any more work on the hinging and controls -- but the spring back is still there.

I would love to have controls like this but so far using ball links is seems imposable due to the friction in the heim joint of the ball.

I meant before you get the linkage hooked up.  If the linkage isn't connected and you have enough friction, springback, or whatever that the surface won't just flop down, then you've got problems that need to be addressed.

Edit: Or you need to build heavier surfaces?  Wait... 

I can tell you this they are that free no pushrod attached. The flaps and elevators are the very last pieces to be glued on. I DO Glue the horns to the flaps. There is no need for slop. I have built at least 50 planes with tapered trailing edges and have done it this way for years. Its nothing new and it works just fine.

Brett congrats on your accomplishments but the need to always throw it in everyones face is not becoming. If you and the 99 others didn't want the info used you should not have let it out.

  I am not throwing it in anyone's face, how do you get that?  You were the one invoking third parties to prove a point - again. The effect is perfectly well documented in this NASA history publication:

http://history.nasa.gov/SP-4220/sp4220.htm and in particular, this chapter:

http://history.nasa.gov/SP-4220/ch6.htm

   Read about high effective sweep angles. Also, I hardly kept it to myself:

http://www.spacebanter.com/showthread.php?t=41486

  and I hardly "don't want the info used", I was just diagnosing the problem. It's hardly a secret, it's well known and we have talked about the same effect, to a different degree, here, on numerous occasions. It's one reason why many competition stunt planes have slightly swept wings.

  The problem was also later acknowledged as the problem by Rutan himself. Rudder application lead to extreme roll response, which couldn't be controlled in the portions of the flight that had weak aerodynamic controls, but still enough to prevent the use of reaction control. They didn't straighten it out until it was near the peak by using RCS fuel.  It's why SpaceShipOne is in a museum, not flying.

    My point was SS1 was hardly the result people coming in from the farm, scribbling some stuff in chalk on the hangar floor, and then hacking it together, all from a clean sheet of paper. All those "engineers" that spent endless hours and literally hundreds of billions of dollars doing research, test flying writing papers, doing wind tunnel tests, for many decades,  laid all the groundwork. Without all that, none of it would have happened.

    Brett

So experience counts for nothing? Do I have to have a degree in aerospace to do this? I would rather have a person who has done brain surgery doing mine than someone who has taught it. Here is a third party name Billy Werwage who as far as I know has no aerospace degree but has build hundreds of models of this design with no problems. I think I would believe his studies over a book. This is my point. All the books in the world don't compair with experience.

It's OK Robert.  We all know you build good airplanes.  That's not what's under discussion.

What you have brought into question is the need to do solid engineering work before you launch a technical project -- and neither Brett nor I are talking about toy airplanes on strings.

Until you have engineering experience, and you've seen a few projects go down in flames (perhaps even real ones and not just metaphorical ones), accept that "negligible for this application" and "none" are two very different measures of smallness.  Accept, too, that mistaking "negligible for that application" and "negligible for this application" can be two very different things, and in engineering practice neglecting to do adequate (and adequately paranoid) analysis have led to things like the exploding X-1, the exploding X-2, the exploding space shuttle, Boeing 787's whose batteries burst into flames, DeHavilland Comets breaking up in midair, the Tacoma Narrows bridge falling down, Chevrolet Vegas needing a quart of oil every 500 miles ("fill it up -- and while you're at it, check the gas"), Microsoft Windows, and the list goes on.

Well, like my dad used to say -- if you put an idiot through school, what you get out the other side is an educated idiot. 

Besides, unless the project structure is one guy working alone, you have to remember that any engineering project is done by a committee, many members of which can throw barriers in the way of good design without ever having to face any music at the end of the process.

That engine compartment you cite probably left a number of very frustrated engineers in its wake who knew exactly what was going to happen, and at least one happy "money guy" who got a year-end bonus for saving some pennies.

I guess I was lucky mine was a Cosworth Black and gold   

I went back and found a picture of the controls of the Juno pictured above. This was before ball links and I drove the flap horn from the rear so as to not have a high angle on the  pushrod going to the flap horn. Also it takes the load off the flap horn as I use to wear them out at around 700 flights
Ed

A more detailed analysis could give some insight as to how the "rigid" flap horn installation would work.  Once I can get bind-free flaps on an airplane with a straight TE, I might look into it. 

Robert will be pleased to know that the 777 had a very influential Chief Mechanic who had a posse that was intimately involved with every aspect of the airplane's design and was empowered to reject designs that were silly or that couldn't be maintained.  The result was a really good airplane. 

I'd like to say that when I worked as an aircraft mechanic with 737-200/300 and some airbus a300, we sometimes wondered if engineers ever held a wrench or followed the steps in the maintenance manual themselves...

Marcus

I've also built several I beamers with swept forward trailing edges and can also certainly tell you that is is most definitely possible to use a single control horn with out making the ends move back and forth in the flaps...Like a lot of others i've done it many times.

The simple answer is "slop".  We, or at least I use relatively loose fitting brass bearings on the control horn to flap trailing edge wheras a full scale setup would use very close fitting bearings.
The slop in the bearings ' coupled [with] free movement through enough travel to operate our control surfaces with freedom.

Randy Cuberly

Randy,

Based on my experience with single horns and swept hinge lines, I completely agree with what you wrote.    I built some airplanes with split horns and tried any number of ways to minimize the flex that automatically results with split horns.  Then I read that Bill Werwage and others were having success using a single horn with the swept hinge lines.  Intuitively, this should not work.  But, as you suggest, there is enough tolerance in the bushings, and in addition, I think enough flex in the hinges and how they are mounted in their balsa structure and enough flex in the hinged surface surfaces themselves (like sheet flaps) that allow the controls to be deflected the +/- 30^o to 40^o before any binding or resistance to control movement is encountered.

Now, there is some practical limit to how much sweep/dihedral can be accommodated in hinge lines for these model airplanes of ours, but I do not know what that limit is.  I have built several scale models with highly swept elevator hinge lines and another with a V-tail where lucky boxes were deemed necessary and used.

I tried using a horn made from a speedometer type cable mounted on bushings.  It worked on the elevators of the scale airplane, but it would offer way too much flex to be satisfactory on a stunt ship.

Then there is the method that Al Rabe used to use where he actually lashed his single horn, made from 1/8" music wire, to the trailing edges of the wing to adapt to the hinge line sweep of his super semi scale stunt ships.  One would think there would be excessive drag and friction to get the flaps to move.  Surprisingly, this worked fairly well.  There was no "springiness" when the flaps were deflected and very little resistance to  the controls.  The controls were not totally "loose" but the control forces needed were totally acceptable.

Keith

Just looking at it geometrically, dihedral will act differently than sweep, and should be worse.  The amount of bend needed in the horn with just sweep is going to be proportional to something like (1 - cosine) of the flap angle times the sweep angle.  The amount of bend needed in the horn with dihedral is going to be more like sine flap angle times the dihedral.

I'd sit down and figure it out for sure, but once I get past figuring out the amount of bend necessary (which I can see a direct path to, after lots of trig), there's so many variables to get from there to "will it bind unacceptable" that if I really need to know for a build project or something the best thing to do is probably just mock things up and try it out.

Tim,

Maybe I was not clear in my explanation.  No "bend" is necessary.  In fact, it would not even work if a single horn is used.  Maybe I do not understand your comment, but a single horn can be used with a swept hinge line or with dihedral of a combination of the two.  

It has been my experience with a number of semi-scale stunt ships that have a "moderate" amount of dihedral coupled with a "moderate" amount of flap hinge line sweep, a single control horn can be utilized without any "detectable" or "noticeable" binding when the flaps are moved through their +/- 30o to 40o range.  This is on several semi scale models of my design as well as the Rabe Bearcat that I have flown for a few years.  The dihedral is not large and the hinge line sweep is fairly moderate on all of these models.  As I mentioned, there is probably a practical limit of combined angles in whatever combination where a single horn is practical.  For most semi-scale stunt ship designs I have seen, these angles do not seem to approach such a limit.

Your idea is valid that if in doubt, make up a 3-dimensional mock up to check it out before investing the time and effort to utilize a single horn and then find out that it is not satisfactory for your given configuration.  I did the tail assembly with the lucky boxes to check out how they would work before I built the Swee' Pea for OTS with its 40o dihedral under each elevator to fly it at this last VSC.  Elevators/lucky boxes worked fine, but the airplane was too heavy to even do a loop.

Keith

Maybe I was not clear in my explanation.  No "bend" is necessary or would even work if a single horn is used.  Maybe I do not understand your comment, but a single horn can be used with a swept hinge line or with dihedral of a combination of the two. 

I was only saying that of the two, I expect that dihedral will have a stronger effect than sweep.  I certainly wasn't suggesting than anyone run away screaming from a design with either dihedral or sweep.

I have several big stunters with dihedral in the wing.  I have made it "au rigour" to use "lucky boxes" for the flap horn.  They are so easy to build and install.  There is never any question about horn binding.

F.C.