SMALL - espritmodel.com SMALL - Telemetry SMALL - Radio
Reply
Thread Tools
Old Jul 07, 2010, 12:04 PM
What could possibly go wrong?
nickchud's Avatar
Market Harborough
Joined Apr 2006
3,702 Posts
Don!

I really appreciate all the work you put into the explanations you give. And you might not believe it, but it's not wasted on me. I think I recognise what you say about
Quote:
spiral instability (the plane wants to steepen up the bank angle by itself until it works itself into a "graveyard spiral").
When I built the wing for my LongEZ, I built it flat on its back so that the crown of each rib was in the same flat plane. I thought that this would give a uniform dihedral all the way from root (29mm thick) to wing tip (8mm thick). Somehow, perhaps due to the different sweep of the strakes vs the outer panel, I think the geometry worked out so that I got anhedral in the outer panel. It eyeballs that way. My Starships have never exhibited that spin and I did not build them upside down. Trevor, who saw the crash, suggested that it was not helped by the fact that mass is concentrated at the tail (motor) and at the nose (battery et al.) Most of the time she flew beautifully but, once that spin started, I was too close to the ground to pull out. Damage is only to the nose and forplane, but if I rebuild her with the same dihedral/anhedral issues, I'll probably get the same results.

As Einstein said: "Lunacy is trying the same thing twice and expecting different results."

And I enjoy your interesting extra details too.
Quote:
The original prototype VariEze (the one with the VW engine) did have just the elevons on the canard
. That is a surprise!

Thanks!

Nick
nickchud is offline Find More Posts by nickchud
Reply With Quote
Sign up now
to remove ads between posts
Old Jul 07, 2010, 12:13 PM
What could possibly go wrong?
nickchud's Avatar
Market Harborough
Joined Apr 2006
3,702 Posts
PS. Another example of spiral instability comes with my Catalina. It's a big, slow plane, so it's easy to manage. But, If I use 'bank and yank' to turn the turn soon starts to tighten. She has no dihedral and a pretty large fin area. Much better turns come from a little roll, then use the rudder and return the ailerons to neutral, or even reverse them, to complete the turn. I think that Ivan Pettigrew, who designed it, would say that's just what he intented.

cheers

Nick
nickchud is offline Find More Posts by nickchud
Reply With Quote
Old Jul 07, 2010, 03:55 PM
Registered User
Glendale, AZ
Joined Jun 2009
215 Posts
Don,

Thanks for the detailed response. I tried the "canard-aileron" setup today. They don't even budge the plane. I will use the servo I installed and run rods to the wings and try conventional ailerons.

Tom
trobinson3003 is offline Find More Posts by trobinson3003
Reply With Quote
Old Jul 07, 2010, 08:00 PM
Registered User
Joined Jun 2005
2,463 Posts
Delta Duck Build

The main wing was turned over and completed with the addition of sheeting, cap strips and servos. The motor will probably be the AXI 2217-12 with 1380 kv driving a 9-4.5 APC from a 3s-2700 mah lipo. RPM should be 11,250 with 285 watts input. The motor weighs less than 2.5 ounces. The wing now weighs about 7 ounces with it's heavy leading edges, solid ailerons and metal gear servos. The front wing weighs a respectable 37 grams ready to cover.
Charles
canard addict is offline Find More Posts by canard addict
Reply With Quote
Old Jul 08, 2010, 08:43 AM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,050 Posts
A couple more comments.

Nick, the VariEze/LongEze series has several degrees per side of anhedral in the wing, to offset the dihedral effects of the sweep in the wing, and from the winglets.

Sweep acts like dihedral, but the exact details are complicated. The old "rule of thumb" is that three degrees of sweep are equal to one degree of dihedral, but that is a gross (even dangerous) oversimplification. The amount of dihedral effect varies in proportion to the lift coefficient. At high speed/low G the effect goes away, while in slow flight/high G flight it increases, reaching its max effects near stall.

This means that the same plane that has spiral instability (from not enough dihedral effect) at high speeds can also have violent Dutch roll problems (from too much dihedral) when slowed down for landing! And no, this is not just theory, I've experienced the problem firsthand.

Fortunately, the amounts of sweep we usually see in models are usually not enough to make these issues significant. Below about 10 degrees (measured from the quarter-chord line along the flying surface) the effects are pretty negligible, and you really need about 15 or 20 degrees before they become really noticeable.

On the 'Eze series (full scale at least), the plane does exhibit wing rocking at stall (or at least as close to an actual stall as those planes get) that is very similar to Dutch roll. As far as spiral stability, the 'Eze's are extraordinarily good, because of the winglets and the sweep. In a turn, the outside winglet has a higher airspeed than the one on the inside of the turn, so it develops more inward lift. This yaws the plane towards the outside of the turn, which then couples with the dihedral effects of the sweep to roll the plane back to level flight. Note, the net effect of the slight anhedral, plus the sweep, plus the winglets, is equivalent to a normal amount of positive dihedral.

Note, winglets on a conventional layout with a straight wing do not have this spiral stabilizing effect, because the winglets are longitudinally at or near the C/G, so they have little or no yawing effect (although they do act like extra dihedral). However, on a plane with a very long tail moment arm, the curvature of the airflow due to the turn can result in the relative wind at the tail blowing inwards, yawing the plane to the outside of the turn, which then couples with the wing's dihedral to roll the plane level again, again providing positive spiral stability.

As far as why you saw what appears to be spiral instability on your models, I suspect that the incidence angles of your winglets might be involved. If they are simply acting as tip fins, not developing significant amounts of inward lift, then you won't get the rolling effects from them.

Going back to the spiral stability/Dutch roll balancing issues, as I mentioned before, too much dihedral can cause Dutch roll, while too much fin can cause spiral instability. Ideally there is a "sweet spot" of combinations of dihedral and fin that avoid both problems. The size of this "sweet spot" seems to be related to the mass distribution in the airplane. The more mass there is in the extremities (like what your friend noted), the smaller the size of the sweet spot. Mass in the extremities also degrades dynamic stability. If you have too much weight in the ends, it is possible that the sweet spot can become negative in size, i.e.: there is no combination of fin and dihedral that does not have spiral instability, Dutch roll, or BOTH! The mass in the ends is what contains the kinetic energy in an oscillation that the plane's dynamic stability has to damp out, and if there is too much mass in the ends, it could be more than the planes dynamic damping effects can overcome. Dutch roll is a form of dynamic instability involving a combination of roll oscillations and yaw oscillations, so mass in the nose and tail (which would increase the pitch and yaw inertia) could influence Dutch roll.

One more comment, on scale models of twins and other multi-engined aircraft, spiral instability from too much fin is a significant possibility. On single-engined aircraft (other than grossly overpowered aircraft like the P-51) the size of the fin is dictated mostly by general handling requirements. However, on a high-powered twin, the engine-out characteristics dictate the amount of rudder authority required. As a result, this class of aircraft tends to have about three times as much fin as an equivalent single-engined aircraft would need. Additional dihedral can offset this to some extent, but there's a limit to how far they can go with that because the extra mass of the engines out on the wings also makes the plane more prone to Dutch roll. The net result is that such aircraft often have to be set up a little more to the spiral instability side of the dihedral/fin balance.

If nothing else, all of this should help illustrate how changing even the most insignificant-looking detail can have ripple effects throughout the rest of the design.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Last edited by Don Stackhouse; Jul 08, 2010 at 08:48 AM.
Reply With Quote
Old Jul 08, 2010, 10:02 AM
What could possibly go wrong?
nickchud's Avatar
Market Harborough
Joined Apr 2006
3,702 Posts
Thanks for that Don

I was pointed to this link by a contributor to RCGroups called NMasters. I expect you know him. He agrees with you, you'll be pleased to hear:
Quote:
So the winglets on the Long-EZ pull inward toward the fuselage really hard. Much harder than you might imagine...
.

I did make an effort to give the winglets on my Long-EZ an airfoil, low pressure side inwards, but they're very thin and the wing is flimsy so I doubt that had much effect. Possibly, they would bend the outer wing upwards in a turn.
Quote:
I suspect that the incidence angles of your winglets might be involved
I will try giving them a bit of "toe-in" and see if that helps. I have been careful not to overpower it, so I don't have much room to use a smaller motor. Though I do happen to have a larger 2 cell battery, which would allow me to site it closer to the centre of the plane.

Anyway, it sounds as if I have the all-clear to patch up my model and try again. Just take a but more care flying it, more altitude and stop showing off.

The EZyCat has the workbench at the moment. Part of the theory behind that was to move as much weight as possible into the middle by using a tractor motor with twin booms / floats / skids to support the canard. I don't have winglets this time, just twin fins aligned to the floats, much more robust and they give me the opportunity to extend the rudders into the water and save some construction time. More photos to follow.

cheers

Nick
nickchud is offline Find More Posts by nickchud
Last edited by nickchud; Jul 08, 2010 at 10:12 AM.
Reply With Quote
Old Jul 08, 2010, 11:44 AM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,050 Posts
Quote:
Originally Posted by nickchud View Post
...I was pointed to this link by a contributor to RCGroups called NMasters. I expect you know him. He agrees with you, you'll be pleased to hear: .
Yup, I know Norm, we're both on a flying wings e-mail list. Nice guy, and very knowledgeable.

The link is long-winded, I could quibble with a few details and maybe add a few, but it's essentially correct.

Quote:
I did make an effort to give the winglets on my Long-EZ an airfoil, low pressure side inwards, but they're very thin and the wing is flimsy so I doubt that had much effect. Possibly, they would bend the outer wing upwards in a turn.
I will try giving them a bit of "toe-in" and see if that helps.
The big problem with winglets is that they tread a very narrow line between the induced drag they recover from the wing and the drag they create themselves. If you're off even a tiny bit in their design and setup, you end up better off without them. To make matters worse, it's usually difficult or impossible to make them perform well outside a fairly narrow range of flight conditions. The setup you need at one airspeed is different than what you need at other airspeeds.

Quote:
I have been careful not to overpower it, so I don't have much room to use a smaller motor. Though I do happen to have a larger 2 cell battery, which would allow me to site it closer to the centre of the plane.
The other option would be to move the motor forward and run a driveshaft aft to the prop, but that opens a whole Pandora's Box of other potential problems.

One more possibility is to reduce the size of the canard, which moves the required C/G location aft, closer to the motor, and also allowing the battery to be further aft. This also increases the aircraft's efficiency since more of its weight is carried on the wing, which is a more efficient lift producer.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jul 08, 2010, 02:45 PM
What could possibly go wrong?
nickchud's Avatar
Market Harborough
Joined Apr 2006
3,702 Posts
Quote:
One more possibility is to reduce the size of the canard, which moves the required C/G location aft
Now, with a smaller motor and battery, I could do away with the fuse altogether and have a flying wing weighing probably less than 1lb. That would be an interesting experiment and it could be fun but what I want is a small EZ model. How small can the canard get?

Andy Lennon's book suggests that 4" (100mm) could be minimum chord at the sort of speeds I have in mind. In fact I'm already down to 90mm. The real EZ has a long narrow canard and I don't want to get short and chunky. Before I had the crash, I was getting good performance, with super small, nodding stalls, inverted flight, loops and rolls. It was just the turn without adding enough power that caused the trouble.

Pilot error!

Nick
nickchud is offline Find More Posts by nickchud
Reply With Quote
Old Jul 08, 2010, 04:55 PM
Registered User
Don Stackhouse's Avatar
United States, OH, Bradford
Joined Jun 2005
4,050 Posts
Quote:
Originally Posted by nickchud View Post
...How small can the canard get?
Zero, but at that point it becomes a flying wing.

Quote:
Andy Lennon's book suggests that 4" (100mm) could be minimum chord at the sort of speeds I have in mind....
With the right airfoils, 4" is still way bigger than the minimum. The thickness and camber does need to go down as the Reynolds numbers go down, but much less than half that is no problem if you do it right.
Don Stackhouse is offline Find More Posts by Don Stackhouse
Reply With Quote
Old Jul 08, 2010, 10:03 PM
Registered User
Joined Jun 2005
2,463 Posts
Delta Duck Build

The fuselage and rudder drawing details were added. The 1/8" balsa sheet fuselage sides were cut out. The AXI 2217-12 motor, the Jeti Spin 44 ESC, and the TP Pro Power 3S-2700 battery were ordered. With the APC 9-4.5E prop, it will pull about 25 amps per the data. The fuselage measures 36.5" long,1.75" wide and about 4" high at the tallest point. The wing spans 30 inches, has a 16.5" center chord and a 9" tip chord. The rudder is 7 inches tall and will be built with 3/16" square balsa. I am anxious to see the fuselage go together.
Charles
canard addict is offline Find More Posts by canard addict
Reply With Quote
Old Jul 09, 2010, 01:38 AM
What could possibly go wrong?
nickchud's Avatar
Market Harborough
Joined Apr 2006
3,702 Posts
Charles!

Those are really good drawings. I hope they inspire some more people to try your Delta Duck. 250 watts, 30" wingspan: that should give you some exciting times.

Nick
nickchud is offline Find More Posts by nickchud
Reply With Quote
Old Jul 09, 2010, 11:17 AM
Registered User
Joined Jun 2005
2,463 Posts
Thanks for the support, Nick. Do you figure that it will fly? What could go wrong? I look forward to mushy three point landings and rudder spins. It will have to be red and yellow to accommodate my vision especially with the possible high speed, high being over 50 mph. With the small front wing it will probably act like a delta with built in elevators.
Charles
canard addict is offline Find More Posts by canard addict
Reply With Quote
Old Jul 10, 2010, 06:46 PM
Registered User
Joined Jun 2005
2,463 Posts
Delta Duck Build

The tail assembly and one fuselage side were built. The power system arrived and I was surprised to see how tiny the 2.45 ounce motor was. It is smaller than a Speed 400 and will deliver a claimed 300 watts vs about 80 for the Speed 400. The fuselage build is simple and will go fast.
Charles
canard addict is offline Find More Posts by canard addict
Reply With Quote
Old Jul 11, 2010, 01:14 AM
Fly, Crash, Repair, Repeat
Red Runner's Avatar
United States, CA, Fresno
Joined Sep 2007
135 Posts
X-Wing Interceptor

Here is the X-Wing Interceptor from Future Flight. The X-Wing is a 1989 design by Rollin Klingberg. I bought the plane in ~1991 already built as a nitro model. It was designed for a .051. I put an OS .10 on it then a Super Tiger .23. It actually flew one flight but needed the nose gear reworked & ended up under the workbench where it obtained lots of hanger rash. I have dusted it off & converted to E. WS 34" AUW 37 oz, 200 watts, Prop: APS 10x5 E

The X-wing had its maiden flight on Sat 7-10. The plane flew & flew well. Just have to remember to keep some speed on landing so the canard doesn't land first. The 200 watts was plenty of power, most of the flight was using about 150 watts.
Red Runner is offline Find More Posts by Red Runner
Last edited by Red Runner; Jul 12, 2010 at 12:13 AM.
Reply With Quote
Old Jul 11, 2010, 02:00 AM
Registered User
John235's Avatar
Sydney, Australia
Joined Mar 2006
1,315 Posts
Nice X wing! I remember seeing these kits back in the day..

I think this is a interesting design for a couple of reasons. 1) There is very little vertical fin area, so it relies on the vertical component of the wing for yaw stability. 2) The canard area is probably less than 13% of the main wing area. Despite this, it can achieve pitch stability by using a long fuselage. I'd say the symmetrical airfoil profile is an important aspect of the design too.

I noticed there isn't much ground clearance on the wing tips so please keep it level on the landings.
John235 is offline Find More Posts by John235
Reply With Quote
Reply


Thread Tools