Can shimming at the trailing edge can compensate for a bit of lost washout?

Hi Joe,

Last week at our contest I had some trouble with my Chrysalis. It seems that when I snugged up the covering I didn't quite get the washout put in correctly. As a result I didn't do well the first round. I have since added some additional WO (about 1/8") and it flies great again. In a contest situation do you think that shimming at the TE can compensate for a bit of lost washout?
THanks
Ray

From Joe Hahn:

Hi Ray. I'm forwarding your question to Don, the guy in our company who can best answer that question. Meanwhile I'll give you my "take" on the question. Let's see---shimming the trailing edge will increase the nose-down attitude of the ship, similar effect as putting in some down elevator trim. She should fly faster, the launch may suffer some from less climbing, depending on the amount of shimming. Extreme case would be "tucking"--a scary thing--the ship dives after launch rather than climbs. Better be quick on the sticks and have some spare boxers ready if this condition comes up.

From Don Stackhouse:

Ray,

Incidence and washout are two distinctly different things, perform two entirely different functions, and are definitely NOT interchangeable.

Incidence has NOTHING to do with the angle the wing flies at. That is strictly a matter of wing design, airspeed, "G" loading and weight. The wing and Mother Nature will decide between themselves what angle (i.e.: "angle of attack") the wing will fly at relative to the flight path through the air. AFTER that is determined, the incidence (shimmed or otherwise) will help determine (in conjunction with the wing's angle of attack) the angle of the FUSELAGE relative to the flight path. That's right, for a given airspeed, changing the incidence will have no, nada, zero effect on the angle of attack of the wing. It will merely influence how well the fuselage is lined up with the airflow at that speed.

When you change wing incidence by shimming the fuselage you also change the angle between the wing and the tail. This means that the airspeed at which the elevator is at zero deflection is changed. To put it another way, if the model flew at 15 mph before with the elevator set at zero deflection, now you would have to add some elevator deflection to fly at 15 mph (in fact, exactly the amount of elevator required to generate the same lift force at the new stabilizer angle of attack that the zero-elevator-deflection tail generated at its old angle). Meanwhile, as I said before, even though the fuselage and tail would be going through all sorts of changes, the angle of attack of the wing at 15 mph would be the same in both cases.

Note also that nowhere in the above discussion do I mention anything about spanwise lift distribution or tip stall margins. That's because incidence changes have no effect on those parameters. Those items are in the jurisdiction of wash-out and wing design.

Wash-out and wash-in refer to twist in the wing between the root and the tip. Wash-out means the tip is at a lower angle than the root. This means that in level flight the root is more likely to stall before the tips. If the wing has wash-in, then the tips are at a higher angle of attack than the root, and one of them will most likey stall before the root. This would be a bad way to fly.

Let's assume the wing is constant chord (so the Reynolds numbers are constant along the span) and that the airfoil at the tip is the same as the root. Let's also pretend that there are no influences on stall angles due to the wing's tendency to have an approximately elliptical lift distribution, even if it has a constant chord (not a valid assumption, I know, but we're talking theoretical here). In this case, if the wing has no twist (wash-out = zero), then in level flight the tip would stall at the same angle as the root, so the entire wing would stall at the same time. Now if we put in 2 degrees of wash-out (the tip is twisted 2 degrees "nose-down" compared to the root), then when the root stalls, the tip is still 2 degrees below its stalling angle of attack. We call this "two degrees of stall margin".

If we now put this same wing in a tight turn, the "inside" wingtip will be at a lower airspeed than the "outside" wingtip, with the airspeed at the root about halfway between the two. Since we require the model to maintain a constant bank angle in this case (we're thermalling here, not doing rolls), then we have to keep the lift force at one wingtip exactly equal to the other. If the inboard tip is going slower, then the only practical way to keep the lift equal to the outboard tip's is to increase the lift coefficient relative to the outboard tip. In the case of a poly wing, the model is slightly yawed toward the outside of the turn, which increases angle of attack (and therefore lift coefficient) of the inside tip, and decreases the outside tip's angle of attack. This may or may not require an input from the pilot to the rudder, the curvature of the airflow due to the turn and its influence on the tail may in some cases provide the correction naturally. If we have ailerons, a bit of "top" or "outside" aileron can help make the necessary extra lift, by changing both camber and angle of attack simultaneously.

The only catch here is that since we've increased the inside tip's angle of attack, we've also used up some of our stall margin. If we make a sharp enough turn, we may use up ALL of our stall margin, at which point we can experience a tip stall on the inside wingtip. In the case of the Monarch 'D' we've designed it with sufficient stall margin for the inside tip to turn a 3' diameter circle without tip stalling. The Chrysalis doesn't have quite that much tip stall resistance, but it's close.

If you change the incidence of the wing on the fuselage, you have not made any change at all in the twist of the wing, and therefore you have not changed the stall margin between the root and the tips.

Even before we actually stall a portion of the wing, we experience changes in the lift and drag characteristics of the different regions of the wing, so wash-out settings can influence model behavior over a wide range of angles of attack.

So we should all go out and put 10 degrees of wash-out in all of our wings? Well, not exactly. If you have too much wash-out your launch height will suffer. In a good throw, the model may be going 70 mph or more as it leaves your hand. At this speed the overall lift coefficient for the wing is essentially zero. If the wing has 2 degrees twist like our example above, then portions of the wing (the root in this case) will be making positive lift ( with its accompanying drag), and the other parts of the wing will be making negative lift, with some more associated drag. At the point where we should be getting almost zero drag due to the production of lift (otherwise known as "induced" drag), we have gobs of induced drag because the tips are fighting with the root, thanks to the wash-out. Because of this, when I design a wing I try to provide at least some of the required stall margin through other means, such as airfoil and planform selections.

Trying to get smooth, stable thermal turns from a model with insufficient stall margin can be an exercize in frustration, as you learned at the contest. Getting decent launch height from a model with too much wash-out can also be very difficult. How much stall margin you need depends on your own flying skill and personal style. This is why we suggest a variety of wash-out settings, depending on your experience level, in the Chrysalis instructions.

One of the nice things about the un-sheeted leading edge type of wing structure we use on the Chrysalis is that it's easy to adjust wash-out, even after the wing is finished. If you don't like the way your model behaves, then change the wash-out and try it again. The negative side of this is that it's easy for the wash-out to change by itself, due to weather changes or leaving the wing in a parked car in the sun all day. Get in the habit of re-checking the wash-out periodically to make sure it hasn't changed.

Don Stackhouse @ DJ Aerotech

Don,
Thanks for the great response. My question was prompted by a comment from one of the local `experts'. I was a bit bummed when after tightening the covering a bit and adding about 3/32" washout, my Chrysalis would sink. It was almost as if `spoilers' ( which it does'nt have) were deployed. When I add about 3/16-1/4" washout it seems to fly pretty well. I also had this problem when I was first trimming it out and almost gave up. It is balanced in the range on the plans.

Ray, I think you're trying to fly it too slowly. The washout is there for tip stall resistance. As soon as you have enough to prevent that, you don't need or want any more. Joe and I have found that the turning stability and tip stall resistance is excellent with just 3/32" washout on each inboard panel, and zero additional washout on the outboard panels i.e.: build and cover the outer panels flat, zero washout, and mount them to inner panels that have 3/32" washout. This give each entire outer panel a constant 3/32" washout relative to the root.

The airfoil on the Chrysalis likes to be flown fast. This is likely to be the case on any good modern hlg. If you slow down an hlg, the Reynolds numbers quickly deteriorate, and the performance goes down the tubes with them. Even in a turn you want to stay fast. I watched a couple of beginning-intermediate Monarch 'D' owners at the last club contest with the same problem. They were dragging around the sky with the fuselage about 3-5 degrees nose-up, coming down like parachutes. I told them that the key on the Chrysalis is to keep the fuselage level with the ground at all times, even in a turn. All of a sudden their flight times (and their enjoyment level) went way up!

When you crank massive amounts of washout into the wing, what you're also doing is making an incidence change for everything except the root area of the wing. You're used to seeing the model with a certain fuselage attitude, and you trim and fly it in that fuselage attitude range. When the majority of the wing is twisted with additional washout to a lower angle, the majority of the wing is now flying closer to the lift coefficient and airspeed for which it was designed. Unfortunately the root is still being dragged around at too high an angle of attack, so the model is still not flying as well as it could, especially with regard to launch height.

I would recommend that you re-set the washout to exactly 3/32" on the inner panels, zero additional washout on the outer panels, remove all shims you may have added between the wing and the fuselage, and then set your elevator trim to whatever it takes so that the model flies with the fuselage level with the horizon when hands-off. Now work on resisting the temptation to pull the nose up! Concentrate on keeping that level fuselage angle relative to the ground during the entire flight. You will find that you need a small amount of "up" elevator in steep turns to counteract the curvature of the airflow due to the turn, but you need to avoid pulling the nose up, just keep the fuselage exactly level with the horizon. Concentrate on keeping you control inputs small, smooth and very deliberate. You might even want to practice flying with the trim knobs when you're going to be flying in the same mode for a while, like in a steady turn, or flying in a straight line between thermals. I think you'll find that your Chrysalis starts performing a lot better.

Don Stackhouse @ DJ Aerotech