|
You may remember (if you could keep your eyes open long enough) that in the last Sloping Off I posed a question about scale airfoils for scale models. In the hope of getting a better answer, I approached the RC Groups Canard Thread aerodynamics guru, Don Stackhouse, for his opinion.
Hi Don,
I've been a largely silent lurker on the Canard thread for a while now and reading your comments on airfoils, and your CV, may I go off thread to ask about airfoils on scale models?
Basically I'm asking if a scale-like thick airfoil (Go 632 for example) would be a sensible choice for a 100" scale flying boat instead of a glider-type airfoil like S 7055.
Regards,
Mike
Mike,
For starters, I at least partially disagree with your other two respondents to your question. (He read the article in the last Sloping Off) However, if you ask a dozen experts the same question, you will get a dozen answers. If they are really GOOD experts, you will get two dozen answers. The airfoils they are recommending as being successful on scale sailplanes are "last generation" sections that were popular before folks had their eyes opened regarding the evils of too-thick airfoils. We demonstrated that in the summer of 1992 with our "Monarch" 1.5 meter RCHLG, and showed that to the world when we entered production with it in 1993. That summer Frank Weston at WACO saw what we were doing and tried it on his larger sailplanes, and demonstrated that thinner airfoils were the secret to a quantum jump in performance for the larger classes as well. Today the only folks still building sailplanes with thick airfoils at low Re's are the ones who don't know any better. Of course if your model is so huge (like some of those 6 meter scale sailplanes) that their Re's are high enough, then it isn't as much of an issue for them.
First of all, what Reynolds number range are you operating in?
What's your target flying speed, root chord and tip chord? Re at std day sea level conditions is chord in inches times airspeed in MPH times 778. It tends to go down a little at higher altitudes.
The air molecules don't care whether it's a scale model or not. However, a 100" model with powered aircraft aspect ratios might have enough chord to get out of the worst of the Re problems. At Re = 250K you have a lot more leeway on thickness than you do at 100K, or 60K.
OTOH, most scale modelers are concerned about "scale speed", which at best tends to be far too high unless the model is feather-light, but you're still trying to get as close as you can. And, having an airspeed that is too high will hurt your model's visual impression more than the visual distortions from airfoil thickness issues.
Yes, I am familiar with this problem, and have had to deal with it on my own designs. And, as I indicated above, the fact that it's a scale model counts for exactly nothing as far as the air molecules are concerned. In addition, the fact that it has a bunch of extra drag from a big fuselage and a bunch of scale details hanging out in the breeze just makes the problem that much worse. You're already flying with a handicap even before the choice of airfoils comes up.
As I said, if the model is big enough, and your Re's are above about 200-250K, you're in a much better position to tolerate a 10-12% thick airfoil. However, your tips are probably going to be more in the 100-150K range or less, and there you're not only looking at more drag, but also less max lift coefficient, just when you really need more.
The real culprit here is the slope of the aft portion of the upper surface.
At low Re's, especially in the sub-100K range, the flow would rather be laminar than turbulent. That is, laminar and separated than turbulent and attached. Even if you turbulate it, the flow may try to revert to laminar. Because it's laminar, there is no mixing between layers, and so those bottom layers have only the kinetic energy they started with to push their way to the trailing edge, which they are slowly losing as they go, due to skin friction.
On the front part of the airfoil, back to the high point, the flow is accelerating, which (thanks to Bernoulli) means the static pressure is decreasing. An air molecule in that flow looks downstream, and sees a lower static pressure than where it is now. This is a "proverse pressure gradient", and it helps push the flow along. However, after the high point, the flow begins decelerating, the static pressure starts to rise, and we now have an "adverse pressure gradient". And the flow's own kinetic energy is all it has to muscle
its way aft against this rising pressure. That same kinetic energy that is getting weaker due to skin friction. And, if this flow is laminar, there is no mixing between layers, so no fresh infusion of energy from the upper layers to the lower layers to help re-energize them. Those bottom layers are on their own, and the closer they get to the trailing edge, the weaker their ability to continue the fight.
If their energy gets too low to fight the rising static pressure, the flow separates, with a big loss of lift and an increase in drag.
The intensity of that adverse pressure gradient depends on the slope of the airfoil surface in that area. For the critical upper surface where this problem tends to be worse, more thickness, more camber, and/or more angle of attack tend to increase that slope, and worsen the adverse pressure gradient. Also, having the high point too far aft tends to concentrate the decelerating into a shorter portion of the chord, which also increases the adverse pressure gradient. That's a problem with some of those airfoils your other experts recommended. And note, ideally the slope near the trailing edge should be LESS steep than the areas forward of there. However, most conventional airfoils (such as the Clark Y) tend to be convex over their upper surface, so the slope is steepest at the trailing edge! This is fine (within reason) if you have flow that's transitioned to turbulent by then, but if your Re is low enough that the flow back there is still laminar, that could be the "last straw". Goodbye, attached flow!
And it's the tips that see this situation first and worst, so we're not just looking at lift loss and drag rise, this is also the beginnings of stall, and tip stall has caused the demise of a great many scale models.
At this point a partial solution might be tickling at the back of your mind. The aerodynamic problem is especially bad at the tips, but the visual problem is more of an issue at the root. So, use something not too awful but shaded more in favor of the visual issues at the root end (for example, not scale thickness but somewhat thicker than you would normally use, but keep the high point a little further forward, which will make the forward part of the airfoil look thicker than it actually is, without hurting the adverse pressure gradient as badly), then transition to an aerodynamically more optimum shape further outboard. In your flying boat's case you do have the problems outboard with the intersections with the nacelles, but you can still try to do the best you can with those.
BTW, the Eppler 205 has some aerodynamic problems with flow separation on the aft upper surface at higher angles of attack, but still well below stall. It can be used as a sort of "spoiler" for glide path control on landing if you're a really good pilot. It can also cause you to fall out of a thermal like a brick before you realize what's happening if you let it sneak up on you (which is very easy to do, the onset of it is very subtle, with little or no warning). For a scale model, it would tend to make your model "spin out the bottom" of the turn onto final approach, which would be really bad. Soartech 8 discusses this problem with the E-205, along with a Selig section that is an E-205 that has been "improved" to fix this problem.
There's probably a lot more we can add to this discussion, but this should give you a few ideas to think about.
Don
|
