This week I completed the first flight of my scratch built wing on the GeekStar. It has different lift properties that required a larger rudder to be added to maintain control of the aircraft, and after increasing the rudder size, it flew just fine.


That officially marks the end of the scratch built mission I've been on for the last year or so; being able to create a RC Airplane from scratch that can meet or exceed a kit's performance. 


Next week will be all about testing this wing against the easy star wing to determine if it really does perform as well or better.

Views: 963


Moderator
Comment by Michael Pursifull on November 19, 2011 at 5:03am

Too much wing surface in dihedral?


Moderator
Comment by Michael Pursifull on November 19, 2011 at 5:04am

I (re)modeled the HiLStar (Bixler clone in X-Plane) many, many times, trying to approximate the performance, I will be following your wing build closely.


Moderator
Comment by Michael Pursifull on November 19, 2011 at 5:10am

OK, nevermind, I was commenting while watching. Let me know if you want to talk over what I did for the HiLStar wing. It is a virtual model, but I spent hours looking at that wing and airfoil.

Comment by Don Cramer on November 19, 2011 at 5:29am

I think you should try cutting some ailerons into that wing:) I'm betting your performance will benefit from that mod big time.


Developer
Comment by Ryan Beall on November 19, 2011 at 8:08am

Not enough dihedral in comparison.  Rudder only airplanes only turn with dihedral.  Cl_beta is the stability term you are looking for.  The effect of dihedral on an airplane is Cl_beta.  Where C is just a convention for coefficient, l is the moment around the x axis (roll), and beta is sideslip on the aircraft.  You create sideslip on the aircraft by yawing. So roll due to yaw is a secondary effect.  You yaw, yaw creates sideslip, and sideslip creates roll.  There are different ways of creating dihedral other than actual physical dihedral.  Sweep etc.  The apparent dihedral on the ezstar is obviously greater than yours because of this.  A rudder only plane with no dihedral (physical or effective) has a very low if not negative cl_beta stability coefficient.  That is exactly what you are seeing here.  You can create more sideslip (bigger rudder), increase your physical dihedral or effective dihedral sweep aft - harder to build), or create your rolling moment through a first order effect via ailerons!  You will like ailerons a lot better I think even though they are harder to build.  

 

Hope this helps!  Let me know if you have any questions!

-Beall  

Comment by bmarshall5253 on November 19, 2011 at 8:18am

Trent, love your documenting of learning aerodynamics!

1. Moving the clevis out one hole on the rudder horn reduces the rudder throw, the lengthened rudder may have fooled you, moving out one hole on the servo horn would increase throw.

2. Quick way to decrease the angle of incidence on your Clark Y wing would be to move your wedge that fills in the gap in the fuselage in the back to under the wing from on top.

3. The swept up wing tip sections on a high lift airfoil act like increased dihedral of the whole wing,  if you reduce the dihedral of the outer panels you'll decrease the stability of the wing and your rudder won't have so much force to over come to turn.

 

Keep it up, you are doing great

 

Burt

Comment by bmarshall5253 on November 19, 2011 at 8:24am

Okay Trent it's up to you to resolve who is on the right track :-)

Ryan thinks you haven't got enough dihedral and I think you've got too much.

You have gone from a semi symmetrical airfoil on the easystar to a flat bottom airfoil on the Clark Y

so we look forward to your next experiment :-)


Developer
Comment by Ryan Beall on November 19, 2011 at 10:03am

Dihedral effect causes the plane to roll away from sideslip.  So left rudder creates sideslip right of the nose therefore a left rolling moment is created.  The more dihedral effect you have the more left moment you would generate.  Burt's "too stable" idea sort of counter intuitive actually.  Look at gliders for example.  They have very little dihedral and therefore usually have a harder time turning with rudder.  So Burt is right Adding dihedral stabilizes the spiral mode of the aircraft (wing leveling tendency) but because the wings now tend more away from side slip because that mode has a greater stiffness, when you create sideslip via rudder input, a greater rolling moment is created by using the rudder.

 

It gets kinda confusing beings it is a second order effect but that's how it works.  So Dihedral is two fold:  Increase stability of the spiral mode as well as increase the rudders effect to create a rolling moment.

 

If you think long and hard about it, The rudder itself as a first order affect creates a rolling moment opposite of the control input but because of dihedral effect the secondary effect of sideslip it will overcome that smaller moment. This smaller first order negative moment from the rudder comes from the fact that the center of pressure is above the CG for aircraft where the vertical stab points up.  Another cool way to increase your rolling moement from rudder is hang your rudder off of the bottom of the plane.  This is mostly impractical for obvious landing issues!  But none the less Cl_delta_r  or rolling moment due to rudder deflection would swap signs for an under slung vertical stab


Developer
Comment by Ryan Beall on November 19, 2011 at 10:13am

Not only is this plane a high wing (increases dihedral effect) but look how much physical dihedral this plane has (which turns great with rudder!).

Is your wing flat on the bottom minus the dihedral on the wing tips?  I'd guess that's probably just not enough specially beings you are more of a mid wing and have zero sweep (all reduce effective dihedral).


Moderator
Comment by Michael Pursifull on November 19, 2011 at 10:32am

I might have the symptoms wrong, they are based on my imperfect understanding of how it flew. It was my initial impression that you had trouble affecting lateral attitude change, and my initial feeling was that you were experiencing too much lateral stability. I wasn't clear about how much lateral "leveling" you were getting, and figured that your comment latter one was that, while too stable laterally, it was, over time, rolling, and you were struggling to keep it level.

 

I have less aerodynamic knowledge than almost anyone who would offer an opinion, so take this for what it is worth (not much) but I have spent a week's worth of time modeling and re-modeling this particular wing in X-Plane. 

 

Here are some of my still-in-testing impressions/tips, which, as I have suggested, may be totally wrong.

 

My first reaction to the wing is that it has too much wing service surface in dihedral. The various curves of the Bixler (in my case) or EasyStar wing tricked me this way, and in the end I have a very different wing design than I started with. 

 

Another thing that I believe "fooled" me between my original wing and my current wing is something I think you also experience... the angle of incidence. 

 

Here is how I handled both. 

 

In the dihedral, I found the wing to be much flatter than it appears. I added increasing sweep from mid-aileron (bixler) to the tip, and reduced the angle of incidence slightly, before returning it again to that of the reference for the wing. I would be surprised if you can easily reproduce the minor variation of AOI that I modeled, but it is worth mentioning. In the wing as a whole, I have a 1 degree dihedral along much of the length, before I introduce a 0 degree at about 3/4, and then begin the more radical wing tip curve a few inches outside of the aileron. All of this is detailed in the X-plane model itself if you want my current working numbers, and I will be refining them.

 

Now, here is a key part of my current "secret-sause", and you might consider it more of an artifact of Plane-maker than a wing feature ... that is, you might just consider this a part of the airfoil, but I handled it differently. While I found the AOI at the fusealog to be ~3.7 degrees (to the centerline/pushrod, not the angle of flight,) I believe that is not correct. Instead, I reduce the angle of the airfoil itself, and then add a fixed flap with a 4-6 degree downward angle at about 24-28% chord. This creates the impression of a steeper AOI along the length of the airfoil, when, in my thinking, the main portion of the airfoil is much flatter than it appears, because of the flap at the back.

 

I used a lot of different dimensions to get here, and I might not be communicating the current dimensions, as I can get confused easily with so many versions, but these and many other issues documented between my HiLStar17F developers guide and the X-Plane model itself (in Plane-Maker, you can open the model and examine each section of the wing in great detail.)

 

However, and this is important, the HiLStar17F has not yet been "tuned" to really simulate the EasyStar/Bixler yet. I don't have a field to fly the real thing yet, maybe next week, so I have no idea how it feels to fly. Which is why I am asking for more Bixler/EasyStar fliers to try it out in the simulators, and provide feedback, data, etc. All this is discussed in the documentation, if you are patient enough to skim through it. 

 

Please let me know if you are interested, and I can e-mail you the docs. They are too large to attach here, and I am trying to keep tabs on the versioning. I don't want to post them just yet, and have someone working with an old version, and thereby have any lost effort. 

 

 

Comment

You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2014   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service