Go Back  RCU Forums > RC Airplanes > Aerodynamics
Reload this Page >

airfoil difference guestion?

Notices
Aerodynamics Discuss the physics of flight revolving around the aerodynamics and design of aircraft.

airfoil difference guestion?

Old 10-16-2002, 01:30 PM
  #1  
edge 3644
Senior Member
Thread Starter
 
Join Date: Apr 2002
Location: Gabs,
Posts: 149
Likes: 0
Received 0 Likes on 0 Posts
Default airfoil difference guestion?

Hi,

I have built two almost identical 90 size fun flies. Each has over 7 sq. feet wing area and 6"wide ailerons. In one of them the airfoil
used is 20% Naca and in the other same airfoil, but from main spar to washout I made airfoil flat. First one rolls much faster than second with same aileron diflection. Why?
Thanks for any advice.
Robert
Old 10-16-2002, 05:02 PM
  #2  
DICKEYBIRD
Senior Member
 
DICKEYBIRD's Avatar
 
Join Date: Apr 2002
Location: Collierville, TN
Posts: 2,749
Likes: 0
Received 0 Likes on 0 Posts
Default Re: airfoil difference guestion?

Originally posted by edge 3644
First one rolls much faster than second with same aileron diflection. Why?
I have been told the reverse should be true. Do you have exactly the same radio gear? Ie: servos, linkage and tx setup?

Beats the heck outta me leutenant!
Old 10-16-2002, 05:56 PM
  #3  
BMatthews
 
BMatthews's Avatar
 
Join Date: Oct 2002
Location: Chilliwack, BC, CANADA
Posts: 12,422
Likes: 0
Received 22 Likes on 19 Posts
Default airfoil difference guestion?

This is very interesting as I've wondered the same thing myself when it comes to funfly models.

My theory said that the curved top airfoil WOULD roll faster or have more lift with flap coupling and your findings seem to confirm this.

There's actually 2 possible (probably 10 but I can only think of 2 )causes for this.

First theory says the flat rear portion is too abrupt a change just aft of the high point and it's creating it's own turbulent bubble wake of sizable proportions. This wake could easily be masking the first 5 or 7 degrees or aileron movement. Not totally mind you. I gather it still obeys small deflections but the response could be very soft around neutral thanks to the turbulence. Along with this there would be fairly high drag. Not a big deal on a fun fly I grant you but if the heavy turbulent boundary idea is right you probably also notice that the model with the flat rear section doesn't speed up in a dive as quickly and perhaps it acts like it has a built in exponential response compared to the NACA model.

For the second possible cause lets look at what the "upper" or low pressure side of the airfoil looks like with the aileron delected. This would be the top surface for positively deflected aileron and bottom surface for negatively deflected aileron. With the aileron deflected to semi serious angles there would be a sharp point at the hinge line of the flat rear airfoil with 2 flat segments and the sharp hinge line. Hardly a good surface for air to flow over. The NACA with has a fuller section curve before the hinge line which will help direct the air onto the upper surface of the aileron reducing drag and directing the air into a stronger "down" wash or flow line (the correct term escapes me at the moment) resulting in greater lift. Even when using smaller movements there may be more air passing closer to the aileron surface with the NACA wing as there's likely a smaller separation bubble, if any, and less turbulence around the aileron. This will show itself as a stronger response to a given deflection compared to the flat rear wing.

BUT this is a fun fly model. Drag can actually be good when we aren't worried about speed. And with the large control throws they use a little softness around neutral can be a good thing when trying to place the model precisely for some tasks.

Which one do you like the best for feel and why? I'm actually quite curious about this now. I've been thinking of a fun fly model and my built in desire to optimize everything leads me to use the NACA type but I can see the other side of the coin too.
Old 10-16-2002, 09:33 PM
  #4  
Ollie
Senior Member
 
Join Date: Dec 2001
Location: Punta Gorda, FL
Posts: 958
Likes: 0
Received 0 Likes on 0 Posts
Default airfoil difference guestion?

Here's another contributing factor. The flattened airfoil produces more drag and therefore the plane flies slower for a given power. The lower the airspeed, the less the aileron response. This effect is inversely proportional to the square of the airspeed so a small difference in speed produces a large differense in response.
Old 10-17-2002, 06:30 AM
  #5  
edge 3644
Senior Member
Thread Starter
 
Join Date: Apr 2002
Location: Gabs,
Posts: 149
Likes: 0
Received 0 Likes on 0 Posts
Default airfoil difference guestion?

Thanks guys for your responses.
The fact is that flattened NACA provides softer center aileron control. What is also interesting is that with this airfoil the increase in aileron deflection from 25 to 45 deg has rather minor effect on roll rate. Both models have Hitec 945, 10kg servos to operate separataly each aileron, therefore deflections are rather true even at higher speeds.
Model with modified NACA is slightly slower than other one. What is also unusual is that in very tight loops at high speed the loops become of fairly regular poligon shape instead of circular, behaving like the laminar airflow on the wing was disturbed with kind of regular pace. This is really funny effect, especially that in the other model to achieve this I should really work hard on the elevator.
Otherwise both models behave very similarly.

Robert
Old 10-17-2002, 03:53 PM
  #6  
BMatthews
 
BMatthews's Avatar
 
Join Date: Oct 2002
Location: Chilliwack, BC, CANADA
Posts: 12,422
Likes: 0
Received 22 Likes on 19 Posts
Default airfoil difference guestion?

So I take it that you're recomending the "proper" full NACA airfoil option? It sounds like they are fairly similar but the full NACA has a higher roll rate.

The extra, and desireable, softness around neutral could then be achieved with exponential on the TX.

Sound about right?


As for that funny looping stuff I had a model that looped like that too but the stalls and recoveries didn't happen at the same point in each loop. It just looked like it was galloping around. If your's in a regular and repeatable shape then it must be because you've got JUST the right size loop and speed to match the stall period.

Just tell them you're doing it on purpose. No one needs to know........
Old 10-17-2002, 06:19 PM
  #7  
Oryx
Senior Member
 
Join Date: Jun 2002
Location: Pretoria, SOUTH AFRICA
Posts: 198
Likes: 0
Received 0 Likes on 0 Posts
Default Re: airfoil difference guestion?

Originally posted by edge 3644
Hi,

I have built two almost identical 90 size fun flies. Each has over 7 sq. feet wing area and 6"wide ailerons. In one of them the airfoil
used is 20% Naca and in the other same airfoil, but from main spar to washout I made airfoil flat. First one rolls much faster than second with same aileron diflection. Why?
Thanks for any advice.
Robert
I ran this in XFOIL (I assume you mean flat surface from the main spar to the trailing edge?), at a Reynolds number of 500,000 and the results look very bad for the modified section. As an example, at zero-degrees angle of attack, the normal 20% NACA does quite well. There is a very small laminar separation bubble at about 50% chord that reattaches at about 57% chord. There is not much of a pressure loss across it. After that the flow reattaches and the BL stays quite thin almost all the way to the TE. The CD is 0.00898 for that Reynolds number and angle of attack.

The modified profile with the flat surfaces on the back look much worse. There is a severe separation bubble stretching from 58% to 68% chord with a very bad pressure loss across it. The flow re-attaches in a thick turbulent boundary layer. The CD is a staggering 0.01506 - an almost 70% increase in drag!

I expect the results may look slightly better at higher Reynolds numbers, but you will probably always loose a lot of performance.

Further, I tried deflecting the last 25% of the airfoil to simulate an aileron. It stalls almost immediately on the low pressure side, which explains the low roll rates you were seeing. Also, the thick boundary layer over the aileron also explains the soft aileron control near neutral...

The effect of flattening the surfaces would have been less severe on a thinner airfoil, but on a 20% NACA airfoil, the flat aft section of the airfoil is very detrimental.

Regards,
Bennie
Old 10-17-2002, 07:48 PM
  #8  
Ollie
Senior Member
 
Join Date: Dec 2001
Location: Punta Gorda, FL
Posts: 958
Likes: 0
Received 0 Likes on 0 Posts
Default airfoil difference guestion?

Nice work, Benny! As technically minded modelers see what X-foil can do for them it is bound to become an essential performance prediction and design tool. How much of your time did it take to do this analysis, if I may ask?
Old 10-17-2002, 08:40 PM
  #9  
Oryx
Senior Member
 
Join Date: Jun 2002
Location: Pretoria, SOUTH AFRICA
Posts: 198
Likes: 0
Received 0 Likes on 0 Posts
Default airfoil difference guestion?

Originally posted by Ollie
Nice work, Benny! As technically minded modelers see what X-foil can do for them it is bound to become an essential performance prediction and design tool. How much of your time did it take to do this analysis, if I may ask?
Hi Ollie

It took about 20 minutes or less (most of it to quickly generate the modified NACA profile), but I use XFOIL almost on a daily basis. I have been a member of the UIUC applied aerodynamics research group for the last three years, so as you can imagine I get a lot of practice - it is our main analysis tool when designing new airfoils. However, XFOIL is not very difficult to use. The best way to learn it is just to play with it until you see what it can do for you. I think someone that doesn't know what "e^n" means and the significance of ncrit will still be fine just running it without messing too much with the more subtle parameters. His answers may not match experimental data exactly, but they are usually quite good for relative performance and it is much better than no data at all. If the test data is available it is of course always better to use experimental data.

Another thing that may be a little discouraging is that there are cases where it takes some time to get a problem to converge. Also, some published airfoil coordinates give very wavy velocity distributions and you need to do a little smoothing to get the inviscid solution acceptable before going to viscous analysis. There may be a bit of a learning curve but it is not outside the capabilities of any technically-minded modeler with some background in aerodynamics.

Interpretation of the data may need some more experience depending on what you are interested in. Lift, drag and pitching moment coefficients are of course straightforward enough but I had to look at the actual chordwise data to deduce from the Cf values where separation and reattachment occured. XFOIL does of course tell you where transition occur and it comes with a nice plotting program that you can use to make up a complete polar after running at a range of angles of attack.

Best regards,
Bennie

Ps. I think writing this post took longer than doing the analysis in the previous one...

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright © 2022 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.