Thermal Wing Design, Input Appreciated - Page 3 - RC Groups
Thread Tools
Mar 05, 2010, 01:48 PM
Registered User
miniphase's Avatar
Quote:
Originally Posted by wardo78
I really like your flap idea
Norm's saying it'll pitch up, I had a wing where the pitch change from deploying flaps was uncontrollable, so it kinda defeated the idea.

You maybe need to get someone to crunch the numbers (not my speciality) on best flap size/location
Sign up now
to remove ads between posts
Mar 05, 2010, 01:48 PM
Registered User
wardo78's Avatar
OK, it is 40 degrees out and KAVU. I am going to fly the 84 for a bit. Have a great weekend Norm!

Wardo
Mar 05, 2010, 01:50 PM
Red Merle ALES VI
Curtis Suter's Avatar
Quote:
Originally Posted by wardo78
Curtis,
So a few questions on your post. Firstly, how far out should the transition from semi-symmetric to symmetric occur? Are the winglets also symmetric? How do I select an airfoil? I am likely to simply do what other, more informed designers have done before me, but I would like to have some sense of the decision process and the results applicability to my specific application.

Thanks,
Wardo
The Panknin formula is designed for two airfoils that have a linear transition from root to tip. Anything else and you're in new territory!

Lots have been written about tiplets, way more than I know. The TLAR version works best. However, assuming you're able to get the proper airfoil, area, sweep, tow in/out etc.... an airfoil tiplet might be best. But I'd just go with a sanded flat plate, that's what I used.

To select an airfoil you'll need an airfoil that has good thermalling behavior for the center i.e. lifting section and either an airfoil that's fully symmetrical or slightly reflexed for the outer section to counteract the negative pitching moment of the inner airfoil.

So I did what everyone else was doing. I used the SD7037 as it's a proven airfoil at the higher Reynolds number than what you see in the TD/F3J airplanes now and the SD8020 for the outer section. The 8020 is fully symmectrical.

If your overall flying weight was 50 ounces or less I'd tend to think a Drela airfoil would be better, but at the higher weight you'll be flying at the 7037 is a fine airfoil.

These are just my opinions and what worked for me. Tinamou sure flies nicely. However, too slow and she will tip stall, maybe I'll work on that this summer.

Curtis
Montana
Mar 05, 2010, 01:51 PM
Red Merle ALES VI
Curtis Suter's Avatar
A keel also helps in keeping the flaps from dragging the ground on landing!

Curtis
Mar 05, 2010, 03:22 PM
Registered User
wardo78's Avatar
I just came in from a fantastic flight. The wave behind my house was going off. The wind is blowing (25kts@3k') consistently out of the NW, which hits the ridge behind my house and makes a very reliable wave. The ridge was the site of the 1934 Vermont State Ski Championship. There are still parts of a Model T motor up top, it was used to power a rope tow. I've noticed that if you get below the wave, there is major sink, but if you stay on the bubble it just goes and goes. It is pretty high up, maybe 1000' agl, and over dense forest. I specked out in a few minutes after my first climb out. There was also some thermal activity about. I ranged out and found some nice lift. The air was choppy and the wing would occasionally get serious buffeting. I did two powered climbs in 45 minutes, after which my hands weren't working up to par. Here is a Google Earth image of the site. My house is in the foreground. I stand at the curve in the driveway and fly from there.

(Picture deleted by author)

In the winter this is a nice option, but it is seasonal. Hence the desire for a TD wing.

Regards,
Wardo
Last edited by wardo78; Mar 14, 2010 at 08:48 PM.
Mar 05, 2010, 05:25 PM
Registered User
wardo78's Avatar
I wont finalize things for a bit as I want an opportunity for as many people to weigh in as possible. At this point if I were to send the order in for cores, here is how it would look. Questions are collocated with the relevant data:

- Length of wing panel (half of the wingspan)
Panel 1: 10"
Panel 2: 50"

Will this give me a true 120" wingspan, or is this the diagonal length?

- Root airfoil name
Panel 1: SD7037
Panel 2: SD7037

I am not sure what the machines capabilities are as far as a transition is concerned. Can the transition happen immediately on panel 2?

- Tip Airfoil name
Panel 1: SD7037
Panel 2: SD8020

- Root Chord Length
Panel 1: 18"
Panel 2: 12"

- Tip Chord Length
Panel 1: 12"
Panel 2: 6"

- Setback of tip airfoil (amount of sweep back at leading edge)
24 degrees both panels

- Washin or washout (both root and tip if applicable)
minus 7 degrees at tip (washout)

I am still confused about what number to give the cutter for geometric twist, mostly because I am not sure I input the correct data. This is what I think is called for. Is this correct?

- Dihedral specification
No dihedral

- Trailing edge thickness
Fully tapered to zero

-Additional cuts for wing spars, wing tubes, etc.
I will do these

-Type of foam
(1# EPS, 1.5# EPS, 2# EPS, 2# Blue, 1.3# EPP, 1.9# EPP, or Spyder)
I want reasonably dense non-beaded foam. Can you quote both Spyder and Pink?

What is the best choice for this application?

As I said, by no means am I set on things. But, having heard no major objections from the esteemed readers, I am thinking I am probably not too far off here.

I looked at my 84" wing, and I don't think it would be difficult to temporarily rig up spoilers on the top with one servo. If I get the time I might do some testing on this front.


Regards,
Wardo
Last edited by wardo78; Mar 05, 2010 at 10:38 PM.
Mar 05, 2010, 06:50 PM
Registered User
wardo78's Avatar
A quick perusal shows that Hobby Lobby has a 5mm coupler, here's a link:
http://www.hobby-lobby.com/5mm_5mm_s...r_546_prd1.htm

Boca Bearings (http://www.bocabearings.com/) has a WIDE selection of 5mm ID bearings.

I don't have a machined 5mm shaft, and it might be required for the sort of runout tolerances allowable at high RPM. I do have what "appears" to be a perfectly straight 5mm shaft. Once I have everything in hand I'll likely build a test bed before thinking of putting it in the air. Better to test to destruction on the ground then aloft.

Wardo
Mar 05, 2010, 06:58 PM
Registered User
wardo78's Avatar
Quote:
Originally Posted by cliffo
http://www.youtube.com/watch?v=11BzyS5CcMs

Drela foils with lots of sweep.

Find a destroyed Supra, for some cheap wings, and give it 45 degress of sweep and winglets. Woo hoo!
That is a wing of beauty!

Wardo
Mar 05, 2010, 08:32 PM
Herk
HerkS's Avatar
You seem to be asking a question about twist – you wonder - have you got it figured out correctly.

This is kind of detailed because your question is pretty general. If this comes across as in any way patronizing, please forgive me, I just wanted to cover what I’m guessing is on your mind.

Design twist depends on several things that we take into account, and some that we usually don’t.

Once the wing layout is defined, then the twist you need depends mostly on the zero lift angles and pitching moment coefficients of the airfoils you’ve chosen, the static stability margin that you intend to fly with, and the flying speed that you have chosen as your design point.

With that information you can use analytical tools to come up with an approximation of the geometric twist you will want cut into your wing cores. You have used Curtis’ spreadsheet – that should work for you pretty well.

You have tentatively chosen the SD 7037 and SD 8000 airfoils to work with . The SD 7037 has a zero lift angle of about minus two degrees and a zero lift pitching moment coefficient of minus 0.08. (You will find different values in different sources. These are from Michael Selig’s wind tunnel tests.) The SD 8000 is symmetrical, so it has a zero lift angle of zero and a pitching moment coefficient of zero. Having chosen those two airfoils for your model you have aerodynamic twist (washout) of about two degrees.

Positive static margin means that the CG is ahead of the neutral point, so that creates a nose down moment. You need twist to counter that. Also the blended airfoil has camber and that creates nose down pitching moment too. You need more twist to counter that.

Once you have figured out the design geometric washout that you must add to the inherent aerodynamic twist, you need to recognize that it’s good for only one flying speed. When you fly slower than the speed that the plane is optimized for, you need up elevator to compensate, and when you fly faster than the design speed you need down elevator to hold that speed.

A good way to look at this is to see that the twist you get cut into your cores is just an attempt to get the plane to be most efficient at the speed where you expect to want the best performance.
Mar 05, 2010, 10:31 PM
Registered User
wardo78's Avatar
Quote:
Originally Posted by HerkS
You seem to be asking a question about twist Ė you wonder - have you got it figured out correctly.

This is kind of detailed because your question is pretty general. If this comes across as in any way patronizing, please forgive me, I just wanted to cover what Iím guessing is on your mind.

Design twist depends on several things that we take into account, and some that we usually donít.

Once the wing layout is defined, then the twist you need depends mostly on the zero lift angles and pitching moment coefficients of the airfoils youíve chosen, the static stability margin that you intend to fly with, and the flying speed that you have chosen as your design point.

With that information you can use analytical tools to come up with an approximation of the geometric twist you will want cut into your wing cores. You have used Curtisí spreadsheet Ė that should work for you pretty well.

You have tentatively chosen the SD 7037 and SD 8000 airfoils to work with . The SD 7037 has a zero lift angle of about minus two degrees and a zero lift pitching moment coefficient of minus 0.08. (You will find different values in different sources. These are from Michael Seligís wind tunnel tests.) The SD 8000 is symmetrical, so it has a zero lift angle of zero and a pitching moment coefficient of zero. Having chosen those two airfoils for your model you have aerodynamic twist (washout) of about two degrees.

Positive static margin means that the CG is ahead of the neutral point, so that creates a nose down moment. You need twist to counter that. Also the blended airfoil has camber and that creates nose down pitching moment too. You need more twist to counter that.

Once you have figured out the design geometric washout that you must add to the inherent aerodynamic twist, you need to recognize that itís good for only one flying speed. When you fly slower than the speed that the plane is optimized for, you need up elevator to compensate, and when you fly faster than the design speed you need down elevator to hold that speed.

A good way to look at this is to see that the twist you get cut into your cores is just an attempt to get the plane to be most efficient at the speed where you expect to want the best performance.

HerkS,
That was very helpful. I hope I haven't come off as offended in any way by everyone's help. The internet is a finicky communication mode. The guidance has been essential to getting my head around the complexities involved. I understand the various elements to a degree, but your explanation of their interrelationships is extremely illuminating. I used those numbers, and the results are in line with what I expected from previous discussions I have read. I am not familiar enough with the impact of varying the static margin, so I stuck with 7.5% which was the default setting.The numbers look like:

Reynolds Number
Speed 25 mph
Result 195000

Airfoil Data
Root Zero Lift Angle -2.000
Root Cmo -0.080
Tip Zero Lift Angle 0.000
Tip Cmo 0.000
CL Manually Calculated 0.60
CL Automatic Value 0.600
Static Margin 7.50 %
Aerodynamic Twist -8.74 degrees


REQUIRED WING TWIST
Geometric Twist -6.74 degrees
Geometric Twist 0.71 inches

I understand that at this twist the wing is optimized for one speed. that is exactly what I am looking for, the best thermaling wing I can achieve.

Thanks,
Wardo
Mar 06, 2010, 07:13 AM
Registered User
wardo78's Avatar
I woke up thinking about prop orientation and the center section design. I calculated the prop size last night using the specs from the inrunner/battery combo I have in mind. The best results were found with a 13x10 prop. The power wasn't awe inspiring, but it was sufficient. Certainly anything with the sort of climb I would like is easily going to overspeed the airframe, so I'll need to learn to be cautious with velocity.Taking HerkS previous comments into account, it occurs to me that I'll have a whole bunch more spinning prop to avoid if I use a pusher set up, and launch as I currently do my 84".

If I were to mount the prop in a tractor configuration, I could fab a composite tube and motor mount reasonably easily. It would solve the rearward CG concern, and be safer. I am wondering if the 13" prop would need to clear the wing completely. To the naked eye, folders seem to open instantaneously. If it needs to clear completely, and my assumption is it would, then the prop would need to be a LONG way out front. Is my assumption correct? I think I have the issues for a prop shaft extension sorted in my mind, so it is doable. Having said that I continue to prefer the aesthetics of a pusher configuration.

With either a tractor, or pusher prop on a prop extension, I am freed up to play with the center section trailing edge. I like the look of a "batwing", where the center chord is longer and the sweep is reversed for the center section ala Horten designs. Although nothing I build is going to be remotely as beautiful as Lothars work, something akin to this, is the idea:



I am looking at extending the center chord of panel 1 three inches rearward. Other than aesthetic appeal, what are the practical considerations? Are there significant aerodynamic consequences?

Thanks,
Wardo
Mar 06, 2010, 08:55 AM
Red Merle ALES VI
Curtis Suter's Avatar
Hello,

Great thread!

Herk is the master. He's right that you'll find different data for the ZLA for the SD7037. I used -3.34 for Tinamou. I got the data from XFLR5. At this link if you scroll to the bottom you'll see the same value.
http://webspace.webring.com/people/a...l/COOKBOOK.htm

So what would I use? I think it's a mute point as Herk says it's just a design consideration and will only work at that one speed flown anyway.
It only changes the Geo twist from 7.42 to 8.76 degrees or 0.14 inches. I think I'd go with the larger value.

Also, to make sure I'm clear, are you building your wing with the first 10 inches of span all SD7037 then the last 50 inches will change linearly from the SD7037 to the SD8020?
If so that should work fine.

Also, In post #40 I'm getting some different values for Aero and Geo Twist. See attached photos.

What am I missing Wardo?

Curtis
Montana
Mar 06, 2010, 09:21 AM
Registered User
wardo78's Avatar
Curtis,

The difference is attributable to the fact that in post #40, I was still using the data from the center chord being 18". My understanding is that the change to 21" means the twist needs to be increased to counter the change in the static margin location. I'll change the order to reflect this if I use the longer center chord. Any thoughts on how the longer chord would effect the flying characteristics?

"Also, to make sure I'm clear, are you building your wing with the first 10 inches of span all SD7037 then the last 50 inches will change linearly from the SD7037 to the SD8020?"
Exactly what I am thinking. I assume the core can be cut as such.

Today is feeling like the first day of Spring, it is forecast to be 45 out, with the sun shining. I hope to get some flying in.

Thanks,
Wardo
Mar 06, 2010, 09:26 AM
Red Merle ALES VI
Curtis Suter's Avatar
Wardo asked:

Any thoughts on how the longer chord would effect the flying characteristics?
<<I really don't know, I'll leave that one to the experts.>>

"Also, to make sure I'm clear, are you building your wing with the first 10 inches of span all SD7037 then the last 50 inches will change linearly from the SD7037 to the SD8020?"
Exactly what I am thinking. I assume the core can be cut as such.
<<Easily. OH, and yes what you are planning will give you the 120" span, I forgot to comment on that.>>

Today is feeling like the first day of Spring, it is forecast to be 45 out, with the sun shining. I hope to get some flying in.
<<It's been cracking 50 degrees in Montana, still pretty muddy from the snow melt>>

Curtis
Mar 06, 2010, 09:51 AM
Herk
HerkS's Avatar
Lothar's thread has been fascinating. Awesome model, but really heavy - 3.7 kg - 8+pounds.

He uses bungee launch and motor start after release. I have used that for several planes too. In my experience it works really well and is a good idea - particularly for a large plane - regardless of whether it's pusher or tractor.

Hand launching of a large flying wing can be tricky. The long wing has high yaw inertia - so if you get it yawing when you toss it, you need a lot of aerodynamic damping to stop the yaw and get it flying straight. The short coupled verticals sometimes just aren't effective enough to provide that. The bungee pulls it straight and provides excess flying speed for a safe launch.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Edge wing design published G_T Hand Launch 2 Mar 03, 2010 12:13 PM
Discussion Planning stage, pby, input appreciated Behr Scale Kit/Scratch Built 8 Oct 08, 2009 06:12 PM
Discussion Input to new winter project appreciated! nick taylor Scale Fuel Planes 1 Oct 23, 2007 09:16 AM
Help! Quick Question, Any Input Would Be Appreciated! Minimole Mini Helis 8 Mar 31, 2006 07:34 PM