Cantilever Internal Braced Fabric Covered Wing

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,489
Location
Saline Michigan
Regarding the structure you posted, the diagonal wire bracing is for drag/anti-drag, and does nothing for preventing the wing cell twisting in the vertical plane, i.e: chord-wise.
And if you over tighten the diagonal wires, the wing will snap to a shape with compression ribs alternating between nose up and nose down as you go along the wing. Not making this up, folks have applied too much torque to the system and "pop" the wing.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,865
Location
Memphis, TN
Really? Sienfeld couldn’t say it strong enough. Why show the flying wires when you ask what’s inside? If you don’t have a skin or geodesic straps, you don’t get torsional control. Internal Starduster picture. 6D2083C5-10AF-4E51-A5FD-9E001E520515.jpeg
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,343
Location
NJ
shown....drag anti drag doubled up and from top to bottom of front to rear spar.......double set of compression members top and bottom .....thinking it might be putting unwanted out of plane forces on the compression members but figured would show it anyway as regards drag anti drag wires with anti torsion features

hbawire.png
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,489
Location
Saline Michigan
So, you can see how much torsional stiffness the wing has. Start with torsional stiffness of the spars, then compute axial stiffnesses of the elements, put in a small twist, compute the changes in length, then the torque on the system. The torque divided by the twist is torsional stiffness from the new elements, and you can compare it to the torsional stiffness of the spars themself. And you can also see how much strain and stress and load is in the new elements to see if they are acceptable at your actual torques. Then once you have iterated your design to within acceptable levels, you can see if you added more weight than the more conventional schemes, like a structural D-tube or full skin or D-tube plus Warren truss ribs....

Have fun.

Billski
 

ragflyer

Well-Known Member
Joined
Apr 17, 2007
Messages
356
In my experience calculating deflections reliably are complicated except in very simple cases. Your two practical options are FEA or testing a scale model. The former generally is not recommended for amateurs but the latter may often be a good choice.
 
Last edited:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,865
Location
Memphis, TN
Prop’s thread on testing is brilliant. He knows how to do that. He found out though that cutting everything close is hard to do.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,343
Location
NJ
put in a small twist
calculating deflections
for wing at limit .....the loads are pretty straight forward....a tilt of 15 Degree is recommended for a load of 25% in the forward direction......one could also measure twist at max aileron load of 9 PSF for my speed. ....but it is going to twist no matter what and how much is too much seems to be my problem.....

The torque divided by the twist is torsional stiffness
have I missed this spec. somewhere in all the design guides (part 23, Glider Criteria, etc.)

edit: apparently not .....at least with this search
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,343
Location
NJ
Well, you know when you are one the edge when you give right aileron, and the plane rolls left...
I could grab the tip and crank in a couple degrees of washout or wash-in with relatively little effort,
I distinctly remember large opposite aileron deflection required to maintain the bank in the 1-26.....is this what you are talking about and is it normal in other gliders?
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,489
Location
Saline Michigan
have I missed this spec. somewhere in all the design guides (part 23, Glider Criteria, etc.)
Spec for torsional stiffness? Not usually specified directly. There are several things to be taken into account:

First piece is the wing needs to have a factor of safety of 1.5 (for wood and metal and metal cables) or 2.0 (for synthetics like Dyneema) while under limit loads which includes carrying pitching moment as well as shear and bending moment. With a structural wing skin, this is usually no big deal - the a structural wing skin that is robust to build and will not cripple contributes mightily to torsional strength. In a fabric covered wing, we usually have a pair of lift struts plus root fittings that are usually more than strong enough once they will not buckle under limit negative g;

Second piece is adequate wing stiffness in torsion to avoid aileron reversal due to wing twist and then to give good aileron response. Work the aileron loads into the wing and show that in total the wing's deformation is small enough to get rolling moments in the right direction and of adequate rate to fly the bird comfortably.

Third piece is the wing needs to have enough stiffness-to-mass ratio to not flutter in any of the full wing modes with bending and torsion usually being the critical modes - 1st bending mode is tip flapping up and down, 2nd bending mode is tip up and midspan down, then reversing, 1st torsion is tip rotating leading edge above and below the rest angle, 2nd torsion mode is tip rotating one way, mid span rotating the other way. Generally, if you get the first couple modes above your Vdive, the rest are ay high and of no big worry, unless one is an integer multiple or integer fraction of some other input. These must all be precluded within the the intended flight envelope. Vdive is what we are protecting for. Preventing these modes takes high enough stiffness to mass ratio in each of bending and torsional modes. Then if either mode lands on its resonance with another mode, things can get really interesting. Usually with structural skins or two lift struts in our little airplanes, this is not much of an issue. Go with a soft wing structure that you are beefing up until just adequate in torsion, and it can design your systems.

This second and third ones are why the faster an airplane is, the stouter its wings and tailplanes tend to be. The higher your Vdive, the stiffer everything has to be to drive flutter above Vdive.

Now flutter prevention is a requirement. The faster your Vdive, the more you have to learn about prediction of stiffnesses and vibe modes. FEA with an Eigen solver is really helpful here as it will predict the frequencies of as many modes as you want, and is usually pretty accurate at it. As to minimum frequency you need to protect up to, well, you have a bunch of learning on flutter ahead. Have fun.

Billski
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
15,355
Location
Port Townsend WA
I believe the old FAA glider standard does give a stiffness criteria that relates the twist/deflection to 200/Vd^2.
Right. FAA Basic Glider Criteria page 79.
How would this torsion criteria, in general, compare with the comment from Boku in post 16 or other ultralight gliders?
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,343
Location
NJ
When I do the calculation for 61mph Vd...I get .053 which I am going to have to figure what that means exactly.
 

ragflyer

Well-Known Member
Joined
Apr 17, 2007
Messages
356
When I do the calculation for 61mph Vd...I get .053 which I am going to have to figure what that means exactly.
The glider manual explains this. You apply a torque (any reasonable amount that results in measurable twist) close to the tip of the wing. You then measure the unit twist (radians per ft lbs) at multiple points (say 4 points) across the span where the ailerons are. Unit twist is the twist you measure in radians divided by the torque you applied. You then plot a curve - x axis is aileron span and y axis is UnitTwist*chord^2. You then measure the area under this curve. If that area is less than 0.053 you meet the criteria. Usually for 61mph Vd airplanes, assuming conventional structures, you should have no problem with flutter, but with as with all things flutter you can never be sure.
 

cluttonfred

Well-Known Member
HBA Supporter
Joined
Feb 13, 2010
Messages
8,699
Location
World traveler
I suspect one of the most efficient and amateur-accessible ways to make a cantilever, internally-braced, fabric-covered wing is the geodetic approach used in the Vickers Wellesley and Wellington bombers and also the simpler wooden Thalman/Yates/Greenwood etc. series of homebuilts.

Welleslystructure338.jpg thalman t-4 cutaway.jpg

 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,343
Location
NJ
Usually for 61mph Vd airplanes, assuming conventional structures, you should have no problem with flutter,
I had not been concerning myself with flutter for that very reason......

not being experienced at all as regards the torsional stiffness as regards control stability....do you have a feeling as to if meeting this specification will make the glider stiff enough for neutral stability in roll?
 
Top