Load testing for unknown wing.

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,109
Location
NJ
Yeah building two sets of identical wings and testing one to destruction is the only way to know for sure what you have. Big commitment.
if you are going to sell plans, kits, or completed aircraft, testing to ultimate or destruction is would be a very good idea.....but for a one-off personal airplane I only am testing to limit.
 

Lendo

Well-Known Member
Joined
Feb 6, 2013
Messages
715
Location
Brisbane
propastie, where do you think you got the 7725 and the 7715 Glass weaves from - their a copy of European weaves developed in Europe and Burt Rutan had them copied. I'm led to believe.
George
 

Lendo

Well-Known Member
Joined
Feb 6, 2013
Messages
715
Location
Brisbane
Atypicalguy, rereading some of the posts, I wouldn't test to failure, but I would design and build a little over your SF*G load. If wanting Aerobatic 6G it's a BIG load for the wing to carry - the numbers add up really quickly. My suggestion if your wanting 4.4 G design for a little over, maybe as much as up to 5 and test to 4.4. If building in Composites use a min of SF 2. Mind you I have seen some composites build that should have had a SF of 3. Still they flew.
George
 

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
I have done a fair bit of composite work, but this plane is wood. Wing is wood truss ribs with plywood skins, single strut. Problem is a bit of increased span and increase in gross weight from 1450 to 1750 along with a large speed increase from perhaps 200 to 270mph on the same spars. On the design side I figured it should pencil out to over 6G. The spar definitely does not. So then I am left to analyze the contribution of +/-45 ply skin. Plate buckling etc. 3/32 ply wing skin actually contributes more to the second moment of inertia than the spar for spanwise bending. Anyway never thought I would be hanging my hat on buckling analysis of 3/32 ply for wing integrity at 275mph, but life is what happens while you are making other plans.
 

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
Atypicalguy, rereading some of the posts, I wouldn't test to failure, but I would design and build a little over your SF*G load. If wanting Aerobatic 6G it's a BIG load for the wing to carry - the numbers add up really quickly. My suggestion if your wanting 4.4 G design for a little over, maybe as much as up to 5 and test to 4.4. If building in Composites use a min of SF 2. Mind you I have seen some composites build that should have had a SF of 3. Still they flew.
George
The wings are already built.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,109
Location
NJ
here is a trick I use to get the Moment of Inertia....let the Cad calculate it for you.

1622677586142.png
 

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
Or just get it from XFOIL "BEND"

.GDES c>

XFOIL c> bend

Area = 178.19264
Slen = 97.633995

X-bending parameters:
centroid Xc = 20.680401
max X-Xc = 27.319599
min X-Xc = -20.680401
solid Iyy = 22250.422
skin Iyy/t = 20108.883
solid Iyy/(X-Xc)= 814.44904
skin Iyy/t(X-Xc)= 736.06067

Y-bending parameters:
centroid Yc = 1.0320668
max Y-Yc = 2.9893732
min Y-Yc = -2.4677467
solid Ixx = 316.08572
skin Ixx/t = 406.86563
solid Ixx/(Y-Yc)= 105.73646
skin Ixx/t(Y-Yc)= 136.10399

solid J = 1210.0881
skin J/t = 1300.8837
 
Last edited:

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
Abbots has some nice spread sheets
I think modeling the skin as a thin cylinder is problematic. The part of the skin contributing most to bending resistance is the thickest part of the wing, because it is furthest from the common centroid. That part is pretty flat. All those cylinder calculations get pretty wonky if the shell is not a perfect cylinder. So I will just model it as a bunch of small slices.

Anyway the critical sigma is pi^2KE/12(1-0.09)*(t/b)^2
b (loaded edge length) is a bit arbitrary,as some of it is in front of the spar (18"), while some of it is behind the spar (21") and some of it is stuck down hard to the spar caps and truss ribs going over the spar (which will not buckle at all).
In any event,
K (buckling constant which varies with aspect ratio of panel) is >14, possibly much greater, as the aspect ratio is very small with a free span along the load direction of 5.5" and a panel width along the chord of 20" (b-if we take the part that contributes the most to the second moment of inertia).
E is 557k
t is 0.102
b is maybe the middle 20" of the chord of the top skin
So you get something on the order of 150psi for critical buckling stress.
Taking the 20" x t is 2.04sq in, times the 150 is about 300 pounds before it buckles
Plus the little bit of skin that stays in axial compression atop the spar itself; that acts sort of like a mini spar cap. The caps are interruped by the ribs, so not contributing much except to space the skin out away from the spar 0.25" and hold the skin in column.
Anyway I will have to put this stuff into the spreadsheet with the spar calcs and see how far it goes loadwise before it hits 150psi in the top skin.
After that it deflects, but still contributes because deflection is less than elastic limit.

Almost this is very similar to what the OP should do for their wing. I will say that 1/8 ply should resist buckling well but that it depends quite a lot in the rib spacing, which defines the panel size. Also, start with the spar calcs because if the spar holds your target load then you can be pretty confident that the skin will add quite a bit.
 
Last edited:

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
That makes sense. But in order to know what limit load is, it makes sense to figure out what load the design scantlings suggest it should support at ultimate. I find it hard to believe that the limit load is 10G, because that would put ultimate at 15G. And no one can really even tolerate 10g to begin with. I am guessing the OP wing pencils out to 10G ultimate.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,109
Location
NJ
sorry but it would be way too much work and time for a poser as I to understand and check your calculations

I think modeling the skin as a thin cylinder is problematic.
I calculated the curved panel buckling between each rib and the spar on the LE. My aft skin is fabric but if it were load bearing that would be another panel calculation. I had to add z stiffeners most of the wing to meet limit....past limit I expect wrinkles in the skin.


start with the spar calcs
My spar in the up down was fine but not so fine in fore aft....when loaded at the 15 deg. angle my compression cap failed at a location inboard the LE .....see build log. Bad design on my part ...that you have an existing design except proof of higher strength.....hopefully you will not have dumb ignorant bad design...as I had.
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,109
Location
NJ
You will see skin Buckling before failure.
my failures mostly Euler buckling failures were catastrophic. The only buckling I saw was 2ud from last wing failure when at 40% I had oil canning and wrinkling ......I added z stiffeners to the LE but the wing failed at 50% the next load test....perhaps the oil canning and wrinkling was besides weak panel buckling strength of the skin also because of stress on the spar that I did not see......One of my problems is that I I can not accurately measure deflection on my rubber tired tractor.
 

atypicalguy

Well-Known Member
HBA Supporter
Joined
Jun 8, 2020
Messages
79
sorry but it would be way too much work and time for a poser as I to understand and check your calculations
No worries. I would not expect that. I am also a poser. I just thought people with actual experience might have done wing skin buckling calcs before. Seems like that is true, which is cool.
 
Top