Have I got my spar bending / shear diagrams right?

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Foundationer, Jun 20, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jun 29, 2019 #21

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    9,118
    Likes Received:
    5,925
    Location:
    97FL, Florida, USA
    The profile of the lift curve across the fuselage is something that I have had questions about for years. As Bill points out, different fuselage and fuselage/wing geometries certainly must have significant influences.

    Other than the approximations in texts, are there any definitive NACA or other reports that provide measured data?

    Thanks,


    BJC
     
    wsimpso1 and mcrae0104 like this.
  2. Jun 29, 2019 #22

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    11,547
    Likes Received:
    2,157
    Location:
    Port Townsend WA
    The Facetmobile is both "all wing" and "wingless". (according to the designer. And he can correct me if I got that wrong:))
     
    Last edited: Jun 30, 2019
  3. Jun 30, 2019 #23

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    289
    Likes Received:
    45
    Location:
    Brisbane
    Billski,
    I would have to agree with all those points, however with all the variables that could/would be encountered, I think it's better (for me anyway) to accept that conservative estimate. I can't help but feel this may also include any losses at the fuselage to wing interface, although this wasn't expressly stated.
    George
     
  4. Jun 30, 2019 #24

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    289
    Likes Received:
    45
    Location:
    Brisbane
    Billski,
    At what stage of the calculations can I exclude the wing weight as the wing is supposed/ reported to support it's own weight ( at least) in flight.
    When I look at HT down load and Wing weight, they do appear to balance-out (on some designs).
    George
     
  5. Jul 8, 2019 #25

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    289
    Likes Received:
    45
    Location:
    Brisbane
    BJC,
    There is nothing in any of the Texts I've got other than what I've given by Sadraey. I like the way he writes as he was Australian educated, now works in the USA. Considering the diverse shapes and sizes, it's nice to have that very conservative guide.
    George
     
    BJC likes this.
  6. Jul 9, 2019 #26

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,673
    Likes Received:
    2,923
    Location:
    Saline Michigan
    David Lednicer was my mentor on this. He was a student of Professor Lesher, was an aerodynamicist for Sikorsky, with John Roncz' company, then AMI West. He did a series of articles in Sport Aviation in the early 1990's with some of the plots published there, and has shown me a bunch of others too. The lift carries across the fuselage, usually a little lower pressures but spreading fore and aft a bit too. Doing eyeball summations of P*dA, they pretty much look like the fuselage carries the same lift (or close to it) that the wing would have carried if it were there and no fuselage were present... You might get conservative and multiply the standard lift by 90% and then iterate the pressure distribution height with a multiplier until the total equals your airplane times g's plus your tail download.

    Love that photo. It makes the lift look pretty much elliptical over the entire span on a low aspect ratio highly tapered and swept wing... That was one of the other things Dave used to point out to me: The spanwise lift distribution almost always ends up looking elliptical.

    Have fun guys.

    Bill
     
  7. Jul 9, 2019 #27

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    9,118
    Likes Received:
    5,925
    Location:
    97FL, Florida, USA
    I met Ed Lesher once, back in the 1960’s. Nice guy.


    BJC
     
    wsimpso1 likes this.
  8. Jul 9, 2019 #28

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,673
    Likes Received:
    2,923
    Location:
    Saline Michigan
    Let's look at the big picture. The wing carries a bunch of loads you can put on your beam diagram. Let's just assume a one-piece wing, elliptical lift distribution, attached at the centerline to a fuselage... Yeah, not practical, but you can get real world after we get all the other stuff figured out.

    The wing lifts. The fuselage and everything in it has weight (pounds, Newtons) and is operating at some positive g. There is also a tail out there, and we will just assume its weight is thrown in there with the fuselage, it is of conventional design with the tail in back pulling downward to null out the pitching moments. What else is going on?

    Well, the weight of the wing and everything in it is also out there... That weight times the g's also has to be carried by the wing. But since the weight of the wing is distributed along the span, its contribution to things like bending moment and shear in the spars is usually pretty small. Then comes the fuel. I do not know about you guys, but my fuel weighs more than my wing, and the fuel is not spread along the entire span, just from BL54 to BL120, so I gotta put fuel in there too. Since the fuel is in there, might as well do the wing weight too.

    Easy way is to setup the wing in an Excel spreadsheet. First column is BL dimension, running from the centerline all the way out to the tip. Then you have columns for fuselage weight*g, tail download, wing weight per station*g, fuel weight per station*g, drop tanks*g, ordinance*g, engines*g, landing gear*g, etc. Each must be thought about.

    Fuselage weight - In our little airplanes, this is the fuselage, cowling, engine, people, bags, tails, all landing gear parts attached to the fuselage and everything in them. Yeah, you can scare the canaries into flight in their cages, but unless they leave the airplane, their weight times g is still in the fuselage... This weight pulls down on the station where wing is attached to the fuselage and you gotta multiply it by the g's you are pulling at limit;

    Tail download - most of our little airplanes have a horizontal tail pulling down in order to null out pitching moments. The total download from the tail is added in here. This is usually a function of weight, CG position, and g. Coming up with download as a function is helpful here;

    Wing weight - If your wing is straight and so is your spar, you just divide your wing weight by the number of spanwise stations and that is your station weights for the wing. Most of us taper at least one thing and sometimes several. Tapered wings will have less weight per foot of span as you go outward because there is less skin. Control surfaces will also taper as you go outward. The flaps my be sturdier than the ailerons and thus weigh more per foot too. And the spar - any self respecting engineer working on a cantilever wing will be tailoring the shear web and spar caps as you go outward. So, you will have weight per foot or weight per inch changing as you go from root to tip. Wing weight times g goes in the table;

    Fuel weight - Hmm. If fuel is all carried in the fuselage it goes up there in the fuselage. If it goes in the wings, well, you gotta distribute fuel along the span per your fuel tank arrangement. Remember it is weight times g's;

    Drop tanks and ordinance and engines on the wings, let's skip those...;

    Landing gear - if you hang your gear and latch it up with hardware on the wings, well, those stations get weight times g's; If some of it is hung on the wing, but some is on the fuselage, spread it accordingly;

    All of the weights and downloads contribute to your shear diagram in a downward direction. And lift contributes upward. Now you have your shear diagram. Check that the numbers at the fuselage walls make sense. Then you can do a numerical integration and get bending moment. If you want to see if you can skip one thing or another or if your tail load offsets your wing weights, you can do it.

    I just put up an example. If you do the whole thing, distributing weights that are on the wing, you need a certain moment. If you simplify by skipping fuel and gear, your laziness will drive you to build more weight into the bird. For the case I pulled out of the air, I got an 18% overbuild on bending moment and 23% overbuild on shear load. That can matter.

    When do you do it? on every iteration. As you refine the weights of your spars, skins, internal structures, controls, control surfaces, and fuel volume, gear, and so on, you update the numbers and iterate things like spars as needed.

    Welcome to the Monkey House.

    Billski
     

    Attached Files:

    Vigilant1 likes this.
  9. Jul 9, 2019 #29

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,673
    Likes Received:
    2,923
    Location:
    Saline Michigan
    I did an approximation of my bird with the wing attached to the fuselage wall, and just putting the gross weight on the wing results in a 43% overbuild in the spar caps. That can be substantial weight.

    Seriously, do your beam diagram, include the weights of stuff in the wings, it will reduce your spar shear and bending loads, and allow a lighter bird. Drive weights down by closely calculating and tracking the real loads on stuff, apply appropriate FOS, design the structures, get good estimates of the real weights, put it all together, and iterate the design until the weight stops changing. If it is not light enough, time to consider different design schemes, materials, etc.

    Billski
     
  10. Jul 10, 2019 #30

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    289
    Likes Received:
    45
    Location:
    Brisbane
    Bill,
    I had a look at a mock-up of the Blackshape prime Light Sport Aircraft at Friedrichshafen Germany Air Show and the Spar (both main and rear) looked an extremely light C channel construction. It was tapered in Vertical width (thickness), Horizontal width and Material thickness, so assume it had the necessary calculated strength and not just something mocked-up for the show. So considering that the wing carries it's own weight I worry about reducing the weight required to be carried by the Spar, too much! So I worry about including everything in the wing, like Fuel.
    George
     
  11. Jul 10, 2019 #31

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,673
    Likes Received:
    2,923
    Location:
    Saline Michigan
    Do not know. If it was light sport, had a fairly deep wing, and it was all graphite, it could get by with a pretty light spar. Or it could have been designed the way I have been warning folks about, and it is then undersize...

    Why? Do a downtown job on the load estimates, design your pieces, do a downtown job on the weight estimates, then total weights and where it all is, then repeat. The dimensions will quit changing after a couple iterations, and you should have an safe sturdy airplane. Going in with load estimates 20-30% high, then designing around that, finding stuff is heavier than you thought, weights on everything cascades up, pretty soon the airplane is overweight.

    If instead you start with a pretty thorough approach to the loads, then weights, then iterate the design, you end up being able to reduce weight in each iteration of the total, which can drive another round of review and iteration.

    Worry over doing a good job of estimating loads and then designing good structures and pieces that are strong enough for those loads and the FOS for that construction...

    Billski
     
    Foundationer likes this.
  12. Jul 11, 2019 #32

    Lendo

    Lendo

    Lendo

    Well-Known Member

    Joined:
    Feb 7, 2013
    Messages:
    289
    Likes Received:
    45
    Location:
    Brisbane
    Billski,
    I'm hearing you, just conservative by nature. I'm going to redo my spread-sheet on the Spar as you suggest.

    On the Blackshape Prime (BSP), Yes! it was a Light Sport and let me try and describe it further, in addition to what I explained before.
    1. There was no foam in the WEB that I could see - or imagine. At the most the Carbon was about 3/16 inch thick at the root. They probably thought, as the foam provides stiffness, with sufficient Carbon BID in the WEB, it was stiff enough - I assume!
    2. The outer layer was Bid, but it wasn't a full wrap, just cut off at the trimmed edge of the C channel. I thought that was a bit sloppy.
    3. The C channel cap width was a good size at the root at about 3 inches. I didn't see any Uni but would be hard to see under the Bid, but naturally assume it was there.
    4. The WEB root was about 8 inches.
    5. As explained before everything was tapered to the tip.
    6. The BSP aircraft has a false floor and below that there was a Tubular (Chrome Molly I suspect) space frame, that the one piece spar was attached with appropriate Through Bolts and sleeves in the Spar.
    7. The RIBS were Molded (dish shaped) Carbon BID (maybe 2 to 3 layers) and no foam.
    8. The RIB Caps were about 7/8 inch wide (if that) and glued between the Spars front and back (no evidence of tapes). I assume they used a good quality (gap filling) Glue as it extruded a little around the edges
    9. The rear spar was also a light C channel with what appeared the same construction as the main spar.
    10. There was no Top skin (on this display) and the bottom skin extended out to where a leading edge might be attached and glued on. The bottom skin attachment was in the same manner as the Ribs - glued.

    Initially I though that this display was just a mock-up, but on reflection it probably was strong enough for the design goals, with minimized weight.
    What I would do differently would be to close out all the edges of the spars.
    Hope that description helps.
    George
     

Share This Page

arrow_white