Sparless spar attachment

Discussion in 'Composites' started by Hot Wings, Jun 5, 2015.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jun 5, 2015 #1

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    One of the options I'm considering for my outer wings is to make them without spars using a sandwich for both the upper and lower skins. I'd also like the outer wings removable for trailering but the conventional way of doing this is to use spar stubs that slid or bolt into place. This means gathering up all of the loads from the skins and transferring them to the attachment point. This adds weight and complexity.

    Below is a sketch of an idea I have to avoid this and use a distributed attachment that can be sandwiched between the skins with just local reinforcement - similar to the way the wing attach fittings are bonded into the EZ center section. The little blue thing is rotated to release the wing. Quick calculations indicate that friction of aluminum to aluminum would be enough to keep the joint together without having to positively lock it together. The slots are to allow the fitting to be milled flat and then follow the curve of the airfoil.

    Has anyone ever tried something similar?

    Edit: change aluminum to steel on steel.
     

    Attached Files:

    • end.jpg
      end.jpg
      File size:
      14.1 KB
      Views:
      247
    • iso.jpg
      iso.jpg
      File size:
      28.6 KB
      Views:
      201
    Last edited: Jun 5, 2015
  2. Jun 5, 2015 #2

    Topaz

    Topaz

    Topaz

    Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,419
    Likes Received:
    5,134
    Location:
    Orange County, California
    Clever, but I don't like that they effectively end up in single-shear. That's a no-no in mechanically-joined high-loading joints. Perhaps I'm just reflecting my limited knowledge of structural engineering, but I can't see why that caveat wouldn't apply here.
     
  3. Jun 5, 2015 #3

    DeepStall

    DeepStall

    DeepStall

    Well-Known Member

    Joined:
    Dec 26, 2010
    Messages:
    136
    Likes Received:
    44
    Location:
    Colorado
    I wouldn't recommend trusting friction for a safety-of-flight critical joint like this. Especially where vibration and play can act to loosen the joint. How does this design carry compressive loads, or shears along the joint line due to torsion (eg aileron deflection)?

    The F-4 Phantom (and others) use a multi-lug folding joint that might be usable while keeping with your stressed-skin concept.
     
  4. Jun 5, 2015 #4

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    It would. There is a lot more wrong with the sketch than just that! I didn't plan it well and didn't want to start over just to get the idea across to see if this might have already been attempted somewhere. If Aircar were around he could probably cite year and model. :gig:

    Loads on the fittings would only be in the 2000 pound range - Thick flying wing, so a large FOS wouldn't add too much weight. Also aluminum should be changed to steel. Just got in a hurry because the plants needed to be covered due to hail.
     
  5. Jun 5, 2015 #5

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    Torsion would be by pins.
    Compression loads by wing skin to wing skin.
    Taper of the "teeth" would tend to tighten the fitting under vibration due to spring tension from the flanges. Also could compensate for minor wear over time. The friction on the mating surface, if the angle is steep enough, is enough to lock it together during the tension portion of the vibration.

    All just theory at this time. Will look for F-4 pics.

    Edit:
    Found a picture. This is similar to my original idea to use a piano hinge like we do for engine cowls. That would be way to "fiddely" to try and assemble in any kind of wind. Must ponder this variation.

    Thanks for the tip!
     
  6. Jun 5, 2015 #6

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,537
    Location:
    Rotterdam, Netherlands
    You want this kind of connections to be statically determined, so no more than 3 point contacts. Of course, concentrated loads in composites are a bear ánd a nightmare.

    In theory, the carbon "piano hinges" might work. It would scare me though, given unpredictable concentrated loads.

    I think, realistically stub spars are your only hope. Note that there is no requirements to have a stub SPAR, just a shear web and continuing in the middle of the wing skins would work too. You need a wedge at the terminus of the foam core and a slight taper, but that seems well do-able?
     
  7. Jun 5, 2015 #7

    Himat

    Himat

    Himat

    Well-Known Member

    Joined:
    May 5, 2011
    Messages:
    2,754
    Likes Received:
    633
    Location:
    Norway
    The idea look to me to be good, to execute it with some sort of elegance is another matter. By the time it is proper thought out it may have got overly complicated. The trouble is to make a positive attachment.

    Have you considered to have a rib at the end of the inner panel and a rib recessed in the outer panel?
    Positive attachment can then be done by a device/devices that clamp the two ribs together with the skins aligning the joint and taking up the shear forces.
     
  8. Jun 5, 2015 #8

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    I probably haven't given this aspect of it enough thought. Each 25mm wide "tooth" would be strong enough for the full load if made from 1.6mm 4130 for the flange. Figuring out how to distribute that load over half a meter or so of skin is the tough part.


    I'd considered something similar but your words helped me envision it slightly differently - and it might work.


    ===============
    It's been a very productive thread for me in a very short time. You have shown me some problem areas that need more study and a few options I'd not yet imagined.

    Thanks!
     
  9. Jun 6, 2015 #9

    flyoz

    flyoz

    flyoz

    Well-Known Member

    Joined:
    Jul 30, 2008
    Messages:
    149
    Likes Received:
    5
    Location:
    Near Sydney
    Why not something like the Mead Adventure ?

    I have done some preliminary sketches and calcs and instead of joining the spars with plates they could overlap and join in the center with one pin similar to the varieze taper pin system

    This would go directly through the upper and lower spar caps ( with small steel taper collars ) - essentially one wing would be slightly higher than the other

    That way you have a stub spar ( half the fuselage length ) and you still join both wings upper and lower spar caps

    Flyoz
     

    Attached Files:

  10. Jun 6, 2015 #10

    BJC

    BJC

    BJC

    Well-Known Member

    Joined:
    Oct 7, 2013
    Messages:
    9,096
    Likes Received:
    5,903
    Location:
    97FL, Florida, USA
    Not certain that I fully understand your drwaing, but it looks as if it would be difficult to "unlatch" the wing extension for removal.


    BJC
     
  11. Jun 6, 2015 #11

    JamesG

    JamesG

    JamesG

    Well-Known Member

    Joined:
    Feb 10, 2011
    Messages:
    2,408
    Likes Received:
    753
    Location:
    Columbus, GA and Albuquerque, NM
    The blue thing is an excentric cam that runs the length of the latching "fingers". When the cam is rotated it forces the hooks of the fingers apart allowing them to slid apart.

    @ Hot Wings:
    That's quite clever, but... In practice, to reinforce the skins of the attachment points sufficiently will add just as much weight as conventional stub spars. Also integrating the interlocking flanges, cams, and what not would be at least as "fiddily" as fabricating the spars and sockets, perhaps more so because it will have to trace the shape of the airfoil contour instead of just being straight shafts.
     
    BJC likes this.
  12. Jun 6, 2015 #12

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    I'm not so worried about "fiddily" with regard to construction - that's only done once. I'm wanting an easy to remove, not fold, outer wing for transport.

    However the more I think about what Autoreply mentioned about evenly transferring the load the less I like this idea. The unit will require some kind of elasticity or the center portion that is furthest from the chord line will be carrying more than it's share of the load.

    If I get time tomorrow I'll take another look at a stub spar. The loads I need to transfer are miniscule compared to what a normal glider stub has to transfer so my minds eye is probably envisioning a much more substantial structure than is really needed.
     
  13. Jun 6, 2015 #13

    Victor Bravo

    Victor Bravo

    Victor Bravo

    Well-Known Member

    Joined:
    Jul 30, 2014
    Messages:
    5,804
    Likes Received:
    4,649
    Location:
    KWHP, Los Angeles CA, USA
    Look at the DC-3 outer panel attach, and the AT-6 outer panel attach. I believe all of the loads are transmitted through bolted joints around the skins. This is apparently a very good way to do the attachment for spar-less wings, however the DC-3 and AT-6 use a large number of bolts and nuts to accomplish this. Definitely not a "trailer home after a day of flying" kind of thing.
     
  14. Jun 6, 2015 #14

    henryk

    henryk

    henryk

    Well-Known Member

    Joined:
    Mar 8, 2010
    Messages:
    4,573
    Likes Received:
    422
    Location:
    krakow,poland
  15. Jul 4, 2015 #15

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,664
    Likes Received:
    2,918
    Location:
    Saline Michigan
    My concern with this scheme is that in order to unlatch you have to get all of the fingers along each of the skins to clear. That means you have to achieve enough rotation of the wire that is actuating the cams of the last one, a long ways down the wire. You could still do it, but to know that it can work, you might need to do some math. Here's how.


    • Excel is your friend - all design dimensions and other numbers that you will be setting and/or adjusting should have a cell at the top of your spreadsheet, a label for what it is, and units;
    • Using cantilever beam deflection equations and your proposed flexible joint dimensions, set up the equations for figuring out how much force will be needed to lift the tab to disengage. May as well calculate stress in flexible part of the tab and in the wire at each tab so you can check FOS;
    • Set up the equations to calculate the spring rate (force/movement at the force) of the tab;
    • Set up the equations to calculate how much torque and rotation is needed to make the force and travel of the tab;
    • Set up a column for each tab;
    • Big step, put a space in each column for each of your equations - lift height, lift force, cam rotation, cam torque, wire torque, wire rotation, wire size, wire J (for round wire, PI*d^4/32), length from the torque input to THIS tab;
    • Next you get to accumulate load and wire twist from the furthest one out, all the way back to the one by your lever.

    Now here is the kicker. The tab closest to the place where you apply torque to the wire will take the torque required, and the rotation required. TL/GJ=theta where T is torque, L is distance from where you put your lever to THIS tab, G is 11.5 mpsi, and J from above. Rotation of the wire between your lever and the each cam is theta in radians. 2*PI radians in 360 degrees... Calculate rotation and torque for each cam. Now, lets remember be that the rotation and torque at the cam furthest away has to be big enough to open it, the torque and rotation in the wire at the 2nd furthest away has to start with the furthest one, and then add the second one, and so on. If the wire were rigid - no twist - you would only have to come up with cam rotation and torque enough to open one tab times the number of the tabs. But because the wire will twist as a torsion bar, the second one will have to rotate far enough to open the first one plus carry enough torque to rotate the second one, and so on down the line. You might need a lot of rotation at your lever to get all of them open.

    I will suggest that your cam might spend the first increment of rotation achieving lift, and the rest (largest portion) at a fixed lift. This way the closest one is open early, the second one follows with a bit more rotation, and so on down the line. This also allows the torque that accumulates from the furthest one back to your lever to be reduced. Is this a lot of math? Not to me it is not, but I am not doing your invention. You are. Set up the model so it represents what is happening, and adjust your design until it works. Way faster to do it with electrons than to build one and find out it does not work, then try to iterate your design when you do not have any idea if making it 50% bigger is enough or 200% bigger won't be enough or maybe that it will only work when it is big enough to make stub spars look light...

    Bill
     
  16. Jul 4, 2015 #16

    Victor Bravo

    Victor Bravo

    Victor Bravo

    Well-Known Member

    Joined:
    Jul 30, 2014
    Messages:
    5,804
    Likes Received:
    4,649
    Location:
    KWHP, Los Angeles CA, USA
    The idea and "outside the box" aspect of this is interesting, I applaud the thought process. However, at the end of the day, if you are looking for a five minute disassembly that is safe and lightweight, you will eventually have to begrudgingly accept that the sailplane manufacturers have done this engineering and come up with the best method. The Polish SZD "Diana" is supposedly a spar-less wing, so however they have done their quick disassembly is probably worth looking into.

    https://en.wikipedia.org/wiki/Diana_Sailplanes_Diana_2
     
    henryk likes this.
  17. Jul 5, 2015 #17

    Riggerrob

    Riggerrob

    Riggerrob

    Well-Known Member

    Joined:
    Sep 9, 2014
    Messages:
    1,137
    Likes Received:
    311
    Location:
    Canada
    May I suggest a simplified version of the wing-fold mechanism installed in the Grumman Tracker (carrier-based, anti-submarine patrol airplane)?
    Each Tracer outer wing panel has 8 tongues along the top skin and 8 along the bottom skin. These tongues are locked to the wing root by pins that slide vertically. Pins are loaded in shear to transmit outer skin loads directly to wing root skins.
    Trackers' hydraulic system has a huge parts count. If you could stick a human arm inside the wing root, you could vastly reduce the parts count.
    Homebuilders could reduce opportunities for errors by adopting the same build sequence as Thatcher's CX-4: build the wing roots first. Then attach the fold mechanism hinges and lock pins. Last build the outer wing panels.
     
  18. Jul 5, 2015 #18

    henryk

    henryk

    henryk

    Well-Known Member

    Joined:
    Mar 8, 2010
    Messages:
    4,573
    Likes Received:
    422
    Location:
    krakow,poland
  19. Jul 5, 2015 #19

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains

    Thanks for the thoughts and the time to reply!

    The math won't be a problem, but between being a bit lesdyxic and typos I do seem to have a hard time getting Excel to do my bidding without a bit of work. :ermm: I've had a chance since my last post here to put some numbers to Autoreply's suggestion to use a more conventional spar stub and it appears that the weight difference isn't all that much. The loads involved are very small - on the order of 5K ft/lbs and a 10 inch thick rib. I've still not ruled out this experiment, but it's pretty far down the options list at the moment.
     
  20. Jul 5, 2015 #20

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,074
    Likes Received:
    2,053
    Location:
    Rocky Mountains
    Looks like a variation of the F-4 fold mechanism another poster suggested. It is very close to my original thought and it need not be hydraulic.
     

Share This Page

arrow_white