Slow delta wings?

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by deskpilot, Jul 21, 2009.

Help Support HomeBuiltAirplanes Forum by donating:

  1. May 30, 2010 #61

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    We're apparently both in the same boat here. I'm not at all sure if my design will be dependent on delta lift or not. I have not settled on an airfoil yet and I guess that's a vital part of the equation. However, whatever I choose, the internal structure needs to have some 'form' in my mind. Although I started with the D.D. as a basis, it's wing design does suit my needs, it barely has a spar of any sort.

    I've spent some time today making a simplified model of what I have in mind. My question would be, will spars of this configuration be able to prevent any twisting? bearing mind that the outer wing is removable. I'm thinking of wooden spars and foam ribs on 3-400 centers. The spars will won't be as high as in my model and will be boxed where they enter the stub wings.
     

    Attached Files:

  2. May 30, 2010 #62

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    Your design is much more likely to benefit from vortex lift than mine is, for a few reasons - the main one being your strakes. But that becomes a much harder analysis problem.

    Without the exact geometry it's hard to run the numbers, but I'd suspect that you're looking at an overall wing CL_max in the neighborhood of 1.0 (0.8 to 1.2, call it, ignoring vortex lift). This is hugely dependent on wing geometry, and amount of reflex at stall; it's somewhat less dependent on actual airfoil (as long as it's sane), washout, etc. in such a low-aspect ratio wing.

    Now, I don't have the numbers for your design; but say that you consider it absolutely critical that you hit 45 kts stall, so you want to be able to do that with a CL_max of 0.9 and no vortex lift. Further, suppose your MTOW is 900 lbs. You must have 131 ft^2 of wing area. Of course, as any of those numbers (MTOW, wing area and geometry, stall speed) vary, the others will as well.

    So, if that's about what your design is drawn for, and 45 kts is your only stall requirement, just make sure that you'll get at least that CL_max without vortex lift, and more on. If you were hoping for a significantly smaller wing, something's got to give, or you have to do enough analysis to estimate and count on the vortex lift you'll get.
     
  3. May 31, 2010 #63

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Cheers Addaon, I'll have to re-run my numbers as the wing dimensions have changed during all my alterations. I much appreciate your guidance on matters that I have no experience.
     
  4. May 31, 2010 #64

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    Just keep in mind I haven't built one of these things either (yet), and that there are enough differences between our designs that some of my off-the-cuff remarks might be badly wrong. Before making decisions based on what I'm saying, double-check!
     
  5. Jun 1, 2010 #65

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Latest rendition, but missing it's rudders...ooops

    [​IMG]
     
  6. Jun 1, 2010 #66

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    I'm concerned by the channel between your verticals and fuselage. Seems like worst-case for interference drag. Even if you're planning on staying with twin verticals, I'd mount them on the fuse.
     
  7. Jun 1, 2010 #67

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Already played with that, looks terrible. Mounting on the fuse side also leads to poor air flow. The only thing I would do is move them a bit further out, say to the edge of the stub wings. They would then have to be removed for trailering. Of coarse, on the real thing, I would include base fairings and, although not initially planned, the fuse to wing join could be faired as well. As they say, all designs are a compromise.
     
  8. Jun 1, 2010 #68

    addaon

    addaon

    addaon

    Well-Known Member

    Joined:
    Feb 24, 2008
    Messages:
    1,686
    Likes Received:
    100
    Location:
    San Jose, CA
    Yep. Or just make the fuse wider.
     
  9. Jun 2, 2010 #69

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    No, no ,no. I'll stick with a wider stance, although on second thought, I think this might be too wide.

    [​IMG]
     
  10. Jun 2, 2010 #70

    Starman

    Starman

    Starman

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    2,011
    Likes Received:
    61
    Location:
    High in the Andes Mountains
    I think those tails are still too close to the cabin sides and will still cause interference. It would probably be better to put them on the cabin like Addaon said, but then they would cause interference with each other.

    There seems to be another problem and that is the contour of the rear of the cabin will conflict with the contour of the rear of the wing airfoil. You can make it all better by having very curved and faired surfaces like the SR 71 has, not just little fillets. Also, why not make the rear of the fuselage go further back behind the wing? Why cut it off so short? If it was longer it could better match the the contour of the wing airfoil. In a tailless aircraft the rear of the wing is pushing down but the rear of your fuselage is lifting up so you have a serious battle going on there, wasting energy, control effectiveness, and fuel.

    If you want to keep the fuselage short and stumpy like that it would actually be better to put the tails on the cabin sides, that way you can isolate the areas of conflicting pressure from each other.

    If this was my design I would round it all off like an SR71, extend the fuselage, and use a single vertical tail further back.
     
  11. Jun 3, 2010 #71

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Thanks Starman, I had no idea about conflicting pressures. Like I said right at the beginning, I'm not an aero-engineer, just some-one trying to design something different and easy for like minded folk. To that end, I want to stay away from composite curves. My fuse may end up being fabric covered so a reflex surface is out of the question.
    I can easily lengthen the fuse and have no problem with the look but don't want to go single fin. Might as well build a DD in that scenario. A longer fuse might make it easier to mount the fin on the fuse sides, I'll have a look into that, it would certainly give me a stronger mount and easier control runs.
    I'm also toying with alternative control devices. Most tailless aircraft seem to have more or less conventional ailerons, presumably for quicker roll-ability. The DD wasn't intended to be thrown about so the inboard elevons are adequate. I have 3 alternative choices, a full length elevon, or a split system. This would negate the need for a mixer assy and therefore, less complications. The 3rd idea, and I don't know if this is workable, is a split system with (stealing Addaon's idea) a split flap under the inboard elevators. This does, however, take it out of the quick and easy build classification. Any comments?
     
  12. Jun 3, 2010 #72

    Starman

    Starman

    Starman

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    2,011
    Likes Received:
    61
    Location:
    High in the Andes Mountains
    I'm not an aero engineer either, but I'm pretty observant and I've been 'observing' aircraft design for a few decades, too many decades :ponder: so I think of myself as world's best seat of the pants designer :)

    Here are some things that come to mind: A mixer with two surfaces can be extremely simple, simpler than a standard control setup.

    You can't put flaps on a tailless aircraft because they will cause it to pitch nose down and it could end up doing outside loops.

    You could move the tails further back so they are half way off the back of the wing, or more. That way you would have less interference drag.

    You could make the fuselage get a little wider as it goes to the back in addition to making it longer and that will lessen the pressure conflicts with the wing.

    You could have the control surface go all the way across the back, including the back of the fuselage. That will help make the back of the fuselage work with the wing rather than against it.

    You could read my first thread, titled Starman's Plane, which is for a pusher delta, to see how my design evolved :) If you can get past the first forty pages of name calling.

    The thing that broke the camel's back on that one is that I wanted the pilot in a reclining position and I found that it is evidently unsafe in a crash. That plane was all about style (and speed), and an upright pilot would ruin the style.
     
  13. Jun 3, 2010 #73

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    11,721
    Likes Received:
    3,320
    Location:
    Memphis, TN
    I dont think flaps would work because of such a long cord to span ratio. The Northrop flying wings had flaps, but I bet with the long wingspan and the sweep, it centered the pressure of the flap. You can see them on the Chino plane and here is some grainy video of the big one YouTube - Yesterday's Stealth
    I think a center vertical stab and the tips would be plenty of vertical plus ease the engineering of mounting the verticals; also you could set it back farther on the fuse., on a "stinger."
     
  14. Jun 3, 2010 #74

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    A couple of quick notes here: Overall the configuration being discussed (deskpilot's delta) is looking reasonable but will of course need tweaking to be functional. Much of the above discussion might have application if this was a 300 mph aircraft however, given the modest speed range it looks to be heading for, many of the items like interference drag are for the most par secondary so not as critical as the discussion seems to make them out to be. The position of the verticals is probably fine in either location although the outboard position is probably better since they're not as likely to be blanketed at higher angles of attack.

    What you call "delta" lift would normally be called vortex lift - in simple terms this is caused by the formation of a separated vortex flow off the fore strake, which creates a significant low pressure field over the surface area over which it is formed. The resultant lift can be significant but it does require the aircraft to be at a fairly high angle of attack. It also requires a fairly good amount of speed since if the flow doesn't have sufficient energy the flow off the leading edge will simply separate, acting more like a stall rather than vortex flow. How much is required though would require a bit more advanced analysis.

    The amount of lift generated can be quite substantial: In a high G turn, the F-16's strake alone generates about 70% of the airplanes lift at that point. But herein lies a problem: That vortex substantially modifies the pressure field distribution so trim of the aircraft can change quite dramatically. Trim plots for strake geometries often look like abstract sting art so it's a technology that's generally not recommended for GA applications since an optimum configuration would take some time (and money) to optimize.

    One good point above talks about the mingling effect of the pressure distribution over the fuselage, versus that of the wing. That too can lead to some interesting trim requirements. At this point probably the best thing to do is to work at rounding the canopy and turtle-deck so that it forms a more aerodynamic blend with the wing. That would not only help a bit with the pressure distribution but would also make it prettier. Failing that, one place you may want to look is at the Dyke Delta and see how his fuselage height distribution is distributed over his wing.
     
  15. Jun 3, 2010 #75

    Starman

    Starman

    Starman

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    2,011
    Likes Received:
    61
    Location:
    High in the Andes Mountains
    I just want to add here that when I mentioned the pages of name calling in the other thread that it was a joke and it should have had a smiley face added to it. In fact I find that this forum is quite peaceful and mature.
     
  16. Jun 4, 2010 #76

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Hi guys, I just love this site, so many diverse opinions, so much banter. Thanks for starting my day off well.
    Now, the come-backs.
    Lengthened the fuse yesterday, good out come with a slightly lowered 'hump'. Tried moving the fins to the side walls but just didn't like the look and still had a channel between the fin and parts of the wall. So the fins stay where they are at the moment. Moving them further back only complicates the whole thing in that they have to go over the elevons, or I have move them to the edge of the stub wings, and that is too wide in my opinion. I'm going to go with the old boating saying that 'if it looks right, it probably is right'. To me they look right where they are.

    Spent some time thinking about option 3 and basically came to the same conclusion as Starman. Outside loops are not in my bag of tricks. I'm not adverse to taking to control surfaces across the rear of the fuse' bit is it really necessary? As Orion points out, I'm (we're) not redesigning an F16 here. I'm in two minds as to which of the other 2 options I will finally settle on.
    In response to TFF, I don't think we can compare flying wings to deltas, or truncated tapers as some prefer to call my wing. I thank you for your input all the same. Actually, the only Delta that I have actually to have flaps, is the Hawker Javelin, but that also had a high mounted tail-plane. BTW, what is a 'stinger'?
    As to widening or curving the fuse, out of the question. As I've previously said, I want to keep it simple (I know, in many respects I'm going against that criteria). Besides, I believe that I will benefit from the additional 'wing area' the the fuselage top surface gives. At present, the wings(including strakes) give me an Aspect Ratio of 2.3. Adding in the total top surfaces of cowl, cockpit and fuse', that figure drops to 1.9 . I haven't taken the fins or winglets into account even though they will provide a small amount of lift.

    On the subject of Vortex lift, I'm not knowledgeable enough to say although, in the early stages I had a hard edge with an undercut to create the required vortex. Thing is, would my landing speed be fast enough for it to work. I doubt it now so my current strakes are symmetrical in profile, much easier to blend into the main wing as well.

    I want to thank you Orion for keeping us centered on the 'Slow Delta wings' concept. It made my day to read your remark 'looking reasonable', sometimes I need to be re-assured. Re fuselage contour, I'm not overly concerned with the overall looks of my plane, just as long as it's 'different' from all others.
    And finally, Starman, I have seen some of your thread and will return to it when time permits. Thanks for the heads up all the same.

    Have a good day everyone.
     
  17. Jun 4, 2010 #77

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    Regarding the vortex lift, I'd still be very cautious with the fore strake - even if it doesn't form the full vortex flow, it will most likely still play havoc with your trim plot and unfortunately that's not an easy thing to calculate or predict and due to issues of scale, nearly impossible to model.
     
  18. Jun 4, 2010 #78

    BDD

    BDD

    BDD

    Well-Known Member

    Joined:
    Apr 2, 2005
    Messages:
    388
    Likes Received:
    3
    Location:
    WI
    I don't think that the spar configuration you have shown would be very efficient at carrying wing torsion loads or pitching moments. As shown, the entire torsion force would be resisted by the top and bottom spar cap connections to the fuselage. This is related to the reaction forces and how they are resolved. For the spars themselves and how they carry this load, they would be better used if they were spaces further apart from each other along the chord. This is true for the entire span but more important at the wing root. That way, VEERRRRY widely separated front and rear spars as is possible in a delta configuration would resist torsion loads rather well in differential bending. The long "moment arm" between the spars would make the vertical loads onto each spar form torsion smaller. I see them as acting like a force couple to resist this force. I also like how spars can also be very deep with a long chord configuration like this. Buckling becomes more of an issue with the deep, thin shear webs and they have to be stiffened.

    There are some natural structural advantages to a delta wing. I would make the most of them for a lighter, yet strong structure.
     
  19. Jun 4, 2010 #79

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Sounds like a case of suck it and see, Orion. Might have to be a case of making 2 large RC models a see if any differences can be seen in their performance.

    BDD, thanks for your comments. I put the post in as I wasn't sure which way to go. In a conventional wing, the spas are at right angles to the fuse' and it's easy to to make a straight 'plug-in' type fitment. With the Delta, or any highly swept wing, the spars usually follow the line of sweep to some degree. If that is to be the case in my design, only one spa could be inserted at the angle it forms with the fuse'. My suggestion was to create a straight forward 'v' shaped tongue which inserts into to stub-wing carrier box at right angles to the fuse'. There would, of coarse, be location pegs/pins at the L.E. and T.E. locations to prevent twisting along the join faces. I'm still not sure. Perhaps you'd care to expand on 'natural structural advantages of the delta wing'.

    In the mean-time, I was doing some research into finding the C.G. of a double delta and came across this:

    The double delta wing is divided into two panels:
    Inboard Panel: the leading and trailing edges are extended to the center line of the wing and tip chord span station is fixed at the break formed by the discontinuity in the sweep of the leading edge of the composite edge.
    Outboard Panel: the leading and trailing edges are extended to the midpoint between the center line and the break formed by the discontinuity in the sweep of the leading edge of the composite edge.

    Damned if I can make sense of it. Any-one care to add some info on the attached image.
     

    Attached Files:

  20. Jun 4, 2010 #80

    deskpilot

    deskpilot

    deskpilot

    Well-Known Member

    Joined:
    Jun 30, 2009
    Messages:
    1,096
    Likes Received:
    185
    Location:
    Morphett Vale, South Australia. Just south of Adel
    Whist you lot slept, I've tried to work it out. It may not be 100% accurate but looks about right to me. Now I'm going to bed. :)
     

    Attached Files:

Share This Page



arrow_white