Horton wings?

Discussion in 'Warbirds / Warbird Replicas' started by conestogaman, Mar 5, 2009.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Sep 30, 2009 #61

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    The pitch effect of sweep is due to the the larger chord part getting into chord dominated GE first. since your 1/4 chord line is swept forward I'd expect it to pitch down but it should be a very small and slow change that can be easily trimmed out by the pilot.

    Hmm... That would mean pulling back more. One more little detail to keep track of in order to avoid bouncing those landings

    If you can find data on how much lift increases relative to hight in chord dominated GE you could figure out the additional pitching moment due to GE by figuring the hight above the surface and the lift increase at a few span wise locations with their longitudinal distance from the CG

    I know you don't like hearing this but I think you're obsessing over significant digits. Remember back in the days of slide rulles they lopped off most numbers after three decimal places because anything beyond that just adds to the computation time without improving accuracy that much. Yeah, it left some unknowns but the planes usually got through their test programs without killing the test pilot. If you don't drop the significant digits you'll be looking at some expensive computer time.
     
  2. Sep 30, 2009 #62

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    Actually, the 0.25c line on my 'plank is straight. No sweep at all. (See two-view below) What I'm seeing in my numbers is a small lessing of the pitching moment in ground effect. Wing CL to maintain level flight in ground effect is less than that aloft, so my notion goes that the CM goes down, too, barring sweep effects. With the elevator on my design, it requires about six degrees less elevator deflection to hold trim in ground effect than out of ground effect, for the same conditions.

    If you can find data on how much lift increases relative to hight in chord dominated GE you could figure out the additional pitching moment due to GE by figuring the hight above the surface and the lift increase at a few span wise locations with their longitudinal distance from the CG


    Oh, you're right. And I wouldn't be trying to nail this Jello to the wall if my design were 'tailed'. Building an airplane is a significant investment in time and money, and I want to not only have something that won't crash, but that I'll enjoy flying as well. You're the last person I need to tell that tailless designs require more development to get satisfactory flying qualities, I'm sure. What I don't want happening is to spend a couple of years building something, only to find that while, yes, it can fly, the handling characteristics are such that I'm scared or frustrated all the time trying to fly it.

    I'm actually satisfied with my ground-effect studies for this design, and the method I've developed. I was just hoping to find a method with some test data behind it, if perhaps you had one. The trim plot for my glider is attached below. My only remaining area of concern is dynamic stability characteristics. A "pecking" airplane (to use Nickel's term for it) wouldn't be fun to fly at all, and as a sailplane pilot, I spend my flying time going looking for turbulence. I don't know how to reassure myself on the dynamic stability score without immersing myself back into a bunch of math I've allowed myself to forget.
     

    Attached Files:

    Last edited: Sep 30, 2009
  3. Oct 1, 2009 #63

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    Re: Horten wings?

    You're getting a 6+ degree difference in elevator deflection for minimum speed in and out of ground effect? :whistle: That seems like a lot. Could it be that you're getting a big enough lift boost from chord dominated GE to require that much of a decrees in AoA to maintain level flight? I don't remember any detailed discussions of chord dominated GE in any of my books. Probably because even low wing airplanes don't get very far into it and have often already flared by then. I vaguely recall something about WIGs having a pitch change when they leave GE but they have never been a major interest of mine so I don't have any books to look that sort of thing up. There are some WIG groups on yahoo. I'll join one of them and raid their archive. If that doesn't work I'll ask
     
    Last edited: Oct 1, 2009
  4. Oct 1, 2009 #64

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    Re: Horten wings?

    It does until you see how limited in span my elevators are, and how close fore-and-aft to the CG. If I made them full-span (neglecting the taper, of course) the difference in deflection would be less than a degree. When you think about it, if the change in CM is proportional to the reduction in CL required to maintain level flight in GE (my supposition), then this looks about right. There's a very noticeable ground-effect with a span and aspect ratio like this.

    In fact, just for giggles and grins, I changed the taper ratio input in my spreadsheet to 1.0 (rectangular wing), and the deflection required for trim at fwd CG (no GE) jumped from 13° to over 25°. The lack of taper moves the elevator closer to the CG and reduces its effectiveness. I tweaked the elevator proportion of span to get the same deflection as with the tapered case, and it jumped to almost 50% of span, up from the 33% it's at now. It's that sensitive.

    If anything, my method probably overpredicts the CM change and elevator deflection anyway. Especially the latter since I'm not taking into account the trailing edge extension you can see in the drawing, when I calculate all the moments. I just did the exercise to see if the change in elevator deflection was into the range I'd already set aside (it is), or if it required more deflection than the fwd CG case (it doesn't). If I'm overpredicting the effect, I just don't move the stick as much in the real airplane. It's all a 'feel' thing when you're flying anyway. In any event, my mechanical elevator deflection range would probably be something like -20° to + 12°, so there should be extra power if it's needed.
     
  5. Oct 1, 2009 #65

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    When does a sailplane become a class A WIG?

    Yeah... the Cm of part of the wing is fighting the Cm of the rest of the wing and, unlike with a conventional tail or swept 'wings, the lift change doesn't do you any good.

    Yeah, again:) I suspect that this is going to be a pilotage issue and not something you can design out.

    I've been digging through the old messages of the yahoo WIG group (you don't need to join to do that). One guy mentioned in 2007 that he's doing his doctoral thesis on ground effect. That should be ready pretty soon:gig: I also saw mention of a file titled "Flight at the height of " in the "ESKA-1" directory of the files section. I'll get that latter, I'm mentioning it now as a reminder to myself.

    And of course there are lots of references to the WIG page so I looked at it. There's some math on the WIG stabillity page that may be what you're looking for. Here's a snip from that page:

    Could the sentence I've highlighted be the real reason that the Fauvels have a reputation for bouncing?

    There are also some pressure plots of airfoils in chord dominated ground effect on the WIG aero page. Basically they say don't use too small of a leading edge radius and avoid having very much curvature on the lower surface. I think you're using something that fits those restrictions.

    I also found a technical report about wingships on the Dryden tech report server that could be helpful.

    And last but not least, that pitch problem that I said I vaguely remember is a pitch-up when leaving ground effect. The cause of this is that the lift vector is way back in chord dominated GE. Most of The WIG designers deal with this by putting the center of gravity at 35 to 45% MAC and using a huge stabilizer. :emb: I'm sure I saw reference to a tailless WIG model in one of the group messages but I've lost it so I'll try to find it latter.

    BTW this is way off topic
     
    Last edited: Oct 1, 2009
  6. Oct 1, 2009 #66

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    Re: When does a sailplane become a class A WIG?

    I know. However it seems we're the only ones in here. Anyone? Anyone at all???

    Yes. I'll still get the lift "benefit" (i.e. - floats like heck...) of ground effect, but the CM change is the thing of interest to me with a plank.

    Well, it's possible that it contributes, but it's extremely clear to me why AV-36's are bouncing, especially upon landing. Two parts to this:

    Pancaking: Nothing new here. Well known phenom, but pilots still have an unconscious tendency towards small, sharp pitch maneuvers when close to the ground. Pull back at six inches above the ground and suddenly 'bang', you're on it.

    Landing gear geometry: This is the biggie, in my mind. Every time I watch those two 'modern' AV-36 videos you posted, I see it again and have it confirmed in my mind. The nosewheel (or skid in other examples) is touching down first and forces the nose up sharply as the airplane settles. That increases AoA suddenly and the airplane lifts into the air again, whereupon it slows from the added drag and starts to settle. The pilot, sensing the uncommanded pitchup, pushes the nose back down and the nosewheel touches first again. The cycle repeats, a classic airframe-initiated PIO. It's clear as day on those videos. The landing gear geometry of the AV-36 needs to be changed so that you touch down on the main wheel first (which should be a little closer to the CG, IMHO), and then the AoA reduces as you put the nosewheel (or skid) on the ground. That would cure the 'bounce' once and for all.

    I know that the nosewheel needs to be longer so that the airplane, with its limited pitch authority, is at the proper geometry for takeoff. But that's the opposite of what you want for landing, and normal sailplane technique is not a full-stall landing. Perhaps a nosewheel that retracts upon takeoff, leaving a skid exposed for landing that's much shorter, allowing the main wheel to touch down first.

    A lot of the older AV-36 examples used a takeoff dolly (Jack Lambie speaks of this on his own example), which probably took care of the problem: It held the aircraft at the proper takeoff geometry, then fell away, leaving a landing gear configured to allow the rear of the skid to touch down first, causing the airplane to pitch down upon ground contact. When people went to wheeled gear on the AV-36, they "froze in" the takeoff geometry, not realizing what it was going to do to landing - or thinking that they'd just land full-stall at the same attitude. Put that airplane into the hands of a 'regular' sailplane pilot and he'll approach and land well above stall, putting the nosewheel (or front part of the skid) down first. Bounce.
     
  7. Oct 1, 2009 #67

    conestogaman

    conestogaman

    conestogaman

    Active Member

    Joined:
    Mar 5, 2009
    Messages:
    34
    Likes Received:
    1
    Location:
    Sack-O-Tomato, CA
    Off topic or not, I don't mind. I enjoy reading thru all your posts. After weighing the boat, I have determined it is too heavy for the Mitchell wing anyway, so that's out.

    Thanks,

    Jeff
     
  8. Oct 1, 2009 #68

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    If you're interested in doing a flying wing, Jeff, there are better ways to approach it. "Cut and try' just isn't very successful for tailless airplanes.

    The 'right' ways require a little math and analysis, but nothing more than algebra. If I can learn it, surely you can. ;)
     
  9. Oct 1, 2009 #69

    bmcj

    bmcj

    bmcj

    Well-Known Member HBA Supporter

    Joined:
    Apr 10, 2007
    Messages:
    12,833
    Likes Received:
    4,797
    Location:
    Fresno, California
    Re: When does a sailplane become a class A WIG?

    Could this possibly be attributed to another factor rather than the pilot overcompensating? Perhaps at the lower landing speed, the reflex is insufficient to sustain the nose high attitude caused by the nosegear/skid, thereby letting the nose drop down to (and even through) the trim attitude.

    Bruce :)
     
  10. Oct 1, 2009 #70

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    Watch this video starting at about 3:00, and tell me you don't see a classic PIO.

    Nice stable approach, nice gentle flare, nosewheel touches down first, radical pitchup, airborne again, radical pitch down, rinse and repeat...

    I'm not blaming the pilot - he's doing what comes naturally, especially for a trained pilot. In this case, it also looks as if the rear skid (used on this particular example instead of a main wheel) is getting into the act after the first cycle, bouncing the nose back down. Regardless if its the pilot or the rear skid, if the airplane had a main rear tire just aft of the CG and that main gear touched down first upon landing, the airplane would reduce the AoA to get the nosewheel down, and this motion wouldn't have happened.
     
  11. Oct 1, 2009 #71

    bmcj

    bmcj

    bmcj

    Well-Known Member HBA Supporter

    Joined:
    Apr 10, 2007
    Messages:
    12,833
    Likes Received:
    4,797
    Location:
    Fresno, California
    Agreed. The initial error appears to be from the pilot flying it on rather than stalling it on, but I suspect that the bouncing is a natural follow-on response of the landing gear (having landed too flat). I wish we could see the elevator motion, but the oscillations looked a bit to quick and limited to me to be PIO (at least for most pilots).
     
  12. Oct 1, 2009 #72

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    Re: When does a sailplane become a class A WIG?

    What I was getting at is that if your elevator position for minimum controllable airspeed is 14 degrees out of ground effect and 8 degrees in ground effect then the first bounce, regardless of cause, is going to throw you from the zone of maximum GE clear out of chord dominated GE faster than a human being can react (or even be aware) so the plane will be reversing its pitching moment during the bounce


    Back on topic:

    Jeff, The landing gear position that Topaz mentioned is just as important to swept wings but the other two phenomena we've been talking about are specific to planks. Dr. Nickel tried to observe pancaking in a prone position H-III and said that even with the wheel a few inches of the ground the pitch response appeared to be immediate and positive with no momentary drop like planks do.
     
  13. Oct 1, 2009 #73

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    And this one. Yeah it's a 1/4 scale model:ermm:
     
  14. Oct 2, 2009 #74

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    'Flying it on' is not an error in sailplanes. That's SOP.

    I agree that once the motion gets started, it's sustained by the aircraft bouncing back and forth between the nosewheel and the rear skid, rotating about the CG.
     
  15. Oct 2, 2009 #75

    Topaz

    Topaz

    Topaz

    Super Moderator

    Joined:
    Jul 30, 2005
    Messages:
    13,623
    Likes Received:
    5,275
    Location:
    Orange County, California
    Okay, I see what you're talking about. Once the airplane is 'tossed' out of ground-effect, suddenly it has insufficient elevator deflection to hold pitch attitude and the nose drops...

    Wouldn't you think that this effect would require the 'bounce' to be a substantial faction of the MAC? I'm not seeing that much altitude change in the videos posted. Seems as if he 'bounces' no more than eighteen inches, if that, although I'm sure it feels more violent inside the cockpit. ;-)
     
  16. Oct 2, 2009 #76

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,436
    Likes Received:
    894
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    It did it a little PIO/bounce on the takeoff too.

    Shannon.
     
    Last edited: Oct 2, 2009
  17. Oct 2, 2009 #77

    bmcj

    bmcj

    bmcj

    Well-Known Member HBA Supporter

    Joined:
    Apr 10, 2007
    Messages:
    12,833
    Likes Received:
    4,797
    Location:
    Fresno, California
    You know I know that, but this plane seems to need a bit of re-think in the landing attitude based on what we see in the video.

    I think Norman raises a good question, but given the overall proximity to the ground and the relatively small altitude changes, I don't think that this is a case of fling in and out of ground effect.

    Bruce :)
     
  18. Oct 2, 2009 #78

    Norman

    Norman

    Norman

    Well-Known Member

    Joined:
    Nov 29, 2003
    Messages:
    2,880
    Likes Received:
    918
    Location:
    Grand Junction, Colorado
    Re: Horten wings?

    Oops, I should have watched the video again before shooting off my mouth:emb:. You're right he doesn't go very high and chord dominated ground effect is relative to the local chord so the pitching moment would be most strongly affected by the root i. e. when the plane is high enough that the tips are out of CDGE the root is still in it.

    That's why swept-back 'wings pitch up even though the lift vector moves aft
     
  19. Oct 6, 2009 #79

    Starman

    Starman

    Starman

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    2,011
    Likes Received:
    61
    Location:
    High in the Andes Mountains
    Re: Horten wings?

    I spent a lot of time researching and designing WIGs. In fact the aluminum I'm going to use for my airplane is scrap from a WIG boat hull I had made. WIGs exhibit a very strong pitching up force when they leave ground effect! This is also why unlimited hydroplanes will instantly flip over backwards if they get into the air.
     
  20. Oct 6, 2009 #80

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    The issue of WIg stability is a complicated one and the design parameters are strongly a function of the class of WIG that is being developed. Optimized, it is true (but not necessary) that Class A WIGs may exhibit pitch instability however, Class B and C WIGs are required to maintain full controllability and stability in ground effect and in free air.

    Regarding hydroplanes, while ground effect does play a minor effect in flip-overs, the primary mechanism of the motion is totally different and a bit beyond this medium for a full description.
     

Share This Page

arrow_white