carbon/wood spar considerations

Discussion in 'Composites' started by Rienk, Aug 23, 2010.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Aug 23, 2010 #1

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    Hey Orion, I have a couple of questions about making spars for such a design as we've been discussing in this thread.
    I intend to waterjet out all the various parts, including the wing ribs and spar web - using CF pultrusions for the caps.

    Let's see if I can explain this clearly...
    Right now, I have the design so that spar web has a step tapered pattern cut into it, to allow for the lengths of the required pultrusions. This allows the top and bottom of the spar to be straight, so that they align with the each of the ribs perfectly.

    The spar web also has rectangular holes cut into it for which the ribs have tabs that fit into - laid out in such a way that only certain ribs can go into certain BL locations (don't want builders gluing in the wrong ribs and messing up the wing kit).

    First question:
    I've been thinking about finishing off the top surface of the spar with a strip of wood, to which thin plywood would be glued to on each side, glued also to the central spar web geodesic design, essentially creating a thin box spar. The main idea behind this is to better encapsulate the CF, and hopefully help prevent delamination due to buckling, etc.
    IS THIS A GOOD IDEA?

    Second question:
    The other thought I had was this: rather than mounting the CF pultrusions to the top of the plywood spar web, having them mount to the side face of the plywood web instead, and then having the ribs properly notched to allow for the varying thickness of the different numbers of rods used.
    By doing this, the various CF rods are being stressed in their largest dimension, not relying as much on the bond joint between rods to keep from buckling (yet still captured between plywood spar cap and ribs).
    IS THIS A BETTER IDEA?

    Third question:
    What is the simplest/best way to have automatic control surface connections/engagement on a light wing?
    I've seen how various gliders do it, but most of these are not simple to fabricate. I thought about a basic torque tube for flaps and/or speed brakes, and push-pull cables for the ailerons (I know this will add some friction, but I'm not sure that is a bad thing on such a light plane).
    maybe even UT's proportional actuators?


    Feedback from yourself and others would be greatly appreciated.
    Once I have come to terms with these options, I may likely build a POC wing and see how it does.


    Bonus question:
    With the precision and ease of waterjetting out ribs and spars, I'm very intrigued with the idea of building a complex wing, using both leading edge slats and drop hinged flaps (true fowler seems a bit overkill).
    We can easily frame up these components with ply ribs, spars and skins - and it would be a fun exercise in high lift systems for very light aircraft (like the Suitcase Airplane I have).
    • DO YOU HAVE ANY ADVICE FOR HOW TO DO THIS? WHAT AIRFOILS TO USE, ETC? (or specific reasons why not to do this, beside being more parts and more complicated?)
    • CAN THE SPRING LOADED SLATS BE ON JUST THE INBOARD HALF OF THE WING, OR DO THEY NEED TO GO ALL THE WAY ACROSS? IF ACROSS, DOES IT NEED TO BE ONE PIECE/SYSTEM?
    • WHAT SHOULD THE SLAT PROFILE LOOK LIKE?
     
    Last edited: Aug 23, 2010
  2. Aug 24, 2010 #2

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    So far so good. the trick is in attaching the carbon rods - make sure to calculate all the shear flow in order to make sure the carbon does not separate from the wood.

    I guess I'm not sure what you mean when you say your web will be a geodesic design. In the previous paragraph you seemed to have indicated that the web is made of ply. But in general, the more contact area for each pultruded element the more stability it will have. Personally I probably would not finish the top off with the ply - I'd just bond the skin there instead. As far as fore and aft face sheets are concerned, yes those will provide some stability however keep in mind the rather substantial difference in moduli. It may be that if the carbon is about to fail and buckle, the wood veneer may not provide sufficient stability unless fairly thick. But then without seeing any numbers, I'm just guessing.

    Yes, this actually does sound better. I might even go so far as to lay maybe two layers of a light glass or graphite over the faces to enclose the pultruded material and coat the web face.

    But at this point I should also say that I haven't done much with this structural type. Most folks refer to Alex Strojnik for this type of spar design. The wood/graphite spar i did for our concept utilized a box where the fore and aft faces were the thin webs and the caps were mounted in channels that were cut into the top and bottom of each web face. This thin ply cross-piece formed a channel at the top and bottom of the box into which the uni graphite was laid. But this could also work well with the pultruded material.

    Personally, I think the simplest is digital/proportional electro-mechanical actuators. But if we're talking more conventional, i use pinned push/pull tubes using rod end bearings at the ends. A simpler system uses ball links - these can be permanent joints or there is also a model that is easily removable. The added friction from cables is generally negligible so i wouldn't really worry about it but, if you do, push/pull linkages are better. I would tend to stay away from torque tubes (unless they're short) since they could introduce a bit of springiness into the system, which might be a potential source of flutter.

    Personally I would not recommend this unless you have a lot of time to waste - given the complexity, I don't think you could ever make it work really well in a homebuilt environment. No, with enough effort it's not impossible but given that simplicity has been your goal in all the discussions, this seems like a needless amount of complexity, with only relatively minimal payback in performance, except of course the gee-whiz factor. But to your questions:

    Just about any section will work of course but given the volume necessary for the components, obviously it would be better to use a thicker profile so you have a bit of room to work in. Obviously this depends on the chord length but I'd say the thicker the better - 15% should be about right.

    Properly designed, the slats don't need to be spring loaded. Look at the slats on the Helio Courier.

    The slats should go all the way across or only on the outboard section. Putting them only inboard is a no-no. And whether they're in one piece or more will depend on the flexibility of the wing. It will also depend on the design of your track mechanism and/or linkage.

    Your last question is pretty much impossible to answer. It would have to be designed for the mission and section selected. Again, for ideas look at the Helio.
     
  3. Aug 24, 2010 #3

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,540
    Location:
    Rotterdam, Netherlands
    The main problem with push-pull connections is that they're prone to non-connecting. That has caused the dead of a significant number of glider pilots and I'd stay away from them if I could (even with the Wedekind-sleeve)
    For lightly loaded actuators you can simply use a pin to connect two surfaces like in the outer surfaces of a duo Discus which then acts with the "main" surface, or only in one direction (less adverse yaw)

    For highly loaded ones I would go with the new type:
    [​IMG]
    And fourth picture
    Here, the "receptor" pivots around the front axis and the wing protrusions, connected to flaps and ailerons fit in and turn around almost the same axis. That's fool-proof and quite compact.

    For slightly sloppy (but very easy) connections, a torque tube with a fitting groove for the control surface works too, but that's usually too sloppy for ailerons (flaps might work)
     
  4. Aug 24, 2010 #4

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    I'm assuming autoreply is referring to the ball-links in his comment about disconnecting. I've heard of that but have not seen it. But there are aerospace grade units that should not have these issues.

    Rod ends used for push-pull do not have these issues but the are not as easily disconnected.
     
  5. Aug 24, 2010 #5

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    6,308
    Likes Received:
    2,266
    Location:
    Rocky Mountains

    I like the method Sonex uses for their Onex. Part count is high but with CNC this isn't a real problem. It also needs adjusting or built on factory jigs.

    2:00 min

    [video=youtube;OTxIC7MTNIk]http://www.youtube.com/watch?v=OTxIC7MTNIk[/video]see-http://www.youtube.com/watch?v=OTxIC7MTNIk

    If routed well a Morse cable has very little friction and there is no need to disconnect anything.
     
    Last edited by a moderator: Mar 20, 2019
  6. Aug 24, 2010 #6

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,540
    Location:
    Rotterdam, Netherlands
    In fact, the aerospace grade ones had those issues (in fact two, wear and, while it looks locked, it really isn't locked) and still have them. Most glider pilots consider them unsafe and avoid them as much as possible.
    A friend of mine had a rod end disconnect in flight because it failed. He came down safely (with one aileron), but even they aren't fully fool-proof, nor are they completely safe from inflight failure. Rumor was that new gliders wouldn't be approved any more if they used any of the above methods because of the too high risk of failing in flight and/or erroneous connection when putting the wings on.
     
  7. Aug 24, 2010 #7

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    Interesting - what's the alternative then?
     
  8. Aug 25, 2010 #8

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    Thanks for the feedback.
    I am going to reply to several posts in a row, and ask a few more questions.

    That of course is my main concern, and I'm not sure how to go about calculating it (since I'm a 'fake' engineer)

    Yeah, that's a bit confusing; but if you take a look at the way we designed the main spar on the TS-2, it is cut out of plywood with most of the material removed (I was trying to describe that as geodesic, but it essentially is a bunch of diagonal bracing). Though it is cut out of solid ply, it effectively ends up as 1/2" cap strips connected by 1/2" intercostals and cross bracing.

    With the pultrusions we are using (.470"x.40") there is a fair amount of contact between the rods, as well as surface area glued to the top of the wood spar.

    I was contemplating adding the wood strip to cover the CF pultrusions because they are so thin. At the root, the thickest part is only about .32", and at the tip it is .040". Of course, this is glued down to the top of the ply which is a virtual wood cap as well, but by adding some wood above, it gives that much more surface area for the thin web ply to glue to. I didn't think that gluing the additional .030" plywood webs would have enough contact area with the .040" thick rods.


    I really don't want the builder to have to do any real composite work; gluing together the CF pultrusions and then to the wood spar cap is about as difficult as I want it to get (except for fabric covering - and I'm even considering ONLY wood covering for the whole thing).

    This is kind of what I'm proposing. The 1/2" plywood center spar (most material removed between 1/2" spar caps) would then be sandwiched with .030" plywood; those outer plys would extend past the center wood caps, capturing the CF pultrusions.
    What i am considering is adding another piece of wood on top, having the thin outer webs extend further to these as well, and completely encase the CF rods.

    I have no experience with digital proportioners, but it sounds very cool.
    However, what I was referring to regarding the push-pull cables is the Morse or Teleflex style - this is what we are using in the Solo... very little friction loss, and extremely easy to install. In the Solo, we are using differential ailerons, which is simple geometry at the stick pivot - the only other connection is at the aileron itself.
    Of course, with two piece wings, we may need a center cross bar, but maybe we could just have a foot of the cable hanging loose when the wing is pulled out.


    For the most part, simplicity is one of my main objectives. However, for these POC wings I thought it would be kind of fun to try to incorporate the high lift stuff, just to pay with.
    Actually, the Helio IS my reason for asking about this - and it is my understanding that the Helio slats ARE spring loaded... it is quite disconcerting at times to hear them slam in and out at high AOA when the springs push them out and fight with the wind wanting to push them in.

    I think some of these devices are pretty cool - especially when I see the stuff put on the 12' wing span of the 'Suitcase Airplane'... triple slotted flaps, leading edge slats, spoilers, double slotted ailerons - an amazing combination (and sophisticated geometry) for a plane that was essentially built with HAND tools!

    The slats would not end up on the production plans (if this design ever got that far)... again, just something to monkey around with.
     

    Attached Files:

    Last edited: Aug 25, 2010
  9. Aug 25, 2010 #9

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    Okay AutoReply, your turn.
    (BTW, I'm kind of Dutch - even learned to walk in wooden shoes and grew up wearing lederhosen - really!)

    I really can't make out the picture to see what's going on, and I didn't see the link (?). I have seen the method being used on the HP-24

    [​IMG]


    I can't tell if what you are proposing is similar. If this is the way to go, I'd be tempted to make an injection mold for it, because fabbing it any other way would be tedious.

    Isn't that how Van's does the flaperons on their RV-12? Is it really that sloppy?
    That is what I figured I'd do with the flaps, although there are other systems I've seen as well.
     
  10. Aug 25, 2010 #10

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    I believe that is the same method as used on the Corsairs and most folding wing fighters of WWII. Yes, I've thought of those too, but a lot of extra weight.

    I'm not sure what the best answer is yet... more discussion might help (me).

    And yes, Morse cables is essentially what I was originally referring to when I said push/pull cables (as opposed to push-pull rods).
     
  11. Aug 25, 2010 #11

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,436
    Likes Received:
    894
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    Autoreply,

    Do you have any web links to accident reports with the L'hotellier link failures that weren't caused by incorrect assembly or excessively worn links? All the ones I have seen hadn't been connected properly when the glider was rigged.
    The reason I ask is that I am using these in my CriCri (although they are in a good location for visual inspection.

    Shannon.
     
  12. Aug 25, 2010 #12

    orion

    orion

    orion

    Well-Known Member

    Joined:
    Mar 3, 2003
    Messages:
    5,800
    Likes Received:
    135
    Location:
    Western Washington
    As far as i can recall the Helio slats are not spring loaded - the track geometry allows them to be pushed back by the wind in level flight but at low speed and higher angles of attack the leading edge suction pulls them out. On the ground you should be able to push them in with one finger. The bang you hear when they deploy is from them being sucked out, not from any internal springs. I got to fly a Courier up in Alaska and years ago i also had the privilege to work with the gentleman who designed the mechanism.
     
  13. Aug 25, 2010 #13

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,540
    Location:
    Rotterdam, Netherlands
    The excessive wear is the problem. I heard rumors that (with vibrations of an engine) some people worn them down in only a couple of days of flying) Since that wear isn't that visible (you have to measure it) nobody noticed until it was too late. I measured a couple of links that were worn down in a single year of flying (about 200 hrs) from new to over the limit.
    If you don't put it together frequently (keep wings attached) simply measuring them regularly is sufficient (according to the AD).

    If you put it together every flight... good luck. In gliding we have the pilots that have had a disconnected surface in flight, those who are going to have one and those with safe automatic connectors.

    Sooner or later you will forget one, or it won't be connected safely. I've seen a fair share of people getting back to earth alive and a couple who dug into the ground.

    Lederhosen are German ;)
    It looks similar. I absolutely failed to find any good pictures of this type of connection. The only way to go is have a look at any modern glider (Discus, Ventus, Nimbus, Duo, LS4b/6/7/8/10 ASW-24 and higher) Alternatively you could inform by DG Flugzeugbau, Alexander Schleicher or Schempp Hirth for some more info.

    Here are some nearby gliders located:
    http://www.onlinecontest.org/olc-2.0/gliding/airfieldRanking.html?rt=olc&sp=2010&st=olc&c=US&sc=12

    No idea whether it's identical. It consists of a hollow pipe that can torque. It has two V-shaped grooves, opposite each other.
    The wing has a slightly smaller pipe, with a much smaller rod, drilled through the pipe.
    Because of the V-shape it automatically slides in the correct position and can't be forgotten.

    I'd only apply for the two systems discussed above, despite their drawbacks.
     
  14. Aug 25, 2010 #14

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    Yes, I grew up on the border - actually spoke more German than Dutch - but now can't speak either :ermm:

    But to get back on subject, this thread is primarily about the spars and wing construction.

    Going back to your comment on the other thread - I need to nail down the airfoil shape before I redraw all the bits and pieces (it takes time to build these jigsaw puzzles).

    I'm also interested in how to best make speed brakes.

    And then there is the issue of wing tips. I am currently intending on just making the straight wing for now, with a means of attaching various wing tips (different lengths and styles), but recommendations now would be appreciated.
     
  15. Aug 25, 2010 #15

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    I'll probably save this challenge for another day.

    Right now I need to focus on trying to get wing parts cut - hopefully to take to to the ESA Western Workshop in Tehachapi, so we can try to put it together while there :)
     
  16. Aug 25, 2010 #16

    drake

    drake

    drake

    Well-Known Member

    Joined:
    Dec 23, 2007
    Messages:
    77
    Likes Received:
    0
    With full flaps down on a helio and then the slats come out there is a bunch of drag and without power the helio sinks like a rock. In that full flap slats out power is needed, a bunch. You may look at backcountry cub they have a slat used on a usa35 b foil that is fixed except it tilts or weathervanes with the angle of attack which opens and closes the back gap of the slat. The foil shape compared to a helio is very close i have templates for all if needed. The down side to the slat on a cub is in order for it to do anything noticeable stol wise you cannot see over the cowling so in bush operations it is useless in my opinion. but just playing around on a sunday with it you can fly it around in some really steep angle and turn around very steep at slow speed with no stall spin characteristics.
     
  17. Aug 25, 2010 #17

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    i would love to get the coordinates and/or geometry you refer to.
    It won't be usable on this particular project (glider) but i would like to explore it for a lightweight pusher design... great visibility and get in/out of anywhere like the Storch would be awesome. Maybe even on an ultralight!
     
  18. Aug 25, 2010 #18

    drake

    drake

    drake

    Well-Known Member

    Joined:
    Dec 23, 2007
    Messages:
    77
    Likes Received:
    0
    on the fixed style slat i have made mold to build graphite slats which are a top skin, bottom and left and right ribs with attach fitting built into the rib. and a mounting design to fit a ribblett 66 style wing which can be taken off in minutes using nut plates in the LE of the wing. That will give you options, one main reason i like it is i operate in the bush and i always hit small brush ,ect with the LE and this does not dent and gives you a so called buffer between the wing and brush. PM me and i can get an address to send info.
     
  19. Aug 25, 2010 #19

    autoreply

    autoreply

    autoreply

    Moderator

    Joined:
    Jul 8, 2009
    Messages:
    10,732
    Likes Received:
    2,540
    Location:
    Rotterdam, Netherlands
    Isn't too hard ;)
    Decide beforehand whether you want elongations possible so you can account for the required spar strength.

    Make sure you take into account the abuse of putting the glider together, because when you do so, you lift half the wing's weight at the tip!

    I'd prefer winglets over extensions. They increase directional stability, give better stall and on such a design, comparable performance. Their design is rather complicated though, especially aligning them with the airflow.
     
  20. Aug 25, 2010 #20

    Rienk

    Rienk

    Rienk

    Well-Known Member

    Joined:
    Oct 11, 2008
    Messages:
    1,364
    Likes Received:
    190
    Location:
    Santa Maria, CA (SMX)
    Thanks for the suggestions.
    Yes, I've done the spar cap analysis for doing the extensions, and it definitely makes a difference.
    But since I'm limiting the length of everything to be under 19 feet, some people may want more span - or in my case - be able to lower the wing loading to capture lighter lift/thermals.
     

Share This Page

arrow_white