Revisting thicker skins for laminar flow

Discussion in 'Sheet Metal' started by LHH, Jan 17, 2020.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Feb 9, 2020 #61

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,167
    Likes Received:
    3,468
    Location:
    Saline Michigan
    I know that this level of smoothness can be achieved in wings and tails on sailplanes and powerplanes that are filled with dry micro and then carefully profile sanded with long sanding sticks. That means cored composites and maybe plywood skinned stuff too. But in aluminum? Really skeptical... Even if you could build the wing accurate enough and with small enough chordwise waviness, you still have the flexibility of the skin...

    Billski
     
    stanislavz likes this.
  2. Feb 9, 2020 #62

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,167
    Likes Received:
    3,468
    Location:
    Saline Michigan
    Still working to understand what goes on in the flow around an airfoil. And relized that I missinterpereted the above quote from Mr Billski. Still having trouble thinking it through. Here goes:
    As described air flows around the wing, pressurising slightly on the "bottom" and creating a bit of vacume on top;
    Meanwhile on the inside of the wing, these pressure differences acting on the wing skins, push the "bottom" skin in and pull the "top" skin out, creating an appearance of inflation;
    I think that there isn't a relationship between the aerodynamic pressure of the flow around the wing and the actual air pressure in the wing.
    Still messes up laminar flow.[/QUOTE]

    We gotta do some more basics on this. My wing stuff comes out of Theory of Wing Sections by Abbott and von Doenhoff, and a good half of the book is a catalog of NACA data on airfoils. Inter-library loan works, and so do the bookselling services. I specifically cited page 371 for my post above, and I included it here.

    Appendix I shows normalized velocity data around a bunch of airfoil thickness distributions. All foils using this thickness distribution will have these velocity profiles at the indicated Cl's. On the bottom of the plot is the shape of the foil, above that are three lines. The middle one is the velocity profile with the foil at zero AOA, which also means zero lift for an uncambered foil. Notice that at x/c = 0, velocity is zero, this is the stagnation point. Right there, a single molecule might go stationary. Everywhere else the air is moving aft. We will get into lift later, right now let's just stay with the zero lift zero AOA part of this. The pressure of moving air on a stationary surface at any given density is less than the pressure of the same air when stationary. When we wrap this thickness distribution along a camber curve, the velocity curves go right along with it...

    A guy named Bernoulli figured out (in the 18th Century) what the math was for relating this when flows are relatively smooth, and even allowed changes in static pressure density while including kinetic energy, and potential energy. In its simplest form, with density and height staying essentially identical, the change in pressure is 1/2*rho*v^2, which we know as q or dynamic pressure. Set a velocity V over the wing, look at a spot on the wing, say .2C, read off the (v/V)^2 of 1.359, multiply V by 1.359, and that is how fast the air is going by is. At standard temperature and pressure, rho (density of the fluid) is 0.002378 slug/ft^3. Plug it into the equation and we know how much pressure the flowing air imparts on the foil skin at that point.

    Big picture? All of the skin except a little bit right around the stagnation point is moving and pushing on the skin less than if that same air was stationary. The top skin is being pulled upwards, the bottom skin is being pulled downwards, and we assume that they are being held in shape to each other and connected to the wing. With the wing at zero AOA, there is no lift.

    Change angle of attack enough to get a Cl of 0.22, and you get the velocity profile of the top line on the upper surface, and the velocity profile of the lower line on the lower surface. You can compute pressures at all points along both surfaces. The bottom skin is now making less force downward than the upper skin is making upward. The difference between summed forces on the bottom and the top is EXACTLY the lift being generated. And both wings are still being pulled away from each other, with the difference between them being the lift made.

    Increase the Cl, and the top and bottom curves will be further from the middle line. At some point you will be able to predict (v/V)^2 going negative. Since even a negative velocity when squared is still positive, we know that the (v/V)^2 can not become negative, but the stagnation point does slide down from the foil nose onto the lower skin as AOA increases, putting the little area of increased pressure on the bottom skin near the leading edge. Nonetheless, at high but unstalled AOA, a very large fraction of the wing keeps operating at lower pressure than static, and the differences become bigger and bigger between top and bottom surfaces as the AOA is increased and as overall velocity is increased. Let's stay away from compressibility effects until you get the basics under control.

    The bottom of the wing is generally not at increased pressure, just less pressure reduction than on the top of the wing. At big positive AOA, you can get some of the bottom skin around the front of the wing at increased pressure, but a big fraction of the foil is still wrapped in lowered pressure...

    Now let's get into air inside the wing... I talk about stationary air inside the wing and moving air outside the wing. The air inside is usually not stationary, because the wing leaks. Air is coming in some places and leaking out elsewhere, but the velocities are usually quite low. Since pressure change is a function of v^2, low velocities produce pressures very close to that for stationary. So I call it stationary air. The moving air presses on the outside of the skin less than the stationary air presses on the inside of the skin. If the air inside the wing is at the same pressure as the static air before passage of the foil, then we can know pretty much what the pressure is on both sides of the skin and at any spot along the wing. Not only that, if we add or remove pressure on the air inside the wing, we can figure out the new pressure difference too.

    If the air inside the wing is the same pressure as the static air elsewhere around the wing, the pressure inside the wing is higher than the moving air around the wing and we are actively inflating the wing with a small pressure, bulging the skins. If we were to seal up the wing interior and draw a vacuum on that volume, we could decrease the inflation effect or even cancel it. Neat idea. But look at the curves again and see that it would be complicated to cancel pressures on both skins over much range of Cl and chordwise position. Then if the vacuum source causes more drag than you will save, well, you have a heavier more complicated airplane that does not fly as well as it did when lighter and simpler...

    Back to the world of reality.

    Many of us store fuel in portions of the wings - and usually forward portions where we might want laminar flow. The fuel is frequently in contact with the wing skins and with a forward facing vent line slightly pressurizing the top surface of the fuel. Then the column of fluid adds more pressure as you go through its depth. Add some g's and the pressures inside can be significantly higher than the static pressure I was postulating at the beginning, so the wings get even more "inflation" effect where you have fuel.

    The sailplane guys use "hollow" wings too. Since maintaining laminar flow is tantamount to religion with the sailplane guys, you had better believe that they do not allow "inflation" to cost them a single pound of drag if they can help it. The sailplane guys with fancy birds flying on days with strong lift will store lots of water (denser than fuel) in many places with the water in contact with wing skins, etc. What do they do to maintain long runs of laminar flow over their wings? Carbon fiber facings on both inner and outer sides of foam for high skin bending stiffness. That, right there is what I call a Not So Subtle Hint.

    I hope that this long note helps with understanding...

    Billski
     

    Attached Files:

    • 024.jpg
      024.jpg
      File size:
      47 KB
      Views:
      7
    Last edited: Feb 10, 2020
    dog, stanislavz and Pops like this.
  3. Feb 9, 2020 #63

    LHH

    LHH

    LHH

    Active Member

    Joined:
    Aug 1, 2013
    Messages:
    41
    Likes Received:
    6
    Location:
    IL
    Yes, very helpful, the consensus is composites are superior( and true for the most part), but factory composite sailplane planes rarely meet the standard for laminar flow without some work. These factories have years of experience and struggle with this standard. What factual data can be found seems to suggest that laminar flow is going to be very hard for a homebuilt plane even if you go with composites. Both aluminum and composites designs are going to require a lot of extra work/planning to come close to this standard. The biggest educational part for me knocking down the notion that if you went composite you were guaranteed better results.
     
    Last edited: Feb 9, 2020
  4. Feb 9, 2020 #64

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member Lifetime Supporter

    Joined:
    Jan 24, 2011
    Messages:
    4,416
    Likes Received:
    2,072
    Location:
    US
    I'm surprised anyone would say or surmise that a composite structure "guaranteed better results.".
     
  5. Feb 10, 2020 #65

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    7,352
    Likes Received:
    2,119
    Location:
    North Carolina
    Don't forget, laminar flow is easier to achieve on the underside, the pressure gradients tend to be more favourable. Look at Billski's 652A015 graph. The lower surface has a rising velocity until 50% chord. That makes it much easier to maintain laminar flow.
     
  6. Feb 10, 2020 #66

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,167
    Likes Received:
    3,468
    Location:
    Saline Michigan
    None of us have implied that composites are guaranteed better results. But the caution is appropriate for the uninformed.

    Skin stiffnesses of cored composites will mean that if the waviness at rest meets requirements, it will also be stiff enough to stay within requirements in flight. So extensive laminar flow with composites will require filling, fairing, and profiling. So will aluminum, but it will also likely require careful design to keep deflections small enough to maintain laminar flow.

    Billski
     
  7. Feb 10, 2020 #67

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    10,242
    Likes Received:
    6,964
    Location:
    97FL, Florida, USA
    Has anyone calculated the theoretical difference in top speed of (for example) an SX-300 as built verses with laminar flow?


    BJC
     
  8. Feb 10, 2020 #68

    dog

    dog

    dog

    Well-Known Member

    Joined:
    Dec 29, 2019
    Messages:
    105
    Likes Received:
    44
    Location:
    lat:45,long:-64
    We gotta do some more basics on this. My wing stuff comes out of Theory of Wing Sections by Abbott and von Doenhoff, and a good half of the book is a catalog of NACA data on airfoils. Inter-library loan works, and so do the bookselling services. I specifically cited page 371 for my post above, and I included it
    Let's stay away from compressibility effects until you get the basics under control.
    but a big fraction of the foil is still wrapped in lowered pressure

    I hope that this long note helps with understanding...

    Billski

    Again Thanks
    That cleared up a few,(slightly embarsing) points for me, and put me in the advantagious position of knowing some of what I dont know,and the need to know it.
    I am taking notes,and working up a list of books to
    aquire.
    Here we are ,centuries in from Bernoulli and its still
    one brain at a time,kinda like shingling a house.
    David
     
  9. Feb 10, 2020 #69

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    10,242
    Likes Received:
    6,964
    Location:
    97FL, Florida, USA
  10. Feb 10, 2020 #70

    dog

    dog

    dog

    Well-Known Member

    Joined:
    Dec 29, 2019
    Messages:
    105
    Likes Received:
    44
    Location:
    lat:45,long:-64
    Yup.
    Keep going back to the stickies.
    I would add that the search function is invaluable for answering specific questions,if you know what to ask....
    That bieng said
    Living in a remote location,with a single device(phone) for internet,(only other option is dial up,not kidding) and a limited budget.
    Some of the online quoted prices are realy high.
    Others are going to be facing greater hurdles,even here in Canada,let alone countrys with "problems" importing things.
    So the real value of the information provided on this site is very high to some of us.
    And very much apreciated.
     
    Last edited: Feb 10, 2020
    BJC likes this.
  11. Feb 10, 2020 #71

    lr27

    lr27

    lr27

    Well-Known Member

    Joined:
    Nov 3, 2007
    Messages:
    3,581
    Likes Received:
    542
    I may have messed up my calculations, but I get that if you reduced the wing profile drag coefficient by .003 at sea level at 230 mph, you'd save 18 thrust horsepower, which I'm guessing wouod mean 22 or 23 shaft horsepower. The wing is only 70 square feet. .003 is just a conservative guess based on a change of airfoils. If the existing wing isn't as precise as a laminar wing needs to be, then the difference couod be much more.
     
  12. Feb 10, 2020 #72

    lr27

    lr27

    lr27

    Well-Known Member

    Joined:
    Nov 3, 2007
    Messages:
    3,581
    Likes Received:
    542
    After looking at Theory of Wing Sections, I think a difference of .002 is more like it, assuming we had an SX-300 with a "turbulent" airfoil as a baseline. As it turns out, according to the Incomplete Guide to Airfoil Usage, the SX-300 already has a NASA NLF section. I think it's supposed to be a metal airplane, though, so I don't know how much difference that makes. I'm sure that on a slower airplane with a larger wing, the difference between a laminar and turbulent airfoil would be more significant.
     
  13. Feb 11, 2020 #73

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    10,242
    Likes Received:
    6,964
    Location:
    97FL, Florida, USA
    The SX-300 is metal. It has a stretch formed leading edge. IIRC, the LE is 0.40” before forming.


    BJC
     
  14. Feb 11, 2020 #74

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,167
    Likes Received:
    3,468
    Location:
    Saline Michigan
    Assuming 270 hp, saving 23 hp is saving about 2 gph. Or if you use the same power settings you can go 3% faster or 6 mph faster.

    This shows us how modest the effect of laminar flow on a plane with high wing loading (small wings) is. There are much bigger fish to fry around a power plane if you are going for drag reduction.

    Go to an airplane with much lower wing loading, no engine cooling, smaller fuselage cross sections, and the drag reduction of getting extensive laminar flow become much bigger and much more important.

    Billski
     
  15. Feb 11, 2020 #75

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    10,242
    Likes Received:
    6,964
    Location:
    97FL, Florida, USA
    That closely aligns with a (Boeing) computer analysis of a Glasair III with an ideal airfoil (LS(1)-0413, Modified) verses a postulated airfoil that was 1/8” low for about 10” centered over the spar.

    Point is, extensive laminar flow may be important for same airplanes, but for the popular fast HBA such as the Glasair, Lancair (both molded sandwich skins) and the SX-300 (stretch formed aluminum wing leading edge skin) it is not terribly important.

    Now for a race plane or unlimited sailplane ....


    BJC
     
  16. Feb 17, 2020 #76

    lr27

    lr27

    lr27

    Well-Known Member

    Joined:
    Nov 3, 2007
    Messages:
    3,581
    Likes Received:
    542
    Hmm.. I suspect that extensive laminar flow on the fuselage would make a big difference. At least if separated flow has already been taken care of. I'm guessing fuselage drag is greater than wing drag. And then there are the tail surfaces. Seems like, to go fast, you have to play whack a mole with sources of drag. Get rid of one and the next worse one rears up before the speed gets much faster. OTOH, if you reduce power needed by 23 hp on the wing, AND 23 hp on the fuselage and tail, that's starting to add up. If engines weigh 1.75 lbs per hp, you can use a smaller one and save 80 lbs or so. Plus the cowling can be smaller and less draggy, and you'll have LESS cooling drag since you need less cooling air. And less fuel will be required, meaning a lighter airplane and smaller wings.... Then you can try to optimize the cooling system for less drag. I understand that for many light aircraft, cooling drag is a significant fraction of the total.
     

Share This Page

Group Builder
arrow_white