Foil Cavitation at Lower Speeds Than Expected

Discussion in 'Hydrodynamics and Aerodynamics' started by Doug Halsey, Aug 11, 2015.

  1. daiquiri
    Joined: May 2004
    Posts: 5,371
    Likes: 258, Points: 93, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    For heaven's sake, Doug, we are not interested in those. :p

    Ok, jokes apart - you seem to be thee right person for the following question, which is my personal curiosity:
    How would you objectively judge the physical (or psychological) effort required by this type of sailboat, compared to classic ones, either mono or multihulls of the similar size? We are used to see them sailed by young professional top athletes, so I would really be interested to hear an opinion from a not-quite-a-teenage next-door guy - though you really do seem to be very agile and in a remarkable physical shape in these videos. :eek: :) For comparison, I am 43 and already have big problems with my back, which have made me abandon my former big love - windsurfing.

    Cheers
     
  2. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    I'm not sure I'm really the best person to answer that question, but I'll give it a shot. I was an avid dinghy sailor all throughout the 60's & 70's & part of the 80's. I raced Moths for 10 years & Windsurfers for 10 years, but I also sailed FD's, Thistles, Windmills & other things. They had great competition, but by today's standards, they were not really very demanding physically.

    Then in the mid-80's, I quit sailing altogether & did other things until I started building Broomstick in 2002. My physical condition was OK compared to most people, but you would never mistake me for any kind of athlete.

    In that sort of condition, I found sailing Broomstick pretty easy. I couldn't just kick back & have a beer, but body positioning, sail trim, etc. were not especially critical. The hardest part was carrying all the pieces to & from the water & rigging & unrigging.

    When I got my Mach2 last fall, it was another story altogether. I had never sailed a boat that tippy, where the correct angle of heel was so important (not just for speed, but for staying upright), the mainsheet has to be trimmed continually & it's critical to maintain concentration all the time. Nothing I had ever sailed before required so much to sail it. I've worked hard to gain weight, strength & conditioning just to be able to sail the boat casually, but I'll probably never be very competitive if I try to race it.

    But there's another Moth sailor that I frequently sail with who makes it all look so easy. He's only sailed the boat for about a year, but he's incredibly smooth. Matt Struble. He recently won the A-Cat North Americans (winning 8 out of 10 races), in a borrowed boat, never having sailed a foiling A-Cat at all before. According to him, that boat is much more difficult than the Moth.

    So, to sum it up : I still think the Moth is very demanding, but apparently the foiling cats are too (at least some of them). However, boats like Broomstick are a piece of cake.
     
  3. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    You have a very acute angle between your foils which produces a locally accelerated flow near the junction. That will be cavitation city, no matter what section you choose.

    You can think of the velocities from each foil being superimposed on each other. Where the local velocity is high, you're going to get a double dose from each panel. There are several ways to attack this problem.

    One way is by making the junction wider. A rounded junction will help, as will a wider apex angle. This moves the panels away from each other, so their interference is less.

    Another way is to tailor the section shapes toward the junction. The maximum velocity on the suction side of the section needs to be lowered, as though you were designing for a higher cavitation speed. There's a limited amount that you can do this way, but it helps.

    Another thing that's been used in the past is a Coke-bottle shaped fairing. The idea is the fairing has a dumbbell shaped pressure distribution that rises to a peak near the fat parts and then has a higher pressure/lower velocity valley in between. By arranging the thick parts of the fairing to be near the leading and trailing edges of the foil, you're superimposing the fast parts of the fairing on the slow parts of the foil, and the slow parts of the fairing on the fast parts of the foil. The goal is to smooth out the pressure variation from nose to tail through the junction, and to reduce the superposition of the fast velocities.

    Something else you can try is to stagger the panels. This falls under the heading of "everything shouldn't get fat (or fast) at the same place." If the stagnation pressure at the leading edge of one panel were aligned with the peak pressure of the mating panel, or the increased pressure toward the trailing edge were aligned with the fastest parts of the mating panel, then you'd be reducing the superposition of fast on fast.

    As to which panel to move forward and which to move aft, I'm thinking the leeward foil is the most critical one. Given that the foil is loaded in side force as well as vertically, the outboard panel is more heavily loaded than the inboard panel, and in many ways the inboard panel is acting as a winglet for the outboard panel. My intuition says it would be better to move the inboard panel aft compared to the outboard panel. But I don't have anything to back that up.
     
  4. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    Thanks for your inputs Tom.
    You're not the 1st one to point out the problems that the steep angle of the V could cause, but markdrela did a remarkable job of coming to my defense :
    To that, I would add the following :
    If I used a shallower V, but kept the foils the same width, then the clearance of the hulls above the water would be smaller than I wanted, unless I lengthened the vertical portions at the upper tips of the foils, which would sacrifice the rigidity of the triangular shape. I could have made the foils wider, but the boat was already oversquare (15' length & 17' beam) & I didn't want to make it even more so. I could have moved the foils inboard on the crossbeams, but then my waterstays' attachment points would also have to move inboard, possibly requiring a heavier crossbeam. And so on. Lots of compromises go into a boat's design. The 1st photo below that gives a little different perspective on sizes of the foils & how everything fits together.

    And of course, at that time I thought that cavitation would only happen at speeds well in excess of what I would ever accomplish.

    The suggestions on how to improve the flow near the apex are all very interesting. As far as I know, all of the recent catamaran designs that use uptip foils have opted for rounding off the apex. That seems to be a very elegant solution that I have already been considering. The figure below shows a family of hyperbolas (with a V as a limiting case) that all have the same width & depth. I'm trying to decide how far I want to depart from the V. More round reduces the chance of cavitation (as discussed above), but there will also be some loss of heave stability when the boat is flying high. More compromises ...

    The other methods that you mentioned would no doubt be simpler for an amateur builer & would probably have to be used if we were dealing with T foils.
     

    Attached Files:

  5. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    I'm not questioning your choice of a steep angle. Every design decision has its advantages and its drawbacks. The narrow V is evidently working for you, except for the cavitation issue.

    Perhaps the simplest option at this point would be to cut off the bottom of the V and add a flat piece connecting them together, changing the V to a trapezoidal U. You would lose a little of the depth, but it's probably not a part of the foil that is really working well for you at present. The junctions between the added piece and the foil panels would be an obtuse angle, with much less interference.

    You could start with cutting the foils just above the fillet in the current junction, which would minimize the loss of the tip. If that didn't work, then you could cut it again a little higher. You could experiment with just the foil on one side, and when you got that one working the way you wanted, then modify the other side to match. That would save you some effort in doing repeated modifications.

    You could still have the vertical extension shown in the photo. Just add it to the outside junction with the horizontal piece.
     
  6. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    Other Factors To Consider :

    In Post#7, on Page#1 of this thread, I mentioned that I had only experienced the presumed cavitation shown in the video a small number of times, even though the boat had reached much higher speeds. I then discussed how varying the incidence of the aft foil determines the amount of main foil that is in the water, which has a significant effect on the likelihood of cavitation.

    My current aft-foil control is adjustable while sailing by twisting whichever tiller extension is being used (there are 2 of them), but the threads in the screw mechanism are too fine to be able to respond to short-term speed variations. It is difficult to estimate the immersed foil area very precisely from the video, because of how much it changes from frame to frame, but my opinion is that I was flying higher than usual for the speed I was going.

    I also mentioned earlier that I'm currently sailing a Mach2 Moth. Its screw mechanism for aft-foil adjustment has a much coarser thread, allowing much faster & more frequent use. I've already taken the mechanism apart & seen how I can adapt it for use on Broomstick.

    Something that hasn't yet been mentioned in this thread is the effect that the surface condition of a foil has on cavitation. In order for cavitation to be initiated, it is necessary that minute nuclei of vapor and/or other dissolved gases be already present in the fluid. The irregularities & crevices associated with roughness of the surface can greatly increase the concentration of these nuclei & lead to earlier cavitation than on a smoother surface. At the time of the video, Broomstick's foils were in worse condition than at any other previous time. They have since been recoated & polished to a much higher standard.

    I really believe that improving these 2 aspects (smoothness & aft-foil adjustability), by themselves will be a big help in delaying cavitation & I'm hopeful that I'll soon be able to exceed 30 knots (if I can tear myself away from the Moth long enough). But I'll keep trying to improve the basic foil design, too.
     
  7. markdrela
    Joined: Jun 2004
    Posts: 307
    Likes: 30, Points: 28, Legacy Rep: 324
    Location: MIT Aero & Astro

    markdrela Senior Member

    For the symmetric airfoils that you have on your foils, cavitation onset depends very strongly on the CL, not just on the boat speed. And the CL depends on the foil submergence depth (via foil incidence), on the point of sail, and also on how much the sail is sheeted in. So whether you have cavitation depends on many factors in addition to the boat speed.

    For a cambered section like the H105, the cavitation onset will still depend on CL, but to a much lesser degree.
     
  8. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    Yes, I should have worded it more carefully. Regardless of all the factors though, at any given time while sailing there should be an optimum combination of both sail trim & foil trim. And I don't think my foil trim in the video was very close to optimal.

    I do like the H105 section & will consider using it for my next set of foils. I want to be careful to avoid overcambering the section though. An interesting aspect of the NACA experiments that I referred to in Post #4 was that the foil probably cavitated on its lower surface at small values of CL. Because of the test setup, the lower surface wasn't visible, but the operators saw bubbles after they got downstream of the trailing edge. The foil being tested (NACA 16-509) was optimized for CL=0.5, but wasn't designed with any drag bucket to speak of. The H105 looks better in that regard.
     
  9. revintage
    Joined: Nov 2016
    Posts: 414
    Likes: 101, Points: 43, Legacy Rep: 10
    Location: Sweden

    revintage Senior Member

    Hi Doug, picking up this interesting thread. Would you mind sharing the profile section of your main foils. I really like your V-concept due to the inherent mechanical stability. Being retired without enough theoretical knowledge, I would like to to take a shortcut and try a working concept as yours to get going.
     
  10. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    The cross section was intended to be NACA 0012, but ended up with a blunter leading edge due to my impatience while building it. Later repairs only made it worse, since the leading edge was the most vulnerable region.

    A thinner section would have had less drag, but 12% was about right for foils of this size, loading & construction.

    A cambered section might have been better at very low speeds, but I was more interested in higher speeds where getting more lift was definitely not a problem. Also, having zero lift when the foils were straight down made it easier to find the correct angles & understand what was happening better.
     
  11. revintage
    Joined: Nov 2016
    Posts: 414
    Likes: 101, Points: 43, Legacy Rep: 10
    Location: Sweden

    revintage Senior Member

    Thank you Doug, so I understand that you never changed to an unsymmetrical profile. Really like the idea of rotating the V around the beam to find the best angle.

    I have actually found a commercially available aluminium NACA 0009 profile with a corda of 135mm, that might be possible to widen it to 0010 or 0011. This means using your geometry that the "Horizontal Projection at Bottom of Hull" would have to be 25" instead of 22,5" to keep the horizontal projection equal.

    Question is how much higher the lifting speed would be and if it even is possible to use a 0009 as it will be more critical wrt stalling? Newly retired at 67 I need a real challenge, collecting parts to build a tri-foiler around the single 24kg A-cat hull I have.
     
  12. tspeer
    Joined: Feb 2002
    Posts: 2,319
    Likes: 303, Points: 83, Legacy Rep: 1673
    Location: Port Gamble, Washington, USA

    tspeer Senior Member

    A NACA 0012 is not a particularly good section with regard to cavitation. The reason is it's a front-loaded section, with high velocities/low pressure near the leading edge and comparatively low velocities/high pressure for most of the chord. Blunting the leading edge only makes it worse. Since cavitation occurs when the local velocity exceeds a threshold, what you'd like to see is the velocity come up to just below that threshold and stay there for much of the chord before dropping down toward the trailing edge. This is known as a "rooftop" characteristic to the pressure distribution. The NACA 6-series sections were specifically designed to have this kind of pressure distribution. In the case of the 6-series, they were they were trying to stay subsonic, so the reason for the threshold was different, but the effect on the section shape was the same.

    A better choice for raising the cavitation speed with a symmetrical section would be something along the lines of a NACA 64012 to a NACA 66012, depending on how far back you wanted to extend the rooftop. A longer, lower rooftop will have a higher cavitation speed, but may have more issues at low speed.

    These sections have a more narrow leading edge than the NACA 0012, and as you have found out, the leading edge shape is critical. So you will want to use templates to profile the leading edge as accurately as you can. That means starting with an accurate template. These days, there are lots of people that have laser cutters, and outfits like Monzano Laser Works can cut your templates for a very reasonable price. I typically use cheap utility plywood (5mm thick). You can get a small sheet of 1/8" birch ply at hobby shops that would work well, too. When you design the templates, make them in a J shape that wraps part way around the leading edge on one side and extends back to around maximum thickness on the other side. This makes them easier to align with the foil. Given that you have constant chord foils with a symmetrical section, you really only need one template.

    The acute V shape of your foils leads to a lot of acceleration of the flow near the apex of the V, lowering the onset speed for cavitation. One thing you might try is to stagger the legs of the V so things aren't all getting fat at the same place. Inserting a short straight segment, so the junction is more trapezoidal in shape, would help, too. You could join the foils in a V, insert the straight piece between them, and then cut off the triangle below the insert, leaving the shape you want. That would be easier than trying to jig three pieces in place to join them end-to-end.
     
  13. revintage
    Joined: Nov 2016
    Posts: 414
    Likes: 101, Points: 43, Legacy Rep: 10
    Location: Sweden

    revintage Senior Member

    Tom, not a bad idea to use a more effective foil. The workshop I have access to has a 150x250cm bed NC milling machine(behind my newly built catamaran beams), so making the foil legs in two light wooden halves to join, would not be the hardest task. Then cover them with fiberglass like Doug did. Skipping the 0009 aluminium profile at 1,5kg/m would also save considerable weight.

    IMG_3247 2.jpg
     
  14. Doug Halsey
    Joined: Feb 2007
    Posts: 640
    Likes: 212, Points: 53, Legacy Rep: 160
    Location: California, USA

    Doug Halsey Senior Member

    I think your aluminum NACA 0009 sections would be an improvement over my wood & glass NACA 0012 foils, without trying to widen them.

    The shapes would be more accurate & easier to keep smooth. The strength & stiffness of aluminum would allow the thinner section to be used, so the drag should be less. Smaller maximum lift would not be any concern because it is never advantageous to operate at very high lift coefficients.

    I also wouldn't worry about any loss in span that could be attributed to the change in foil thickness. I think the more relevant factor is the span of the basic foil skeleton, which wouldn't have to change.

    One suggestion I would make would be to extend the outer segment of each V a few inches beyond (i.e.- inboard) the intersection with the inner segment. That way, when the boat is flying very high & only the tips of the foils are in the water, they are still somewhat wing-like & not the shapeless blobs that the apexes would otherwise become.

    Otherwise, a moment like this: 4-15-09_0057.jpg
    can easily be followed by this: 4-15-09_0081.jpg
     

  15. revintage
    Joined: Nov 2016
    Posts: 414
    Likes: 101, Points: 43, Legacy Rep: 10
    Location: Sweden

    revintage Senior Member

    Great input Doug, the alufoils are tempting as they will be easy to assemble. On the other hand, building foils the way you did it, will be so much lighter and lets you choose the section you want, as Tom had a few alternative suggestions to the 4-digit NACAs. Must talk to my friend who runs the NC-mill how hard it will be.
     
Loading...
Similar Threads
  1. S V
    Replies:
    19
    Views:
    2,797
  2. Maarten88
    Replies:
    5
    Views:
    1,003
  3. alan craig
    Replies:
    5
    Views:
    1,212
  4. B.NARENDHIRAN
    Replies:
    3
    Views:
    1,106
  5. Alan Cattelliot
    Replies:
    75
    Views:
    6,944
  6. hashtag_laeuft
    Replies:
    20
    Views:
    3,029
  7. dustman
    Replies:
    54
    Views:
    4,833
  8. MacktheYounger
    Replies:
    155
    Views:
    9,038
  9. S V
    Replies:
    0
    Views:
    1,876
  10. Mikko Brummer
    Replies:
    34
    Views:
    5,294
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.