Foil Cavitation at Lower Speeds Than Expected

I would respectfully (indeed very respectfully) disagree. At 0:33 it is IMO clearly visible that there is a cavitation sheet starting at the leading edge of the foil, and extending to at least 70-80% of the chord. It seems to be triggered by the too-sharp apex of the foil's Vee.

IMO, the main problem here is the excessive Vee of the foils. It forces the hydrofoil to work at higher loading in order to produce the required vertical lift, hence it cavitates more easily.
Since it seems impossible now to change the foil configuration, you could consider another option - putting a small Gurney flap along the trailing edge, in order to decrease the working AoA of the foil and the peak pressure at the leading edge. That could help, IMO.

 

Don't you mean that the Angle of Attack (AOA) of the foil is too high?

CL stands for the lift coefficient, I believe.

 

CL is the direct measure of lift at a given speed, the angle of attack is not.
The foil has to produce a given amount of lift in order to support the weight of the boat. Hence, it will set at some CL. The resulting AoA depends on many things, including the geometry of the hydrofoil, the interaction with the free surface etc.
So, in this case and for this technical analysis it is more correct to talk in terms of CL, than in terms of AoA.

Cheers

 

I wouldn't put it that way.
The Vee angle is a design choice. A steep Vee just means that you need more submerged area to avoid separation, and it also means that the point of the Vee gets more problematic. But making the Vee shallow will promote ventilation, and will have less ability to react against the sail sideload. Clearly there's a good Vee chord/angle/airfoil combination which balances the lift and sideforce requirements, all within the cavitation and stall constraints. I suspect that Doug's first-try Vee foil is far from this ideal.

 

Either one. For a given foil geometry and submergence depth, the CL and the AoA are uniquely related. (The Froude and Reynolds numbers also matter, but much less so).

For design purposes it's better to think in terms of CL because it is a direct indicator of the surface Cp values, and hence it is a better indicator of stall and cavitation than AoA.

 

What airfoil do you have on your V-foils? From your photo it looks like the LE is a almost semicircle.
Makes sense for a quick home-brew foil made in the basement, but not so good if avoiding cavitation is important.

 

Heh.
One thing I've learned is a "quick kludge" that ends up sort-of working is forever.

 

My original intention was to use NACA 0012, but the shaping was so tedious that I got impatient & just ended up with something only vaguely similar, with the biggest difference being the larger LE radius.

I wanted a symmetrical section because I was toying with the idea of using negative lift on the windward foil to increase righting moment. I gave up on that possibility for several reasons : 1-My VPP calculations indicated that it would be detrimental for windspeeds less than at least 25 knots; 2-The extra loads involved would mean I would have to make the structure much heavier ; 3-when lifting down, a V foil wouldn't be stable anymore without something like an external wand.

I also thought a symmetrical section would be better because the CL is very low under most foiling conditions, assuming my estimates of the optimum immersed foil area are used. Also, the relatively small span of the foils means that I shouldn't make the takeoff speed too small, because the induced drag would become excessive.

For my next one (if there is one), I probably wouldn't want much camber either (maybe 2%?), and I would certainly have a sharper leading edge. A big question is what would be the optimal sharpness, both for avoiding cavitation, ventilation, as well as all-around conditions?

By-the-way, when I got my Mach2 Moth, I was really surprised how sharp the foil leading edges were, but there are other factors involved, so I won't get into that now.

 

Ain't that the truth!

 

What you see here is exactly analog to a ventilating rudder on a heeling boat (note that Doug seems to have used a symmetric profile as well). There is a strong downwash between the suction surfaces, producing a vortex with a downward peripheral movement along the suction surface. The ventilation comes "from behind", just as on a rudder when stalling due to ventilation.

There are some sequenses showing the process; @38 sec there is a downwash at the aft part producing a bubble cloud moving forward along the SS, and finally opening a path forward to the surface. But note that the air is mixed into the vortex flow. @43 sec the cloud is blown backwards and disappears. @45 sec a new event is starting witht the foil emerging.

With the foil shape used here, cavitation could certainly happen, but with the operating depths in this case, the low pressure zone is more likely to induce ventilation from the surface in the downwash region.

I do agree with the Gurney flap/interceptor proposal; it could work with a height of ~1 to 2 % of the chord, but first a suction side fence about a chord from the vee tip.

May I suggest that you check the ventilating Moth rudder on youtube (Google search for ventilating rudder). There are some events where you can see the ventilation "cloud" beeing generated at the Surface and then still survive when the vent path is cut off.

The ventilation depth (h) at a speed (V) is determined by the Froude number: Fnh=V/(g*h); where the critical Fnh is a function of profile shape and angle of attack. As an example, a truncated symmetrical circular arc profile with thickness/chord =0.15 has a critical Fnh of ~3.2 at aoa=0. At aoa ~15 deg Fnh is ~1. To get an idea of the order of dimension of the numbers, let us assume an operating depth of ~0.4 m and a speed of 10.3 m/s (20 kn). We then have a Fnh of 5.2, which tells us that this profile is most likely ventilating, unless special measures are taken. This also means that, without anti-ventilating measures, there can not be any vapour cavitation down to a depth of about one meter at this speed and with a critical Fnh=3.2.

Now if the ventilation path is cut off by a fence, then the local static pressure may be further reduced below the fence, resulting in vapour cavitation.

Some further info on surface piercing struts and foils is found in "Hoerner, S. F.: Fluid Dynamic Drag 1965", freely available as a PDF on the net. It is a veritable goldmine for empirical data, in spite of its age.

 

Ventilation

Here is the Moth video you refered to, Baeckmo:

https://www.youtube.com/watch?v=s58KsP9GvHE

 

With all due respect, I don't understand how this is useful. If the critical Fnh is a function of the profile shape and aoa, then how can you conclude anything about one case from the critical Fnh of another ? Or is there some theoretical way of calculating it that hasn't been mentioned? And what would that imply about the previous belief (K.L. Wadlin) that a low pressure is not, by itself, sufficient to cause ventilation, but that boundary-layer separation is also required?

 

Doug (Halsey), I have found this other video of yours: https://vimeo.com/34164014 and have really enjoyed it. I think it is very instructional, besides being very involving for the viewers.

The numerous sudden ditches of leeward foil are also a very educative example of a phenomenon which aeronautical engineers cal aeroelasticity (guess the hydro guys call it hydroelasticity?). All of a sudden the static equilibrium of forces and moments around beam root breaks, the torsional elastic energy stored in the horizontal beams decides to get released and inverts the AoA of the foil. Negative AoA (possibly even divergent), downwards lift, crash stop of the boat.

A more in-depth analysis of how beams and foils are made and joined together, and also whether the cavitation which we have discussed here plays a role in setting off the torsional divergence, before proposing a solution to the problem. But it would be a nice exercise for the engineers and aero engineers here present.

Cheers

 

It is useful in that it gives an idea about the limiting boundaries ("what is possible?"). The truncated foil at zero aoa represents a low drag, base vented foil that requires a high speed in order to ventilate to a given depth and it is comparatively resistant to suction side ventilation when operating at an aoa up to ~15 degrees. With increasing aoa you reach the limiting case of 90 degrees, where you have a transverse plate. This configuration requires the lowest speed in order to ventilate to the same depth. Using the Froude depth number makes comparisons possible; your case must fall between the two extremes and as a logic consequence vapour cavitation can be ruled out, no matter the details of your foil concept.

Since cavitation and ventilation are requiring different countermeasures for control/avoidance it is relevant to find out what is happening; it's not just academic trivia.

The statement about bl separation still holds, since you have a very strong secondary flow in the suction side downwash. Compare the rising spray line on the pressure side with the falling profile of the fluid inside the vee! The 3d character of this flow means that the bl separation does not occur simultaneously along the foil span (as it would in a parallell flow), but gradually. The bubble cloud that is the result of the corkscrew motion of the downwash is kept stationary by the vortex induced by the reentrant jet that comes from the reattaching boundary layer. This effect is often seen with ventilating propellers with traditional "aerofoil" profiles; the bubble cloud rotates relative to the blade and sticks to it for long time (hundreds of revolutions), long after the ventilation path is cut off.

The downwash intensity is increased (and with it the lift) by the V geometry, which gives the "foil waterline footprint" a toe-in, heel-out configuration when you increase the aoa. When tha aoa gets negative, on the other hand, you have an efficient brake, as seen in the video......

 

Daiquiri : I'm glad you enjoyed the video, but sometimes I regret having included some of the more entertaining portions. In spite of the design & construction deficiencies brought out in this thread, the performance hasn't been that bad. It's broken 25 knots on numerous occasions & reached a peak of 28.7 at one point. It's a lot of fun to sail (but not very challenging compared to my Moth) & I think it's taught me a lot.

I don't think I'm just deceiving myself to say that the little crashes didn't used to happen as much. The condition of the foils deteriorated gradually over the years until, by the time of these videos, the fiberglass on the leading edges of the main foils was completely gone. I had sanded down to bare wood & hadn't properly recoated them. So, often the boat would sail very nicely in the early part of the day & then do worse later when the leading edges got rougher because the moisture made the wood swell.

The day of the cavitation video is a good example of that. Here's another video of the boat sailing earlier that same day. https://vimeo.com/104854385
This one was taken from off the boat, instead of using the GoPro. I don't think there were any crashes during that part of the day. The cavitation video was taken a little later & then a 3rd video, showing a crash causing a capsize was taken a few minutes later. The link for the capsize video is at https://vimeo.com/104982325

I don't think there are any other videos floating around (except for some private ones ).