Foil Cavitation at Lower Speeds Than Expected

In the spirit of showing my boat's negative features, as well as its positives, I decided to post the following video : https://youtu.be/n7UjSdY3XTg

It shows the leeward foil of my trimaran cavitating (loudly) at a boatspeed of about 20 knots. The video is in 2 parts : the 1st 23 seconds show the wide-angle view from my GoPro & the next 23 seconds show the same footage, zoomed in on the foil's apex.

There are several comments I could make about what's happening (& some questions, as well), but I wanted to get a few of your comments first.

 

Doug, I haven't seen cavitation enough to know for sure but wouldn't the foil have lost lift if there was cavitation?

 

That's what I would have thought too, but if the pressure inside the bubbles is low enough to keep the water vaporized, maybe not.

Also, there doesn't seem to be any obvious effect on the drag since the boat doesn't decelerate when the cavitation starts. It's hard to tell, though, because we don't know if the boat would have kept accelerating.

We do know that there would have been large effects on both lift & drag if this had been ventilation instead of cavitation, since the pressure would be atmospheric instead of the vapor pressure of water.

I'm wondering if cavitation often triggers ventilation & that's the reason for its worst effects. Maybe it's not so bad as long as there's no ventilation.

I'm hoping to hear from people who know more about it than I do.

 

Here's a page from an old NACA report (NACA L-758) showing experimental data for a horizontal submerged foil. The curves are CL versus Speed and CD versus Speed, for several values of angle of attack. Each of the curves includes an arrow, marking the point where the experimenters first noticed cavitation.

I would say that the effect of cavitation on the lift doesn't seem too extreme, but its effect on the drag is larger, especially at the largest angles of attack.

 

It means that you're operating at an excessive CL for the airfoil sections you have on your foils. An excessive CL produces a strong LE pressure spike where cavitation will start even at relatively low boat speeds.

The simple fix you can try for your existing foils is to reduce the foil incidence angle so that it runs deeper, and thus with more lifting area and a correspondingly reduced CL. If the reduced incidence does not enable you to take off, then you need either
* larger foils (larger span, or chord, or both)
* variable incidence which can be changed while on the foils (start at large incidence, reduce when at high speed to suppress cavitation)

If you make a new set of foils, then I'd make sure you have an airfoil designed for delay of cavitation. Tom Speer's H105 is a good example. This will probably have a significantly higher cavitation speed than something generic like a NACA4412 or Clark Y.

 

Looks like you seriously need fences on those foils too.

Hard to tell in the video..but the noise could also indicate 'signing'.

 

This is all very true, but in 9 years of foiling I only experienced this obvious cavitation a handful of times. And I reached much higher speeds many times without cavitating.

The difference, of course, was in the foil settings. The main foils are set (on the beach) at an angle that allows them to provide sufficient righting moment at low speeds. (I use amas with very low volume & often don't use the amas at all). So, I don't adjust the main foils at all while sailing. There is a reduction in angle as the boat speeds up & the nose pitches down, and depending on the incidence of the aft foil, this is usually enough. But I am continually adjusting the aft-foil incidence, searching for the optimum amount of foil in the water. Sometimes I get it wrong & the boat flies too high. Even without cavitation, this would cause the CL to be too high & could lead to a stall & crash.

When I built the boat, I expected to be making new foils very frequently, but as it has turned out, I still only have the originals. (That's life!). When I make new ones, I will certainly use something better (maybe the H105), but I will also round the apex slightly like all the current uptip foils seem to do. This has advantages for both viscous & inviscid reasons (but may sacrifice a bit of the inherent stability of the V).

 

I thought the bottom was rounded -- hard to tell from the video.

Yes, you definitely want a decent radius at the bottom of the V. The mutual foil-leg interference at the sharp point will certainly promote both early stall at low speeds and early cavitation at high speed. Rounding the point dilutes this interference significantly.

 

I did use fences a few times early-on & I hated them! All they ever seemed to do was create little separated areas. (& make it much harder to sand the foils).

The main use for fences is to suppress ventilation, which isn't generally a problem - thanks to the large dihedral angle (60Deg) & forward cant angle (usually ~7Deg). I don't think there's any sign of ventilation in the video.

I assume you're saying that because of how high the spray goes up the foil. If so, do you think that there would be less drag if the spray was deflected by a fence before it could get too high ? I would be willing to put a fence back on, if it was high enough to usually be out of the water.

Sorry, I don't know what that means ???

 

Only a little; not nearly enough.

 

Well, it is going to make a difference, how much...ahhh..that's the big question. May be very minor indeed to the point of nothing worth while, but it wont do any harm either. In the absence of any data hard to say for sure.

I'd still place one below the WL too.

It is merely the noise the trailing vortex makes. The vortex has a frequency of vibration that is dependent upon the size of the immersed body, the foil. It may be calculated using the Strouhal number from the Reynolds number of the foil.

To cure it, one can simply slice off the trailing edge at a sharp angle, in simple terms.

 

I doubt that you actually have cavitation here; the dynamic head is ~6 m which is far more than the operating depth. With such strength of a sink in the vicinity of the surface, there must be a transverse movement towards the sink. In the view showing the outer port foil, there is clearly a strong downdraft along a line slightly aft of the leding edge. The basic medicine for this disease would be fences, better foil profile and lower aoa.

Within the semiclosed "pocket" at the vee tip (collection of boundary layers from both foil arms), you will get a stable recirculation zone with a more or less fixed air bubble. The elasticity of this bubble may well produce a resonance system that is reinforcing the Strouhal "singing" that Ad Hoc is referring to.

I could not hear the sound; but "pure" vapour cavitation will manifest its presense by a sudden increase of "white noise" in the kiloherz region. Lower frequencies indicate ventilation.

Btw; what is the load carried by each foil section, and what Cl is needed for this, considering the sharp vee angle?

 

For years, I ignored the possibility of cavitation on my boat, because I had read that it wouldn't be a problem until about 40 knots. I would read or hear of people claiming that their foils had cavitated & would secretly scoff at them, thinking they didn't know the difference between cavitation & ventilation. Even after experiencing what's shown in my video a couple of times, I didn't believe it was cavitation. However, coming across this GoPro footage made me consider the possibility & subsequently doing some calculations finally convinced me.

In order to see what's happening more clearly, I extracted 15 seconds from the video & slowed it down to 25% of the normal speed. This slow-motion video is at https://youtu.be/lHu7ePEw7YM

In the 1st few seconds, it's clear that the white area originates near the apex of the foil & spreads upward to a varying extent, as the foil rises & falls. Sometimes it reaches the water's surface, but when it does,the chordwise extent of the white area is always a very small amount near the leading edge.

If this were ventilation, instead of cavitation, the patterns would be just the opposite : a depression of the surface getting sucked down the foil. Can you really look at this video & not agree that you're seeing cavitation ?

Additional support for this view has been given in earlier posts :
1-If it were ventilation, the lift would have been drastically affected
2-Theoretical local flow velocities at the apex of a V foil (or at the midspan of a wing with large dihedral) are considerably higher than those at the midspan of a horizontal foil (or wing), so cavitation is possible (or probable) at a much lower speed.

If this really is cavitation, then I don't believe that fences would help. In fact, I believe that in some cases, they would cause small areas of flow separation, which in turn would trigger ventilation.

 

A flat fence just above the waterline which blocks the spray won't help -- it will merely dissipate the kinetic energy in the spray water. To extract this energy the fence must have camber, with its leading edge turned downward, so that it has a forward force or "thrust" when it turns the upward-moving water towards the streamwise direction. This turning can't be done cleanly, because some of the water will squirm out sideways out of the cambered fence. Also, the cambered fence will then have excess drag when it's submerged. Probably better to leave it off entirely.

I suspect that spray drag can be reduced by using an unusually small LE radius near the waterline, but that's not practical on a V-foil where the waterline can be anywhere along the foil. Better to preserve the airfoil shape to minimize cavitation.

 

This assumes that the airfoil is close to the ideal shape and CL so as to delay cavitation as much as possible. With a "bad" airfoil and CL, getting cavitation at 20 knots is certainly possible.

Cavitation is expected when the minimum Cp on the airfoil drops below the negative of the cavitation number. Assuming a near-zero vapor pressure and a small water depth, this is:

Cp < -p_atm/(0.5 rho Vinf^2)

For Vinf = 20 knots = 10.3 m/s, this criterion gives Cp < -1.88 , which is easily achievable (well before stall) on a foil which is undersized and/or has an inappropriate airfoil. Most likely, this negative Cp will first occur close to the leading edge.