Foil Cavitation at Lower Speeds Than Expected

I've done some more XFOIL calculations & they show that camber alone goes a long way toward reducing the cavitation. The 1st figure below shows Cpmin values for the NACA 0011 & NACA 63-012 sections with & without camber (and also the more carefully designed H105). The next 2 figures show the lift curves & drag polars for the same sections.

You could probably use the Windknife Foil Nose + G10 sheets to produce a fairly decent cambered shape, if you wanted to go down that route.

 

Picked up the NF and the 1mm G10 today. Will epoxi two sheets of G10 to the NF with the rear end open. Adding camber should be no problem. What camber f/c and location xf/c would you recommend? Something like 4312? Will run them in Javafoil.

As a laymen I have some problems about areodynamic basics. Actually don´t even understand the Cp curves, even if I know how they should look to be better or worse ;-) .

 

I'm not sure how much camber would be best, but the case I ran had about 2.8%, and shifted the zero-lift angle by a little more than 2 degrees. That's probably close to the maximum you would want, but I need to think about it more.

The max camber location is approximately 50% chord. Your camber shape will be affected by having a symmetrical nose (although it can be tilted slightly). So we need to wait & see what shapes you manage to produce.

Here's a screenshot of what the section & the camberline look like. I'm also sending an Excel file with the actual coordinates.

 

I´m a little confused as your profile seems to be close to 1,4% and 50% when I import the coordinates into Javafoil.

Edit: Just a matter of how it is defined, your 2,8% is the same as Javafoil 1,4%.

When I guessed 4312, this was because of the almost flat underside being easy to achieve ;-) .

The image shows what happens when I compress the 0012 NF to 18mm at 45mm cord length,corresponding to a 150mm 0012 profile. Question is where it is being deformed?

 

You're right about the camber. It's a little over 1.4% (not 2.8%). Without thinking I multiplied by 2, probably because I had just been looking at the thickness coordinates.

I'm a little confused about whether you think you want 12% thickness, or 11%. Also, whether your chord has to be 150mm. Using the G10 sheets should allow you to vary that to some degree, I would think, and skip trying to deform the NF.

 

About 12% the nose foil is profiled like that and I have to accept.

150mm/6" is what you used and it seemed to work. Any suggestions to make it even better?

Will try to superimpose a known profile over thecompressed nose in the image above, to see what has the most resemblance. Working on that.

 

The nose foil is 12% for only a specific value of the chord. But using the G10 sheets, you can make the chord larger or smaller (as long as you keep the junctures smooth).

Broomstick's foils are actually 6 1/4" = 158mm, but there's nothing sacrosanct about those particular numbers. Being slightly larger would probably help take off earlier in lighter winds, and help delay cavitation to higher speeds. Being too much larger would possibly hurt performance in the mid-speed ranges, but I think anything up to about 175mm would be good overall.

And if your all-up weight is larger than Broomstick's, then the foils should probably be larger too.

Windknife's website says the FN's width is 60mm, but I'm not sure whether or not that includes the lip at the juncture. And it doesn't say what the max thickness is. Do you have those numbers?

I think it would be better if you didn't have to compress the nose. Keeping the compression consistent over the full length of the pieces seems like it would be very difficult (& time-consuming).

 

The nose foil is specified as 0012. So it seems logic to go for that.

Not sure the nose foil slightly compressed like I modified min is 0012 any longer, but it indicates a 168mm corda of a 0012 profile. Less work if we could accept that. Initialy it is 22,5mm wide at 45mm indicating a corda of ca 200mm.

Also my calculated total weight at 175kg incl. crew will be 1,07 times higher than Broomstick, this means more projected area needed( 1,07^2?). The projected area of the prolonged(+20cm?) outer foil, of the V below the joining point, that you recommended, will of course add.

Another question arises:
What do you consider being the optimum camber for the 0012 in this application? 1,4% seems rather small.

 

There's always a bit of subjectivity in decisions like this. The boat has to operate in a variety of conditions, and the optimum is slightly different in each case. But I'll stick my neck out & recommend the section designated "CLL=0.3" in the attached figures.



The figures show XFOIL results for the NACA 0012 with 5 different amounts of camber, ranging from 0 to about 2.8%. Each camberline belongs to the NACA a=1 series (0.99, actually), with the CLL parameter indicating the theoretical optimum CL in thin-airfoil theory. As before, I've included results for the H105 section as a benchmark. (If you want, there's much more that could be said about the details...)

The foil corresponding to CLL=0.3 is shown below:

and its particulars are given in the attached Excel file.

 

You might want to go backwards through your cavitation speed calculation. Instead of calculating the Cp_min for a given speed, calculate the cavitation speed for a given Cp_min. Then you can plot cavitation speed on the X axis and lift coefficient on the Y axis, as shown in this example. If you're given the speed and the lift coefficient, you also know the foil loading, so you can add lines of constant foil loading to the graph. That helps to size the foil so you are operating within the section's cavitation bucket.

 

Thanks for helping.

Will go through this material during the weekend.

This is the input page of Javafoil where you easily can vary camber f/c and location. You can also import a *.jpg, will try adding the nose foil.

 

I'm familiar with Javafoil. It's the only free airfoil analysis code I know of that can handle multi-element sections. I think it's OK in the linear range, but I think Xfoil has a better boundary layer code. And, of course, it's not as capable as MSES for multi-element analysis.

 

As a layman Javafoil was the easy way out to get going ;-) .

By the way, as this a low-budget project, I want to compare an over rotaded round 70x2mm mast tube with A-cat sail to a regular 60x145 mast section with A-cat sail, to get a coarse indication of how much more drag will be induced in the premier.

Is this doable in X-foil or Javafoil?

Starting up with the round tube at 80 bucks + mast track is within reach.

 

Please advice, at what CL is the proposed cambered profile with Cll=0,3 supposed to work on Broomstick, compared to the old angled 0012?

 

What about adding a bend to the 135mm 0009 foil making the rear a fixed angle flap?