#4 Warning: Could not find interpolants

[Italo Franchini]

Hi Dan, nice work you have here! Hey I'm getting this warning when it's reviewing the constraints: Could not find interpolants. And then shows me the last value of x coordinate and the previous one, in this case x:1.000 xmax=0.997.

Any idea of how can I fix this?

Thanks a lot!

[Dan]

Hi Italo,

Could you post the seed airfoil file and input file you are running for this case? That will make it easier to tell what is going on.

Dan

[Italo Franchini]

Of course, I totally forgot about that. It's the NACA2415.

[Dan]

Can you post it as an attachment, along with your inputs.txt file?

[Italo Franchini]

Sorry I was distracted, finally here it goes

[Dan]

Okay, thanks. I will look into it. When I run your input file as is, it doesn't give that error. However, if I switch seed_airfoil to 'four_digit' instead of 'from_file', it does give the error. If by chance you have actually been using the 'four_digit' option, just switch to 'from_file' for now. When I get a chance in the next couple days I will try to figure out what's causing the interpolation error when using the 'four_digit' option.

Dan

[Italo Franchini]

I have tried from_file and still have the same error. How did you run it? I'm just executing the following in the cmd: xoptfoil.exe inputs_n2415_2.txt naca2415_opt
Any idea of what am I doing wrong?

Thanks for your help!

[Dan]

That's how I ran it too, except I didn't specify an output file. That shouldn't make a difference, though. I just used the airfoil file you attached in the earlier message. It's strange that it would give the error for you but not me. I guess it could be something related to numerical precision (which might be different on different machines) throwing off the interpolation routine. Hopefully when I get to the bottom of it for the four_digit version on my computer it will fix it for you too.

[Dan]

Italo,

I believe I fixed the issue. The problem came in the thickness checks, where the lower surface y-coordinates are interpolated to the top surface x-coordinates so that the top and bottom surface can be subtracted to get the thickness. The issue was that, in some cases where the bottom surface TE point was further aft than the top surface TE point, it couldn't be interpolated correctly near the TE. I changed it so that the code first checks which point is further aft (bottom surface TE or top surface TE) and then chooses the correct interpolation direction. In your case with the NACA 2415, the top surface y-coordinates need to be interpolated to the bottom surface x-coordinates, instead of the other way around, to make it work. The code should now figure it out automatically in every case. Let me know if it works for you.

Dan

[Italo Franchini]

I see, good job! I'll check it out. By the way another quick question, in one of the optimization routines I got the following messages:

BL array overflow
Increase IVX to at least 338

I don't know what IVX means.

Thanks a lot! And amazing job you did here!

[Italo Franchini]

Everything works great now, thanks for your help!

[Dan]

Okay, I'm glad it works for you! IVX is a parameter from XFoil. It is related to the number of points on your airfoil. If you are getting that warning, it is probably because you have more points representing your airfoil than you need. If you really want to get rid of the error, you can modify the source code and recompile. You will want to increase the value of IQX (the maximum number of airfoil points) in xfoil_inc.f90, or you can modify the value of IVX directly. I don't really recommend doing the latter, though, because the relationship between IQX and IVX comes from the original XFoil code and I don't know if changing it might mess something else up.