How to use xfoil

[raptor22]

I have xfoil, but have no idea how to get airfoil files, or create them, or do much of anything in it, really. I've tried several times withou achieving anything.

Since I'd have to guess noone wants to write me an xfoil manual, does anyone have alink to some form of instruction website or something?

I looked at the yahoo xfoil group, and did searches, without actually finding much. And I don't want to start a thread asking everyone everything.

--Alex

[CloudyIFR]

http://www.profili2.com/eng/default.htm

[raptor22]

I know about profili, and am thinking about buying it. But I still want to know how to use xfoil.

[spatial]

the profili site has some on airfoils good information so don't dismiss it out of hand.

you could also try:
http://www.aae.uiuc.edu/m-selig/
http://www.profoil.org/
http://www.mh-aerotools.de/airfoils/
http://adg.stanford.edu/aa241/AircraftDesign.html
http://raphael.mit.edu/xfoil/xfoil_doc.pdf
http://raphael.mit.edu/xfoil/sessions.txt
http://www.b2streamlines.com/
and
http://www.b2streamlines.com/RCSD.html
There is an excellent series of articles by Dave Register starting in the October 2000 issue and running intermittently to Feb 2003 which cover airfoil selection, interpreting polar diagrams etc etc. I'm reading my way thru a stack of printouts of these at the moment and can highly recommend them.

cheers
Paul

[HELModels]

The manual is pretty good. Also, try "NACA 4412" ; "oper" ; "visc" ; "re"; "cl"
those will at least let you look at NACA 4 digit and see alpha, CL, Cm, etc. and plot of foil in pltlib. Following the flow chart in the manual, reading the manual will help you when creating/modifying your own. Learn to "load" a foil and save. It should save to whereever the executable directory or specify when naming save file. The manual works.

Harry

[BMatthews]

There used to be an Xfoil group on Yahoo Groups. If you google xfoil it shoudl turn up some references to that group.

Also I seem to remember that there was a fairly large user manual file that came with the Xfoil download. Or was that from the Xfoil group.... it's been a while.

[spatial]

the xfoil yahoo group is worth joining, even if it's just for access to the files section. there are a number of case studies, pdfs and scans of some very relevant articles, including a piece by Mark Drela entitled "Elements of Airfoil Design Methodology"....

cheers
Paul

snipped from the xfoi_getting_started.txt:

2) Print out the reference manual and read it carefully. It will help you understand all the the features of this program

3) Look at the case studies in the files sections of this group. It gives you some real world examples of how to correctly use XFoil

4) Look at the frequently asked section at the end of this document if you encounter problems. More then likely, you will find a solution to your problem there.

5) If you are into learning how XFoil actually does what it does, take a look at the technical papers in the file section. But be ready to polish up on your math.

[orectolobus]

Raptor22.

Also check out Airfoil Design Workshop at
http://www.tdmsoftware.com/afd

This program also has a graphical interface to feed airfoil data to XFoil.

Terry

[joemadeus]

I've been experimenting with XFoil v6.94 for a while and found it to be pretty interesting -- and useful. As you discovered, it's not pretty, but it definitely is functional. In general, you'll spend your time in four different areas:

a) The main menu. This is where you load the airfoil data files, save changed airfoil files to disk, and go out into other XFoil functionality.
b) 'ppar': This is where you adjust the panelling for the airfoil. XFoil breaks up the surface of the airfoil you load into small segments called panels. You can adjust the number of panels here, how many are allocated to the trailing and leading edges, etc.
c) 'gdes': This is where you can make some basic adjustments to the geometry of the airfoil. Typing in 'tcpl' in the 'gdes' screen, for instance, gives you the thickness, camber and leading edge radius for the airfoil. You can then type in 'tset' to set the thickness and camber. Very useful for playing with the basic properties of the airfoil.
d) 'oper': This is where your analysis takes place. Typing in 'a [alpha]' will compute the Cl, Cd and Cm for the airfoil for the alpha you give it. Typing in 'as [start alpha] [end alpha] [interval alpha]' will compute a whole series of alphas, starting at [start alpha] and ending at [end alpha], in [interval alpha] increments (replace [start|end|interval alpha] with numbers and leave off the brackets.)

You can get a list of commands by typing in '?' anytime.

Lately I've been playing with Selig's et al SA7024 airfoil, as found in Gopalaratham's el al paper in AIAA in 2001 ("Design of Low Reynolds Number Airfoils with Trips"). Here's what I did this evening:

1) Load the airfoil datapoints. Make a directory in the XFoil install directory called 'Airfoils' and put a file from the UIUC site (go here) (http://www.aae.uiuc.edu/m-selig/ads/coord_database.html) into it. Try one of the ones that begin with 'SA' or 'AG' first -- those are developed by Selig (the head of the team that did a lot of wind tunnel testing of low Re airfoils) or Drela (the author of XFoil) and I know they load correctly. To load the file, type in 'load Airfoils/sa7024.dat' (if you grabbed the coordinates for SA7024 from the UIUC site, and strip off the single quotes.)
2) Adjust the panelling of the data to something more reasonable. I usually find that there are too few points in the coordinate files (which come from all over the place, btw.) Too few points mean lots of difficulty for XFoil -- the more points, the smaller the angle between two points on a sharp curve and the better your results (and the faster XFoil computes them). Type in 'ppar' to go into the panelling parameter page, then 'n [the number of panels]' to tell XFoil how many panels you want to use, and then hit the return key a couple times to go back to the main menu.
3) [I loaded some polar files I'd computed earlier, adjusted the airfoil in gdes, then went back into oper. Since you're just getting started, I'll spare you the details. Just get some results first.]
4) In 'oper', I switched to viscous mode by typing in 'v', then gave it a Reynolds number by typing in 'r [a number, like 100000]', and adjusted the number of iterations to make to 500 by typing in 'iter 500'. Then I kicked off an analysis run by typing in 'as -4 15 .1', to compute polars for all the alphas between -4 and 15, at 0.1 degree increments. My computer is a Pentium III @650MHz, btw, and this run took about 30 minutes -- not that I need this much detail, I just happened to be making dinner and had CPU cycles to spare :). My Pentium-M @1.6GHz laptop at work does the same thing in less than 10 minutes, depending on the panelling. You might want to go with one degree increments if you're just getting started, since having results right away is important :) The first image I uploaded below has the results you typically see from this sort of run.
5) Still in 'oper', I typed in 'psor' to sort the polars I computed, then 'pplo' to display them. The second image I uploaded below has the results for several runs, all of SA7024 at varying Reynolds numbers, and all with a trip at 20% of chord. The cyan line is a run I executed with a camber of 3.0%, vs. the usual 3.25% camber for SA7024.

There're some things to watch out for:
* If the RMS error (when you're doing a run, the number right after 'rms: ...' in the output) goes really high for long periods of time, above 1.0x10^5 for instance, I find it's time to reload the airfoil data points. I don't know how it gets itself into such sitations, but it does.
* If you get an error message that says 'Compressibility calculations invalid|too high', it's time to stop XFoil and start over. Again, not sure why it gets into such situations... apparently it has something to do with sharp trailing edges (from what I've read in the Yahoo! newsgroups), but I'm not sure.
* The plot screen will pop into the foreground whenever your cursor its it. I don't know who thought this was a good idea but I, for one, think it's the most annoying thing I've ever seen. If a window is behind others and I haven't clicked in it, I want it there, if you ask me.

Hope this helps!


-j

[raptor22]

Thanks alot guys, the info has been alot of help. I have been able to do simple stuff now, and its a start. I am going to purchase profili,a nd continue to experiment on the actual xfoil interface.