Search the 1638 airfoils available in the databases filtering by name, thickness and camber. Click on an airfoil image to display a larger preview picture. There are links to the original airfoil source and dat file and the details page with polar diagrams for a range of Reynolds numbers.
Included below are coordinates for approximately 1,600 airfoils (Version 2.0). The UIUC Airfoil Data Site gives some backgroundon the database. The airfoils are listed alphabetically by the airfoil filename (which is usually close to the airfoil name).Answers to frequently asked questions are postedhere Airfoils FAQ and for the mostrecent changes see theupdate history.
When I looked at the 2D data of the several airfoils I found some unusual high values for the cl slope over alpha (e.g. dcl/dalpha>3*PI/rad at alpha0 for DU40). That is very surprising to me to have a slope 1.5 times higher than in thin airfoil theory for such a thick profile in the linear area and only looking at 2D data. I know you did not measure the data by yourself but maybe you have some informations which could help understanding?
AeroTwst - Rotation of the chordline of the airfoil relative to the zero-twist location, such that the blade pitch angle plus AeroTwst is the angle between chordline and the rotor plane (neglecting structural deflection such as elastic twist).
StrctTwst - Rotation of the principle elastic axis for edgewise bending of the airfoil relative to the zero-twist location, such that the blade pitch angle plus StrctTwst is the angle between principle elastic axis for edgewise bending and the rotor plane (neglecting structural deflection such as elastic twist).
I am working on hydrofoils and windsurfing fins. I am using airfoils from the UIUC Airfoil Data Site. What I have been doing is using the .gif images, converting them to .jpg to import into Sketchup, then tracing them. But I have trouble getting it precise and clean enough, and I get flat spots on my foils. I am trying to just import the data files, instead.
Foil23by551659900 302 KB
Little addition: the airfoil data can be downloaded from the database as ascii files (extension is .dat). Rename it to .txt and it can be opened with Javafoil etc. To convert the airfoil to x-planes .afl format, use FoilTrans.
If you really want to calculate some accurate subsonic polars, become familiar with MIT's XFOIL program. It will not do supersonic airfoils..... but for compressible subsonic its the best out there! its all through the command prompt, so it takes some getting used to, but it is a GREAT and FREE program it even has reverse calculation of airfoils..... you define the pressure distribution and it will return the geometry that corresponds to that distribution.
Ex: Outdated, you should not use these airfoils in high performance designs
S: Flying wing airfoils are characterized as reflexed, in German called: "S-Schlagairfoile"
F: Airfoils developed for the application of flaps or working well with
new or edited airfoil
MH45-8%
Somewhat faster than the original 45er, but a lack of clmax. Therefore stall is coming like a shark during high speed turns for example. A very fine training for your flying skills but no insurance for a long life...
It has not been one of my best ideas modifying the airfoil this way. To get an idea of the influences: the lack of camber is compensated by higher thickness. Lower thickness together with higher camber works well too. But a lack of camber and thickness causes trouble on this specific kind of reflexed airfoils. E.g. 1,5/10 (camber%/thickness%) and 2,5/8 works well too. But 1,5/8,0 can cause trouble!
MH60
The LOGO Team often flew this section. You can believe me: they were damn good! It was no fun to fly against these guys that time! Surely a fine design of Martin Hepperle, perhaps somewhat better than the MH45. This difference seems to be more induced by the resulting Tailless design than by the airfoil itself. The pitching moment (cm, cm0.25, cm25, cm1/4 however called) of the MH60 is lower than that of MH45. So for same basic cl you need more wing washout when using MH60. This causes more differences in handling than the airfoil itself.
Attention: Some pilots have had trouble with a marginly negative pitching moment! E.g. LOGO Team have had a model (HOLON) with a design cl around 0.1 (bad idea!!!). Stabile flight in clean wing configuration with MH60 had been upside down!!!
TL54
Thorsten Lutz has developed the airfoil series TL54, TL55 and TL56 using Eppler Code. The airfoile series is already some years old, but has been hardly known in tailless scene. The difficult relations concerning stall behaviour on Tailless using different airfoils does normally not concern to the requirements of competition RC-Models. The potential benefite using different airfoils is minimal due to the flight behavior. Often it is so worse that just Tailless speed models use different airfoils. But even F5D designs normally use just one airfoil because cd optimization means reducing cd where cl requirement is small. And where is cl small? @Wingtip. What means stall @wingtip? Ok, the influence of reduced cd on stall behavior should be quite clear now. RC pilots normally don't like digging their models...
Therefore we do not talk about teamwork performance of this TL series, just about possible application using one airfoil for wingdesign.
The main attention during design phasis aimed for long laminar run on airfoil lower surface at high speed (cl=0.1 .. 0.0). This is the basis to achieve low airfoil drag at highspeed. Inspite of 2,41% camber TL54 reached this design goal. We have here an airfoil that provides a lot of lift (clmax) despite very low drag at low CL! The direct consequence is an extremely wide area of application! Now the F3B designer should become poorely soundproofed among us, because the secret of the MH-32 is reasonable just this...
TL-55 is not suitable for F3F because upper surface laminar run is not as long as on other airfoils. The MH-60 might be here better choice. In negative flight figures (upside down) the TL55 will be much better than the usual suspicious . Therefore, as section for the ZAGI it would be certainly interesting, beside Sipkill, a special section for Foamies.
Moreover, it knows how to polish in contrast to EH1,5/9 laminarer run with more than 95% on the airfoil bottom. On account of these qualities the airfoile resistance is much lower opposite the EH1,5 / 9 in all areas how one recognizes in the comparison-polar. In addition, the TL56 has the nose in front on account of the very about formed airfoile nose in the stunt flying. Only in torn figures the EH1,5/9 will engage better and make possible faster figures , in all other the TL56 is partly even very clearly consider.
RS 004A
One of the newer creations for Swept Wings. Design by Sielemann/Hans-Jürgen Unverferth. Due to negative pitch moment a lot of washout is needed. I suppose same handling characteristics (something strange!) as tailless designs using SD7003.
This RS 004A has been used on CO8 and this model have had handling problems they never get rid of (Source: Sielemann). This was probably induced by the airfoil. They never managed their problems in flight stability around pitch axis. For a first attempt in low-cm pitch down airfoils it is not a good choice I think. SD7003 is a better choice to do this for the first time I guess.
I suppose that the real cm is lower than calculated. Having too low cm in design and later on correcting this by using Flaps/Elevon is always extremely critical!!! Often you never get rid of stability problems around pitch axis. I know this problem from several designs of myself!
S5010
Quite well but something too thick. Due to this it is something gust sensitive. Direct flight comparison between S5010 and S5010/8% showed this on a quite heavy day. Wind around 20-30kts, gusts up to 40kts. The model with the origin airfoil had some problems in holding course and there was an interesting effect on height axis: when hard gusts came in, the model swang for 1-2s! The thinner airfoil in same conditions (parallel runs) was faster and did not show this swinging. Both models speed was around 100-150kts, quite fast. But these are the conditions in competitions where this can decide about victory or not. This was the final reason for using the thinned airfoil for my F5B contest model. In the meantime I know that a thickness about 9% does not cause these problems on this kind of model.
Never forget that these experiences are tight related to the test model HS40 (F5B): Electric Engine 1.7kW (2.1hp), climb rate 50m/s (60-70kts), wing loading 75g/dm FAI (=24,58oz/ft) and weight around 2kg (4 lbs). That means: For a F3B or allround tailless design this has not much to say. But this example shows that you can run into trouble in flight dynamic induced by airfoil beside cl/cd and velocity distribution diagrams. Ok?! Not more! Ok, back to the unmodified airfoil: clmax is high enough, high start works well. Testwing for this airfoil had been HS40V1 "Sexxy".
(This airfoil is also known as S 5010-098-86)
S5010-8%
Does not have the before mentioned problems in gust sensivity. It is a quick and very agil airfoil, some people would say nervous. I like this but it is not everybodys thing. Than use the S5010-9% and it is ok. Tailless in F5B-600 can use this section. For F5B it could be a good choice. But therefore I recommend a S5010 (2,4/8,5) because of high speed turn radius. Back to S5010-8%: in all weather conditions it works well. For slope soaring it is perhaps one of the best allround sections I have ever seen.
MH45 is better at lower wing loadings (15-40g/dm = 4,9-13,1oz/ft), S5010-8% around 20-80g/dm, S5010 35-100g/dm, S5020 is a good choice for 25-90g/dm. Just to get an idea what you can do with these sections in allround conditions. This data given is not to discuss about 1 or 2g/dm where a section works or not. But for small up to midsize models (1-4m) this could be helpful. No, I don't want to talk about Rn-numbers, because handling characteristics have much more to do with model's dynamic behaviour and its airfoil than actual Rn. So I think it is more helpful for you to talk about wing loadings...