[AIRCRAFT] AIRFOIL SPECIFICATION

Hi Matt,

(here is a link to the Aerofly Wiki airfoil page: https://www.aerofly.com/dokuwiki/doku.…aft:tmd:airfoil )

When you use airfoiltools make sure to select a proper Reynolds number range. For slow and small aircraft you can start at 100,000 but for faster and larger aircraft (think Learjet and bigger) I'd keep it at 1 Million or greater. After ticking the checkboxes press "Update plots".

pasted-from-clipboard.png

On the Cl/alpha plot you can extract the Cl0, AttachedRange, AttachedCenter and StallRange parameters.

• Cl0 is easy, just follow the 0° line up until you hit the curve, then read it on the left.
• The AttachedCenter is at 50% between the stall in negative angle of attack and stall in positive angle of attack. Sign is flipped for some reason. So about 1-2° in this case, which comes to 0.017 or 0.035 radiant, let's set 0.025.
• The AttachedRange is the angle difference from the AttachedCenter point to the beginning of the stall, about 10° or 11° or 0.17 to 0.19. Let's use 0.18 for now.
• StallRange is the angle difference from beginning of stall to fully developed stall but this has to be adjusted later in the sim. The plot stops prematurely, so we don't know when the stall is fully developed, I'd go with 8 degrees for now, so 0.14 or so.

pasted-from-clipboard.png

From the Cd/Alpha plot you can extract the Cd0 and get an estimate for CdAlpha

• Find the lowest drag coefficient, that is Cd0
• The formula to extract CdAlpha is quite long, so I won't go into much detail. Basically if the drag increase as angle of attack is high then I'd go with a value between 0.5 and 0.65 for CdAlpha, typical for rounded nose with thickness at 10-30% of the airfoil chord length. If it's a high performance glider profile and the thickness is further aft then I'd go with a lower value of 0.3 to 0.35 or so. Default is 0.4. In the sim, if you fly at high angle of attack and the sinkrate is not enough or you use too little power then this value should be increased. If the drag is too much at high angles of attakc then you need to lower the CdAlpha.

pasted-from-clipboard.png

The CmAlpha plot often looks more like a U-shape but the basic airfoil in Aerofly only simulates a linear function. The airfoil2, used in our latest aerodynamics can simulate it but that is still work in progress. This is where some guess work and trial and error comes in.

• Cm0 is the value of the moment coefficient at 0° angle of attack. For most airfoils with positive camber the Cm0 is negative.
• CmAlpha is the gradient of the moment coefficient as angle of attack increases. An airfoil for flying wings with an s-shape tail will have a negative CmAlpha. Most other airfoils with positive camber have a CmAlpha above zero and a negative Cm0.

In this example I would go with a Cm0 between -0.05 and -0.08 and CmAlpha of 0.033 per 6° or 0.033/0.105 or 0.31.

It is important to set the incidence angle of the aerowing correctly first. Then you can play around with variations in Cm0 and find the Cm0 at which point the aircraft flies in a trimmed out configuration. If Cm0 is too negative the aircraft is going to nose dive and may pull up very late, if at all. If Cm0 is not negative enough then the nose will pull up and the aircraft might stall. This is one of the best values to calibrate a known real world pitch trim setting to a given airspeed. So if you know at this and that CG with this and that mass you need this much trim you can use Cm0 to fine tune this value. But the aerowing incidence has to be set correctly first to make this possible.

pasted-from-clipboard.png

Aerowing geometry, including dihedral can all be done with the geometry coordinates in the aerowing. No need to split it into multiple sections if the dihedral changes, just increase the Z coordinate of the sections. This is where the program tmEdit comes in handy, just pull the leading and trailing edge onto the 3d model geometry, then fine tune the incidence values until the leading edge and trailing edge both are aligned with the model and you should be all set. If you do this manually you need to define the stations by hand, which takes longer but can be done.

• In your wing in the TMD clear all the station list entries.
• Pick the 3D point that is the leading edge, enter it's coordinates (StationX, Y and Z) into the tmd. Pick the 3D point at the trailing edge at the same y-coordinate (same left/right position). Work out the angle from the chord line to the horizon - that is your incidence value
• Then proceed to the next point, add stations at each point where either the geometry has a kink or where the flap changes
• Then fill in the flap fractions and station flaps