Hi,

I'd recommend using the frontal projection area at zero Angle of Attack. For the length scale I'd suggest the largest diameter at zero Angle of Attack.

cheers!

I agree completely with Mr. Sander.

thank you Mr. Sander and Mr. Peter for replying,

regarding the characteristics length, i forgot to mention that the major axis of the elliptic cylinder is parallel to the flow for zero angle of attack. so for other angles of attack, should i use the projected width of the cylinder or the major axis for the length scale?

thanks.

I think you will have a more accurate Reynolds number if you used the projected width at the angle of attack in question.  Reynolds number is ratio of inertial to viscous forces and the inertial forces will change with angle of attack.

thanks Mr peter,

one more doubt, as u suggested to take the frontal projection area at zero angle of attack for non-zero angles of attack also, will this be valid for the the 90 degree case also? i am studying different configurations from zero to 90 degree.

Well, technically you can use a different projection area and length scale at every angle of attack. However, If you want to compare angles of attack (for instance in a polar diagram with drag and lift coefficients) length scale and projected area should stay constant at all times. As a reference you can take a look at Abbott & von Doenhoff (1959) or at hoerner (1965)

Hoerner is a good reference. But, any reference area is good so long as you are consistent. Don't vary that with angle of attack, because you will confuse yourself and others by doing that. Just use the major axis for the length scale.

Making a parallel to what one usually does for a wing at incidence, the reference area for the calculation of forces should remain constant. For the cylinder it looks reasonable to employ the cross-sectional area across the axis of the cylinder (a rectangle). It should not change as the angle of attack is changed. However, it is also usual, for the calculation of other non-dimensional quantities, such as the Strouhal number for vortex shedding (which eventually will occur in this case) to consider the projected area in a section perpendicular to the flow direction (i.e. taking the sinus of the angle of attack) because, in this case, the physical phenomenon is actually directly depending on that scale. When accounting for aerodynamic forces it is a different situation and one definitely needs a fixed reference area (though different choices can be made) to be able to make a sensible analysis of the results.