Hae Goo Kang asked 11 months ago
1. Background
I found that unrealistically strong downwashes and the induced drag rise in the wing-fuselage juncture area in the wing+tail+fuselage analysis of T2 type resulting from the obvious virtual unrealistic vortex similar to the wing-tip vortex due to the sudden cutout of the center wing area included inside the fuselage, which resulted in unrealistically very high drag. (Refer please to the image files attached.)
As I know, the wing-fuselage joint has an adverse effect as follows:
1) Increasing wing-profile(form) drag by below around 25% and not for the induced drag since the induced drag due to the center wing portion located inside the fuselage disappears at least gradually in the wing+fuselage assembly.
2) Decreasing gradually the lift due to the cutout of the center wing portion located inside the fuselage.
2. Request
I think aerodynamics of the wing-fuselage systems treated in the classic aerodynamics references such as “Aerodynamics of the Airplane” written by Hermann Schlichting et al
should be incorporated to reallistically and practically predict lift and drag of the wing+fuselage assembly model.
Thank you.
Sincerely yours,
… Hae Goo Kang (+82-10-7543-7796) …
<Attachments>
Thanks for the feedback and suggestions. Unfortunately the image is not linked.
The connection of the wing to the fuselage is one of the trickiest parts to handle numerically and has been a difficulty since the beginning of xflr5. The main reason is that locally the pressure gradients and the velocity can be quite high, which in the absence of viscosity creates large pressure forces, downwash and drag. This is a characteristic of panel methods. Possible work arounds in flow5 are:
– increase the vortex core size, but this is tricky because if it’s too large it causes numerical errors witrh the small panels
– use the VPW; however there is still need for a buffer of wake panels which can be the cause of numerical interactions between wing and fuselage
– exclude the fuselage’s contribution from the calculation of moments.
I’m reluctant to add modifications from external sources because they would likely work for one model and less for another. Also they would need a lot of validation beefore I’m confident that they can be added to flow5 and there isn’t a lot to go on.
All in all the jibe 2 case presented in the validation section of the online doc is not too bad, even if there is a slight overestimation of lift and drag in the case of thick panel methods.
André