Questions regarding to the wing tip vortices modeling

[Eric Wang]

Hello Rob,

I tried to use OpenVSP to investigate the performance of OHS (outboard horizontal stabilizer) configurations. The configuration generally looks like Fig.1. The horizontal and vertical stabilizers are behind the tips of the wing, harvesting energy from the tip vortices like static vanes. 

When I tested a variant as shown in Fig.2, sometimes I got pretty odd results of span efficiency, CDi and CDt depending on the angle of attack and the span wise/chord wise location of the horizontal/vertical stabilizers.  For example, Fig.3 shows jumpy span efficiency looping through AOA of 0 to 10 deg. I guess part of reason could be relatively bad convergence as shown in Fig.4. However, sometimes I got weird results even with improved convergence. (I used 10 to 15 iterations)

When tried the configuration with only vertical stabilizer  + wing or only the horizontal stabilizer + wing in the model, the convergence improved a lot and the "jumpy" results also went away. 

One of my speculations, the observed odd behaviors are somewhat reflecting the physics that multiple tip vortex systems are interfering each other causing relatively bad convergence and singular results. However, I am not 100% sure if VLM code can be used for this kind of study. I think it could, but I might be wrong. If you can give me your two cents on the problem or some general guidance, I would appreciate. 

BTW, I found NASA papers about the wing tip vortex flow field experiments. I will do some modeling V&V and keep you posted on that. 

Fig.1: A common OHS arrangement

Fig.2: A variant of Fig.1. Note the span wise location of the vertical and horizontal stabilizers relative to the tip of the wing is pretty sensitive to the problem stated in this post  

Fig.3: Span efficiency of Fig.2 case. The delta is big between steps of AOA. E is really high, but it could be due to the selected reference span of the wing, not the total span of the wing + horizontal stabilizer. 

Fig.4: Convergence of CDi, CDt, and E of Fig.2 case at AOA=10 deg

regards,
Eric