techwinder Staff asked 3 years ago
This thread is opened to provide insight and answer questions about the wake models in T6 analyses.
As a reminder, the wake model is an essential building block of the analysis, and there are several models which can be selected in flow5 depending on the type of analysis.
In the case of the LLT and VLM, the wake is represented by straight vortex lines extending in the direction of the x-axis, which with the assumption of small aoa is also the direction of the freestream flow.
In the case of thick surfaces and panel methods, the wake can be represented either by a flat panel wake (FPW) or by a Vortex Particle Wake (VPW). Historically the FPW came first and is the method implemented in xflr5. To my limited knowledge, the VPW was first proposed by D. Willis in 2006.
However the more I use the VPW, the more it appears to be a breakthrough development and the future of panel methods: it is a very elegant solution to the problem of wake relaxation, and it is tolerant to issues of intersections of wake panels with wing and fuselage panels. On the opposite, I tend to view more and more the VLM and other FPW models as legacy methods useful mainly for fast preliminary analyses and T7 stability analyses.
A T6 control analysis with a FPW aligned with the x-axis is technically possible but it would essentially be a T12 analysis without the viscous loop or the sideslip. I’m reluctant to add the feature because I feel that it would add more complexity and new risk of errors to the current situation.
The current options may be confusing and could be improved, but I don’t see a way to do it without adding more complexity to an already overloaded interface.
With the benefit of the experience so far, my recommendation is to use the T12 analyses for quick preliminary designs and the T6+VPW in the final stages when more precision is required.
In all cases the preferred method is the triangular, uniform, thick surface panel method for its numerical robustness and versatility.
All feedback and suggestions are welcome.
André
Szymon Szyszko answered 3 years ago
Hi Andre,
Thank you for opening this thread.
My personal opinion, biased by the use-case (optimisation), is that although VPW wake is great, it is computationally expensive. It is useful for better accuracy, but in my opinion, the gain compared to cost is for most of cases not worth it. It is great that flow5 offers alternatives for better accuracy, and I cannot wait to see more great stuff being added as time passes, but equally important for me is to have the “quick-and-dirty” method well supported and giving me all the necessary analysis options. For example, there is currently no sensible way of getting beta to sweep analysis for flat panel wake model for aicraft with the fuselage as recalculated wake in T6 will likely intersect with fuselage panels (making results so inaccurate that it cannot be assumed correct). Results for flat-panel wake could be quite comparable with VPW in beta sweep analysis otherwise, for fraction of the calculational time!
To summarise, I think flow5 is a great tool, with even greater potential. Low fidelity tools without fuselage consideration and higher fidelity tools such as T6 + VPW + viscous loop work great and do their job super well. Unfortunately the middle ground, the compromise between those two could get slightly more attention! Not sure if this is only my personal view, or others would share my opinion, but it would be great to see some improvements for non VPW analysis!
Cheers,
Szymon