NASA Drag Buildup Workbook

[Bruno Guerra]

I am currently on an Aerodynamics team working on the design of an aircraft.
We have a few questions with regards to the "NASA Drag Buildup Workbook" file and would be immensely grateful if you could shed some light on it.

Link OpenVSP: http://openvsp.org/wiki/doku.php?id=dragworkbook

We understand that the Wing Parasite Drag (CD0,wing) is calculated using the "EDET Advanced Airfoil" form factor equation, as per "Delta Method, An Empirical Drag Buildup Technique, Nasa Report".
According to said report, this form factor only contributes to the friction drag (CDf,wing), meaning that the form drag (pressure drag due to separation) is not being taken into account. This is seemingly in conflict with
most references we have access to, including the OpenVSP documentation itself, which state the contrary.

In the same vein, we struggle to understand why the value for the Wing Parasite Drag (CD0,wing) is summed with the Profile Drag (CDprofile=Cdf+Cdp) since this means that the Friction Drag is, presumably, being accounted for twice.
From this perspective, it would make the most sense to ignore the Friction Component from the Profile Drag (Cdf) and factor in the Pressure term only (Cdp),
resulting in CD = CD0,wing + CDi + Cdp (Profile Pressure Term) + CDm (Miscellaneous terms).

Best regards, 

Bruno Guerra

Aerodynamics Engineer at CEiiA, Portugal

[Rob McDonald]

Bruno,

The file that fed the Drag Buildup Workbook has not been properly supported since OpenVSP v2.  Since that has been deprecated for many years, we aren't actively supporting the tool anymore.

What led you to that link on the Wiki?  While the page has not been removed, I thought all public references to it had been removed.

As a replacement, OpenVSP now has a built-in parasite drag buildup tool http://openvsp.org/wiki/doku.php?id=parasitedrag that should meet your needs (and perhaps will have similar questions).

Rob

--
You received this message because you are subscribed to the Google Groups "OpenVSP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openvsp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openvsp/eeb3cc96-6a67-4de5-97fa-844c2068d2fan%40googlegroups.com.

[Bruno Guerra]

Thank you for your answer, Rob.

With respect to your inquiry, a search for "Drag" in the main wiki page turns up an entry named "dragworkbook", through which you have access to the files at hand.

As for the matter of the built-in parasite drag tool, we have indeed been experimenting with it, and with that in mind, we'd like to pivot to the newer tools and recontextualize the question.

According to the original NASA report, referenced in the prior exchange, the formulation in use by the Parasite Drag Tool pertains only to the friction component of drag, seeing as all the components are calculated in the same way. Should we, however, assume that the correct interpretation is the one presented on the wiki itself, whereas the form factor provides the contribution related to the pressure drag independent of angle of attack?

And assuming that the induced drag obtained through the VSPAERO code is fully lifted dependent, what would you advise as the best way to account for the form drag, separation, due to variations in angle of attack, or if it should be neglected.

Thank you for your time,

Bruno Guerra

[Neal Pfeiffer]

Bruno,

Excrescence drag is very important on small aircraft.  It is typically calculated by using a detailed buildup from a careful inspection of the configuration.  I've attached a paper from 2014 that has various references and an example.

..... Neal

https://groups.google.com/d/msgid/openvsp/ff4e6c0a-c25f-444d-96ae-8f707d567230n%40googlegroups.com

.

[Rob McDonald]

Bruno,

I suspect the documentation for the NASA spreadsheet was written somewhat informally, I would not apply the most strict interpretation to their words.

Typically form factors account for form drag.  Form drag is a pressure drag -- but that is viscous in origin.  It relates to the thickness of the boundary layer at the trailing edge.  While it is often treated as independent of lift coefficient, a more thorough analysis will reveal that it is actually a function of aoa or cl (even in 2D).  If you look at the data in Theory of Wing Sections, you see largely parabolic behavior for 2D drag data -- it is not induced drag, but lift-dependent form drag, which is a viscous drag.

VSPAERO has a simple form drag model in it.  It is based on some empirical data for NACA airfoils.  It is sensitive to local Reynolds number, thickness to chord, and local lift coefficient for wing-type bodies.  It is sensitive to finess ratio and Reynolds number for non-lifting bodies.  You will see it output as a separate term -- the term can vary with lift coefficient.

This calculation will only include contributions of components included in the VSPAERO Set.  So, if you throw out some components, they certainly will not be included.  This model can do a good job -- particularly if you have a wing with a lot of taper (large Reynolds number variation) and other things like that.  I really like it for rotating wings -- it is the best way to get a parasite power contribution on props and rotors.

OpenVSP's Parasite drag tool uses form factor and wetted areas to do a traditional drag buildup.  The user can select their preferred method at every step of the way.  The user can also input interference factors, etc.  There is an additional sheet for including excrescences.  The lifting surface contribution is not a function of alpha (cl).

I see these tools as complimentary.  At a minimum, you could use VSPAERO's inviscid induced drag term along with OpenVSP's Parasite Drag tool for a simple buildup of the remaining drag.  For an aircraft with props and/or rotors, I would suggest using the VSPAERO parasite power.  Then, as you want, you can include the VSPAERO parasite drag term and replace the corresponding terms from the OpenVSP drag buildup.

As always, these things take engineering judgment and some experience.

Rob

To view this discussion on the web visit https://groups.google.com/d/msgid/openvsp/ff4e6c0a-c25f-444d-96ae-8f707d567230n%40googlegroups.com.

[Bruno Guerra]

We will take a look at the attached paper. Thank you, Neal!

Rob,

First of all, thank you for your answer. We think many things became clearer, and the NASA spreadsheet will be treated very carefully.

Thank you for your time,

Bruno Guerra

[Brandon Litherland]

Bruno,

To be fully transparent, the drag buildup workbook was intended to provide a common method of accounting for other drag sources outside of VSPAERO.  Although partially customizable, it took a fair bit of understanding of the individual entries to get a reasonable buildup even back during the Version 2.X days of VSP.  Since the Parasite Drag analysis in OpenVSP, this workbook has not been supported as Rob mentioned.  Perhaps it's time to take it down completely to avoid confusion.  In the VSPAERO Polar file, you'll find that CD0 is lift-dependent in newer versions which is the form drag that Rob talks about.  If you run an alpha sweep, you'll find a minimum of this value which I tend to remove from the VSPAERO results as the lift-independent parasite drag.  The lift-independent component of drag is added back in from the Parasite Drag tool results as CDo.  So for a relatively simple breakdown you'll have induced drag both from lift (CDi) and form drag (CD0 - CD0_min) plus the Parasite Drag CD0.  If you want to experiment a bit, you can also turn on the stall model estimation using either a 2D Cl_max or the Carlson Pressure Correlation to see if suppressing the peak CL improves your polars.

Hope this helps.

- Brandon

[Bruno Guerra]

Thank you very much for your answer, Brandon.

That does indeed clear up most of the leftover doubts we currently had with respect to the drag component distinctions, and this knowledge will certainly be of aid for the next stages of our work.

Best Regards.

-Bruno