Using OpenVsp for conceptual sizing

[Alexandre Borowczyk]

Hi,

I am working on modifying a model I have into a canard configuration.

I'm trying to size my tail and I am confused on how I can obtain information such as the neutral point or the wings MAC?

Thanks,

Alexandre

[Neal Pfeiffer]

You may want to look at NASA-TP-2382

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19850011615.pdf

Long Yip carried extensive testing on a full-scale Long-Eze airplane.

I survived the Starship program during the same time, and it was customary to use the trapezoid wing area as the reference area, and likewise to define the MAC from the trapezoid wing.  I think that that is what you will find in the report above.

Note that unless fuel is put into the fuselage, it needs the large strake tanks ahead of the wing to put the fuel on the CG location.  That adds considerable weight and wetted area to the configuration.

If you look into it very far, you will find that canard airplanes are inefficient compared to a well-designed, tail-aft configuration...

..... Neal

---

From: Alexandre Borowczyk <borowczyk...@gmail.com>
To: OpenVSP <ope...@googlegroups.com>
Sent: Friday, February 9, 2018 11:30 AM
Subject: [OpenVSP] Using OpenVsp for conceptual sizing

--
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.
For more options, visit https://groups.google.com/d/optout.

[Alexandre Borowczyk]

Thanks for the amazing reference!
Efficient is not currently the main driver for the configuration, well as long as it's not to inefficient. Also, the model I'm working with is RC so it's fully electric which alleviate some concerns.

I'm confortable in hand calculating the MAC and Neutral point. However, I wanted to learn how to get then with vspareo and collaborate my results and see how more complexe geometry may affect accuracy of the "hand"calculations.

[Neal Pfeiffer]

Lee Nicolai's updated aircraft design book has S&C equations for canard aircraft.

https://www.abebooks.com/9781600867514/Fundamentals-Aircraft-Airship-Design-Aiaa-1600867510/plp

I'm sure that Jan Roskam's books would also have them.

http://www.darcorp.com/Books/

Good Luck!

..... Neal

---

From: Alexandre Borowczyk <borowczyk...@gmail.com>
To: OpenVSP <ope...@googlegroups.com>

Sent: Friday, February 9, 2018 12:52 PM
Subject: Re: [OpenVSP] Using OpenVsp for conceptual sizing

Thanks for the amazing reference!
Efficient is not currently the main driver for the configuration, well as long as it's not to inefficient. Also, the model I'm working with is RC so it's fully electric which alleviate some concerns.

I'm comfortable in hand calculating the MAC and Neutral point. However, I wanted to learn how to get then with vspareo and collaborate my results and see how more complex geometry may affect accuracy of the "hand"calculations.

[Rob McDonald]

If you run a 'Stability' case in VSPAERO, the X location of the
Neutral Point will be output in the *.stab file.

It is the result of running two alphas, and then examining the slope
of the CM vs. CL curve. It will work for an arbitrarily complex
configuration (any combination of wings and whatnot) for either VLM or
Panel method.

The value of cbar is an input. Since it is a reference quantity, it
can be anything -- so long as you are consistent.

VSPAERO does not have a concept of the location of the MAC, but you
could back one out that is reasonable. Moments are calculated about
the specified CG location. If you don't specify a CG location, the
Neutral point location will still be correct.

Rob

[Alexandre Borowczyk]

Thanks for the info Rob,

Out of curiosity how do you find the Np from the CM vs. CL curve?

Also what do you mean by cbar?

If I understand correctly, running the stab analysis with only a wing the np obtain should be the wing aerodynamic center right?

@Neal,

I was think of getting this book: https://www.amazon.ca/gp/product/0123973082/ref=ox_sc_sfl_title_1?ie=UTF8&psc=1&smid=AMR50ZI4JZU9U

Do you know it? 

[Rob McDonald]

cbar is the reference chord. Typically also the length (not position)
of the MAC. Like the reference area Sref and reference span bref, it
is a selected value for a design that is used for
nondimensionalization. We nondimensionalize forces by Sref, and
pitching moments by Sref*cbar, and other moments as appropriate.

The definition of the static margin is:

SM = -dCM/dCL = (xnp-xcg)/cbar

I might have some sign errors in this, doing it off the top of my
head. Draw out the statics problem for longitudinal stability and
you'll convince yourself of this pretty quickly.

So, if you run two AOA cases in an aero code and track CM,CL at each,
you can get the slope dCM/dCL. If you know your chosen cbar, and the
reference point for the CM (xcg) then you can quickly find xnp. By
default, you might set xcg=0 and cbar=1. In that case, the math is
easy enough for me to do.

Yes, the aerodynamic center of the wing is the same thing as the
neutral point of a wing-only configuration. AC is a term that refers
to a flying surface in isolation. NP is a term that refers to a full
vehicle.

Rob

[Neal Pfeiffer]

Alexandre,

I have that book as well, but if you want more info on canards, I would suggest Nicolai.  He describes tail aft, canard, and no tail all together so you can see the math relationships.

..... Neal

---

From: Alexandre Borowczyk <borowczyk...@gmail.com>
To: OpenVSP <ope...@googlegroups.com>

Sent: Friday, February 9, 2018 5:41 PM

Subject: Re: [OpenVSP] Using OpenVsp for conceptual sizing

[Alexandre Borowczyk]

Awesome, thx for the help.

One last thing, how the values obtain should be treated in terms of accuracy?

If I understand it's not CFD but what interpretation can be made?

Thanks

@Neal, thx I ordered Nicolai's book!   

[Neal Pfeiffer]

Alexandre,

Vortex lattice, panel methods, or higher-order computational aerodynamics should all give reasonable answers as long as the flow remains attached to the body.  Once separation occurs, all of the methods start deviating from the actual behavior.

Lift and pitching moment will be most accurate, and drag will be least accurate.  If boundary layer corrections are made (potential-flow method or Euler method coupled with a BL code or Navier-Stokes code) the lift curve and pitching moment curve with attached flow will be more accurate.  Even with BL corrections, the drag will have inaccuracies, since there is no simple way to accurately model all of the excrescence items of a configuration.  For a small airplane, the drag from these excrescence items can be 40% or more of the basic wetted area friction drag.  Even for a large airplane they can be 10-15%.  But the trends of drag deltas due to configuration changes can be examined with many of the CFD codes.  Just understand the capability of a given method.

Good luck!

..... Neal

---

From: Alexandre Borowczyk <borowczyk...@gmail.com>
To: OpenVSP <ope...@googlegroups.com>

Sent: Saturday, February 10, 2018 1:37 PM

[Alexandre Borowczyk]

Cool thanks for the information! I definitively need to read more on this.

So I understand it's important to keep the analysis domain far from stall.