Further Explanation on Dynamic Analysis

342 views
Skip to first unread message

Juan Buritica

unread,
Feb 27, 2024, 4:18:28 PM2/27/24
to OpenVSP
Hello Rob,

I've been playing around with VSPAERO stability functionality and I would like to clarify some doubts:

1. If steady is sufficient for static analysis and PQR is made for dynamic analysis, what is the purpose of the "Pitch" analysis?

2. On the OPENVSP documentation website, the layout of the csv files seems different than the way that is currently displayed on version 3.36. I was wondering if you could clarify the syntax of how the coefficients are listed in the csv files containing the results for the PQR analyses. For example, in the attached  picture,  is Alpha_CD equivalent to the derivative of CD wrt Alpha?

Thanks!
image_2024-02-27_155236057.png

Rob McDonald

unread,
Feb 27, 2024, 5:40:52 PM2/27/24
to ope...@googlegroups.com
Pitch is a shortcut analysis for when someone only wants to calculate the aerodynamic center and static margin.

Steady runs seven cases -- pitch only needs to run two.

I believe your interpretation is correct.

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/d88a27db-d725-4709-a649-30580cc9be6en%40googlegroups.com.

Juan Buritica

unread,
Feb 27, 2024, 7:29:52 PM2/27/24
to OpenVSP
Thanks for your response Rob,

Regarding the Pitch case, I've tried running it but I don't know how to use it for the purpose you describe. To determine de Neutral Point, I've used the regular VLM analysis to get the Cm vs Alpha plot and then iterate through different CG locations until the slope of that curve is practically 0. If the Pitch function does this quicker and more precise by all means I'd like to know how to use it properly.

Rob McDonald

unread,
Feb 27, 2024, 8:12:33 PM2/27/24
to ope...@googlegroups.com
All it does is run two angles of attack.

From this, it estimates dCM/dCL which is the static margin.  It uses your CG position and reference chord to determine the neutral point location.

This is reported in the *.stab file -- I don't think we read it back into the GUI.

Rob


Juan Buritica

unread,
Feb 28, 2024, 10:39:26 PM2/28/24
to OpenVSP
Gotcha. I've proceeded to execute the analyses and had a few questions regarding their output files.

1. For the Steady Analysis, the stability results first appear to be with respect to angle dependent variables such as deflections, alpha and beta, however, if you keep scrolling then the breakdown changes on a per coefficient basis (Starts with CD and lists all the derivatives wrt to CD). I find it confusing because the coefficient values from both breakdowns do not match and I was wondering if you knew why they are different (See attached pictures).

2. For the Q and R analysis, is there a reason why the derivatives wrt to q and r are shown as q+alpha_dot & r+beta_dot respectively? I was thinking they were shown that way to convey that at small angles q=alpha_dot and r=beta_dot, but I wanted to clarify that.

Thanks! 
image_2024-02-28_223320304.png
image_2024-02-28_223213899.png

Rob McDonald

unread,
Feb 29, 2024, 12:19:16 AM2/29/24
to OpenVSP
CL and CFz are not the same thing, one is rotated by alpha.  Same goes for CD and CFx.  They will be similar, but there are trigonometric transformations between them.

The Q and R analysis does an unsteady dynamic analysis that can not (on its own) separate the steady q component from the unsteady alpha_dot component.  If you really need them separate, you can subtract off the steady q term from the static analysis.

However, these two terms show up as a sum in the equations of motion, so you typically don't need the q and alpha_dot derivatives separately -- so we just present them as the sum.

Rob

Tim Swait

unread,
Feb 29, 2024, 6:42:44 AM2/29/24
to OpenVSP
What needs to be present to do a stability analysis? If I select anything from the 'Stability Type' drop down in Run Mode then it just flicks back to 'Off' each time. I am doing a hang glider wing in VLM mode. It's just a wing, there is no empennage, no fuselage and no control surfaces defined. Also I haven't defined masses other than setting the Xref point for the CG. Do some of these things need to be present to run a stability analysis? At the moment I'm just manually calculating the slope of CMy versus alpha slope to get pitch stability and haven't started looking at the other types.

Juan Buritica

unread,
Feb 29, 2024, 8:34:44 AM2/29/24
to OpenVSP
Thanks Rob for your response.

Yes I am aware of that rotation, I apologize if I didn't make myself clear. What I was referring to, for example, if you look at the values of CD_alpha on each picture they are different although they belong to the same analysis. That seems to happen with other coefficients as well, I was wondering if maybe it was an unit change  (deg to rad) or something that sort.

Thanks for clearing out my doubts about the q and r analyses!

Juan

Rob McDonald

unread,
Feb 29, 2024, 12:05:13 PM2/29/24
to ope...@googlegroups.com
I'm sorry, I have no idea where that screenshot is taken from -- what part of what file.  Could you please just post the file you're confused about?

Rob


Rob McDonald

unread,
Feb 29, 2024, 12:06:33 PM2/29/24
to ope...@googlegroups.com
Nothing special.  There must be something else about your file.  Can you post your *.vsp3 file?

Rob

Tim Swait

unread,
Mar 1, 2024, 11:23:35 AM3/1/24
to OpenVSP
Hi Rob, there's an example file here. If I open it in OpenVSP, go to VSPAERO, click the Advanced tab, select any of the options for Run Mode>Stability Type then it just flicks back to 'Off' regardless of what I click on.

Rob McDonald

unread,
Mar 1, 2024, 12:05:20 PM3/1/24
to OpenVSP
It is because you have the XZ symmetry option on -- that disables the Stability options.

It should probably still allow them for at least pitch stability.  And possibly Q stability -- but the others involve inherently asymmetrical flow conditions.

Rob

Tim Swait

unread,
Mar 3, 2024, 3:20:25 PM3/3/24
to OpenVSP
Ah, thanks, that makes sense. Although I am only interested in pitch for the time being I will need to include the other axes.
Reply all
Reply to author
Forward
0 new messages