Understanding the Solver for a Senior Design Project

[Thomas Dylewski]

Hello y'all!

I am a senior working on our senior design project in college where we will be designing a small drone that will fly at the end of the year. We are currently working on optimization and simulation, and our department has purchased several Flightstream licenses and we would like to use it to help understand the flight conditions of the drone.

I have created a mesh in OpenVSP as per the User Manual's best practices, and have imported it and followed the tutorial for the 737 to set everything up. When it came to set up the solver and run, I was not quite sure as to what to input for the reference values. When I do run the solver with the same reference values for the 737, I get unusable data at our calculated trim AoA (about 2 degrees).

I am not quite sure how to fix this, so some help would be lovely as we wrap up this semester.

Thanks!

Drone Specs:

Weight: 2.04kg

Span: 1.73m

Wing Area: .5017 m2

Cruise Speed: 25m/s

Chord (rectangular wing): .29m

Trim AoA = 1.32 deg

Here is the mesh file that I am using.

Here is the mesh in Flightstream and the Solver view.

[cory goates]

The reference parameters are whatever you want the output values (CL, CD, Cm, etc) to be non-dimensionalized by. Using the same ones as for the 737 will definitely give you weird results. Typically, the reference velocity is just your current velocity; reference area is the wing planform area; and reference length is the wing chord length, or something similar. But those are just usually what's used.

[daniel.enriquez]

Hi Thomas, 

I reviewed the geometry you sent. 

There are a few issues that make your geometry different than the 737 tutorial. 

1) The vertical stabilizer needs to fully intersect the boom, otherwise the Unite operation will give weird results. I suggest either increasing the boom size or decrease the vertical stabilizer width so that they can be fully there. Alternatively, you may consider removing the booms entirely since they contribute very little to the aero forces. 

2) The horizontal stabilizer does not have a clearly defined trailing edge. This may give weird results in FlightStream. I'd suggest changing the airfoil so that it is a more standard shape. 

3) The vertical stabilizer has a blunt trailing edge. Make sure you apply 2 trailing edges and a base region as shown in this video: link

4) The intersection of the boom/wings and fuselage/wings is not correct. Currently these components are flush. They need to be either fully intersecting, or have a small gap. Either way would work. 

Go ahead and make these changes, and send over the model again. 

- Daniel

[Thomas Dylewski]

Thank you both! I am new to using CFD software such as this and this feedback is very helpful! I will fix the model and try it again.

[Thomas Dylewski]

I have fixed the model and replaced the rounded rectangles I used for the tail with a NACA 0005 airfoil as per the suggestion of my professor and advisor. I was able to import and run the software with the proper reference values and at our calculated trim AoA, I am getting values higher than initially calculated (.17 vs. .107). I believe that initial lift coefficient is derived from calculating the needed lift at SLUF and then using that value to back out a trim AoA using what we learned in our flight mechanics classes. Looking at the distribution of the lift coefficient along the y-axis shows that the fuselage contributes a good bit to the lift with the wing having a more expected lift distribution. 

Attached is the current mesh I am using (I changed the spacing as recommended and have them very slightly not touching). I also made sure to use the best practice values for each piece as discussed in the user's manual. 

Here is the lift coefficient distribution as well along the y-axis. I took care to ensure that each blunt surface (fuselage and booms) had both a base region and a trailing edge.

Attached is the lift distribution of just the wing without any other components for reference!

Thank y'all for all the help! Hopefully I'll be able to show off the final product later on!

 

[Daniel E.]

Great,  happy to help!