SimScale CAE Forum

Sheet Bodies Detected - why?


#1

Good evening.

This might be a simple question, but I can’t seem to find an answer on the internet that really explains the mistake in my project. I keep getting the error

"Sheet bodies detected.
The model contains sheet bodies (surface data). It is recommended to solidify all parts of the model as otherwise unexpected meshing behavior can appear.
"

What am I doing wrong (or not doing) that causes this issue? I created the CAD project in Onshape and would like to do some basic CFD on it (I have no experience in engineering, this is all just for fun).

Link to the Simmscale project here
https://www.simscale.com/projects/rynop_email/soapbox_v1/


#2

I see in the part list of your geometry that you have some surfaces. CFD simulations calculate fluid parameters for elements in a 3D mesh and so all your parts must have thicknesses. Can you explain what the aim of your simulation is? You have chosen the incompressible solver so it looks to me like you want to simulate the aerodynamics around the car but if this is the case, you need to mesh the air around the car (geometries --> add geometry operation --> enclosure), not the car itself. Furthermore, your CAD model has interference (where two parts intersect) which is an absolute no for CFD and if this is an aerodynamics simulation there is no need to model the internal structure of the car, you would be better off importing the car as a completely solid body.

Edit: I see you say this is just for fun and you have no engineering experience, well start on a simpler geometry, in fact start by going through the tutorials on the website. Although simscale is easy to use as far as CFD packages go, it requires knowledge of proper CAD preparation and the fundamentals of fluid dynamics. I think it would be a lot of work for a ‘lay person’ to become proficient because of the background knowledge required.


#3

Hi @roy_g.

Thanks for the feedback.

This project is just for fun. We’re entering a soapbox derby race at the office, and me being a car nut, decided it was time I learn more about these things. Basically I wanted to design the chassis and body in CAD and then run a couple of CFD simulations just to see if there are any obvious areas where we can cut down on drag on the body (I’m expecting to see a large portion directly behind the car and from the rear floor upward as we’ll have separation there. I was thinking about adding a vent/scoop from the floor of the vehicle to divert air to the rear of the car to clear out some of the turbulance at the back, almost like a make-shift diffuser (I know it won’t have a venturi effect without correct shaping). I have a mathematical background so many of the concepts are familiar, I just haven’t worked with any CAD or CFD programs before.

Thanks for the feedback about meshing the air around the car. I thought I had done this by creating a cartesian box around the vehicle, specifically one longer to the tail end than front? Maybe I missed something there. I have made changes and removed the inner structure of the vehicle (I had assumed Simscale would disregard these as the outer shell was encompassing all of it).

Let me try again. I’ll let you know how I get on.


#4

I just saw your 2nd case and yes you seem to have the correct mesh. I will try and give you a more detailed reply latter in the day. Note that aerodynamics is not my area of expertise so although I will be able to help you setup your simulation I probably wont be able to help you optimize your design. Oh and soap box racing does sound very fun. Is it red bull soap box?


#5

Thanks Roy, I really appreciate the feedback.

It’s for the local soapbox race, but if we do well enough there’s no reason why we can’t enter for the Red Bull one next year.

As you can tell, we’re taking it a little bit more seriously than other people probably would, but second place is still the first loser (and I’m learning a lot :slight_smile: ).


#6

So I think the simulation is set up well. The boundary conditions seem correct including choice of slip for the air volume and non slip for the car. I think the design could be improved by having a more gradual slope on the rear end as such a sharp gradient will cause the flow to separate resulting in a low pressure region behind the car which causes resistance. I think that for a hobby project this diffuser idea is misguided and the time will be better spent improving the steering and weight distribution. As I said though, aerodynamics is not really my area of expertise otherwise I would try to be more helpful.


#7

Thanks a million!

Yeah I was thinking of dropping the rear a bit. Just need to get a first draft out so I can check that everything is working as expected. Thereafter the changes should be easy to integrate as the process will be the same.

Thanks for the feedback on the steering and weight distribution. For that I think we’re good. We’ve got a slight tail heavy car, but the weight distribution is almost dead center. The pivot point for the car sits just about where the driver’s hips go, so it should go well, and we have some ballast weight we can use to balance it properly.

For the suspension we’re doing a pushrod suspension that we’ve already designed (I have a Baja racing friend in Canada who’s helped with that).

If it all goes well this thing will fly (in the good sense) . Once the aero results are back I can start trimming the design to cut down on drag