Findlay234
|
| posted on 26/10/09 at 09:31 AM |
|
|
3D computer model of a locost for CFD
Hi there, does anyone have a 3D model of a locost in Solid Edge or Parasolid format? If not does your software allow exporting in Parasolid? Or does
anyone know where this can be downloaded. I want to put together a CFD model of a locost but dont really have the time to build the initial 3D model
from scratch....
|
|
|
|
|
balidey
|
| posted on 26/10/09 at 09:53 AM |
|
|
Gulp, thats one hell of a project.
Company I work for got a commission done for CFD of one of our vehicles. it took us about a month to get the model in a state where they could use it,
then they did the tests, again taking about a month. It cost us THOUSANDS and then we backed up the theoretical results with real-world trials using a
prototype vehicle.
If you want to do it but don't have time to do the initial model then I think you may be missing the point.
This is a BIG task and just like FEA, unless you are already an expert in the field then don't trust your own results.
But after saying that.... good luck. If you manage to get worthwhile results then you have a job for life
|
|
|
Findlay234
|
| posted on 26/10/09 at 10:15 AM |
|
|
Well this isnt going to be an entirely scientific experiment. Im not going to try and find what the exact values are for any part of the car but just
to get a better understanding of the airflow around it (it might also create some pretty pictures).
I wont have any way of validating the model with real data and by the fact that the model will take a long time to run I wont be able to conduct
proper sensitivity and convergence analysis. Even with a 3D model already drawn it will require a lot of time fettling the model to mesh well, im just
trying to reduce the overall time as much as possble. If i can get hold of a 3D model already made I should be able to get a 'first pass'
set of results within a week or so which should present a starting point if more detailed analysis is desired.
I currently work as a design engineer for QinetiQ on very high altitude UAV's and use CFD almost every day for thermal analysis of the aircraft
components. I wont call myself an expert though, my current models arent working as Id hoped and im having to go back and change a few of the models
to mesh better along with conducting further sensitivity analysis to work out where the possible errors are occuring. Oh and one of my assumptions
might be wrong, but hey ho!
cheers
|
|
|
balidey
|
| posted on 26/10/09 at 10:26 AM |
|
|
quote:
Originally posted by Findlay234
I currently work as a design engineer for QinetiQ on very high altitude UAV's and use CFD almost every day for thermal analysis of the aircraft
components.
Well in that case I guess I was teaching you how to suck eggs 
Just wanted to let you know how involved this project woulc / could become. But you seem to have a big head start
|
|
|
balidey
|
| posted on 26/10/09 at 11:51 AM |
|
|
Flak Monkey has this on offer...
http://gtslocost.locostsites.co.uk/cad.htm
|
|
|
Rob Palin
|
| posted on 26/10/09 at 02:26 PM |
|
|
I've got stuff from Turbosquid before for use in CFD:
http://www.turbosquid.com/Search/Index.cfm?keyword=caterham&media_typeid=2&categoryid=20031
There are loads of great 3D models available online, created by people preparing them for video games. Some are a bit ropey, some are based on
die-cast models so have limited floor information, but some are done by people who really should get out more (like a Jag XJ200 which even had ruffled
leather seats!).
A (limited) freebie download of something like Rhino is useful for switching formats of these files to STL or something else more usable.
A warning though is that the models are often created to *look* right when rendered, so can be constructed out of lots of isolated sheets of mesh, not
integrated water-tight volumes. This can mean a fair bit of stitching and cleaning before you can use it.
|
|
|
