Help with Solar Electric Car Workflow and Software Selection

[kenryou]

Hi there, I am working on a big project to build a solar electric car but I do not know where to begin with between aerodynamic and chassis design; and which Autodesk software to use.

1. Should I build a monocoque chassis and run through the aerodynamic or should I use Mudbox to sculpt a body and run through the CFD; followed by slicing it into half and build on the internal geometry?

2. I know that I would need to use Autodesk Inventor and Simulation but which discipline for Simulation do I use as there is a mechanical, multiphysic, CFD.

3. Do I need Alias and Mudbox (for modelling aerodynamic if aerodynamic design comes first)

4. How can I model a nomex honeycomb core with carbon fiber/kelvar layers with Inventor? As they do not have the material selection in 2013 for the core.

5. I know that the student license cannot be used on commercial version. But to what extent? As I would like to show the companies of our work to get them sponsor, etc.

Your advice will be appreciated. Thank you.

Note: There are uncertainties of the size of other components such as engine, electronics, etc.

[flynfrog]

Hi there, I am working on a big project to build a solar electric car but I do not know where to begin with between aerodynamic and chassis design; and which Autodesk software to use.

1. Should I build a monocoque chassis and run through the aerodynamic or should I use Mudbox to sculpt a body and run through the CFD; followed by slicing it into half and build on the internal geometry?

2. I know that I would need to use Autodesk Inventor and Simulation but which discipline for Simulation do I use as there is a mechanical, multiphysic, CFD.

3. Do I need Alias and Mudbox (for modelling aerodynamic if aerodynamic design comes first)

4. How can I model a nomex honeycomb core with carbon fiber/kelvar layers with Inventor? As they do not have the material selection in 2013 for the core.

5. I know that the student license cannot be used on commercial version. But to what extent? As I would like to show the companies of our work to get them sponsor, etc.

Your advice will be appreciated. Thank you.

Note: There are uncertainties of the size of other components such as engine, electronics, etc.

what rule book are you working to?


1: you only need to do aero on the outer shell. No idea what mudbox is. We mostly used Solidworks but inventor would work too


2: depends on what you are doing. You don't "need" to do any sim. It sounds like you will be doing well just to get a car that moves under its own power.

3: No

4: Why would you need to model the actual plies? If you are going to try to sim the composite FEA good luck. Its not something for beginners and you probably don't need it.

5: If you are making money off of the software you cant use the EDU licence.

[kenryou]

It is for the World Solar Challenge and our team is the first batch of our school. I know it is going to be tough as we have no experience. But I believe hard work pays the price.

[flynfrog]

It is for the World Solar Challenge and our team is the first batch of our school. I know it is going to be tough as we have no experience. But I believe hard work pays the price.

I would start with the rule book. What is your fundraising plan? These cars are not cheap. I would suggest you have the same number of people dedicated to funding as you do design. What motor are you planning to use? What tires?

Watch your lead times. Batteries and solar cells can have lead times of 6 months or more you don't want to try to build a solar array or a battery pack the week you are packing for the race.

I would shy away from a Composite chassis if this is your first car. You could build a 4130 Space frame much quicker and safer. There might not even be a weight penalty. This also provides you with a drive-able chassis if you are still trying to get your body done.
Testing is worth pounds of weight that could have been saved. Focus on getting the car built rather than saving that last 10%

The most successful car I raced had over the race distance in testing miles before the race started.

Check out
http://www.opencortex.org/

[kenryou]

We are planning to race in 2015 so there is plenty of time if managed wisely.

I have expressed interest in structural design but as you said, such monocoque design is not easy, that is why I would like to know whether to design the shell or the internal member to shape up the shell.

I will look into space frame design too and if support for monocoque is limited, I will replace with it instead.

Thank you for your advice, flyn.

[flynfrog]

read the rule book. You have to have X amount of clearance around your driver. The drivers eyes have to be at a certain point. I believe now they have set seat back angle. You also have to meet some crash structure. Once you punch these in you get your crash structure. You shell then fits over it. You then get to try to balance packaging with aero, with array area. You also factor in array efficiency a curvy shape starts to reduce the efficiency of your array.

In short start with your driver and build the car out from there.


My advice would be to start with the chassis. Get all of the parts and pieces in place. Then add a body around it. Then you can revisit your original chassis and determine if you can make X smaller will this buy me Y in drag reduction or array space.

[flynfrog]

This book is a must have

http://www.amazon.com/Leading-Edge-Engi ... ading+edge

[Greg Locock]

Point 5 is not an issue, this is a good educational use of software.

If you can find Clive Humphris' old book, Speed of Light I think it is called, it takes you almost step by step through how we designed the Aurora 96 car which won in 98 or 99.

You seem very fixated on software, but it is an engineering project not a software project. In the first steps of designing the 93 car we built a plywood buck of the structure and interior (full size) and then experimented with fitting the driver and all the other gubbins in. No software required, took a weekend. Then I did start using a spreadsheet to keep track of CG, as we were very concerned with high speed stability, and needed to keep the 3 tires evenly loaded as they were only BMX tires and the car weighed 240 kg. The aero was calculated by Clive, fine tuned with a 1/5 scale model on the roof of a car, and then checked with wool tufts and experiments to determine the optimum flying height and pitch. We did zero FEA, it was all drawn on mylar or paper, and we came 5th. We did the early testing at the proving ground as we were concerned with the durability of the aluminium frame, then we did some long drives further north where there was enough sun to really sustain high speeds.

For the 96 car I did some FEA on the wheels because we went to a wheelmotor. The aero may have been done in CFD or the wind tunnel. I don't think the structure ever saw FEA.

[flynfrog]

I second Greg. We did a ton of plywood cockpits. Our cars had a unique steering system that took quite a few hours to work out the ergonomics.

[marcush.]

I certainly would start with accumulating all the requirements you have ,starting with the rulebook and your goals .Then decide on the technology to use based on your budget and connections and partnerships
The set goals will lead you to packaging hardpoints volumes and surfaces you have to supply to make it all work .Then you need to verify the faisibility of your concepts before even thinking of actually how you will tackle things.

so ...based on your concept ,what is the weight of the components of your vehicle? and how does this translate to your needs in terms of structure ,wheels suspension,how much do you need in terms of cornering (tracks ,speed!) or being able to soak up bumps and anad and will decide where to place emphasis and what to keep as simple as possible..

Your complete set of requirements will leed you towards the product ....Inventor or whatever is the hammer or knife to do the final work of shaping ..
I would make sure to have a realistic human body model incorporated in your CAD tool as this will be a great help to avoid major mistakes but still i would build a seating buck to verify it will be ok