Page 1 of 1

Possible future entrant.

PostPosted: Tue May 07, 2019 8:07 pm
by geofff
Hi Everyone,

I'm thinking of designing a car for the formula and wondered if anyone could give me some pointers regarding CAD software? I've looked at the AutoCAD stuff and Fusion 360 can be had on a hobbyist / inventor licence for free. However I'm told it lacks decent FEA & CFD modules? Does anyone know better?

The other option is Inventor from Autodesk, but I'm not sure on the 'student' licence length if I sourced some online training.

Or are there other options? All advice gratefully received. Don't get me wrong, all of these CAD type programs will be new to me so I expect to invest time in training of some sort.



Re: Possible future entrant.

PostPosted: Wed May 08, 2019 8:07 am
by geofff
Also I'm based in Swindon, so if there are any drivers / constructors in the area I'm more than willing to lend a hand even if it's just passing the tea or tinnies.

I've built three Westfield's and even raced an Alfa 33 with the 750MC in Hot Hatch over twenty years ago, so I have provenance!



Re: Possible future entrant.

PostPosted: Wed May 08, 2019 9:20 am
by F1300Tony
Welcome, Geoff. ... raftsight/
I used draftsite which was totally free up to a couple of years ago. It was very good and seemed to be fully compatible with AutoCad, drawings going to and from each software with no errors. I never used it but it went 3D with rotation. I think you can now get it free for 30 days. Probably illegal but with a second e-mail address this might be extended. I also used pencil and paper!
As I always say, my strong recommendation is that you race a tried and tested 750F car before designing your own. Otherwise you will not know what a good car should feel like and you will not know why it isn't quick. Very few new cars work straight out of the box, most take a lot of sorting in terms of spring rates, some take major mods to make them work and some are never quick at all. Of the drivers I have worked with, only one in three could tell me what the car is doing, which suggests only one in three people could successfully sort out balance and shocks. You will design a better car when you know 750F from the inside.
The other thing I suggest is that you look at every car and talk to 750F people at a circuit. They will all help, unless they are buried in car bits or it's 20 minutes before a race. And read the regulations several times!
Best of luck.

Re: Possible future entrant.

PostPosted: Wed May 08, 2019 9:21 am
by BenandOlly
Hi Geoff

There are pros and cons to all the CAD software, and budget is a big consideration. I use Solidworks but the program cost around £5K. If you are a student you can get a copy for about £50 but Solidworks do want to see proof that you are in full time education.

We also use Fusion 360, mainly on the CAM side. It is really good value, and as you say, you can use it free as a hobbyist or very small business. There are a lot of online tutorials for Fusion 360 which help to make it accessible, but companies like Solidworks want to sell training course so they limit what they put out for free. You shoudl also look at Draftsight which has been free up to this year but is now on an annual license. They have a 3D package which I have not yet looked at. Draftsight is made by Dassault who also have Solidworks and Catia, so they should know what they are doing. Of course they want people to progress up their range to the more expensive packages, so they limit the functionality of the entry programs.

It is hard to advise you on FEA and CFD without knowing your background. Solidworks has FEA, as will all the high end packages, of course you need to know what loads you are designing to in order to get meaningful results. Good CFD is less common and from what I understand needs a great deal of understanding and skill in order to get worthwhile information out.

Rhino is good, especially for surfacing work, but I have lost tack of the cost of it now. My sons and I modelled our Racekits chassis using Draftsight for 2D, made a 3D assembly in Solidworks and modelled the bodywork largely in Solidworks, but also some parts in Rhino as it was easier for us to control the surfaces. Olly used Solidworks to run FEA on some new rockers that we made and we also ran FEA on the old rockers so we could see that we had a stronger part than we had before.

If I have not confused you enough, it may also be worth looking at Onshape. It is a cloud based package that has been coming on leaps and bounds in recent years and my other son George is an Onshape fan.

Good luck with whatever route you take. 3D modelling is great but it can be very time consuming and unless everything is modelled it can lead you up the garden path.

All the best


Re: Possible future entrant.

PostPosted: Wed May 08, 2019 9:40 am
by Martin Kemp
As a design engineer, I have used CAD and FEA analysis on and off for many years on a variety of engineering projects, however when it came to designing the Falcon and Merlin race cars I went back to the drawing board and calculator. The reasons were time, effort and cost. I did not have access to a work CAD system so I would have had to get a CAD system I could afford and then learn to use it. It was quicker to draw the design on paper and much quicker to modify it with a rubber and pencil. Where a simple CAD system would have been useful, especially as we have now built several cars, is in providing DXF files to get simple parts LASER cut.
If you want to use CAD , then to get the best from it, look for a system that will allow you to create and edit tubular structures easily as this can be much more time consuming than creating solid components such as suspension uprights. You may also want to create component assemblies so you can see the effects of positioning the engine, driver, axle etc. on mass properties. This is usually a separate module. You can do it treating the whole car as a single component but it is very clumsy.
To use FEA you will need a representative model of each component to be analysed. To analyse a chassis you would need a beam element module that enables you to create a stick model of the chassis and attach the section properties to each bit being analysed. Of course you will also need to know what loads are going to be applied to components when being cornered, kerbed, or crashed. More technically, it is vitally important for the accuracy of the results to specify the correct constraints for the model. Finite element analysis is highly specialised engineering and takes a lot of training and engineering knowledge to get correct results.
I think this makes FEA look easy and racecar aerodynamics are so complex that an amateur would be hard pressed to improve on what they could achieve by reading and understanding books on the subject.
However if you are training to become an engineer or just want to try out the techniques then you can have an interesting time but it is not the fastest or most efficient way to get out on track.

There is other software that is very useful for designing a 750 car:
Suspension calculators will give the roll centre position and stability during braking and cornering. There are many of these including our own 750 Formula sponsor, Racing Aspirations. I would not go back to old fashioned methods for these calculations because they are very time consuming.
Centre of gravity calculators. These are useful if you do not have an assembly module with your CAD system and Racing Aspirations do one of these as well.

The key elements in a 750 formula race car design are, in no particular order; mass, centre of gravity, suspension geometry, chassis stiffness, drag, downforce, and crash worthiness.

Mass/stiffness/crash worthiness/Centre of gravity
For a front-engined car it is pretty easy to produce a car well under the present weight limit, so you can go to town on triangulation and protecting the driver without a need to optimise the mass. The good old fashioned balsa wood chassis model will show up where there is most flexibility in the chassis.
For a transverse car then it is quite difficult to get down to the weight limit and CAD and FEA can be more useful to optimise the design.
For either type it is not hard to work out the C of G on a calculator if you know what the main components weigh but software will make iterations easier and it is more critical for the transverse design as it is certain to have a rearward bias.
For crash worthiness, beef up the forward roll hoop supports and protect the drivers feet from the front wheel and wishbones and make sure the engine/gearbox won't try to join the driver.

Of course I realise that I haven't actually answered your question but I am retired now so whilst my knowledge of engineering won't go out of date, software systems and prices are changing by the hour.


Re: Possible future entrant.

PostPosted: Wed May 08, 2019 2:43 pm
by geofff

Thanks for all the replies and yes it has given me something to think about.

Martin: I've read Staniforth, McBeath and even tried to get my head around some of Milliken's books. The pictures are good for inspiration even if only 5% of the maths is going in. I will take your advice and won't ignore the basics and look at some of the Racing Aspirations calculators.

Tony: I think my problem will be overthinking an issue which then leads to indecision. We'll leave my driving out of it for the moment :lol:

I've decided to give Onshape a shot, I had noted it before Ben's comments but had discounted it due to it's open source nature.

I'm thinking better to jump in with something at zero cost and get a bit of general CAD knowledge before proceeding further.

Hopefully will have a nose around at Cadwell in July, Croft is a bit of a trek.



Re: Possible future entrant.

PostPosted: Thu May 09, 2019 5:45 pm
by GregSwan
Hello Geoff,

You're most welcome to pop up to V Factory in Bicester. An extra pair of hands is always useful! There are many in the 750F worth listening to with far more experience and knowledge than me, but I do have a library you can browse/borrow. For the basics, I always find myself returning to Carroll Smith to clear my head whether designing, fabricating or bolting something together. Huw's Racing Aspirations software is great and will save you some cardboard and string. As ever, the big question is what you're aiming for, rather than the tools you employ. I shamelessly measure up and reverse engineer in my head every car that passes through... especially the ones I know that work. Hopefully it'll save me wasting my time building a clunker if or when when I finally get round to it. We're new to the 750F and run the Ingham as it's simple, it works... and until we find it's limits I wouldn't dream of messing with it. No surprise then that I agree with Tony's "strong recommendation".

Best regards,

Re: Possible future entrant.

PostPosted: Fri May 10, 2019 7:43 pm
by Baggers
+1 for draft sight
I use librecad.
I've used Huws software to help with the geometry.
For CFD you can use open foam. It's packaged with CAE linux, but the learning curve is intense to say the least. I'm going to probably have to learn blender to create the 3D models I need for.

Re: Possible future entrant.

PostPosted: Mon May 13, 2019 7:53 am
by geofff

Greg thanks for the invite, next time I get a free day on the weekend I'll ride over to the V Factory for a look around.

Baggers thanks for the tip, will have a look as soon as I've finished with the Onshape on line training.