Designing a suspension with basically no constraints! 1

[SusTestEng]

Background: A couple of my friends in the automotive industry want to design a car in our free time. The main drive behind this is to learn and get our brains working again. Most of us are in development or evaluation type jobs where the real hardcore engineering is not used on a regular basis. The car we want to design is a small two seater with alot of power(a rich mans go-kart if you will). Not sure if we will ever build this, but at least get it done on paper and model it.

What I was thinking: Basically I think I want to try a push-rod SLA suspension at all 4 corners(but, I can choose anything). I think it would basically be a formula car suspension with some more durability built in to carry 2 passengers. My vision is sort of like the Ariel Atom, if you know what that is.

Problem: In the past I have always worked with some serious contraints that limited the design in some way or another. This time, we have no constraints on anything yet. We can choose any wheel base, any track width, any geometry. The only thing I would call a constraint would be to roughly figuring out where the CG would be.

Question: Where would you start? What would you use as guidelines to help make some initial decisions? What basic softwares would you use(private use)? Are there any "perfect curves" to follow? Are there any "perfect alignment" settings to use?

I've never done this with basically nothing to work with before. I think this would be a good topic to open up peoples mind and get away from dealing with packaging and space constraints we usually deal with. Any thoughts or comments?

 

[Guest0527211403]

I have seen a few Formula SAE teams use SusProg (

http://www.susprog.com)

to kick off their designs. I was interested in using it to design a home-made RC car, but that never really materialized.

Congratulations to the University of Toronto FSAE team for winning the Formula Student Championships in the UK!

 

[GregLocock]

"Where would you start? What would you use as guidelines to help make some initial decisions? What basic softwares would you use(private use)? Are there any "perfect curves" to follow? Are there any "perfect alignment" settings to use?
"

Look in the FAQ for a list of parameters, you'll need to consider almost all of them.

My approximate order of setting things is

roll gain
RCH
camber gain
semitrack gain

Those set the basic front view geometry

Then I set up a Bundorf understeer model, and get some approximate ride rates.

That gives me the desired bump steer gain

Then I do a whole bunch of other things and rinse and repeat.

In the FAQ there is also a list of free, and not so free, software. I use all the free stuff except freecad, and some bits I've written myself.

Basically judicious use of wishbone.bas and carsim and spreadsheets based on milliken would probably get you there.

susprog3d is a fine suggestion if you want to spend a little money.

There are no perfect curves or settings, as so much depends on things like tyres and other details.


Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SusTestEng]

Thanks Greg,

I know you have much more experience in this area than I do. There is alot of work on this sort of project that may bring some more restraints to the table, like width of the powertrain, but I want to start playing and learning something new. I basically do all the development work and very little design work at my current job, so my skills in that area are starting to escape me. We decided to tackle something like this to start learning something again and not allow our brains to flush the things we have learned in the past. And I think it is every automotive engineers dream to build his own car and not have to answer to bean counters and upper management on everything you do.

And if anyone else who reads this has done something similar, it would be great to hear your experiences and troubles with your project.

 

[rpmag]

I am doing something similar in a sense to the OSC project and the plans will be released. However the compromises are always there in my view and I am 100% sure that my knowledge of what is ideal is far more limited than Greg's.

 

[SusTestEng]

"Basically judicious use of wishbone.bas and carsim and spreadsheets based on milliken would probably get you there."

One of the guys that wants to help on the car actually works for Carsim now, so he can use it as a training for himself as well. This should be really fun!

I started some drawings of the basic shape and I think I like the "2 seater(side by side) F1 car shape that I drew.

 

[UnbalancedEngineerin]

Hey guys,

I just stumbled on this great forum. I already reconize a few names from other forums, but this topic caught my interest. I'm starting a similar clean sheet of paper design for a small prototype that I intend to build and compete with in NASA's West coast endurance series.

I too have always modified existing designs so starting from scratch is new for me as well. I've been playing with the demo of Susprog3D and it seems like a valuable tool. Has anyone here used the optimization functionality?

In any case, I'd love to share ideas with others in the same clean sheet of paper quandries.

Jason S.

 

[SusTestEng]

I saw something the other day that got my attention. On a typical new era F1 car, the inboard mounting point of the front lower A-arm touches underneath the front nose. I am not sure if this joint is a common pivot or if it is fixed and uses the flex of the carbon fiber to make up the travel. As I stated earlier, we all are used to working with severe packaging restraints and could NEVER do this on a normal production car. What would be the advantage of this configuration? If I can find a clear photo of what I am talking about I will try to post a link.

 

[SusTestEng]

Here are some good photo's of what I am talking about:

http://randomnothingness.com/albums/userpics/silverstone/Dsc02267.jpg

http://www.photof1.com/s6-toyotaf1-1.jpg

Old (2003) way:

http://photo.gotodon.net/albums/userpics/10003/normal_img_7758_std.jpg

 

[GregLocock]

That's called a single keel. Other teams use a twin keel, where you have the keel running down each side of the tub, rather than in the middle.

There are possible aerodynamic advantages to a twin keel, and it may be easier to integrate the structure into that of the body.

There again it may not be, those who know aren't talking.

The single keel has the advantage that the two lower arms are pulling directly against each other, reducing the load in the body, and I would guess there may be some improvement in aero sensitivity to yaw. I believe most teams are going to single keel.

All the non adjustable body end pivots are carbon fibre leaf springs.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mloew]

Greg,

I was pretty sure that the flexures remained in Ti rather than CFRP. Do you have a citation for the carbon flexures?



Best regards,

Matthew Ian Loew


Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

http://www.gurneyflap.com/Resources/ bar honda 005-347.jpg

but it could be black paint over Ti, certainly the spherical joints terminate in Ti. I had a strong memory that Webber's suspension failure at Melbourne was due to a failure of the carbon fibre leafs. I've certainly seen pictures of metallic ones, some cars seem to use a mixture of black and silvery ones.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sportech]

I have worked on two clean sheet designs and I think that you will find that there are always constraints especially if the vehicle is small and lightweight. The first was a mid-ship Suzuki powered car (1150#) that was completed in 1988 and the second was a Harley Davidson powered front engine car (1120#) completed in 2005. The Suzuki powered car was done with paper and pencil and the Harley in Solidworks. The H/D car was designed by my partner who used 3 separate suspension programs to verify the design. It has oval tube control arms, billet uprights and pushrod/rocker arm shocks. The chassis is 1 3/4" tube bent in large radii, similar to the Atom. The powertrain is a 124 cu. in. V-twin and a rear mounted transaxle with sequential shift. Weight distribution is 48/52 (ft/rr) with driver. The bottom line is that no matter how much design work is done you still have to test and make changes to get the most of your package.

Good luck with your project!

http://www.funcar.com/sportech/index.html

http://www.twintechcars.com/