Designing the suspension for my car

[superdaver]

I'm currently in the design process of building a car, and have been for over a year now. I'm planning on building a one off car in my garage. I am not an engineer, but I have a fairly good understanding of physics and mechanical components.
To give you a better idea of what I'm building you can view some pictures in my new car design blog

http://www.progressiveperformance.blogspot.com

The current renders are to scale, and have been designed to fit an RSX engine and transmission in the rear. Enough Intro tho

I'm having difficulty designing the suspension for my car, as there are functions, and jargon that are a bit above me. I have been reading through many of the posts on this site to familiarize myself with some of the basic concepts of suspension design.

Basically I want to take my car's design, and throw it around a corner at 1G or more. While maintaining some sort of driveability. I decided to go with some bigger wheels, as they ride over bumps a bit more smoothly, thus removing the need for extra travel in my suspension.
I've also been studying suspension geometry of cars that do well in the slalom, but it's very difficult to find information of the very high performance cars, and their suspension setup.
If anyone would like to give some guidance on this topic it would be much appreciated. If you need any Direct measurements from the chassis to do any calculations I would be happy to share them.

 

[GregLocock]

see my response on your blog

Cheers

Greg Locock

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

 

[superdaver]

I decided on 19 or 20 inch wheels
then I'm probably using some Mazda RX8 spindles knuckles
basically all I'm left with is what camber gain do I want, roll center height, and I gotta aim for no bump steer.
I have a height of 950mm
a wheelbase of 2000mm
and a width of 1800mm
most of these can be changed slightly to improve handling characteristics.

 

[hemipanter]

I had a short look at your site, and there is a lot to be discussed to come up with a good answer.
The A-arm on the drawings implies that the steering rod is to be located between them in height. In order to make them to line up with one of the A-arm the front leg of the A-arm should be of less angle. Provided you want the rack to be in front of the A-arm. It is hard to find any spindles or uprights that meet any higher demand of geometry. Sai is usually, 7-10 dgr. Ok, everything may be taken in to consideration, but we end up with a pretty normal suspension system. I like front and rear suspension geometry to be identical.
Ultima may not sell their stuff but you are better off with a design specially made for your car. Avoid using bent tubes in the chassis, also not letting tube attach in the middle of another tube. The only way to tell if the chassis is strong enough is to set it up for twisting, welding in tubes in deflection areas.
I don’t see how the belly pan is to be arranged, or you may not be working with ground force?
Regards
Goran Malmberg

 

[superdaver]

1 So to make for a better steering geometry, I should lower the angle of my front lower control arms? I can see my current setup having a problem with bump steer now that you mention it. I do want the rack in front of the wheels, and there's plenty of room there to put it wherever it needs to be.

I looked into making the front and rear suspension identical, but because my engine is transverse mounted, it makes packaging very hard. There isn't a whole lot of space on either side of the vehicle to fit suspension in the rear. I tried to find parts to fit the engine in longitudinally, but I couldn't find a trans axle short enough to fit in the rear of my car. But if you have enough money there's always a way to make things work. I'm on a bit of a budget, as I'm earning all the money for this project.

The chassis was just drawn in under the skin, and I haven't quite had enough time to figure out where I want to have all the triangulation set up. As for having some tubes in the center of other ones, I accidentally did that while trying to figure out the best place to put extra triangulation. I'll have to replace some of the current chassis renders with newer ones.

I was planning on using a structural tub and bonding it to my chassis using a urethane glue. You can see the tub in the chassis picture section.

2 What are your thoughts on using a structural tub in the passenger compartment?

I was also planning on using the windshield and roll bar from the Lotus Elise. In doing that it will make bonding the windshield to my car easier.

3 What would you suggest doing to remove the need for bent bars?

 

[hemipanter]

1
We must be sure talking about the same angle-modification here. The front leg of the A-arm that has the steering rack in front of it, should be closer to 90 dgr to the centre line of the car. That is so the steering rod and the steering rack should describe a straight line out to the steering arm of the spindle. If the A-arm is like an A, the steering rod must be angle to the rear of the car unless the spindle steering arm is VERY long. The point is to have the steering rod in the same plane as one of the A-arm.

As you see, there is almost impossible to find A-arms and spindles to suit the application of an own designed car, so I suggest designing and fabricating your own stuff, which
is the secret behind not beeing a "citcar" build up.
2
You mean the "chassiswithtub1.jpg"?
Everything that support the chassie is good. It might be as simple as sheet metal floor welded to the frame.
3
Using straight bars and som sort of metal profiles or arms that meet up against the outer body. Of course a bent bar could be "profiled" reinforced to meet the demands.

My experience from twisting chassies is that when we reach some 20000N/dgr, there are NO acceptance for anything but straight tube handling straight forces. Which put high demands for chassie tubing design. There is no easy "tip" here.

Sorry if I am sounding to much "racing oriented" but I am only talking funktion here, and it is up to you to make the street and other compromices.

Kind regards
Goran Malmberg

 

[GregLocock]

Goran, what do you think of the 1g latacc target?

Cheers

Greg Locock

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

 

[hemipanter]

Goran, what do you think of the 1g latacc target?
Cheers
Greg Locock

Well, I have been "skidpadding" a lot, also reading other sports car manufacturers claims about their cars cornering capability, of more than 1g. In one case it turned out not to be skidpad but some average cornering without time, dgr and speed. On a 200 feet circle we will only speed 62km/hr, so goundforce is pretty much eliminated.

Using an accelerometer over a shorter cornering event will
show 1g with a not to bad sportscar using good tires. On the skidpad it will not.

However, the skidpad is a "steady state" situation so the car could be setup for this type of situation, using assymetrical cambers etc. I am more in to using the skidpad for BALANCE reason than cornering capacity numbers.

I better stop now...
Goran

 

[GregLocock]

We were taking some data the other day in an SUV, it reached 0.7 g peak latacc, without doing anything even slightly absurd on public roads.

The same car would struggle to maintain 0.85 on a skidpan.

So, OP, where does this 1 g requirement come from?



Cheers

Greg Locock

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

 

[superdaver]

Goran

1 I understand the angle modification of the A-arm leg now. Thanks for the clarification. It's going to be difficult to use current parts, but my budget doesn't allow for custom fabricated uprights. The first car is going to have to be built on a budget. If it does what it's supposed to and someone wants to buy one after the fact, I'm sure I'll be using more custom fab parts like uprights etc.

2 Yeah I was talking about the chassiswithtub1.jpg You suggested using sheet metal instead?
The reason I had the idea of using a resin and fiber tub, was to help eliminate some sound transfer from engine to chassis to passenger compartment. I was going to use a urethane glue to glue the tub into the passenger compartment of the chassis. My thoughts were that the glue would be slightly flexible giving it some sound resistance.
doing this in turn could help reduce the amount of sound dampening material I need on the floor, as well as giving more form to the cab. It would also reduce the need for a central beam in the console for support, and could actually end up being lighter in the end.

3 As for the bent bars, I probably could use strait ones and weld them together at angles and have metal profiles epoxied onto the outer panels. In building my chassis, I have a feeling that I won't have too much trouble getting it stiff enough. I'm not saying it's going to be easy, but my basic knowledge of structural rigidity should come in handy. I can always add in more triangulation if it's not stiff enough.


On to new stuff.
I recently realized that my car was scaled wrong. Comparing my current dimensions with an elise, and putting them side by side in my modeling program. I decided that the Elise was a very small car already, and to have something smaller wouldn't be a good idea at all.
If you look at my current renders you'll see that I've made some significant modifications to the wheelbase of the car. Extending it 265mm to make it 2265mm in total right now.
I also made some major changes to the roof line.
After finding a top view of an Elise I was able to make the windshield and side windows the correct dimensions.

Let me know what you think

Dave

 

[GregLocock]

It's not the curve of the tube that matters as such, it is the change of direction without triangulation.



Cheers

Greg Locock

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