priority for deciding suspension geo

[vitesse]

i m working on wingeo for designing suspension
geometry.
Can anyone help me out in deciding the priorties of various parameters like
1.Camber change.
2.roll center migration.
3.ride height
4.swing arm lenght
in arriving to final geometry
and should we keep the IC above the ground ?

[vitesse]

i m working on wingeo for designing suspension
geometry.
Can anyone help me out in deciding the priorties of various parameters like
1.Camber change.
2.roll center migration.
3.ride height
4.swing arm lenght
in arriving to final geometry
and should we keep the IC above the ground ?

[C.Zinke]

This mainly depends on your tires. For example: If you have tires with a lateral stiff sidewall, then you will have a lower priority on reducing positive camber in roll.

My suggestion is, that you should develop a overall design concept. This includes a overview how you like to manufactrure you suspension parts (If you have casted uprights, then you will have better chances to realise aggressive kinematics). Then you chose your spring and damper rates, develop the kinematics and do iterations to optimise your design.

[civicsit]

Read Race Car Vehicle Dynamics.

[C.Zinke]

But be carefull with this book. There are alot of suggestions that are not very scientific and therefor not applicable for FSAE.

[J. Vinella]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by C.Zinke:
...develop the kinematics and do iterations to optimise your design.


But be carefull with this book. There are alot of suggestions that are not very scientific and therefor not applicable for FSAE.
</div></BLOCKQUOTE>

Who ever said the suspension system was scientific. Do you know everything about your tires? How can you "optimise" to a system with open variables?

vitesse, RCVD is a great book. It will not answer every question you have, but it is a good guide to lead you to your own design. The IC question you had is a good one...your RC and IC are connected. You will have to decide which one you want to focus on.

Good luck,

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by C.Zinke:
But be carefull with this book. There are alot of suggestions that are not very scientific and therefor not applicable for FSAE. </div></BLOCKQUOTE>

It might not seem scientific, but in the absence of tyre data including the effect of track surface, there's a big limit to how scientific you can actually be. John's right in that regard.

It's very easy to believe a simulation, but unless you've validated the tyre model used in a track test it's impossible to say that the simulation reflects reality:

http://www.mscsoftware.com/alpha/view_article.cfm?volume=9&articleId=76

RCVD is an excellent book. Although on the recommendation of others on a recent thread I'm currently waiting on a copy of the Rowley Engineering book.

Ben

[exFSAE]

#1 - More grip.

[C.Zinke]

Okay, i knew that a comment like mine would raise a discussion with an difficult standpoint for me ;o). Unfortunately i have to go to the university library to take some text passages, that are "not scientific" in my opinion.

I agree with you that you have to make assuptions during suspension design, so that it only can be scientific to a certain level.

But from what i remember RCVD becomes unscientific on a level where it is not necessary. This was in text passages where the author deals with anti-X and spring calculation.

So i will have a look again into this book. And maybe i was wrong. Otherwise i will bring up a new thread with those "unscientific" text passages, ok? BTW, of course its a great book

[Christopher Catto]

well, if you really want a scientific book, then buy

Tyre and vehicle dynamics, Pacejka, Butterworth Heinemann
The multibody systems approach to vehicle dynamics, Blundell Harty, Butterworht Heinemann

First one is pretty tough to read unless you are a research student or have enough time to dedicate to simulation.

Second has more useable knowledge and not too hard to read apart from the chapters on multibody mathematics which are pretty dull.

Stick to RCVD or Olley. At least if you do something wrong its cos you have absurdly high kpa, rather than because your locus plot becomes unstable when l/R becomes higher than 0,05. For gods sake, just enjoy the racing!

I bet Caroll could make a car go faster in a week than the whole of TNO working together. If however you have the resources and brain to tackle more complex stuff, thats the way to go. Thats how the Ferraris and Renaults win at weekends. Constant iteration.

[Mark TMV]

Minimize roll centre migration and optimize the camber change to suit your tires.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mark TMV:
Minimize roll centre migration and optimize the camber change to suit your tires. </div></BLOCKQUOTE>
If we're going to introduce 'roll centre' into the discussion, you need to be looking here:

http://fsae.com/eve/forums/a/tpc/f/125607348/m/47110403711

Regards, Ian

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Christopher Catto:
well, if you really want a scientific book, then buy

Tyre and vehicle dynamics, Pacejka, Butterworth Heinemann
The multibody systems approach to vehicle dynamics, Blundell Harty, Butterworht Heinemann
</div></BLOCKQUOTE>

First off I agree. Both books are great. Here comes the But...

At no stage does any of Pacejka's work cover the interaction of rubber compound and track surface. In that sense most of the complexity in his models is arbitrary without a decent model or track test value of what your friction coefficient actually is.

All nice neat Pacejka curves should be overlaid over the massive hysteresis loop that the fit sails through the middle of. If your tyre model isn't temperature dependent you can't optimise the suspension through all phases of the corner, because the compound changes as the energy input increases. Looking at things in this way (as Chuck Hallum did in his papers) explains a lot of those "unscientific" race engineer rules of thumb you find in Smith and RCVD.

The problem is, if you start learning tyres by looking at a Pacejka model you miss a lot of the crucial realities that were too complex to fit in the model.

Milliken might be "unscientific", but the thrust of their work is always about practical methods that are actually useful. Indeed Damian Harty has a paper called "the myth of accuracy" and also discusses the difference between practical vehicle dynamics and theoretical vehicle dynamics in his book.

Finally a quote from Peter Wight's book Formula 1 Technology:

"History shows that engineering (using an idea without necessarily understanding it) usually gets the drop on science (understanding an idea without necessarily having an idea what to do with it)"

Ben

[Christopher Catto]

The man is spot on!

you can guess which book i never open! Best thing to understand tyres would be to get some BASIC data. Not overly complex stuff, and then have a look at your ackermann and your castor and your camber gain. It takes prolly 2-3 years to get some good grips with a generation of FSAE cars. No good changing from 10inch wheels to 13 and back every year.

[J. Vinella]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Christopher Catto:
...look at your ackermann ... </div></BLOCKQUOTE>

Finally somebody mentioned anything relating to steering geometry(edit: minus all that castor/caster malarky). I lump this into suspension geometry and it is high on my list of importance.

Remember this isn't a dragster, there are a few corners on the enduro course. http://fsae.com/groupee_common/emoticons/icon_wink.gif

[vivy]

just in addtion to above discussion ,, i have been to student 2007 comp. and found that in some cars had really wide and really long wishbones at the rear ,, is it aiming for attaining somethin in the design . y so much wide rear wishbones... ??

[J. Vinella]

When you say wide do you mean wide track, the distance between centerlines of the wheels, or from a front view of the car long wish bones in relation to the fronts?

Or do you mean wide as in longitudinally, as in looking at the top of the car with wish bones of very large angles. If this is what you mean, I bet the cars you are referring to have no rear box. No rear structure behind the engine supporting the suspension points. The goal in this design is mostly to cut the weight of the car by losing a rear structure and pushing the loading points of the rear suspension forward into the engine support part of the chassis.

[vivy]

yeah u r right that cars dont have rear box... ,, actually john is there any consideration of how much wide ur rear wishbones should ar on what factors thus this width depend upon... that is in top view ,,

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by J. Vinella:
The goal in this design is mostly to cut the weight of the car by losing a rear structure and pushing the loading points of the rear suspension forward into the engine support part of the chassis. </div></BLOCKQUOTE>

John, my first suspension design eliminated our rear box, but it was not to lose weight (in fact we went from 426 to above 450) . It had much more to do with inertia and weight distribution.

[J. Vinella]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by flavorPacket:

...(in fact we went from 426 to above 450) . It had much more to do with inertia and weight distribution. </div></BLOCKQUOTE>

I can only make these assumptions, you pulled weight from the rear, the weight of the car went up, all for some inertia gains. That doesn't sound like a good design decision to me. My point to vivy is that is one major reason to cut the rear box is to lose weight. The weight you add with the suspension tubes being longer is far less then the structure you lose in the rear.

To answer your question vivy, if the two upper points are ahead of the rear axle along with the lowers then the member see some "interesting"� loading. Nothing you cannot design around but there none the less. We run an F4i and the perfect top forward point is a nice spot on the engine, however it was further than I would like from the center line of the vehicle.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by flavorPacket:

my first suspension design eliminated our rear box </div></BLOCKQUOTE>

My first suspension design eliminated our rear box too, but for us the negatives out weighted the positives.

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by J. Vinella:
That doesn't sound like a good design decision to me. </div></BLOCKQUOTE>

The suspension leader can only control the weight of the suspension. If somebody wants to add a turbo, there's not much I can do if it's well-justified. At least the turbo package lowered the cg...

And I think it'd behoove you not to call other people's decisions, which you know nothing about, bad or good. FYI, I added between 6-8% to the rear, depending on driver.

[J. Vinella]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by flavorPacket:

And I think it'd behoove you not to call other people's decisions, which you know nothing about, bad or good. FYI, I added between 6-8% to the rear, depending on driver. </div></BLOCKQUOTE>

I did say I only made these assumptions. It was bad form on my part to assume and pass judgment on that assumption. I apologize.

I however disagree that the suspension leader can only control the weight of the suspension. Harking back to the topic of this discussion "priority for deciding suspension geo" The weight of the car plays a major factor in any suspension design and thus the overall geometry. If the turbo was well-justified and it overcame your argument, then so be it. But if you convinced the team from an overall performance point of view that a turbo would hurt suspension's performance and in suit raise lap times. Then you have effectively controlled the weight of not only the suspension. You could change this example to anything and I feel my point still applies.

[flavorPacket]

The suspension can certainly influence the design of the car, but it's up to the other systems' designers to make them light. I agree that certain things can have a direct effect on weight, such as hardpoint location and certainly wheel/tire choice. BUT, most of the other systems on the car do not really concern themselves with susepnsion geometry. For example, the fuel tank designer doesn't really worry too much about where the control arms go, as long as they're not in the way.

Now, if you're saying suspension design can control weight because the suspension designer is able to convince other people to change the configuration of the car, I agree. But that's the person doing the work, not the design. And in my case, our simulations and tests proved over and over that a turbo was the way to go.