Chassis Question (first year team)

[Kenneth]

Hey guys,

First post here, but I've been perusing for the past month or so. I know how important it is to search the forums before starting a new thread, and I think I've done just about that (okay, I lied, I haven't gone through 76 pages of threads!) but wanted to ask a few questions regarding the beginning of Chassis design. Just a bit of an introduction: I'm from a Canadian College, and because we haven't been officially okay'ed by our Dept. of Eng. I won't state which college I'm from, but I assure you, I've got a good group of guys here that want this, and are willing to putting in the effort to make this project happen. I've been appointed project lead and plan on having a competitive car (oh the confidence!) ready for 2012. I have decided to take up the task of designing the chassis among other things.

So here goes...

From reading a couple books, and from reading numerous posts, it seems like the general consensus is to start with the basic parameters of the car, wheelbase, track length etc., then moving onto suspension/points before even touching the chassis.

What I wanted to know is, How do you guys come up with that initial frame design? I've been thinking about this for the past few days and to get that initial design down, I would want to sketch it all out. Get approximate dimensions. Input into CAD (SolidWorks), test on CAD, make adjustments and make sure it meets the requirements set by the SAE.

Are those initial dimensions for the most part guesses based on the drivers / research on the optimal setup? I'm thinking, please tell me if I've got this all wrong, yep. Once it's in Solidworks, given the type of tube we'll be using, (4031 for the main structure/roll hoops, 1020 for bracing/triangulation...this might change as we go along) we'll test for torsional stiffness and make the appropriate changes from there on.

Any help I get would be a great help to the team and I would greatly appreciate it.

Thank you!

[Kenneth]

Hey guys,

First post here, but I've been perusing for the past month or so. I know how important it is to search the forums before starting a new thread, and I think I've done just about that (okay, I lied, I haven't gone through 76 pages of threads!) but wanted to ask a few questions regarding the beginning of Chassis design. Just a bit of an introduction: I'm from a Canadian College, and because we haven't been officially okay'ed by our Dept. of Eng. I won't state which college I'm from, but I assure you, I've got a good group of guys here that want this, and are willing to putting in the effort to make this project happen. I've been appointed project lead and plan on having a competitive car (oh the confidence!) ready for 2012. I have decided to take up the task of designing the chassis among other things.

So here goes...

From reading a couple books, and from reading numerous posts, it seems like the general consensus is to start with the basic parameters of the car, wheelbase, track length etc., then moving onto suspension/points before even touching the chassis.

What I wanted to know is, How do you guys come up with that initial frame design? I've been thinking about this for the past few days and to get that initial design down, I would want to sketch it all out. Get approximate dimensions. Input into CAD (SolidWorks), test on CAD, make adjustments and make sure it meets the requirements set by the SAE.

Are those initial dimensions for the most part guesses based on the drivers / research on the optimal setup? I'm thinking, please tell me if I've got this all wrong, yep. Once it's in Solidworks, given the type of tube we'll be using, (4031 for the main structure/roll hoops, 1020 for bracing/triangulation...this might change as we go along) we'll test for torsional stiffness and make the appropriate changes from there on.

Any help I get would be a great help to the team and I would greatly appreciate it.

Thank you!

[alumasteel]

It sounds like you've got the general process correct in your head. If you are proficient in designing CAD models that can be updated easily (SolidWorks does make the tubing easy with the "Structural Members" command in the weldments section), then you should be able to evaluate fairly quickly where things can go. Figuring out how to layout a basic suspension will take some time for your suspension team anyway. I would recommend as a first year team you don't get too hung up on tire data. At this point, you won't have the foundation to really use it appropriately anyway. You'll just want to figure out if you have a turn of radius X and the car is traveling at velocity Y, how much load transfer, roll, vertical hub deflection, and steering are required to make that happen. You can then design the kinematics around that while knowing where the suspension pickup points could hypothetically exist in 3D space.

The most important thing to remember is that you are not going to optimize to a perfect design, but rather you are looking for the best compromise. Once you have a fairly decent idea for how the suspension will behave for a variety of cornering scenarios, and you are content that it will be agile enough to handle the range you'll see in an FSAE autocross, then freeze the suspension design. Once you have that in place, you'll know where certain chassis nodes will need to be placed, along with your engine mounts. Then I would thoroughly read the rules and make note of every rule regarding required tubing. It's been a few years since I've read a rulebook, so I don't know if they've added anything, but the biggest items are the roll hoop supports and the front chassis triangulation, IIRC.

From there, it is just connecting the dots. Just do your best to connect them in a way that adds triangulation and stiffness. If a tube doesn't physically hold something up or add to the stiffness, it's dead weight. You can determine this through FEA analysis. As a first year team, I would get the baseline chassis, do an FEA run and analyze that run and make changes accordingly. You may even want to get a willing professor to help with analysis part. But once you've done this, freeze the chassis design as well. Then, start building the car. Remember the rule of pi: if you've estimated something will take you 1 hour to complete, in reality it will take approximately 3.14159265 hours.

[Dash]

Once you pick the wheelbase and track, you have to start thinking of a few other things.

1) Templates - These are going to dictate how small you can go with the opening of the chassis and footwell. Extruding the templates in SolidWorks will probably give you a rough idea of what space you can use for tubing.

2) Packaging - You need to make the chassis big enough to house a lot of things. The biggest things being the driver, the motor, and the gas tank ( most likely ). You need to keep driver comfort in mind here. If the driver can't move his arms enough to turn the steering wheel to lock, then its probably too small.

3) Rules - There are several rules which dictate basic dimensions of other tubes like the main roll hoop, side impact structure, etc. These will also help you shape the car.



Basically, our chassis guy plays connect the dots with the suspension mounting points, main and front roll hoops which are based the the 95th percentile male wearing a racing helmet, and several other critical mounting locations. All the while making sure templates and other rules will be passed. It can be a tough job, especially for a newbie. Good luck!

[Francis Gagné]

One with to remember is the functions of the chassis;

It is basically a huge bracket which enclose a living person. So confort and safety are to look at closely. Tube takes places, don't forget about part access, you'll save precious time when in the shop.

Think about load path, where the suspension loads go, try to make the load go effiently into the frame. For the technique you seem to have it, put the thing that does not move a lot (pilot, templates, bulkhead, suspension point, etc). To put things in, keep it simple and stay aware of interferences and engine removal!

If you want more insight read the Pat Clarke's articles on formula student germany website (read all of them in fact!).

[RobbyObby]

As long as you design around the template rules, your drivers should have no problem fitting in the car. No harm in throwing Todd in to double check though! http://fsae.com/groupee_common/emoticons/icon_wink.gif

Also, while it is true that ideally you begin with suspension hard points and build your chassis to them, there will be compromises. When designing your suspension, be mindful of the chassis constraints or you could get in trouble quickly.

Id also love to know your reasoning behind using both 4130 and mild. If you have access to 4130, why not use it throughout? And if cost is an issue, why not make the whole thing out of mild? Many teams can and do make very solid frames out of mild. I dont know how much you'll really gain by using chromo just for the roll hoops, etc. That being said, it doesn't mean its bad, each team has its reasons, just curious to know yours.

[Kenneth]

Thanks for the responses!


RobbyObby, I'll definitely look into that. I've just been looking/reading around and I've seen a couple teams that have used that combination at some point. Admittedly, I don't readily know the exact properties between the two. Generally however, I know chromoly's used in roll cage manufacturing and is easy to weld (says my welder). A quick Wiki search tells me it's stronger and harder than 1020. But you can bet your current chassis that I will do more research into tubing after I get that suspension and initial chassis design down.

One thing I haven't read before is in Dash's post, Templates. Are you saying, make the front bulk head for example, extrude longitudinally, and that would be the minimum I could build around? That kind of idea?

[Luniz]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">...Templates. Are you saying, make the front bulk head for example, extrude longitudinally, and that would be the minimum I could build around? That kind of idea? </div></BLOCKQUOTE>

Reading the rules sometimes is a good idea! The dreadful Templates dictate how big (read: enormous) the cockpit entrance and the driver's legroom have to be and generally make FASE cars look like whales since 2009.

[125zr]

How to calculate static force attach on the chassis?..How to start?..I am confused because at the certain place of node there is no force.

[Big Mo]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Kenneth:
I'll definitely look into that. I've just been looking/reading around and I've seen a couple teams that have used that combination at some point. Admittedly, I don't readily know the exact properties between the two. Generally however, I know chromoly's used in roll cage manufacturing and is easy to weld (says my welder). A quick Wiki search tells me it's stronger and harder than 1020. But you can bet your current chassis that I will do more research into tubing after I get that suspension and initial chassis design down.
</div></BLOCKQUOTE>


They are the same stiffness, so IMHO the only reason to build some structure out of 1020 is to save on the cost of 4130 tubing. 4130 is stronger, harder, and looks to be tougher due to more elongation at failure.

Another reason for 1020 is to build a weak spot into the chassis on purpose, I think Pat Clarke gives the example of a-arms being made of 1020 so that in a crash they fail before they warp the chassis.

[Kenneth]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Re </div></BLOCKQUOTE>

Yep, I was reading that in one of the books I bought. Actually, I haven't seen it really recommended here but "Motorcycle-engined Racing Cars" is a pretty good primer.

http://www.amazon.ca/How-Build...8223992&sr=8-1-spell (http://www.amazon.ca/How-Build-Motorcycle-engined-Racing-Cars/dp/1845841239/ref=sr_1_1?ie=UTF8&s=books&qid=1298223992&sr=8-1-spell)

I have a couple of papers by Caroll Smith on the way. Hopefully those will help me out in terms of equations for suspension and chassis design.

[Crispy]

You only get the best out of 4130 if it's heat treated. Whenever you weld it, it's going to be annealed near the weld and not much better than 1020. If you don't heat treat 4130 you are basically paying for performance then throwing it away. We make a composite chassis now, but the hoops are 1020.

[dean]

i'm the first year team to get involve in the FSAE..can somebody tell me the link which i can get to know about the calculation in producing a racing car...urgent!!

[RenM]

F=m*a

[Barky]

T + M - SL - MOS = RC

Where T = Time, M = Money, SL = Social Life, MOS = Members of Opposite Sex

(Hint: solution only converges as T and M approach infinity)

[Kenneth]

Thanks for the info guys! Maybe in the next couple of weeks, I'll get the cajones to throw our design on here for a bit of criticism. http://fsae.com/groupee_common/emoticons/icon_redface.gif