Designing New swing-arm

[santeria13]

Hi all,

As part of my final year project at university, I am required to re-design my GS500 swingarm to achieve maximum weight savings. My current plan of action is to design somewhat of a C sectional, double sided swing arm whilst adding solid face material around the axle inserts.

However I am struggling to come up with the values and positions of the forces to input on to my model for FEA analysis. I have calculated the net load from the shock and rider weight in both the x and y direction as well as the maximum rear braking force. I have no idea in what positions on the model I would apply these forces however and am confused on how I would calcuate possible cornering forces or the load transferred when going over a bump etc.

Furthermore, I do not know what constraints or boundary conditions to set as my knowledge on FEA is very limited thus far and did not really get a choice on a project without it's use. I was wondering if anybody with adequate FEA knowledge would be able to aid me in telling me what are the main forces I should be focusing on and what positions they should be set to act on the swingarm model?

[Watcher]

Can't really help you with your technicals, but I know KTM uses sort of a ]---[ shaped swing arm (imagine --- as the axle, perspective from sitting behind the motorcycle) and they have a bunch of reinforcing Xs lining the outsides.  It's sort of like your idea, I think.

[Joolstacho]

You need to get hold of the following book:

Motorcycle Chassis Design: The theory and practice.
by Tony Foale & Vic Willoughby
ISBN 0-85045-560-X

It's the 'bible' on the subject!

[philward]

Although you have limited FEA knowledge you should see the fact that you have project that heavily involves this technique as a positive thing - you're at university to learn not just parrot what you already know.  Those people with 'comfortable' projects that don't stretch them won't get nearly as much out of it.

The other comment that I would make is to not be shy asking your course staff for direction.  In my experience they aren't going to want to spoon-feed you or do the work for you, but if you've had a good think about the problem and had a fair stab at trying to work it out yourself but are still stuck then any reasonable academic would be more than happy to help get you over the stumbling block.  After all you are paying to learn this stuff.

Look forward to seeing what you finally produce!

[Endopotential]

Did you pick the project design yourself?  That's a cool degree program that lets you work on your own motorcycle!

And make sure that once you work it out, start CNC'ng new lightweight swingarms for the rest of us!

Best of luck!

[gregjet]

Not sure a C section will be torsionally rigid enough. Will be fine in a simple vertical and horizontal input but won't it be sloppy in the long axis rotation ,unless you make a very tall C. Even then all the long axis rotation at the single plate axle area will tend to flex more than it should under leaning load.
I applaude the idea of redesigning the ancient box- section form to something more torsionally rigid, but not sure a C section is the way to go , so I will be interested to see how you apporoach the torsion issue.
BTW the KTM Duke swingarm isn't a simple C. It has triangulation ribs internally to help stiffen it AND being forged ( I think) aluminium has inherent stiffness from the volume of material.
Please post you development and any protos and problems found. VERY interested.

[RichDesmond]

You need to get hold of the following book:

Motorcycle Chassis Design: The theory and practice.
by Tony Foale & Vic Willoughby
ISBN 0-85045-560-X

It's the 'bible' on the subject!

+1000. Great book. Tony has a facebook page also, and another great resource is the motorcycle chassis design email list. (For you youngin's, email lists are what we used in the pre-forum, pre-web days of the internet.)

For swingarm design, the most important (and most difficult) situation to model is hitting a bump while leaned way over mid-corner. Do a good free body diagram and work from there.

[santeria13]

Hi all,

Thank you for the replies! Yeah I picked it myself  They tried shoving a project about designing an automatic user-less vacuum cleaner down my throat but I wasn't having any of it. I have ordered a copy of the book recommended too. I have calculated all the forces I require except for the force while hitting a bump mid corner. There would be a reaction force from the tyre which would have a horizontal and vertical component, acting with an increased load on the inner arm depending on the distribution of weight between the two arms in a corner.

So I have calculated the normal horizontal and vertical reaction forces acting on each arm whilst cornering without the tyre hitting a bump. Now is there a certain formula  to calculate the mass multiplier when you assume a certain height of travel? When it comes to going over a normal bump, I simply assumed that the weight value would be multiplied by a factor of 5 and multiplied that by the normal reaction force on the axle. I can't seem to find a good source of information regarding these types of forces and every other swingarm design ive stumbled upon has simply assumed a factor of 5 or something similar, prompting me to do the same.

In regards, to the CATIA part of it, here is the kind of thing I am thinking. Attached is a picture of the original CAD part I made before realising that the arm should be curved to ensure the pivot and axle spacing match up to dimension. In the process of re-making it with a curve now and it's coming along nicely. Will post more pics when done.

So in terms of obstacles right now:
1. How to calculate the exact value of the forces when hitting a bump
2. Where exactly to place my reaction forces and especially cornering forces  on to my fea model in terms of nodes.

Perhaps a good idea would be to create a cylinder acting as the axle in the axle insert and applying an upward force on the nodes on the bottom face of the spindle? In my mind, this would simulate the force acting from the tyre on to the axle. Then for the rear suspension forces I would assume I would do the same thing except apply the forces to the upper face of the rear shock bolt.

As for the cornering forces, I am not sure where they would be placed? From what I have read, they are focused on the mid point joining section between the two arms but would I select all the nodes on this middle face or would it only be a concentrated area? It is very hard to find any similar projects online or tutorials which have to do with anything similar.

[lefty1615]

Keep in mind that while people might say the force is concentrated in the middle of the member this isn't correct in a technical sense. Forces on mechanical members can only occur at connecting points so realistically what your going to have is some sort of moment produced by the forces at either end of the swingarm and that will produce the "force" concentration in the bar. Likely people are using "force" to talk about stress in the bar here which is fine for general considerations but not a realistic representation as far as your FEA analysis goes.

As for the force produced by a bump, I would personally start by calculating it as a falling problem. Idk if that's right and there's probably some short hand calcs that exist in the industry, maybe in that book you ordered.

I gotta be honest, you picked a bear of a project here.

[Suzi Q]

Is this going to accommodate a 200 rear tire? 

Very cool project, best of luck. Also super interested to see how things progress. Will you be building actual prototypes, or is this all done via CAD?