Designing a Motorcycle Rear Fork/Swingarm 1

[HLai]

Hello everyone,

I'm in the process of designing a swingarm for a 250 pound motorcycle. The kind of a swingarm is the traditional types found on cruisers. Wikipedia calls it:

Swinging fork - the original version consisting of a pair of parallel pipes holding the rear axle at one end and pivoting at the other. A pair of shock absorbers are mounted just before the rear axle and attached to the frame below the seat rail.

I'm lacking confidence in my FEA and having problems calculating the forces. Two things that I want to FEA on this swingarm are:

-when driving through potholes, I want to make sure that when the suspension travels in bump that it is strong enough.

To analyzed this, I assumed a weight bias of 0.7 and used a magnification factor of 2 for the amount of force that the mounting point of the rear axle will receive. Is 2 a reasonable factor to use for this application? I feel that it is a bit too low but have no idea how to calculate it.

250*0.7*2=350

Anyways, I fixed the pivot point of the swingarm and the two shock mounts on the swing arm. Then I applied the force of 350 on the axle mount.

-when driving through bumps where the tire contact patch is in contact with the road in a non-centered part of the wheel, I want to make sure that the swingarm is stiff enough when it is being twisted.

I modeled the axle in CAD and defined the contact between the axle and the swingarm. I then offset the force of 350 pounds from the middle of the tire by half the tire thickness.

I feel that I should have learned all this in school but I didn't. Thanks to everyone.

 

[BrianPetersen]

One thing I can see straight away, is that you had better consider the weight of the rider and passenger and cargo! Use the total GVWR in the calculation, not just the weight of the bike.

Another thing is that when the bike is cornering, the force is the vector total of gravity plus the cornering G-force, and in sport bikes, the cornering G-force can be easily 1.0 and possibly 1.2-ish, and on top of *that* you should apply the bump load.

Yet another thing to consider is the force imposed by the final drive (chain / belt pull or torque reactions imposed by a shaft drive). The chain pull can be quite a bit higher than gravity under maximum acceleration. It helps that it is imposed along the length of the swingarm rather than at 90 degrees, but it's still there.

Sport bike swingarms are largely governed by stiffness considerations - just take a look at any late model swingarm and you can see this. No idea what the FEA would say, but it looks like they are designed to be extremely stiff in torsion and ordinary bending, and less stiff in side-to-side bending (going over a bump when leaned wayyy over requires the swingarm to bend sideways slightly). This is a tricky business, though. The swingarm length, pivot height, etc have an influence on the handling.

 

[Bribyk]

It never hurts to reverse engineer some existing parts just to see if you're in the ballpark or not. See what the FEA spits out and then measure some similar ones from production bikes and compare your materials and cross-sections.

FYI: Vehicle suspensions are typically designed to withstand 3 (racecar) to 5 (street) times the normal tire weight for bump situations and, as Brian mentioned, this includes the passenger(s), cargo, fuel, etc...

 

[nobog]

HLai, You can look through history and see the developement of swingarms. A 70's vintage dual-purpose bike, lets say a Yamaha DT250 2-stroke, weights about 250 lbs - you could beat the cr@p out of those things (dirt, street, you name it) and I can't say I have ever heard of the swingarm being a week point or bending, amazing considering the puny tubing they were made of. Now take a new Yamaha 250 4-stroke (about the same 250 lbs) and look at the swingarm - massive compared to the 70's version. Both worked but the new bike can take 30 foot jumps where the old bike would snap - frame and all. So it all depends on application - input (potholes) need to be defined - and thats nearly impossible to model. For your application, basically any 1" square chrome moly tube will work for a low performace 250 lb street bike with 3" of travel. Look at the swing arm on a 550lb Harley Sportster, it looks like it should bend under its own weight but it doesn't. Off road, or high performance could take a major engineering effort - FEA or not. Regards, JK

 

[HLai]

Using the 5g bump assumption and including the weight of driver, passenger, and cargo, I have made some tests on the twist, bump/compression of the swing arm and chassis.

For the swing arm, I had deformation of about 1/16" max but i don't think that matters too much. I think what matters more is the difference in deformation between the left and right leg of the swing arm. Am I on the right path?

The FOS of the swing arm is only 1.6, with the weak spot being at the hole for the axle. I modeled the FEA with the swing arm, bolts and axle parts for the mounting points and assigned contact between them. 1.6 seeems a bit too low to me since I was aiming for something closer to 2-3.

For the chassis, how much deformation is too much? Does 1/32" for the 5g bump test sound like a lot? What about 1/16"?

 

[GregLocock]

5g etc are strength loads - you don't worry about the deflections much. Normally the stiffness of the structure is considered to be a design goal, you need to consider what your objectives are there, as well.



Cheers

Greg Locock

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

 

[patprimmer]

I would have thought a deflection that changes the wheels alignment or sideways motion or inclination would be the only aspect of deflection of interest.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[HLai]

The biggest deflection on the motorcycle chassis is the head tube, the tube that holds the steering fork. This would change the steering alignment quite a lot.

The chassis I'm designing is almost identical to the Norton Featherbed chassis due to packaging concerns. It was not surprising to see the head tube area being the weakest.

GregLocock,

Will the deflection be calculated from the dynamic loads then? In other words, if the dynamic loads show a maximum of 2g, would a static load of 2g calculate accurately the amount of deflection?

Or would it be twice the amount shown for dynamic load? In otherwords, if the dynamic load is 1g * mass, would I use 2g * mass then?

If you know the theory behind all this, it would help very much also. Thank you very much.

Horace

 

[GregLocock]

I don't know what you mean.

Cheers

Greg Locock

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

 

[MikeHalloran]

>>>
-when driving through potholes, I want to make sure that when the suspension travels in bump that it is strong enough.

To analyzed this, I assumed a weight bias of 0.7 and used a magnification factor of 2 for the amount of force that the mounting point of the rear axle will receive. Is 2 a reasonable factor to use for this application? I feel that it is a bit too low but have no idea how to calculate it.
<<<
Search this site for discussion on 'bump stops'. It's been covered. .. Or maybe I shouldn't have mentioned it.

>>>
If you know the theory behind all this, it would help very much also. Thank you very much.
<<<

The theory behind all this is that when it breaks, you will go to court and be skinned alive, figuratively. ... unless the plaintiff was a Hell's Angel, in which case it will be literally.

No offense is intended, but you seem to be in over your head, and deeper in it than you think you are.


Mike Halloran
Pembroke Pines, FL, USA