Finite Element Analysis 1

[kwklein]

I've got a clutch assembly for a performance motorcycle. The goal is to reduce the mass but retain strength and durability. My question is if FEA is the best approach to determine the best compromise for the design of the assembly or if there's a different type of analysis that can answer the question of best design and material use.

 

[Bribyk]

To do this properly and if you're keeping it at as a wet clutch you're getting into the tribology, slip-stick friction, heat generation, centrifugal stress... See if anyone makes a proven replacement clutch.

Coming from a 5-year FSAE vet, there are much better avenues to pursue to improve your car. The stock clutches on any late model 4-cyl are adequate and messing with the engine internals is almost guaranteed to cause engine failure.

 

[kwklein]

Bribyk – The focus is more around the housing than anything. The actual clutch fibers and disks will remain as off the shelf items. We already have a new redesigned clutch assembly with only changes to the primary gear assembly (outer hub), hub clutch sleeve (inner hub), clutch cam set, and the clutch pressure disk. I understand that the OE clutch assembly is adequate but the focus is on the improvement. I think you may have missed my last post that explains the application a bit more.

 

[Bribyk]

My apologies I did miss that particular line in your previous post.

Here's something to think about though, Suzuki superbike teams add rotating mass their engines (crankshaft) to improve the bike's drivability by providing a more predictable throttle response. This shouldn't be as much of an issue on an FSAE car s you have added drivetrain components that a bike doesn't but making the clutch lighter is going to change its engagement characteristics.

As far as materials, if OEMs are using magnesium and such for engine covers, they've probably though about it for a clutch basket (potential fire issue maybe?).

 

[IRstuff]

I would bet that current housings have been optimized, so you'd need a totally different material. Unfortunately, there really aren't that many choices. AlBeMet is one option, but it's expensive, and requires hazardous material handling. Nonetheless, it would significantly lighten the housing. Ti might be another choice. The end result would likely be something that's maybe 20% lighter, but 2 times more expensive.

TTFN

FAQ731-376

 

[GregLocock]

Bear in mind that the machining cost for parts like this often exceed the material costs (it certainly was when I used Ti) so the choice of material is not really a cost based decision. Beryllium alloys are banned by the FIA, at least for some series.

Cheers

Greg Locock

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

 

[kwklein]

Thank you all for the feedback. These are all good conversations and I appreciate the input. The application is strictly focused on drag and land speed racing which means smooth throttle response isn’t necessary. It’s always full throttle/wide open. I do understand magnesium is very much a fire hazard. I was only using that material as an example of the weight difference between the OE material and the final material. My thoughts are more along the lines of aluminum alloy, metal matrix, Inconel, or just using titanium. The machining costs aren’t an issue but material bulk costs can be.
I’m sure we can agree that if the engineers who design the motors were given less constraints some parts may have been designed differently and / or alternate materials would be an option. The hazmat materials may be an issue so I am trying to steer clear of Beryllium alloys but not because of FIA restrictions. They’re not off the radar completely just not a primary option.

 

[EdDanzer]

I think you could design a lighter better part with FEA. The analysis will only be as good as the input data. Contrary to many beliefs, machined and heat treated E4340 can be lighter and stronger than many materials in high load or impact applications. One method is to use a CAD and FEA software that allow many “what if” iterations in a short time. Once you have explored these different designs then hand them to a FEA specialist to confirm the better design.

Several years ago we had to design a heavy duty low speed brake and found the number of design variables small and straight forward. Total torque transmitted is a function of pressure and coefficient of friction of the mating materials, removing the energy (heat) produced can be a challenge.
Rather than use several small springs you might consider a machined spring like these:

http://www.taylordevices.com/machined-springs.html

Combining the spring shape into the parts could provide a stiffer, stronger, yet lighter part.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[GregLocock]

If you really know the loads then consider using an optimising FEA such as Optistruct (there are others).

Cheers

Greg Locock

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

 

[Tmoose]

"The application is strictly focused on drag and land speed racing "

Those strike me as wildly disparate forms of racing, aside from needing big HP numbers.

If the stock part is bullet proof I'd spend the money on a suitcase full of carb jets and needles (or a laptop with cables and software for mapping), some dyno time (5800 ft altitude to simulate Bonneville's air density), and a heap of final drive sprockets, and tell my race buddies the clutch is machined from titanium billet with "chrome-moly" inserts to resist denting from the clutch disk tabs

 

[kwklein]

Tmoose - I almost fell over in my chair with the Ti billet and chrome-moly comment. There are several areas that need improvements one being the clutch. There was a focus on intake/exhaust but the application level was to granular. The other thought was a quickshift trans design but it's been done already and costs a bundle more. Anything to reduce parasitic losses is open territory. Cuurently I only have buy-in for the clutch assembly re-design.
btw, what's a carb? lol.

 

[Tmoose]

Orient Express sells a NINJA 1400 "kit" ( contents unknown) for $543.
ttp://

http://www.orientexpress.com/images//OERProducts/OER-2000-ZX10-SAC.jpg

Muzzy's is 630- $760 and includes a fuel shut off feature for smoother shifting.

http://www.muzzys.com/MZAS-K12/

Still need a compressed gas bottle, etc.

A 6 inch long 8 diameter hunk of garden variety 6061 could cost over $200 before the machine shop even turns on the coffee machine.
(McMaster CArr 1610T75 Multipurpose Aluminum (Alloy 6061) 8" Diameter, 6" Long - In stock at $232.09 Each )

 

[kwklein]

Tmoose - yes 6061 is a common aloy used but normaly of the aged variety. We're leaning tward a 8090 alloy though.

 

[kwklein]

Tmoose - the shiftless trans I had mentioned is a bit more advanced than just a piston to operate the shift lever. If you're interested in reading a bit more about them here are two good sources of information for you.

http://www.zeroshift.com/pdf/RcarN6V15_Zeroshift.pdf

http://www.zeroshift.com/transmissions.html

cheers