Mechanical Systems Lab Idea
 

To start off I am a senior in mechanical engineering. I ride a EC 250 in hare scrambles and enduros. I am in a class called mechanical systems laboratory and have to come up with an experiment to test something mechanical or of interest. I would like to do something on dirt-bikes because they interest me and mostly because it would make for a neat presentation with a dirt-bike sitting at our display table. I was hoping for some ideas on something that can be tested, even if it's been tested before. I had some ideas of doing a dynamic analysis of different flywheel and their effects or measuring the flow of the water pump and doing a heat transfer analysis of the dirt-bikes cooling system. Any ideas would help.

Thanks

Nick

How about hydraulic pressure inside the fork cartridge and how its affected by speed and valving.

Air flow through the engine. That will keep you up for many nights.;)

Just some random ideas...
 

Calculate the fastener torques for various items on your bike and compare to the owner's manual.

Strain gauge a component on your bike (such as subframe) and measure the strains during a ride. Convert the strains to stresses and then estimate the applied loads.

If you have access to accelerometers, you can mount some to your bike and measure g-forces overs bumps, jumps, etc.. This could also tie into what Glenn suggest in his post.

Thermal efficiency of the bike's radiator when riding at 15 mph versus 30 mph. You can even show videos of KTMs overheating on technical rides. :D

Calculate the required spring rates to achieve the desired rider and static sag. Knowns would be rider weight, bike dimensions, and leverage ratios of the suspension.

Create CAD model of motorcycle wheel (rim, spokes, hub, etc.). Perform FEA on the model to predict stresses due to an applied vertical load. Estimate the fracture load and then develop a correlation to an wheel tested in the lab.

WOW, I'm just stay here and lick the glass ....

Thanks for all the good ideas, please keep them coming. I could possibly do most of them as long as I can get sensors and gather the data easily in the lab. I don't know how well I can gather data while riding and running the data acquisition program on a lap top. I like the thermo model at different speeds especially vs. the ktm. Just have to find a ktm to test, I guess its time to make some phone calls before the season starts up.

Why not build a laser-diode based displacement sensor to measure wheel travel under use? You can also integrate a GPS receiver, accelerometer on the wheel and look at displacements and accelerations (you can calculate velocities from those) as a function of location on the track.

Are you using all of your travel? If not, you can adjust the suspension to better utilize what you have.

You can build a system front/rear (using same transducers), and do it for a couple of hundred bucks (and lots of your time). This tool, which is something like race-tech has, would prove really helpful in dialing in a suspension. The ability to look at location-based performance would be invaluable.

If you don't do it, maybe i should!

Well by comparison, my bicarbonate volcano idea really sucks. :(

agreed.. my burn out demonstration would not go far either....but it would fill the lab with some great smells with the right pre mix..

I could come there with a bike and you could do a test on how hard you have to hit a Georgia pine tree to get an exact replica of the bark on my left shoulder. Left my darn chest protector at home..... My shoulder has a perfect bark pattern on it.