Visit the forum instructions to learn how to post to the forum, enable email notifications, subscribe to a category to receive emails when there are new discussions (like a mailing list), bookmark discussions and to see other tips to get the most out of our forum!
Capstone Design Class at Georgia Tech
  • I'm currently a senior Mechanical Engineering Major at Georgia Tech, and taking our final course, Capstone Design. We're still bouncing a few project ideas around, and having trouble sticking with one. I was hoping y'all could suggest something we could work on that has something to do with Open Source Ecology. 
    The only catch is that we need to bring our Professor a problem, not a solution. I realize that y'all have come up with a bunch of different solutions in the form of tools and fabrication techniques. Do you know how I could approach my Professor with a specific problem to solve? I'd be gracious for any suggestions! 
     
  • 4 Comments sorted by
  • Take a look in the wiki, there are quite a few challenges outlined that need to be solved.

    Some wiki pages to look in:

    The main LifeTrac wiki fans out into quite a few pages, some of those may give you ideas on problems that still need to be solved: http://opensourceecology.org/wiki/LifeTrac
    for specific list of issues for LifeTrac check out: http://opensourceecology.org/wiki/LifeTrac_III/Feedback

    Some random things I would like to see:

    - reliable design for front loading washing machine
    - a universal kitchen appliance: combines a kitchen stand mixer, a blender and a food processor. the kitchenaid is: very low rpm, very high torque. and the blender: is very high rpm, decent torque (for ice). and food processor is: high rpm, high torque:

    imageimageimage


     
  • Hey mate,

    Yep there are lots that could get done with the tractor, or you could almost choose any of the ones that haven't even been started and go through the entire design process which could be good projects. 

    For the tractor, as you'll see in the comments there hasn't been any real engineering done on it. Designs were chosen based on simplicity and "gut feel" of people without a ton of machine design experience. Not to say it's not a valid design, it's just that there's no actual data to back any of the claims up.

    Some tractor stuff I'd like to see addressed:

    - Proper modular and assemblable frame design that will take a skid steer loading. Designing for simple assembly while still meeting tolerance requirements. 
    - Proper drive train, analysis of cost, simplicity of design, etc of different options such as hydrualic motors on each wheel vs chain drive and only two hydraulic motors
    - Method of creating a wheel mounting method that is engineered properly and analyzed
    - Create a ROPS cabin

    As eukreign was saying check out the feedback on the LifeTrac III, and I'm trying to work through some of this stuff on the LifeCat 1.0 (You'll see it on the main lifetrac page)

    You could even do any of the tractor accessories. It's changing over to the Bobcat Universal quick attach plate so the tractor design is inconsequencial. So figuring out a good trencher design (http://opensourceecology.org/wiki/Trencher) or something like that would be great and probably more easily done and buildable for a capstone course.

    Give it some thought and I'd be happy to talk more to you guys about it!

    Cheers,
    - Mike

     
  • "The only catch is that we need to bring our Professor a problem, not a solution. I realize that y'all have come up with a bunch of different solutions in the form of tools and fabrication techniques. Do you know how I could approach my Professor with a specific problem to solve?"

    If you want a problem with nothing even close to an obvious solution, I suggest taking on one of the many "universal" or "modular" challenges. For example:

    Universal Rotor Natural motion is linear, mechanical motion is rotary. The GVCS is essentially a giant, distributed mechanism for transforming raw materials (dirt and sunlight) into modern stuff (houses and cars). That means there's going to be a lot of rotary motion. It would be great if all those rotors could use a single standard so they are all interchangeable. This probably wouldn't be a single rotor so much as a standardized set of rotor parts that, when assembled in different ways, cover the entire design space (IE: torque x RPM x moment, or something like that). The real challenge would be designing a set of parts that are all themselves easy to create from scratch (casting, cold rolling, simple welds, etc). 

    Hydraulic Actuator OSE has chosen hydraulic power as its default standard because of its power density and flexibility. A high-end hydraulic drive gets 85% efficiency, while a high-end gear drive gets 95%, but hydraulic power can be transmitted through hoses and is infinitely variable. Ultimately, the GVCS will have to be able to produce its own hydraulic motors, cylinders, hoses and fittings. Can you guys figure out how to meet international standards from scratch?

    Steam Engine Again, not the most efficient choice, but steam engines can be powered by anything that will burn. Liquids, solids, powders, gases...anything. Figure out a way to build a super-critical engine, from raw materials, that will last 50 years.

    Modular Construction A vital part of the GVCS idea is that all the technologies will be based on compatible standards, as opposed to what we have now where everything was developed piece-meal. Starting from scratch means being able to control the designs so they work naturally with all the other designs, even if it's not the best decision for that particular design. A big part of creating a system is to pick a single indivisible unit and built everything on multiples of that. But what unit? What simple parts can go together to make any arbitrary machine? More importantly, is there a way to make the system "dimension agnostic," meaning people could build the machine using whatever they have available. For example, the GVCS is currently designed with imperial units, which means people with access to metric parts have to basically redesign the machine to work with the materials they have. But, designing in metric creates the same problem for people with access to imperial parts. Could the machines be "dimension agnostic" so that it doesn't matter what dimensions the parts are based on?
     
  • Matt, those are all good topics, but I think they would be a bit too undefined and broad with far too many variables for a capstone project.

    Each of those requires serious analysis of all the tools that would be using them to adequately create a proper system. And the problem is that that definition is not quite there yet with the GVCS. So yes they could start work on it and progress some, but I'm not sure it's the time for that yet. For something like the hydraulic actuator, these are not technically advanced projects, they are just projects that require the precise tools to make them. All the data to make them already exist and could be found easily. But actually machining down to 0.01mm of precision is where the capabilities are lacking in OSE currently. this is something that will be addressed in time though with a precise open source lathe/multimachine, but generally pretty expensive and complex tools requiring being able to draw upon years of machine manufacturing.

    I was thinking about it on the drive home and as much as I would want them to work on my pet project like a new tractor frame I think one of the enabler tools that are a bit more concrete, like one of the tractor accessories (seeder, hay rake, trencher, spader, etc) or something like dairy milker would make a pretty cool definable project. Their question then becomes how do i make said tool simply, with local resources, at least half or more savings in cost and only with local resources and labour? And then what tradoffs do you have to make in performance to achieve these other gains? These tools exist which provides a source of the definition and scope, but then require the full analysis and innovative design to make it simple, usable and cheap. 

    I'd be happy to work with you guys eweinhoffer if you wanted to try and set some definition on stuff. I'm a mech eng 2007 grad from university of toronto if that helps justify it to your prof at all too haha

    Let us know either way what you end up deciding! I think getting universities involved is a great idea.
     

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Login with Facebook Sign In with Google Sign In with OpenID Sign In with Twitter

In this Discussion

Loading