Mechanism for CAD practice

[Frenchie461]

Now that it's off-season and I have some free time, and I've spent some time working on CADing a robot drivetrain, so I was wondering what a good mechinism is to practice on?
Any suggestions or a game to work around would be great.

[Hawiian Cadder]

http://www.dezignstuff.com/blog/

this blog periodically puts out solid works challenges. they are more into complex parts however. i had a little fun with klein bottles recently, i find it a good exercise because it works in surfaces, 3d sketches, math, and it doesn't take too long. if you are really serious, then re-design old robots. i re-did our 2011 robot at least 5 times, and i did 2007-2010 once as well.

My best Klein bottle:

[,4lex S.]

Do the calcs for, and design an arm for a previous game. Well engineered arms mounted on the kitbot can win a regional, ask 2056 in their first year.

[MattC9]

Do what every great car company does, reverse engineer a robot!!

Your probably thinking "how do i do this?!" what i'm doing right now is finding robot CAD files on here(CD) and at http://frc-designs.com/html/2011.htm, now remember none of these robots come with mates, so take it apart and learn how it goes together and how it works! Like 148 and 33 have there main robot it 1 assem, with multiple sub assem. Looking at some of these robots is great CAD practice and a great way to learn about common FRC mechanisms!!

[Starke]

Originally Posted by MattC9

Your probably thinking "how do i do this?!" what i'm doing right now is finding robot CAD files on here(CD) and at http://frc-designs.com/html/2011.htm, now remember none of these robots come with mates, so take it apart and learn how it goes together and how it works! Like 148 and 33 have there main robot it 1 assem, with multiple sub assem. Looking at some of these robots is great CAD practice and a great way to learn about common FRC mechanisms!!

As Matt said, FRC Designs is a new website that will be used to host robot system and sub-system designs! Below is the correct link that was posted incorrectly previously.

http://www.frc-designs.com/html/2011.html

Matt

[Bongle]

My team has talked about but never implemented a motion simulator. The programming team prototyped the game->cRio interface (it's really easy), and we prototyped a very small scale motion machine.

What I'm talking about: http://www.youtube.com/watch?v=5E_Hv3jSuc4

The racing sim Live For Speed outputs all of its physics data (car accelerations, positions, etc) as a series of data packets over the network, which can be directed to and read by your cRio. Build a robot that moves a seat like the YouTube video shows, and you've got yourself an excellent recruiting tool. You could probably also use it for flight sims, etc.

Short version:
CAD Practice: Design a a seat/monitor combo with at least 2 degrees of angular freedom, that will have enough angular acceleration to quickly respond to inputs from the computer game. For bonus points, design it so it can easily be packed/unpacked into a vehicle available to your team to take to demo events.

Programming practice: Design a program on the cRio that could tilt/pitch the machine so to trick the driver into thinking he's actually driving. There's actually a bunch of psychology involved: link

Business practice: Market a "motion simulator" kit.

Got lots of money? Build this: http://www.youtube.com/watch?v=9aG28c1aQNk

If anyone is really interested, I can hook them up with the code needed for the game->cRio interface.

[akoscielski3]

Originally Posted by ,4lex S.

Do the calcs for, and design an arm for a previous game. Well engineered arms mounted on the kitbot can win a regional, ask 2056 in their first year.

they were paired with 1114 though. Like usual But they are still good I'm not hating on them

[Nick Lawrence]

Originally Posted by akoscielski3

they were paired with 1114 though. Like usual But they are still good I'm not hating on them

2056 was freaking good in '07 on their own. #1 seed at GTR.

They were NOT carried.

-Nick

[Andrew Schreiber]

Originally Posted by akoscielski3

they were paired with 1114 though. Like usual But they are still good I'm not hating on them

I hope you aren't saying what I think you are.

[Akash Rastogi]

Originally Posted by akoscielski3

they were paired with 1114 though. Like usual But they are still good I'm not hating on them

They seeded above 1114 at GTR I believe

edit: Nick beat me to it

[akoscielski3]

Originally Posted by Nick Lawrence

2056 was freaking good in '07 on their own. #1 seed at GTR.

They were NOT carried.

-Nick

Oops well i wasn't technically on the team yet some im going to guess ur right. I just always remember 1114 picking them in GTR.

[JesseK]

CAD a gearbox with a clutch in it. Source the clutch, try to find a sketch or detailed drawing for it, and put it into the gearbox.

Clutches act as good mechanical safeties for the very likely event your programmers muck something up during the course of sensor software integration. You mostly likely can counter-balance your lift or arm so the clutch isn't always slipping; conversely if the arm/lift gets caught on something then the clutch will slip. The caveat to this approach is that your rotary sensor (potentiometer, encoder) will need to be mounted after the clutch rather than before it -- i.e. you should put it on the rotary joint itself rather than the 1st-2nd stage of the gearbox.