|
|
|
 I like to divide checklists like this into "stuff I have to do before applying power" and "stuff I'll find faster by watching it fail". - EricVanWyk [more] |
 |
|
|||||||
|
||||||||
This is the product of our R&D team. This could potentially find itself in the 2011 competition. A special thanks to Team 118 for the inspiration.
Discussion
view entire thread
07-01-2011 01:03
Chris is me
Re: pic: Team 2220's Swerve Drive Module
Be weary of the amount of chain wrap you have on that upper sprocket. Eyeballing it, doesn't look like a lot.
07-01-2011 01:50
jspatz1
Re: pic: Team 2220's Swerve Drive Module
I would suggest #25 chain/sprockets and an AndyMark aluminum sprocket on the wheel. Much lighter and worked great for us in a similar module and reduction.
07-01-2011 08:02
Jared Russell
Re: pic: Team 2220's Swerve Drive Module
I have some concerns about the support of the module itself. It looks like you are using a single COTS "lazy susan" type bearing. My experience is that these sorts of bearings are great when it comes to thrust loads, but they can suffer devastating failures when subjected to lots of torque loads (imagine if your robot is being pushed sideways, or if it turns at a high speed - there's a lot of torque trying to pry your bearing open). Your pivot is located at the top of the module, so the distance between it and the floor acts as a lever arm and amplifies the torque the bearing will experience. We have used this type of bearing on turrets before, and twice we have experienced devastating failures after a big hit where the bearing cracked open.
If you want to get away with a single, high pivot point, I would point you to the www.team221.com Revolution swerve module design - if you read through their product descriptions, you will find the McMaster part # of the bearing they use for module support. Better yet, you can use multiple pivot points (one high and one low) to prevent excessive torquing at the top.
07-01-2011 10:28
jspatz1
Re: pic: Team 2220's Swerve Drive Module
Jared is correct, the lazy susan bearing cannot handle the pivot function for a top mounted module like this. These bearings are relatively light sheet metal made for thrust load only. With the side loads and impacts you will see in FRC they will fail quickly (the sheet metal races that contain the ball bearings will fail and separate.)
Look at this recent thread: http://www.chiefdelphi.com/forums/sh...=swerve+module
It shows some mounting styles including our own. For modules with top support only, this support needs to be especially rigid and strong. Most module designs use lower support to avoid this difficulty. I notice you have a tall stem feature on top of the module. Could that be captured in a pair of bushings to form your pivot?
07-01-2011 11:10
RoboLuigi
Re: pic: Team 2220's Swerve Drive Module
Chris is me: You concern is quite valid, we may have to change our gearing ratio in the future to fix this if it becomes an issue.
jspatz1: I definitely agree, we plan to switch to #25 chain and sprockets if/when we have a weight issue this season. We used #35 on our prototype to try to avoid chain stretching or breaking, and were going to try #25 next to see if we can reduce weight. Your team's success gives us more confidence that it will work well and remain durable throughout the season.
Jared341 & jspatz1: We had a similar concern with that bearing, in fact the students and the mentors spent an entire night discussing the issue. The main support for the module is the column on the top, which will be inserted through two bushings in the chassis much like you suggested, jspatz1. This will allow the module to pivot within these bushings, and the two points of contact will take most of our torque loads we may suffer from collisions or rapid direction changes. Our bearing therefore should only have to support thrust load. The only issue we have left is to find a more suitable bearing to take this thrust load.
Me and Jedward45 will try to post the other pictures of our swerve system ASAP so you can see our full design for yourself.
07-01-2011 16:55
jspatz1
Re: pic: Team 2220's Swerve Drive Module
If you are capturing the top stem then it should all be good, as long as its connection to the module is strong. Here is the Mcmaster needle roller bearing we used for thrust load. Much smaller and lighter than the lazy susan. Just fit the ID closely and it requires no other guidance.
07-01-2011 18:52
Gdeaver
Re: pic: Team 2220's Swerve Drive Module
This is similar to what we did last year. A page with what we did.
http://wiki.team1640.com/index.php?t...VI_Drive_Train
07-01-2011 19:05
Madison
Re: pic: Team 2220's Swerve Drive Module
Unrelated specifically to the development of swerve, but
http://wiki.team1640.com/index.php?t..._annotated.jpg
is the best thing I've seen in months. Adding the McMaster standoffs to the wheel/sprocket interface is a fantastic idea that we will be stealing from you this year. 
08-01-2011 02:00
steelerborn
Re: pic: Team 2220's Swerve Drive Module
Wow really nice CAD work.
I would look into thrust bearings for the module.
Silverthin has a great selection:
http://www.silverthin.com/
Keep up the great work, posting different revisions up on CD is a great method.
view entire thread
Tags
loading ...