[CRUSH-build] continued planning for Las Vegas

[Gary and Joann Hawes]

Bill,

Garrett and I are going to skip today since we have something to do tonight and don't have the energy for both.  I will come up with a thrower trial for next week.

Joann

  ----- Original Message ----- 
  From: Bill Bennett 
  To: build crush 
  Sent: Thursday, March 16, 2006 1:59 PM
  Subject: [CRUSH-build] continued planning for Las Vegas


  Hi team

  On Tuesday we discussed input that was received from team email, mainly driven by email from Chris and Brian, and contriutions from numerous other team members. Bottom line was that we believe building a conveyor belt collector would take most of the practice day in Las Vegas and this is too high a price to pay. It would remove any time to practice as well as any time to make software work. 

  So we've allocated most of the practice day Thursday to the software team and the player team.

  The main hardware change we will try is to make the current collector design work with reasonable reliability, by preparing for the following changes before going to LV: 
    1.. remake the collector's roller so that it is wider with less shaft stub from each end, thus much less likely to get bent up 
    2.. change to a stiffer shaft (drill rod) 
    3.. fit the roller with a sprocket on each end, two different sizes, offering either the speed we have now or a 2-1 speed up 
    4.. extend the collector ramp from the bottom, and the deflector from the top, so they overlap and there is nothing the ball can hit except those surfaces
  We think with these changes, enabled by eliminating the weight of the basket winder mechanisms (motor, wheel, springs, cable, pulley, etc) we have a chance to make the collector very reliable. A fallback plan is to enclose the collector space and use it just to collect balls that will be discharged into the corner goal by running the collector in reverse. 

  The main prep for this hardware change is in making a new collector roller assembly. Bill will drive this effort next week. Also, we've ordered camera parts to fit FWIPER with a 2006 camera and enable debugging the camera and autonomous code before LV. Both of these efforts will use permitted fix-it periods of the next two weeks. We've pretty much given up our fix-it periods for this week, which is fine (people needed some down time). But if we use it all, we have 4 five hour periods remaining before LV.

  We'll also add dividers in the basket to reduce the jamming. 

  The goal is to get these changes done in the first two hours of the LV practice day, then yield to the software and player team for the rest of the day. 

  At today's meeting we'll unload the trailer. Over the next two weeks we have the chance to improve our travel kit planning (it was not bad but could be better), develop some team spirit stuff (a player dance would be nice, some cheers, etc), and generally plan some of the details of the LV trip that we ran out of runway on before Phoenix. For example, an organized effort to get game video.

  Another thing we should do is some ball throwing trials. Jeremy did a great job, but we didn't really put any time into seeing who else might want to try for this position. I'd like somebody (Joann??) to take the task of designing a human player test and running interested students through it to choose the LV human player. 

  Lastly, who is interested in coming Saturday for coffee, hot chocolate, donuts, and watching regional webcast? Saturday is not available for fix-it period work, but we could watch the play and maybe refine our own play strategy a bit. I'll come if anyone else wants to ....

  Bill 


------------------------------------------------------------------------------


  _______________________________________________
  CRUSH-build mailing list
  CRUSH-build at crush1011.org
  http://crush1011.org/mailman/listinfo/crush-build
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://wizardofaz.whsites.net/pipermail/crush-build/attachments/20060316/bc271f97/attachment.htm