[CRUSH-build] RE: [CRUSH-all] Preparation for Las Vegas

[Green, Wesley A]

All,
My vote is on option 3 - conveyor belt.  It requires a pole with
bearings on the top with the use of the collector system we already
have.  We would also need a wall to climb up (can double as one wall of
the basket) that should be made of polycarbonate (Lexan) NOT acrylic.
I'm not sure how the rules go but the only way I see this option
happening is by building and testing the concept, then tearing it all
off before it ships.  Bring all raw materials and drawings to Las Vegas,
forgo any practice time and just build all day.  Drawings are a definite
must - hand a pack to the machine shop first thing in the morning and
build whatever else we can while we wait for parts.  
I also agree with adding the shelf in the basket.  This proved to
eliminate the bad jams, the minor ones were freed up by the vibrations
in the shooter.  This can be done in the 'fix it' windows.
Another note on the conveyor - we need to gear it down.  Given the
current design it may mean we flip it around to use the 2-1 gear
reduction or make another one to get it even slower.  Another option is
to use the motor and gears from the basket dump... this could be too
slow, I'm not sure.  Any balls that are on the way up could be used to
shoot in 'barf' mode.

A note on the shooter - firing from close range is better but we seemed
to consistently shoot low.  Let's cut it back in the fix it window... we
have a spare in case it creates unforeseen problems.
I also saw the motor slipping down a bit - we need to make stops at it's
highest point.  The less ball squish the less we slow down the motor as
a ball passes through, to a point... after that we don't get enough
traction or speed.

This could only work with a lot of effort on everybody's part.  If
people are burned out or are generally dissuaded by the amount of work,
I can empathize and will be willing to work with any of the options
Chris laid out.

Wes

-----Original Message-----
From: crush-all-bounces at wizardofaz.whsites.net
[mailto:crush-all-bounces at wizardofaz.whsites.net]On Behalf Of
Christopher H Hanson
Sent: Monday, March 13, 2006 1:09 PM
To: all at crush1011.org
Subject: [CRUSH-all] Preparation for Las Vegas


It's time to get ready for Vegas.


Here are some things that I have been thinking about, and would like to
run
by the rest of you. Please respond with additional suggestions, or more
pros and cons.

I see that we have three options on how we can handle robot preparation
for
Vegas:

1) Stay-the-Course.  The robot is functional.
We need to fix the axle for the sweeper, and work on switches and code
fro
the ball dump.

Pros:
      - More time to practice and test code on Thursday.

Con:
      - We won't be much better then we were on Saturday in Phoenix
(which
wasn't bad).


2) Make upgrades that do not require structural changes (no welding or
time
consuming tasks)
Someone on the team (I don't remember whom) suggested that we turn the
sweeper into a low goal shooter.

Pros:
      - Minimal construction time.  The robot already can collect balls
into the sweeper area (it does it now).  We only have to put sides and
extend the  ramp.

      - Low added weight.  We are right at the limit, but the additional
weight would not be very much. (also see next pro).

      - Ball Dumper no longer needed. Since we are shooting low, we
don't
need the parts that make the basket dump. Subtract motor and springs.

      - Autonomus code to run to corner and unload into low goal is easy
to
do, and very effective (remember Fast Eddy).


Cons:
      -  We would no longer be able to feed the shooter after the
initial
supply of 10 balls runs out (see possible mitigation 1).

      - We would be removing the shaking mechanism that helps balls to
not
jam (see possible mitigation 2).

      - Would take some time on preparation Thursday to accomplish
upgrades.

Mitigation 1:
      If we no longer have to dump the basket, it can be taller.  We
could
add some kind of netting and minimal frame to allow our human players to
help  reload by tossing them in. - add weight for wire frame and netting
-
remove weight for existing backstop.

Mitigation 2:
      We could place ramps into the basket the keep balls from jamming.
Two ramps inside might work.  - drawback: according to Bill, we can only
get 6       or 7 balls into the basket with ramps. - mitigation 1 might
allow us to put more balls in.

      Option: Replace shower curtain sides with flexible plastic tubing.


3) Make upgrades that require structural changes (welding or time
consuming
tasks)
Another suggestion has been made to add a conveyor belt to the sweeper
section to pick up balls and deliver them to the basket.

Pros:
      - Reliable way to feed our shooter - unlimited ball supply.

      - Could be used in reverse to score low goals.

      - Removes redundant basket dump (subtract motor and springs)

Cons:
      - Requires welding of frame

      - Untested functionality and more complex parts. Thus time
consuming
to implement.

      - Doesn't solve jamming problem (see mitigation 2 from above).


My Opinion:

Although we played well in Phoenix, I think we could do better with
upgrades to the robot.  The ball dump needs to go.  The concept is cool,
but our implementation is weak.  We are trying to do too many things,
each
with their own mechanism.  Even if they are all working, we have more
points of failure, and we tend to be only mediocre at preforming the
tasks.
We should concentrate our focus on doing something great.  This will get
us
noticed.

We also need to get out of the panic mode of always having to fix
things,
and not have enough time to test our changes out before we go out on the
floor.  This is stress inducing, and makes the experience less
enjoyable.
Every time we came back to the pits, it was a panic to fix something
else.
While the Canadians next to us were out watching matches, going back to
the
hotel, or having fun.

Therefore, I recommend that we proceed with option two.  It gets rid of
unneeded mechanisms, and allows us to concentrate on doing something
well.
And can be accomplished in a reasonable amount of time without welding (
I
hope).

I like the idea of the conveyor belt, and if we had a chance to get
Kareem
back here and work on it, I would recommend that option.  However, we
only
have Thursday to actually built and test it in Las Vegas.  This puts us
back into panic mode trying to get working and tested.


Other things we can do while preparing for Vegas:

Although I like the cart that Mike L. made, it is a real pain to have to
pick it up and put it down.  It is also hard to push without handles.
Can
we make or buy a cart that would carry the robot high so we can work on
it
up, and also have handles to help with steering.  It would also be great
if
the robot wheels were not on the cart, so we can test it without having
to
use blocks.

We should use our experience in what was used in Phoenix, to better help
us
pack for Las Vegas.  Let us take less but bring more.


Please let me know what you think.

Chris Hanson

_______________________________________________
CRUSH-all mailing list
CRUSH-all at crush1011.org
http://crush1011.org/mailman/listinfo/crush-all