[CRUSH-build] build team work before San Diego

[Bill Bennett]

Hi Builders

We took advantage of some of the drive time back from LV to discuss priorities for fixing stuff on Kareem before competiting in San Diego. Of course only part of the team was in my vehicle, and some didn't come to LV, so please feel free to add your thoughts to the summary below.

We thought these were the top issues, in order of importance:

1 - Drive Train Reliability
Drive train problems were a constant headache. Bearings fell apart, chains came off, wheel alignment was constantly out of whack. We need to upgrade this so that the drive train doesn't need to have difficult and time consuming maintenance after every match. 

The IFI wheels were not as good as they looked on paper. Although we got better at it, tread maintenance was tedious and time consuming. And with the 1" wide wheels we used and only 3/8" shaft size, the bearing and axles simply couldn't take the abuse of competition driving and collision. Steps we can take in the right direction would be:
- bigger diameter axles and bearings
- shorter axles
- direct drive some or all of the wheels. With a six wheel design, direct drive the center wheels. With a four wheel design, we could either direct drive the wheels on one and and link front to back with chain, or independently direct drive all four wheels.
- stiffer axle material
- stiffer wheel design

Frame racking may have been part of the issue. When we get Kareem back, we'll study the frame to see what bent and what held up.

The proposal we tentatively stated for San DIego is (following observations about other bot designs that work well) to make single piece aluminum wheels (a fairly simple lathe operation) with a short large diameter hardened steel axle. Keep the six wheel design, change to 4 or 5" diameter wheels (to get the speed right), direct-drive the center wheels, chain drive the front and back wheels, design better chain tensioners. The center wheels will mate directly to the transmission output shaft. The corner wheels will have large bearing cups and mount on large (maybe 5/8 or 3/4) steel axles.

2 - Shooter Accuracy
The shooter worked pretty well but would hit the target more reliably if we shot closer. We should adjust the shooting angle to accomodate this. Also we should adjust our optimum range for shooting from the bottom of the ramp. This allows using the ramp as a simple way for the driver to know the correct shooting position, having the ramp to block defenders on that side, and using the ramp as a partial stabilizer against defenders pushing on us. We should also look into getting the shooter wheel balanced so it doesn't vibrate so badly. The vibrations are probably what killed the first shooter motor, and we don't want it to happen again.

3 - Autonomous scoring 
We got close on this. We should finish it and get some points in autnomous mode. 

4 - Collector yield
The collector worked fairly well, something like 60%, but still failed on too many balls. We should take the time to finish tuning it to get collection percentage high. 

What does anyboyd want to add to this?

See you tomorrow.

Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://wizardofaz.whsites.net/pipermail/crush-build/attachments/20060403/f60c2e8b/attachment.htm