254 Presents: Cheesy Poof Build Blogs

[Cory]

If you read the 2010 blog in it's entirety you'll notice that on day 2 of build we were testing a 8wd drivebase on the bumps.

This was something we were very proud of and thought was very cool. With about 6-8 hours of work we were able to manufacture frame rails and bolt together a frame to accommodate 8 wheels. We were then able to raid our spare parts bins for bearing housings, wheel shafts, sprockets, tensioner cams, gearboxes from the 08 robot, and 4" wheels.

We spent a lot of time prior to that diagramming out the interference between the bump and frame with various wheel sizes and configurations. We were very concerned about making a decision without empirical evidence that it would work, so being able to do this was invaluable.

At a few points we have considered moving away from a 7/16" hex on the wheel side of the shafts and a 1/2" hex on the sprocket side in favor of using hex bearings and a straight hex shaft all the way through. Every time we've shot it down because it makes everything we've made since 2006 incompatible with any future parts.

Having that supply of parts that has remained functionally (if not perfectly) identical and interchangeable since 2006 allowed us the flexibility to throw together an entire drivebase in under a day and test something critical to the design of our robot. This offers a huge advantage if you have the foresight to keep families of parts common.

[JesseK]

Originally Posted by Cory

At a few points we have considered moving away from a 7/16" hex on the wheel side of the shafts and a 1/2" hex on the sprocket side in favor of using hex bearings and a straight hex shaft all the way through. Every time we've shot it down because it makes everything we've made since 2006 incompatible with any future parts.

Having that supply of parts that has remained functionally (if not perfectly) identical and interchangeable since 2006 allowed us the flexibility to throw together an entire drivebase in under a day and test something critical to the design of our robot. This offers a huge advantage if you have the foresight to keep families of parts common.

This is an incredibly huge insight for anyone that it's new to. This is typically how industry moves. Why did XYZ make that inane decision in their latest iteration of product instead of using a blanket 'best practice'? In any complex integrated system (robot, submarine, satellite, cars) it's usually for backwards-compatibility.