Thursday, June 10, 2010

Kalamazoo progress

Yesterday I started forging the mounting braces for the Kalamazoo pump arm. It's great to have the forge running again, but it's going to take time to build up my calluses again. I only got 2 of the 12 ends flattened out nicely, but I did get all of the decorative twists put in at least.

With the pump system under way, I'm working on locking down the design of the drive train itself. Most of the parts (pillow blocks, sprockets and a rod end bearing) are still on their way, but I picked up the differential last month. Made for a lawnmower, it should be perfect for my use here.

Unlike a normal car differential, it is driven inline with the output shafts via the large sprocket. Which really simplifies my design, even if I'm sad not to need a Watt style parallel motion.

The output shafts lack keyways, sadly, and are far too short for my purposes anyway. Currently I'm planning on getting keyways cut and extending them using shaft couplers. I might do that only on one side, but I'm not sure if I can stand that level of asymmetry in the design. I'll need keyways cut for mounting the sprockets anyway -- we're dealing with far too much torque here to get away with set screws!

I was planning on making a nice detailed post here with graphics and everything, describing how differentials work. They're both pleasingly elegant and surprisingly simple. But every time I started outlining that post in my head, I realized I was just paraphrasing a really great video I found once. So instead, I'm just going to post that video. It's from the 30s, but the technology really has't changed much.


  1. Nice video! It reminds me of that video explaining the computing mechanism core of the continuous-update targeting mechanical computer for battleship guns.

  2. Gee, now I feel obligated to build one out of Lego.

  3. woot. Nice video. With motorcycles, even! :) Looking forward to seeing this thing work....