Testing the design

I have the logic for a 16-bit Logical Mill worked out and verified. Now, I have calculated that a 16-bit Logical Mill will require approximately 1,200 rods, and 5,500 bumps. The average rod is approximately 13 inches long, thus the total length of rods would be somewhere around 1,300 feet. That is about USD 500, but that doesn't count the work that goes into drilling the holes.

Dan Meana at NextFab suggested that once I have technical drawings of the rods, I can put them up at mfg.com and get quotes from machine shops all over. I would guess a few thousand dollars would cover the rods.

Dan also suggested that for bumps I could use PC board standoffs. The shape of the sense bump is slightly different from the shape of the value bump: the sense bump is longer along the rod axis than the value bump (which helps in reducing false value changes), and the value bump has diagonal sides (which also helps). Hex PC board standoffs are the perfect shape, and are already threaded, and come in both male and female forms. In such large quantities, I can get these for USD 0.05 apiece, for a total of about USD 275.

I plan on making the casing out of acrylic, so that the inner workings can be seen. A few hundred dollars in acrylic should cover the casing.

Then there is the drive mechanism. After some experimentation, I found that I pretty much have to make it out of brass to reduce friction. I will probably end up needing about 300 feet of 1/4" x 1" brass bar, costing about USD 2,500.

When everything is totaled up, and including some pieces I haven't included (parts of the drive mechanism, mainly, since that is the least-developed part of the design), I would estimate the whole project will require USD 10,000.

And so, Kickstarter.

Kickstarter-logo.jpg

The only way I am going to get this thing funded is by crowdsourcing. However, before I submit my proposal, I have to be absolutely sure that everything is going to work the way I think it does. Aside from simulating the logic, the physics of the device must work.

Therefore, I've written up a list of things I need to test before even considering a Kickstarter project. These are:

  • What is the design of the drive?
  • How quickly can the drive be actuated?
  • How reliable is the drive?
  • Is nylon fishing line suitable for inter-module connections?
  • How reliable is the fishing line?
  • What is the length limit of a rod?
  • What is the force limit of a rod?

Once all these questions are answered, I can build a final version of the two-input three-output logic module, which will prove all the concepts. By setting the input and turning a crank manually or driven by a motor, the module should work reliably, quickly, and repeatedly.

Then and only then will I be able to start a Kickstarter project. Also, I have to work out the donor gifts. Maybe an all-acrylic manual version of the two-input three-output logic module at the something-hundred dollar level. Engraved acrylic pieces at lower levels. Maybe CNC engraved metal pieces at the higher levels.