I've decided to take this project from a loose idea to reality. I have general plans for the major components that I will need and some of the details that will make those parts work. My notes and drawings are ugly and scattered, but I'll at least list and loosely describe the major parts and how I envision their implementation.

• Plywood Sides
  • Two pretty simple and nearly identical panels to hold everything together. The goal is to build it in a way that utilizes a 4' x 4' sheet of plywood perfectly, but we'll see what happens through the design process.
• Handle, Arm, and Print Cylinder Axle
  • This part is still under consideration, but the way I envision everything currently the print cylinder is affixed to this rod during installation, and driven directly by turning the permanently attached handle. It might be nice to have a simple pawl system to prevent rotation in the wrong direction. To uninstall the print cylinder (when changing stencils or ink pads) it would be necessary to fully remove the axle and handle assembly from the system.
• Attachment/Engagement Disks (Print Cylinder)
  
  • I'll describe the print cylinder more fully in a coming section, but put simply it's a one foot section of 6" PVC pipe capped on both ends with these printed disks. They will have simple set screws to fix them to the axle during installation. They will likely be modified in future iterations in order to add features like a chain drive from a motor and syncing with a paper feed system.
• Stencil and Ink Pad
  
  • For the stencil I'd like to take a page from silk screen and use the same materials common in that process, silk coated in a photo emulsion that can be exposed under UV with the positive parts of the design being washed out after exposure. However rather forcing a thick ink through the fabric with a squeegee, I intend on using a piece non-woven material like felt soaked in a thin alcohol based ink (maybe?) under the stencil and printed by pressure from the impression cylinder. This will undoubtedly take some experimentation, but it also offers some interesting options such as variable and custom inks for different looks and opacities, multi-colored ink pads, and printing of things that aren't ink (glues or ???).
• Screen and Ink Pad Holders/Print Cylinder
  • As noted in a previous section, the print cylinder is based around a one foot long section of 6" PVC pipe. As it turns out 6" is nominal and it's actually 6.625" giving us total possible print length of something like 20.8" by the width of our pipe. However, we need a certain amount of space around the edges of the print area to support and hold our stencil and ink pad. I'm going to try some experiments to see if I can hold the screen in place reliably with screen door/window spline cord. Ideally it will hold the screen tight enough that it can also hold the ink soaked pad as well without any additional parts.
  • These holders will be printed cylindrical parts with grooves for the spline cord and support for the ink soaked pad. They will be glued to PVC pipe. They will have to be printed in sections, but that shouldn't be an issue. I may also make some jig type parts in order to ensure uniform installation of the glued pieces. Additionally the printed parts will have some form of registration marks printed into them to help ensure uniform installation of stencils when that is critical.
• Vertical Registration Assembly
  
  • This assembly will consist of a small spring loaded wheel pressed up against a wheel attached to the print cylinder's axle. The wheel attached to the axle will have a detent of sorts so that the print cylinder can be stopped in the same position consistently. It will also have a set screw holding in place on the axle so that it can be adjusted as necessary. In this way, it's possible to make the machine feed paper against the print cylinder in a fairly predictable way and hopefully offer a fairly tight vertical registration of the printed image from page to page.
• Attachment/Engagement Disks/Impression Cylinder and Axle
  • Same deal as on the print cylinder, but less is involved. I see the impression cylinder as being passive and only moving when engaged by the friction from the print cylinder or paper passing between them. The impression cylinder will be made from 4" PVC pipe initially, but I may experiment to see how it's size affects print quality.
• Impression Cylinder Setter
  • This is probably the most complicated part of the whole assembly. I have a few drawings I'll upload at some point to help illustrate the whole thing. Both sides of the impression cylinder will have a mirror copy of this assembly so that each side can be adjusted independently. Each will consist of worm gear assembly to adjust the gap between print cylinder and the impression cylinder. Additionally the assembly will be spring loaded against the direction of the print cylinder so uneven mediums can pass through the gap without damaging the system and it is it less likely to get jammed stuck. It may also help with the un/installation of the impression cylinder if that becomes important.
• Print Cylinder Holder
  • Something that holds the print cylinder almost in place during installation of the axle may be very handy. It may also get scrapped if I come up with a better way to change the print cylinder.
• Paper Feed and Adjustment Tray
  
  • For now the machine will be hand fed one sheet at a time. The feed tray will feature some sliding guides to help align each sheet as it's fed.
• Paper Scraper Fingers
  • These are some thin, springy fingers that gently scrape the print cylinder to remove any printed sheets that stick.
• Print Catcher
  • A simple try to catch the printed sheets.

A few (of the many) potential pit falls, and some general notes:

• finding the right "wetness" for the ink pad, too wet and it's a mess, to dry and you'll only get a few faint prints
• the strength and reliability of printed parts without a significant amount of post-processing, especially on threaded things. Heat set inserts seems great for a lot of cases, but I wonder about long term usage on things like set screws
• Schedule 40 PVC has a tolerance of 0.1" which could be very problematic. It seems unlikely that it would be that bad, but it could. Also small variations could be problematic too. Hopefully the spring in the impression setter will help to alleviate some of that, but it's hard to say until I start playing with it
• screen printing silk, what's the best way to finish the edges to prevent fraying? Will spline cord actually hold it? How to I hold the material while coating it with photo emulsion and during exposure and washout?
• cutting and drilling the plywood sides accurately, jigs will help a lot, but it still may be problematic
• making a large, time consuming, costly, messy "printer" that doesn't work well enough to make me care about making better versions
• I'd like to try and use the same hardware elements on many of the parts so that anyone who wants to make one can just order a dozen 3/8" nylon bushings instead of 2 of this and 4 of that
• It's sometimes hard to decide when to use an off the shelf solution and when to make your own. As I travel further down this road, I suspect I will find things I hadn't thought about. Hopefully I find them while they can still be incorporated into the design.