If we add a tympan and frisket assembly to a Poco, we can achieve precise registration, enabling multi-color and double-sided prints, rather like an iron handpress. Indeed, Neil Giroux was apparently inspired by his experience on a handpress to adapt that approach to his Poco, http://vandercookpress.info/vanderblog/2007/10/01/poco-0-with-tympan-and-frisket/
Below are larger versions of the pictures from his discussion (click them to see even larger versions).
  Note that we're talking about the framework on the left of the picture above (as distinct from the tympan wrapped around the cylinder).
Inspired by Neil's approach, Dave Griffin built his own copy, http://www.windsweptpress.com/pocotym.htm
I wrote, asking for a few more details (how thick are the frame elements?) and he sent along this picture
 with some discussion
I aim to do things a bit differently. Rather than attach the tympan-frisket assembly to a chase (or the bottom frame in Griffin's scheme) and then lock the chase into the bed, I'm planning to attach the assembly to a single piece of furniture and use that piece as part of my lockup at the end of the bed, against the 2 taper pins. This approach avoids building the bottom frame entirely and allows more area for the layout.
It may not work out. Furniture is 5/8" thick, leaving just over 1/4" clearance to the cylinder. It's not clear that I can build adequate frameworks from such thin stock. I'll try it initially in oak or carbon fiber (things I can do myself). If those won't work, I may have to talk to a machinist.
To hold the paper off the type when the tympan is lowered, I'll take a tip from Rummonds' book and line the windows in the frisket with weatherstripping. It's a sticky backed, foam-rubber material that's thick enough to support the frisket-paper-tympan combination above the furniture (or a Boxcar base), but compress when run under the cyclinder. I think it'll work perfectly.
I experimented a bit this weekend with some white oak. 1/4" thick by 1/2" wide seemed inadequate. 3/4" wide seems better (later I realized that the frisket frame need not be as rigid as the tympan frame; perhaps 1/4 by 1/2 would have been perfect).
Another interesting problem is how to hinge all the parts together. 1/4" stock doesn't leave room for standard hinges. I'm going to try building a few sets of knife hinges (Brusso makes beautiful knife hinges, but they're a little large). They aren't that hard (I think) and will make an excuse to use the lathe.
Ah, here are some small ones we can buy. I'll try to finish mine, but it's good to have a fall-back position.
Note that it's important that the hinges attaching the tympan frame to the furniture (or chase, if you build it that way) should be free of slop. Otherwise, registration will be lost.
 Here's the assembly locked in place around a 9" x 12" Boxcar base. In the picture below, you can see one of the quoins; the other goes under the frisket frame.
Which is a problem... The hi-speed quoins are taller than the furniture (11/16" versus 5/8") and the frisket frame won't lay flat with the quoin in place. Ouch, what a mistake! I'm not sure about the best solution. I'm going to explore using Wickersham quoins. From the pictures, it appears that they are no taller than the furniture. Ah, experiments show that they work great.
Of course, this won't be a problem when printing with lead, except perhaps for the largest forms. On a related note, I bought some iron furniture and learned that it's not all created equal. The stuff from Challenge is 5/8" high, matching my wood furniture. The stuff from Morgan and Willcox (sometimes stamped M & W) is a bit taller and makes a clearance problem.
 The picture below shows the assembly locked in place, but unfolded. Of course, I still need to cover both the tympan and frisket.
  A couple of closer shots showing more detail. Hmmm, in the bottom picture, imperfections in my joinery are showing up. I'm afraid they're even more visible in person. Luckily, the tympan covering will obscure the evidence.
 To cover the frisket, I'm using tympan paper (I have a giant roll) held in place with masking tape. It's not especially taut, but I don't think it needs to be. On the other hand, I think it'll prove useful to be able to replace it easily.
I'm going to have to experiment a bit to find an adequate tympan covering. In this case, we certainly want it to be taut and stable. Rummonds recommends Linson paper, but it's not available in the US. On his website, he mentions Permalin, but they've been bought out and their products discontinued. Following the lead of my model-airplane friends, I'm going to try silk. They use it to cover their wings, putting it on wet and painting it with "dope" to shrink it in place. I ordered a couple of yards from Thai Silks, the 15.5 mm Habotai (the heaviest they have). We'll see how it works. As an alternative, I've also ordered some SIG Koverall, a heat-shink polyester fabric that's supposed to be tougher than silk.
 I've decided to try the Koverall first, mostly because I ordered the wrong sort of dope for shrinking the silk (we want butyrate rather than nitrate). But the Koverall went on easily and shrank beautifully taut.
 I covered both sides of the tympan with a layer of mylar (0.01" thick), held in place with transparent tape. On the top side (above), I added a second piece of mylar, hinged along one edge with bookbinder's tape, along with a piece of drafting vellum between them. I'll add any required additional packing here and paste makeready pieces to the drafting vellum.  In the picture below, I'm testing the impression on both ends of the cylinder. You can see the windows I've cut in the frisket along with a scrap of paper taped to the draw sheet. For real printing, we'd use some sort of guides or points to hold the paper in place.
 In the next picture, the frisket is closed. You can see the windows along with two strips of 3/8" foam weatherstripping that I'm using to hold the paper well above the inked type (what Rummonds refers to as frisket bearers). When we run it through the press, the foam compresses and the paper is pressed into the type.
 In use, I expect to have most of my packing (sheets of mylar) on the cylinder, held in place by a single piece of tympan paper. On the tympan assembly, I'll use a single piece of mylar on the front (toward the type) of the tympan along with two pieces on the back. The single piece on the front will give a fine, hard packing. To achieve a softer packing, I'd add another sheet or two of paper on the front of the tympan, betweem the mylar and the paper.
For those put off by all the detail above, here's a description (from John Setak) of a very simple scheme that should work well for many purposes
|