Given that Goudy's method was relatively well documented, surprisingly little information remains concerning his actual procedures (what ATF and the Dale Guild would call, quite accurately, "protocols") for actual matrix engraving. For the most part these must be reverse-engineered from a few surviving photographs.
Neither Bruckner's biography nor Lewis' Behind the Type (1941) include images of matrix engraving Bruckner's view of him at the E&PM Co. pantograph shows him using it for working pattern cutting. The view in Ars Typographica (1934) is too soft-focus to be of use.
This is a big missing section in our understanding of Goudy's method. Goudy's working pattern had alignment information scribed on it indicating the top and sides of the body of the type. The matrix to be cut has a predefined, fixed, "head bearing" and "side bearing" which define the distance from the top and left side of the matrix to the top and left side of the type as it will be cast. Matrix headbearing and sidebearing distances are defined for each style of matrix as a part of the specification of its specification. 
The working pattern and the matrix blank (planchet) must both be placed in the matrix-engraving pantograph in very precise relative orientations so that the headbearing and sidebearing of the resulting cut matrix are correct. The methods Goudy used to do this are not discussed in any of the published sources.
In the day to day practice of metrology as developed in the early 19th century (by Whitworth and others) a distinction was made between "line standards" and "end standards." In using a line standard, you have to match up the position of a line - a process always subject to interpretation (and basic errors, such as parallax in observation). In using an end standard, you simply measure against the physical end of an object. Errors may still be made, but this eliminates a great many errors of observation. So, for example, if you want to make a gauge which will measure one inch, you could scribe two fine lines an inch apart on a longer bar (and thus make a line standard) or you could manufacture a block which is exactly one inch long.
The concept of an end standard also fits well with the idea of "banking": where you locate a workpiece by pushing it up against a fixed stop. The workpiece is said to "bank against" the stop. This is much more repeatable (and faster) than locating a workpiece by trying to position a line scribed on it.
Goudy would seem to have used a mixture of both methods, although the quality of the published photographs is insufficient to make them out with complete certainty. Four items are of interest: positioning of the working pattern and positioning of the matrix blank, both in his work before and after the 1939 fire.
His positioning of the working pattern pre-fire is sown best in Kellerman's film. Here the top of the working pattern banks on a fixed bar at the top of the pattern-table of the pantograph. It is not possible to ascertain how the left-right alignment is made. The view in Typologia confirms this. The view in Advertising and Selling (1939) is not sufficiently distinct to be useful here.
In his pre-fire practice, the matrix blank is held in a small vise fixed to the workholding table of the pantograph. It is shown in both the Kellerman film and in Advertising and Selling (1939). Presumably this vise has been positioned meaningfully relative to the pattern-table. This vise holds the matrix blank in place with a screw on one side; the side of the matrix blank banks on the fixed jaw of the vise. It is interesting that the vise seems to have been reversed left-to-right between the time of Kellerman's film (1933) and the 1939 Advertising and Selling article.  It isn't entirely clear whether or not there is a banking surface for the top of the matrix blank; I cannot discern any. If one is not present, then this presents an interesting puzzle: Goudy seems to be banking the top of his working pattern while aligning its side in an analog fashion, but doing the reverse (banking the side) with the matrix blank.
The working pattern is shown banking against a fixed bar at the back of the pattern-table, as before. The left-right positioning is shown a bit more clearly, though. This pantograph has a centerline running from the operator to the column, and this centerline is drilled with small holes. There is a pin in one of these holes, just below the working pattern. This pin (and the centerline) line up exactly with the scribed vertical line in the working pattern which indicates the left side of the body of the letter 'E' being engraved. So it is clear that at least in post-fire practices Goudy was using a mixture of banking (for the top) and alignment with scribed lines (for the left-right alignment).
In the view of the matrix being engraved, the bottom of the matrix banks against a bar on the back of the worktable (remember, the matrix engraving it is rotated by 180 degrees). This is straightforward.
But this same picture shows an apparent puzzle: it looks, at first, as if the matrix is sideways on the worktable - oriented with its long axis parallel to the long axis of the worktable. This is of course impossible - the matrix engraving process rotates the image 180 degrees, not 90 degrees.
I think that the answer can be seen in a nearly invisible "break" in the front edge of this apparent matrix. What I think is shown is actually not one but two matrix blanks - you can just barely discern where they meet. The one on the left, I think, is fixed "permanently" in the machine and constitutes a fixed banking surface for the side of the matrix blank being engraved. Note that it is held down with a clamp which isn't quite parallel with the table surface, and which is, indeed, slightly bent from excess pressure.  The actual matrix blank (again, I think) is the one on the right side. It is held down by a single clamp which is very carefully adjusted so as to be parallel with the table. Holding a matrix blank with a single clamp is perhaps not as secure as possible, but if carefully done with a workpiece which will receive little force and which is banked against two surfaces it should work.
The working pattern used by Goudy was right-reading intaglio. The Matrix it cut was also right-reading intaglio (as all matrices must be), but on the machine in cutting position is rotated by 180 degrees from the pattern's orientation. As would be expected, these matrices show no lining or fitting information; that is encoded in the definition of the matrix format. Goudy's matrices as shown in the Kellermann film, around 9:40, have no identification or set information on them (but of course this might have been added later).
The distance between the top and bottom lines on the working pattern (which represents the body size) was 2 1/2 inches. The body size intended for the engraved matrix was arbitrary; the pantographic reduction was set so as to achieve it from a fixed pattern size.
1. While in many kinds of typecasting machines the operator can change the effective head and side bearing, this increases the time (and thus cost) of the casting and decreases the chance that types from different casting runs will line and fit with each other.
2. It is unlikely that the photograph in Advertising and Selling was flipped accidentally, because it shows the two E&PM Co. pantographs in the same relationship to the Deckel pantograph as in the photograph in Bruckner. I presume that it is unlikely that the film footage in Kellerman was flipped.
3. These are conventional machinist's strap clamps. They're held down with a T-bolt in the center (the T part is down in the T-slot of the worktable) and their outer sides are positioned vertically by adjusting screws. The adjusting screw for the left clamp is oversize and holding its end of the clamp a bit too high.
All portions of this document not noted otherwise are Copyright © 2013 by David M. MacMillan and Rollande Krandall.
Circuitous Root is a Registered Trademark of David M. MacMillan and Rollande Krandall.
This work is licensed under the Creative Commons "Attribution - ShareAlike" license. See http://creativecommons.org/licenses/by-sa/3.0/ for its terms.
Presented originally by Circuitous Root®
Select Resolution: 0 [other resolutions temporarily disabled due to lack of disk space]