Good product development needs to follow some fundamental principles, and a critical, impartial, design review process (CDR).
Some years back, I was employed by a subcontractor to Hewlett-Packard design and construct a prototype in-office paperback book-manufacturing machine. The intent of this device is to provide a means of assembling relatively small quantities paperback books on demand without being burdened sending this task out to a book manufacturing company. This seems like a sensible thing to do, as it would drastically short the time required to print a book and have it available for distribution.
One problem was how to create perpendicular edges on each booklet section. This problem can easily be demonstrated by taking a stack of, say, five sheets of printer paper, folding it over double and looking at the edge opposite the fold -it will have a V shape. The two pages making up the center of the fold will stick out slightly more than each succeeding page. The proposed solution to this was to trim each booklet page a slightly different amount before attaching them together, by either stapling or gluing.
A large number of motors and a large number of steps would be required to trim each individual page a different amount in the many booklet sections and require not only large amounts of power but also large amounts of computer processor power to keep track of all this. A decision was made early on to incorporate an existing processor, and this processor was not up to the task of controlling the multitude of steps required to fold, staple or glue, and eventually trim each page in a book. These individual sections then needed to be glued to a light cardboard spine and front and back covers laminated over this.
The various mechanisms required were entirely workable, but some basic simplification was needed. Trimming each individual page was unnecessary. This is not done in mass-produced paperback bookbinding, but rather following gluing the spine and soft cover in place, the complete book is trimmed. Why not use the same method in an office-based book-manufacturing machine? Replace a single trim operation for a trim operation that requires each leaf to be cut a different amount, depending on its position within the individual booklet sections. Using the single-page trim method the final product will not have a smooth appearance at the edges due to the tolerances required when working with paper and making many very fine cuts.
One other fundamental requirement was that the machine needed to work with 120 VAC power and draw no more than 20 amps, and no other appliances on the circuit. This would require a circuit dedicated directly to the machine, with no other outlets drawing power from this circuit. Higher amperage and/or higher voltage would require custom wiring which would greatly limit its commercial appeal.
By simplifying the trim operation, the need for higher than available electric power would be avoided and using the inexpensive, available microprocessor to control everything would be acceptable.
This design flew in the face of a basic principle that I hold dear-keep your design as simple as possible (KIS) and add complication only when absolutely required. So, I was very curious as to how the design came to be what it was. I came to discover that the design was being forced upon the subcontractor because of Hewlett-Packard’s system of rewarding original ideas. If someone’s idea made it through to production, a significant cash bonus would be awarded. The manager in charge of this project was the originator of the design concept.
At the time, Hewlett-Packard was going through considerable internal management turmoil, and the project was eventually scrapped. Not surprisingly, another company succeeded in constructing a similar machine a few years later.
Norman T. Neher, P.E.
Analytical Engineering Services, Inc.
Elko New Market, MN
www.aesmn.org