Design and Process Hand-in-hand

[Jim Becker]

I originally posted this in the Design Forum, but Mark suggested I cross post here for more discussion.
----------
While I was reading the latest issue of Fine Woodworking (April 2005) and the article on page 60 detailing a very nice pedestal table project, it struck me that this was an excellent example about how design not only needs to determine the end result, but it also need to be an integral part of how you get there.

For this example the three-point base is prime fodder for discussion. The author, John Zeitoun, uses a trammel-mounted router to do a lot of the shaping on the components, which are mounted on temporary/disposable substrates to facilitate easy handling as well as to allow them to be processed further with other tools later in the game. He obviously had to plan for this as part of the design process...

"How do I make this using what steps, in what order and with what tools?"
"What joinery will create the effect I want while still being 'accomplish-able' in my shop with the tools I have available?"
"What do I need to do to make this process repeatable for x-times so that I maintain consistency throughout my project?"
"How can I do it safely...?"
And so forth.

What experiences have you had with projects that call these things and more into account? What would you do differently if you made the same project again...not just process, but in the design, itself, so that you can get the same or better results more efficiently?
<!-- / message --><!-- sig -->

[Mark Singer]

Jim,

Here we go again!:cool:
This is a very thought provoking and interesting thread. It is the kind of thought process that allows a woodworker to expand his abilities and the quality of his work and not have the design appear as if it were designed around his shops limitations or his own. It is an all inclusive process that relates design and construction early in the process.....think and plan out the project to yeild a finer result and a more satisfying construction process.

It seems that this type of organization is really required for the construction of chairs. Here components are cut ....left front leg ....right front leg....rt apron...left apron... Rear right leg and so on. The pieces must be precisely the same. The joinery must be precise for appearance and strength. If you are making 10 chairs , you will have 20 front legs. 20 side aprons ...10 front and 10 rear aprons. The precision and repeatability is very important. I typically will use "sandwich" templates and a router to make exact components over and over. First I saw them an eigth over on the bandsaw. I will typically use mdf template jigs for mortises. If the aprons are splayed the jigs are in the form of wedges to create the angle. The tenons must be placed carefully with curved parts such as the back. Finally front leg apron sections and rear leg apron sections are assembled. Finally the skewed side aprons join the fronts to the backs. The design and process are really one. Yet there is a great deal of hand shaping and comparing by eye to try to be consistent where hand operations tend to vary...this takes skill and an eye. Finally the chair is assembled and the excitement to see if it is as comfortable as the prototype....By comparison singular pieces like a table or cabinet seem less critical of a refined process...there is much less to compare and there are many ways to achieve a good result. If a door is off on a cabinet...you make one door....on a series of chairs it is a mess to correct a problem. In the end I don't want the chairs to look as if the design is a result of avoiding a chalenging joint . If the aprons don't splay...the chair appears very rigid...if the back is straight and vertical...it looks and feels uncomfortable. It looks then if the limitations of the crafstman limited the fluency of the design and the final product. When the result is fluid and constitutes a well composed "whole" the skills of the crafstman shine and the result is uncompromised. No short cuts were taken in design to make it a simpler piece to construct....This is a very good thread and the road we all must travel to achieve our best work. There is no seperating the design from the process of building....in the end the design should speak and the process....a bit of a mystery...not so obvious and in a subtle way quietly and confidently supporting the design

[Marc Hills]

Jim:

I saw that article (although I didn't buy the issue). But you're hit on a very vital and interesting point; design needs to figure heavily into not just the finished piece, but the work process required to achieve it.

For me this was harshly brought home by a Arts and Crafts style globe stand I started making this past fall. I got the idea from an old Popular Woodworking link. The centerpiece of the design was a horizontal wooden ring in which the globe itself is suspended, made up of 4 splined sections, which I called quadrants.

The article was a little vague in terms of actual plans and even more so on construction details, but did provide a scale template for the four quadrants. I dutifully enlarged the template to full size, laminated it to cardboard, and traced it onto some oak boards. I then proceeded to rough cut the four pieces with a band saw.

Now the thing is, the splined edges of those four quadrants had to absolutely, positively 45 deg relative to each other, and the quadrants themselves had to represent exactly 90 degress of arc, both on the inside and outside curves. Get any of those three dimensions wrong, and the jointed sections would not produce a perfectly circular ring.

With four rough cut quadrants, I had a terrible time getting everything right. I fashioned a cumbersome carrier jig to hold each piece while I trimmed the splined edges on a table saw, and spend way too much time at the OSS trying to creep right up to the curved lines.

Let's just say the results weren't matching my expectations. One thing led to another, winter set in and that project sits, unfinished, on my assembly table.

In hindsight, the obvious process design decision would have been to completely ignore the quadrant templates, just miter and spline together four sections of straight stock slightly wider than the width of the finished ring, and then cut the inside and outside circles with a trammel mounted router.

I don't have a trammel for my router, but I can assure you I could have fashioned one in far less time than that ridiculous carrier jig I made for trimming the quadrants, and it would have produced much, much better results.

But that quadrant template led me astray. From a design standpoint, it was completely inappropriate for the construction process best suited to arrive at the desired, finished result.

[lou sansone]

It has been my experience that the first time through a project that you have to come up with a bunch of jigs and such to do the job. I very rarely build the same thing twice, the exception is when I have taught woodworking. You look at some of the masters and say "wow, how can he/she do such good work", well they have been through that drill before. Engineers love ww because of the steps involved. It would be interesting to take a poll @ SMC to see what % of folks are engineers.

[Perry Holbrook]

It's not exactly an answer to your question, but I have started making mock ups of new designs out of scrap or plywood as I work out design details and construction methods. The pictures I posted last week of 2 tables are good examples. The copper tile table in particular went thru numerous design changes cheaply with plywood until the design was finalized.

Perry

[Jim Becker]

Perry, I'm glad you bring up mock-ups/prototypes as I'm finding them to be essential, if not critical for some types of projects. And they really help work out not just the design, but also the methodology one will use. (It's a great way to use up scraps, too... ;) )

[Mike Cutler]

Jim. The Library shelf that I built last summer, and the wall unit I am currently building, required full sized MDF templates for all the router work. These were essential, and the project could not be built without them to precisely align and index all the parts together. It's safe to say that the templates and drawing took alot more time to construct than the actual piece(s) they were the template for.
Each step has had to planned in advance, there are over 500 M&T joints in the project(s) and they are all dependent on the others to fit properly, or the whole project is just firewood next fall.
I've done some crazy things with the Delta 14-651, and I wish I could say that all the router work was safe enough for a video, but it wasn't. Lot's of climb cutting, and progressive plunging on edge grain while climb cutting.
If I had to do it all again, I'd have to say that I would have chosen another wood. Brazillian Cherry( Jatoba) is a bear to work with, and it is HEAVY :eek:. The sizes and weights of the material itself have identified some of the outer limits of my tools. No way I'd do it again with a 6" Jointer, and a 14" Bandsaw.
I wouldn't change the design for processing efficiency, I'd rather take more time and not have everything look sterile and geometrically formulaic.