Bringing Digital Fabrication to My Backyard Fence

Growing up in Northern California, a place that teems with natural beauty, I've had an uneasy relationship with technology. I've seen the region around the bay area transformed into a mecca for the computer industry. As a kid, I watched my share of television and spent a lot of my allowance at the local 7-11 on video games. 

My father was a classically trained musician and composer. He often observed there were two lives to every experience - the life of the experience as it happens, and the life it has in your mind after it happened, the recollection. I inferred that TV and video games didn't have quite the same "shelf life" as more artistic endeavors. He usually made these remarks when we were outdoors on a walk.

It must have been hard back then for my Mom and Dad and for other parents too. Mine, like many of their contemporaries in the 70's, were part of a wave divorces. Technology was already reaching deeply into the lives of individual families, whatever their status or configuration. My sense was the divorce rate only increased the role technology played in family life.  As a parent myself now, and divorced, I am struck by the theme of "technology-as-surrogate" that has resurfaced for me during the pandemic.

Between the radically changing landscape of what was to become Silicon Valley, rising divorce rates and the more general movement of the country toward the adoption of an information economy, there was, and is, a lot to think about.

Does technology help our relationship with family and friend, or - more generally - the natural world? Does it help us in other ways? Does technology insulate us from nature and make us care less about natural resources that are so worthy of reverence and conservation? 

Fig Leaf

It has been said that Frank Lloyd Wright used a lot of 30 degrees angles in his work because that was the angle of his triangle. If the contemporary equivalent is the computer, how does that technology affect the work of the modern-day architect.  Without always knowing specifically why, I have seen a lot of buildings that look like they were drawn on a computer, sometimes in a good way, often in an expedient (read: bad) way. I don't blame the technology itself, but I do believe we need to be mindful of the insulating effects accompanying technology's use.

Without getting lost in all this, I bring up these questions as the backstory to my design and construction of a backyard fence. After considering several options for demarcating the property line, it felt like I had spent enough time figuring out what my CNC machine could do in the abstract during this pandemic. Eventually it came time to consider what the fence wanted to be, what it could be. We have a really productive fig tree in our backyard and this seemed as good a motif for the yard as anything.

Of course, replicating nature with technology is a lot of work. I found it considerably more labor-intensive than generating a simple orthogonal pattern of some kind. But it felt like a good moment to remind myself that computers can be used for both economic reasons and for reasons of beauty. This fence was an endeavor to pursue the latter. 

Admittedly, it was not the most cost-effective solution, constructed as it was with Heart B Redwood tongue and groove boards.

As an architect, I have read a lot about digital fabrication but I have seen little effective evidence of its implementation in the world around me. Most often I see it in restaurant signs and patterned guardrails on commercial buildings. 

Exploded Diagram of Fence Boards

Like so many other construction tropes, the manifest solutions are often formulaic and can be implemented on a fairly repetitive basis. It doesn't take long for these constructs to look "tired" as they enter our collective unconscious on a large scale.

The fence required the milling of 54 separate boards that were designed to "tile" on the fence posts to create an engraved mural. It was labor-intensive, and honestly, in the middle of it, I was worried I might wind up wasting too much expensive wood getting my head around the process. 

Thankfully, the CNC mill performed as expected and the process has opened up a new way for me to make beauty out of simple natural materials. I don't know that I will always have the time I spent on this prototype, but the option to do some manner of a natural motif that is tailored to a site has piqued my curiosity and I look forward to the chance to explore it again sometime soon.

Close-up

Clamping the Redwood Boards 

Prelude to a Fence

Ever since I have lived at our present home, there has been a very sad fence on the property line we share with a our neighbor. Like some mysterious archeological dig, this fence had history. Near the street, this fence line held a pretense of normalcy. However, as it progressed from the front yard to the back yard alongside our driveway, a series of unfortunate construction details ensued. 

I don’t know specifically why the fence was cast on a concrete curb…I have theorized it was to keep some previous dog or rodent from digging under it. This would all be well and good if the wood that was cast into this curb, was holding up. It was not. In stark contrast to the unusually substantiality concrete curb, the wood fence above it, was precarious. “Teetering”would be a good descriptor. When it came time to purge the yard of this ailing fence, the wood fence itself would go quietly. The concrete curb was another story.

Intermittently, fence posts would penetrate this curb and descend into their shallow and insubstantial concrete footings. The post’s width was essentially the same as the curb, so everywhere there was a post, the curb would be interrupted. So much for using a concrete curb to maintain a separation of wood from soil! To make matters worse, the demolition of this fence was made so much more challenging by the existence of this curb. Removing a rotten wood fence is one thing, you can push it down. Removing a continuous concrete curb with sporadic shallow concrete post holes, is another matter. A trip to Aaction Rents to get a jackhammer is required. A digging bar is required. Ear plugs are required. A flexible back is a plus also. 

If the new fence wanted to be a strong and healthy one, with even post spacings, the curb and its post holes needed to go. While typical yard work might be pruning bushes and mowing the lawn, this fence replacement felt like cruel and unusual punishment.  As you might imagine, the new fence design did not have the same random post spacing as this old fence. As such, it was really impossible to avoid casting new post hole near, or partly in, old post hole. The new concrete holes often took more concrete than the old ones did. 

We are now half way to the back of the property…

After the teetering-wood-fence-atop-poorly-constructed-curb experience we encounter a redwood tree. This is not your Platonic ideal of a Redwood tree. This redwood tree was a miserable redwood tree specimen. Maybe it is a misnomer to call it a tree? It might be more appropriate to say this redwood tree was a bush left over from some far more noble creature that had existed freely. A time before people came along with their appetites for fences and reduced it to the miserable creature it now was. Imagining it now, I would describe the “event” like passing an accident on the freeway. A fence had clearly had an accident with a tree. It was not clear there would be any survivors.

Moving past the tree accident we now encounter the third act. Here, whatever aspirations the fence builders had to make a straight fence, have been abandoned. More small trees arise. Whoever had worked on the fence had, at this point, clearly abandoned all hope of making it to the back property line without resorting to desperate measures. The fence bobbed and weaved around the trees. Sometimes it appeared the carpenter had needed the tree to hold up the fence. Other times it appeared the fence was being pushed over by the unruly trees. The post holes became shallower as roots complicated excavation. The unreliable tree branches became poor substitutes for footings as the fence leaned against these forms with increasing frequency as the fence terminus approached.

As awful as it was, there is something about this found chaos that excites in me the potential alchemy of a new design. This kind of work is far from the trophy projects I am guilty of wanting. On the other hand, if I was honest with myself, it is this challenge that is far more prevalent and has the reliable distinction of being the major substance of design work needed in the world. Let this be a relief. Look at how low the bar is? Witness the previous design train wreck. This is all that there is to surmount in order to call your subsequent project a success. If you simply do this, you will be a kind of healer. If you can make something that tempts beauty, the whole experience will become a kind of transcendent experience that feels reliably good.

There is no better place to implement a design than in these scarred places. As an architect, I believe our ability to reshape our world can positively impact our existence. Making a shelter that fosters human inhabitants in the unkind wilderness is one of the clearest and conspicuously heroic examples of a designers craft. We have all seen the beautiful mountain retreats, cliff top homes and vineyard estates. A handsome structure ensconced in a nature setting is a compelling and lovable image. There is the potential for (excuse the pun) ground-breaking work in this setting. 

New technologies can be used to harvest resources in creative new ways. It is a wonderful gift to have these kinds of projects. But there is also a escapism here that we can lose ourselves in. Are the cliff top dwelling in a design magazine, the civilized rogue of a male magazine model or the wild beauty of a cover girl, really so different? How much of what we see in these images is of any real substance? If we permit ourselves to look closely, there is often a lot less substance there than we might care to admit. This idea of remaking nature has lost some of its nobility. Yes, it is still a wonderful undertaking when done responsibly and sensitively. But let’s face it: It isn’t always done this way.

Few here in California, can reasonably give that compliment to the vast majority of our constructs. Many, including housing developments, are speculative ventures designed with an industrial ethos geared to generate income for someone who is not necessarily living there. In many ways, the fence I had in my backyard was this sort of thing. It was either built by a renter who had not stake in the outcome of the property, a landlord who was a slum lord or a homeowner who, unfortunately, did not know how to build. 

Given all this, it is hard to escape the sense that good design and construction should be implemented here as much as anywhere; in the places where humans had already built but good design and care had been foresaken. Expediency had ruled the day. Let’s skip a few pedigrees and edify a rescue dog!

In many ways, California is America’s side yard. Even if someone gets a piece of land and it feels untrammeled, chances are, if one looks closer, they will see the traces of things that came before, that have not been entirely unearthed. Robinson Jeffers spoke of a “cabin in the woods under spared trees.” Every form of construction is like this; a form of destruction. 

I could speak about the tired and self-evident need for some kind of sustainable design, the truism that we must, in good conscience, leave something better for our children by more being sensitive to our natural environment. Beyond this goody goody language, I believe there is something more primitive and primordial that coconspires with this more sanctimonious and hand-wringing rhetoric:

We want to generate beauty out of natural resources. That kind of sensitivity is both a coping skill and an homage. This ritual inevitably generates a kind of respect for our environment that does not need to be taught. Sensitivity is incidentally learned on the way toward goals that are of a more immediate, and perhaps more self-serving, nature. It’s okay. It might even be healthy.

We can forget about all this when we are afraid. Afraid that we aren’t going to survive if we spend our time making beautiful things. Afraid that there is not enough time or energy to be spent on such endeavors. All of this is perfectly understandable and equally sad. We also know that living in fear, is no way to live. 

“What you can do or think you can do, begin it. For boldness has magic, power, and genius in it.” -Goethe

I myself am certainly no exception to this ethic of expediency. I have done, and will do, many things to “just get them done” and I am grateful to my partner Lisbet and my children for being patient with me through this fence project. It was clearly about more than just putting up a barrier and I know there were moments when it competed for my time with other more social activities.

The fence I built left a lot of room for improvement, but the shortcomings were more the outcome of pushing myself to make something I had not made before and not the shortcomings associated with a rushed process that I was familiar with. I look forward to sharing that fence design in the next post.

 

A Recent Study Model for Sebastopol Residence

A few progress shots of a recent home design we are working on in Sebastopol. The site model is CNC milled laminated MDF with the architecture done out of bass wood.

Fastener-Free Studio Table

The idea for this table was to craft an attractive and usable piece of furniture that was readily reproducible at scale and would be implemented for initial use in our office. We didn't know how many we would ultimately need for the studio space and the notion of something reproducable on-demand, was compelling.

In the past, I have expediently used utilitarian foldable tables like one sees at bake sales and school registrations. Unfortunately, many of the affordable table designs prevalent today are unable to attend an architectural interior as a legitimate piece of furniture.

I've always admired the modernist design aspiration of creating something both utiliarian and attractive and there was a clear call for something with warmth and presence for our office space.

With this basic premise, we made a series of parametric toolpaths for the studio Shopbot on two sheets of plywood. Because these toolpaths were written parametrically, the design can be "flexed" to accommodate different sizes and shapes as needs dictate in the future. 

More information on the making of this table can be found at: 

https://www.instructables.com/Russian-Birch-Table-Without-Fasteners/

Rhombic Dodecahedron Stool

This latest chapter in a fascination with folding has been challenging. For several years now I've been interested in using CNC technology to create folded plywood objects.  Initially this involved applying piano hinges to pieces of plywood to make furniture.  You can see examples of this here and here. 

Over time, this method of cutting piano hinges and screwing together the joints with the hinge leaves felt complicated.  You had the many faces to cut and then you had the many hinges to cut. For something that was suppose to be automated, I felt like I was working on a methodology that was pretty involved.  The table design worked well for the most part because of the glass top that allowed one to peer down into the bristling interior.  I liked the clean star-like radiating lines but when I made the chair, the object felt too mechanical.  It was felt like a Frankenstein chair for a would-be Frankenstein.  It also wasn't all that comfortable.   

It was pretty clear what the simpler solution was...if possible.  I needed to "weaken"(but not break!) the plywood joints enough to steam bend them.  This would capture the wonderful simplicity associated with origami and create more supple and fluid shapes.  Undoubtedly this would come with a structural and mechanical price.  How far could one bend the steamed joint?  How strong would the joint remain after this bending? I didn't know if the thinness required for steaming would also render it too weak to support my weight. Would it break the first time someone sat on it? I'm happy to say it feels pretty strong after all the faces are connected.

The term "thick origami" is a phrase that has been emerging recently and in many ways the piano hinge furniture represents the clear geometric challenges associated with this way of constructing things. The plates, instead of all pivoting off one side of a cutout, pivot off the convex side of the fold.  This complicates the geometry of the cutout substantially. 

But this most recent furniture example is more challenging than objects made with this purely hinged approach. This stool - and hopefully future builds - does NOT exploit the use of fabric hinges or pivot hinging and their clear points of rotation. 

These "pliable sheet" builds hope to account for the anticipated deformation that occurs at steamed joints and compensates for this in the initial flat geometric layout. I've made several efforts to collect data in this regard. More data collection is in the works as I am working toward a strong 90 degree bend.

It is an exciting time to be working on this. Fabrication methods are evolving quickly in ways that strive to take advantage of both the structural and esthetic characteristics of the CNC approach. There are a lot of techniques to get comfortable with. Some examples of this:

What is the most efficient and accurate way to do a flip mill of a sheet of plywood? 

How does one tighten up the tolerances on the CNC machine when accuracy begins to degrade?

In parallel with these challenges there were, and remain, many questions about how to best modify the plywood to accomplish a good fold. Larger angles of folding remain a challenge and there are tradeoffs in seam width and depth with respect to flexibility and strength. This initial project was chosen, believe it or not, for its simplicity.  While the shape might appear somewhat complicated but it has this going for it:  All the dihedral angles (the angles between the faces) are 60 degrees.  

This made the initial folding ritual less complicated and did not impose too much stress on the joints.  Having said this, my first pass at this stool had several joints crack due to inadequate steaming.  I got discouraged and thought about giving up. I want to thank my son Jesse for telling me he could use one in his bedroom if I was able to finish it. This gave the thing purpose beyond the abstract and vainglorious desire to solve some large abstract design challenge.  There have been big improvements from where I started out, but the whole thing is definitely a work in progress.  The challenge of a good reliable steaming mechanism for the joints continues to be a large focus. Both the CNC and steaming techniques can use improvement.

Stay tuned for future pliable geometry.  

Looking Back at a Greenhouse Built Twenty Years Ago...

Joyce Gross was kind enough to share with me some recent pictures from her wonderful garden in San Leandro.  She inherited a greenhouse I designed and built over twenty years ago and she graciously sent me some recent pictures of the building as it stands today, newly renovated. A thank you to Michael McGee as well for doing the much needed repairs. It felt like an appropriate time to revisit the project.  If you get a chance, check out her blog too.  

It was a turbulent time in my life.  I still hadn't taken the architecture exam and wanted desperately to finally design and build something meaningful.  This little project, so modest in scale, remains large in my mind.

At the time, mass customization hadn't taken off yet. I remember driving over to the Simpson Strong Tie plant a few miles from my house in San Leandro. I had a floppy disk with a DWG file on it for the plant manager. They didn't usually work with other people's files but if I wanted to bring it over, he would try to use it to cut out what I wanted.

San Leandro use to be called Cherry City before it became an industrial center and the Broadmoor area, where I grew up with my Dad, Grandmother and sister, still had a few larger lots with vestiges of this older agrarian time.  Our backyard still had a concrete slab where the barn use to be. The barn was before my grandmother - a school counselor and music teacher  - got ahold of it. She kept many of the old trees and added others.  There were cherry, lemon, orange, persimmon, plum, peach, tangerine, fig and apple trees all in her backyard.  I'm certainly grateful for the "advances" in our environment today, but looking back that situation seemed "abundant."

I haven't seen the neighbor, who hired me to do the greenhouse, in many years. He was an interesting man.  A physicist who worked out in Livermore, he was very open to experimentation.  He use to joke that he didn't understand why architects didn't treat building more experimentally.  "Why don't you build a prototype that could be rebuilt after we see the flaws?" he would ask.

I've always liked that observation of his. It exposed a basic difference between architecture and so many other technological undertakings.  Cars, airplanes, bicycles and other devices have such explicit functions but architecture in many ways is constantly being adaptively reused.  It not only tends to exist in time for longer than these other things, it also can change its function over time. It is not unusual for the hypothesis of the experiment (e.g. "let's build a one bedroom house") to change over the span of this experiment we call construction.  So many buildings get additions or remodels over the course of their lives. Cost aside, this in itself, tends to discourage a sense that a rebuild would improve things substantially. You get one crack.

For whatever reason stars aligned.  The neighbor hired me to build a greenhouse for his orchids. I was coming off a painful divorce and wanted a physical task that would keep my mind and body occupied.  My grandmother had a bunch of old panes of glass from a disassembled greenhouse that use to be standing in her backyard so I designed a greenhouse around adaptively reusing these old panes of glass.  We reused all her old glass in this greenhouse design.

Greenhouses are simple and elegant structures comprised predominantly of structure, "stops" (to hold the glass in place) and the glass itself.  Because the budget was so tight I designed an assembly where both the structure and the glass stop were derived from a single 2x4. Similarly, the CNC metal plates cut by Simpson Strong-Tie allowed 2x2 pieces of wood to span the entire width of the greenhouse.  This created a branch-like clerestory that felt sympathetic with a house for orchids. It still allowed good penetration of  sunlight.

Instead of wiring the greenhouse with electrified window operators we used passive solar "autovents" that utilized a mineral wax piston that expanded without the use of electricity.  During the hottest part of the day, the skylights open on their own. In the evening the skylights close and help keep the plants from freezing during the night.

The whole thing was a function of economy.  So often when this term comes up in design it is  associated with "compromised beauty."  "If only we had a larger budget, the design could be so much nicer."

For me, the design experience of this greenhouse stands as a contrast to that logic.  The reasons are hard to articulate but I keep it in mind when I'm up against tough economic design constraints. Certainly there are flaws in the design and the limited budget did require working with crooked material and thin glass, but when I consider the more conventional work I've done, it seems to me the ingenuity that the budget necessitated generated something other work might have lacked.

A big thank you to Joyce Gross who has made such a wonderful backyard environment of which this greenhouse is but one part and for sharing her shots with me.  Check out her blog to see some beautiful plants a few more shots of the greenhouse.

DIY CNC Cabinetry: Cultivating an alternative to the Home Depot/IKEA model.

The Cabinet Carcasses

I've had a small cabinetry project in the works for a while.  It needed to be done on a shoestring budget.  I made the requisite trip to Home Depot and tried to keep things simple.  Three Melamine cabinets in a "U".  One cabinet for each leg of the "U".  The doors would be simple slab doors.  Nothing fancy. For the uninitiated Melamine is pretty much the same thing as "Formica" (or HPL).  Both products are essentially a resin impregnated paper finish covering a "lovely" particleboard core.  Melamine just uses a thinner paper and therefore presents less of an edge at corners.  It is both cheaper and more esthetically appropriate for cabinet boxes and fronts.   Formica is more logically used on countertops for its wear resistance and that is the probable choice for the counters on this project down the road.

Even at Home Depot, these three stock cabinets were $2500.

Cabinetry can be tedious work so I'm not ungrateful for that pricing. But the Shopbot in my studio was the obvious alternative. What could be done with sweat equity for a fraction of the cost?  It also felt like the hardware quality could be improved on. The Blum Metabox system was used and I ordered these components, as well as some good drawer pulls at CabinetParts.com.   This greatly simplify the assembly process and helped to ensure good operability.  So far the project is looking like it will cost somewhere around $700 in parts.

The milled panels

It is worth noting that there is an MDF (medium density fiberboard) alternative to particleboard for the core of these products.  For this application it felt unneccessary and would have sourcing the material more challenging. I still used my friendly local Home Depot for the raw Melamine sheets.

Thus far the cabinet carcasses (or boxes) are complete.  I used Confirmat screws to assemble the boxes and was relieved to see I could attach the panels to each other by hand quickly and easily.  The CNC toolpath writing is primarily useful for locating door and drawer hardware, which can be unforgiving with a lack of precision.

It is also important to get your cutting speed right since Melamine has a real tendency to chip.  A lot has been written about how to avoid this on line.  I found a compression bit really helped to minimize this although it is my single biggest concern on the project. If the drawers and doors turn out okay, the process will be posted on Instructables in the future.

Pollarding a tree? Make a Wattle Hut!

There are a lot of Mulberry trees on our street.  For years, at some point in midwinter, most of my neighbors get out there and pollard them, leaving the few remaining branches to remain with their denuded cauliflower stumps.  Without being entirely sure why we do this but not wanting to be the pariah of my good neighbors I have gone and done it too.  I think the previous owners of my residence were field workers and they did some interesting braiding with the branches. I've always liked this little flair of non-conformity.  That creative branchwork has been something I wanted to riff on further. While there are many shallow justifications for pollarding online I've been unimpressed by their rigor. The whole thing feels a little like cropping a dogs ears or something. Since I didn't really know why we were doing the cutting to begin with, I wanted to imbue this ritual with more meaning.  Today my boys and I got out there and finally made a small wattle hut.  A coupe notes on the making process.

1. To trim the tree correctly you have to cut the branches relatively close to their base.  

2. To do the wattle correctly you have to recut the base of the branch so the associated curved element segment is eliminated.  This allows you to work with a fairly straight "bow-like" piece.  

3. I thought it might require more than one tree to make this structure and I stock piled three trees worth of branches before starting. Ultimately, I think you can do it with just one.

Anyway, if you want to lighten the load to the dump and your looking for something useful to do with your pollarding you might try this (or something else!).

A Grasshopper Routine for Generating Folded Plate Forms

A routine for generating folded plate structures was recently completed here at the studio using the parametric Grasshopper software.  A few examples are depicted above.  This routine is intended to facilitate the creation of folded plate forms for various architectural uses.

These forms are becoming increasingly easy to build due to mass customization and computer milling, but it is my sense that the forms are still largely avoided in the built environment simply because the design process is so tedious.  After recently completing a few origami-inspired furniture pieces, seeing the St. Loup Chapel and enthusiastically reading a recent paper about origami and folded plates (by Hani Buri and Yves Weinand) I decided to take a crack at simplifying this process.

The forms in this routine are generated by simply sketching two curves on two perpendicular planes with the two curves sharing a common point of beginning.  As curve B sweeps along curve A it reverses orientation at each new line segment. This generates a complex array of alternating origami-type "mountain" and "valley" folds.  The geometry of these forms is complex enough that not all curves will be initially deployable but after a bit of editing, the user will quickly develop an intuition for how the curves can be modified to function properly.

Origami Table

Mixing memories of childhood origami with "folded plate" structural ideas, has been a rich vein of inquiry for me.  So much is happening with prefabrication and panelization right now. It feels like a good time to be riffing on this.

A while back I did a series of chairs on the Shopbot. Following the 3B chair, I wanted to attempt the more ambitious task of a table.  From a structural standpoint, the chair was relatively easy but the challenges of taking a thin "paper-like" material and making a table was more daunting.

What follows are some excerpts from the design and fabrication process.  There are also a few shots of the final outcome.  I wound up designing about nine different tables before making the one we see here.  We even partially built one of the previous designs before giving up on it as too "floppy".  In fact, this new table is made up of facets from the previous design.  This way of working is easy with hinges and something that makes up for the heavy computer time that is really required to pull off this sort of fabrication process.

The reality is that this way of putting things together is relatively new and there is a lifetime of work with this tectonic. I'd like to see how this might inform some other objects for a bit. In short, while there are still refinements, it felt like a good time to post the results of the work and see what comes of it.  A big thank you to my two amazing sons Jesse and Niles for their help. I also wanted to thank Steve, at Arrow Glass, for supplying the tempered glass top.