The New Garden
The main character in this blog post is actually a fence, but please read on.
I grew up in the countryside of rural Michigan on the family farm - never fully “immersed”. The crops raised were wheat, corn, navy beans and sugar beets but the job of raising them had been passed on to neighbors leasing our fields before I became old enough to “farm”. However, the family vegetable garden was always there. Over the years it migrated about the yard, expanding to fit the needs of a growing family of six, and providing all of the usual table vegetables with plenty of strawberries, raspberries, green beans, pickles, peas, corn and tomatoes left over for my mother to preserve. That garden grew to its 2-acre crescendo when I and my three siblings came of age. Selling tomatoes, onions, dill and pie pumpkins wholesale to the village farm stand became our summer job during the high school years. Looking back, “doing the garden” was not the drudgery that it seemed at the time and it kept us using our muscles out-of-doors. It also taught us to appreciate gardens, what a good job looked like and, basically, how to “work”. The value of those lessons far exceeded the spending money we earned. But this post is not about that 1970’s garden, nor the two small back yard gardens that were nurtured by my wife and I during our stint at family raising. This one is about the new garden that we are starting at our Harvard home.
The suburban gardening mentioned above was truly a mixed bag experience for us: a few precious vegetables and a lot of varmint control. The foe at our Westfield, New Jersey plot was the gray squirrel. While a chicken wire fence served to keep the rabbits at bay nothing could thwart the determined squirrels from getting to our almost ripe tomatoes - frustrating! Sometimes the squirrels would “dine in” on the vine, but often they preferred “take out”, leaving the half-eaten tomato core for us to find on our picnic table when we returned home from work. I tried trapping the creatures and releasing them in a park 5 miles away (beyond a divided highway) but this “resettlement”, while satisfying, eventually became a Sisyphean chore. In the end we resorted to raising plants that squirrels would ignore: lettuce, pickles and herbs. Our Lexington, MA home bordered a woods and the two raised bed gardens were immediately beset by squirrels, and also chipmunks, rabbits, groundhogs and turkey. Yes, turkey love to peck at pink tomatoes. Fenceless, it eventually flourished as an herb garden. We now share our Harvard yard with all of those critters plus raccoon, deer and bear, but I am determined to go “once more unto the breach”, and this time win!
My foolhardy(?) optimism is based on all of the fences we observe while walking the roads in our neighborhood. Everybody has a garden of some sort and they are all fenced. I figure the fences must be serving some valuable purpose. During the summer of 2020, after completion of the Red Top Workshop, we created our garden comprised of four (4ft x 8ft) raised beds that were terraced to accommodate a sloping backyard. It was a lot of work to pick axe these plots out of the rocky New England “soil” (thanks! again for the help Ben, Andrew, Greta), and we think they turned out nice. The final act, as I saw it, was to surround them with a fence that we would not grow tired of looking at and trust that said fence would also be the barrier needed to grow vegetables in “varmintville”.
Raised bed garden before fencing
Design
Since the garden perimeter is small (12 ft x 16ft) I wanted to keep the overall height proportionately modest. I also wanted to avoid two features found in many backyard garden fences: metal posts and sagging wire. After mulling these requirements all winter, researching in books and comparing the various fence philosophies found online, I arrived at a design. I would use 4x4 pressure-treated (PT) pine posts mounted to the garden frame at the natural vertices and these would support rectangular fence “frames”. The frame members would be fabricated in the workshop from dadoed 2x2 PT pine and they would enclose a sturdy wire mesh made from livestock feedlot fencing, specifically something called “hog panels”. The Pros of this design include: no concrete post footings; no sagging wire; short stature; while the Cons could be: fence openings not narrow enough to prevent invasion; short, 3 foot height could still allow deer to browse overtop. To mitigate the potential deer issue I would plan to put the large tomato plants at the interior of the garden. If the holes of the feed panels proved to be too large I could always back them up with a finer wire mesh and still avoid sag. Of course, squirrels and other climbers can still maraud at will, but that should be the case in all of the neighborhood plots … yet they persist. There are certainly other gotchas awaiting but at least I had a plan supported by a strategy. A simple drawing on graph paper helped me to be sure the proportions and fastening scheme of that plan would work.
The planning to get actual dimensions and dado specs happened in the workshop sketch book while pacing about the garden timbers, tape measure in hand. I show a scan of this book to reveal how many projects are, in fact, made; from crude sketches quickly jotted down as new ideas or solutions are conceived. It’s ugly and sometimes involves the back of an envelope, but it generally works. From there I was able to plan for the lumber and hardware purchases.
Materials
I secured the 8ft long 4x4’s and some 8 and 12ft long 2x4s in time to clamp & dry them for a few days before starting the project. I also got 4 hog panels from a local farm supply store. These unwieldy things are 16 feet long and weigh 32 lbs apiece! Farmers may be the fittest humans on the planet not drawing a 7-figure sport salary, bless ‘em! Add a couple dozen 6 in. lag screws, some 2 in. deck screws, a pair of hinges and a latch, all galvanized, and it turned out to be a $350 investment in materials. Not bad, considering the excitement that is brought about by accumulating a pile of raw material that you know will soon be turned into “something”. Sort of like collecting 64 college basketball teams in early March and knowing that in a few short weeks a champion will somehow emerge. “Hey honey, we should really book that $350 against our entertainment budget this year. What do you think?” (Silence)
Dimensioning
This fence has only three wooden components: posts, frame members and a door. These all serve to support the 4 gauge wire panel fencing. In that sense it is a simple project. It’s really a “production” project given the repetitive nature of turning out 8 posts and over 150 linear feet of fence frame. Making the posts was an easy operation. To provide a 2 in. “knob” above the fence height, each post would need to extend 39 in. above the garden frame. This would be accomplished by cross-cutting the 96 in. long 4x4s in half , yielding a remainder 9 in. portion (96/2 - 39 = 9) at the bottom for fastening to the sides of the garden frame timbers. The plan was to avoid pouring concrete footings and rely on the stability of the timber-framed garden bed to support the posts. To provide two perpendicular surfaces of contact with the timbers the posts were notched to create a 2 in. x 9 in. rabbet on the bandsaw. The 2 in. portion of the post that laps over the top of the garden timbers would also provide an attachment surface for the fence frame. After all the posts were prepared they were clamped snugly together so they could continue drying without “losing their figure”.
The dimensions of the frame members were likewise chosen to be easily obtained by halving a raw material, this time by ripping down the 8 or 12 ft. length of 2x4s on the tablesaw. I don’t especially enjoy using my furniture-making tools for cutting wet, PT wood, but I thought purchasing a contractor’s tablesaw for the occasional backyard project would be extravagant. (Didn’t even bother to ask.) I would just use old/inexpensive blades and clean them off afterwards. The sawdust thing is a bigger issue. Contact with the chemicals used to treat lumber should be avoided but these are not readily rubbed off on hands, clothes or tools as they are pretty tightly associated with the wood fibers. However, that means it is even more important to avoid ingestion of those fibers as sawdust. In practice, PT lumber is usually dimensioned by cross-cutting the planks to length with a circular saw, out-of-doors, and then fastening them to the Project. Thus, there is not a big concern with dust exposure in that setting. My plan was to rip PT material to non-standard widths and then dado slots into these pieces. Both of these operations would be done indoors and both create substantial amounts of sawdust. The vacuum system in my workshop does a good job collecting dust as it comes off the tool, but further measures: frequent floor sweeping; air filtration; and, most importantly, donning an N-95 dust mask were taken. Probably a good idea to give the whole shop a thorough dusting afterwards, too.
Anyway, cutting the damp 2x4s went smoothly using a thin, 24-tooth ripping blade on the tablesaw. It was fascinating to observe the pent-up grain tension in some of these southern yellow pine planks. Sliced along their length the wood in many of these straight 2x4s bowed or crook’d substantially upon being “released” as 2x2s - bother! The next step was to cut a 1/2 in. deep groove (dado) down the center that would be wide enough to accommodate the width of the wire panel fence (5/16 in.). To keep the groove in the center of the frame and of uniform depth, the bent-up 2x2s would have to be brought back into line, at least momentarily, during that 4 inch stretch when they pass through the dado saw blade. This was accomplished by creating a “channel” composed of a rip fence on the right, a feather board on the left, the steel table of the saw below, and a stout piece of wood clamped above the blade. In truth, 3 out of these 4 directions (E,S,W) are constrained in the same manner when making most cuts on the tablesaw. It is the presence of the “bullying” fence north of the blade that was crucial to getting accurate dados in these boards. Once the framing stock was milled it was cut to the proper lengths on the miter saw.
Constraining the wood on all 4 sides was critical to making accurate dados in the warped frame wood. Completed groove shown on the piece at left.
Design and construction of the door was put off until after the fence was erected. I wanted to see how the wire panels “behaved” once slotted into their grooves to determine how much of the door should be allocated to wire versus wood. (I’ll describe it’s construction here, though.) I found the wire panel to be suitably stiff when braced upright by the frame, so it would be used in place of wood for the door “panels”. My concern with the 40 in. wide door was avoiding warp and so I decided to make it “substantial”, using 2x6 PT boards for the frame. With an extra, center stile this 37x40 in. door consumed almost all of two 8 ft. planks. After cutting to length I dadoed 3/4 inch deep grooves along the interior sides of the stiles that would secure both the wire panels and the tenons made at the ends of the top and bottom rails. This is referred to as a “square-stuck” frame and is easy to make. My only modification was to skip the dado on the horizontal (rail) portions where the wire panel rests. My fear was that the resulting 3/4 in. deep “gutter” on the bottom rail would hold rain water and somehow hasten the door’s destruction. I opted instead to drill holes to incorporate the vertical wires of the panel. This made for a remarkably sturdy structure once the wire panels were tapped into place. The door, which resembled those chunky, Big Little books that we read as children, was fitted with spring-loaded “screen door” hinges prior to hanging as the last step of assembly.
Assembly
Assembly for this Project would be done at the job site, itself. (Love that phrase: job site.) To begin, the posts were seated at strategic points about the perimeter, squared against the garden frame, made “plumb” with the help of a carpenter’s level and then secured to the garden timbers using a pair of 6 in. lag screws. In actuality, the four front posts were installed first and then the two, 4 ft. front fence frames were constructed, just to be sure things were going to work-out as planned.
On YouTube, I saw similar frames made on the floor of a garage, then carried to the yard and shimmied into place between two posts. While that would work for this application, since the panels sit on the garden wall it was easier to construct the fence panels in situ. This also provided the opportunity to straighten the frame pieces in the process. I first laid down the bottom member, cut to fit exactly between two posts. For some this was preceded by screwing down a “training” 2x4 board to the garden timber alongside the position of the frame and then clamping them together so as to make the crooked frame member straight for fastening to the timber with 2 in. screws. The two side members, all cut to a uniform length of 33 and 5/8 in., were then mounted perpendicular to the bottom frame by screwing into the posts. These “troublemakers” also required some straightening-out, accomplished by clamping each to a 4 ft carpenters level mounted alongside. This method creates a “U” shaped partial frame into which the hog panels could be slotted into. The panels needed to first be trimmed to size with bolt cutters beforehand. Lots of measuring, snipping and then trimming to get the fit right, but I was impressed at how well-made and “square” these panels were. Good product! The capping frame member, selected from among the straightest of the dadoed 2x2 stock, could then be cut to length, placed on top and secured with screws at each end.
The job went well, methodically fencing the space between posts, one panel at a time. Except … the “12 foot” section at the back of the garden turned out to be precisely 12ft., 2 and 1/8 inches between posts. You guessed it, 2 and 1/8 inches longer than the 12 ft 2x4s that I had already ripped and dadoed. That cumbersome process had involved opening the door of the Workshop behind the tablesaw to accommodate the hind end as it was being milled - no desire to replay that scene with even longer stock. Plan B was to purchase, prep and interdigitate one additional 4x4 fence post in order to use the 12 foot frame pieces, now cut in half. Plan B worked well and breaking up that long run of panel probably made for a sturdier fence, in the end.
Fenced garden before door installation
Door installation and trimming went just like the cabinet-scale work to which I am more accustomed. The screen door hinges re-seat the door with a satisfying ‘clank’ upon release. It’s a farming sound, actually.
View from the deck (note turkey skulking at right)
