The Great White Duck
The one project I had that had been part of that Oakland Hills firestorm was after I moved to Sonoma County. It had been a luxury home of a big time telephone company executive and it had burned completely to the ground. It had been at the very southeast corner of the fire on Grizzly Peak Blvd., at the top of the ridge. The house next door to the south was untouched, and the undeveloped natural landscape across the road had been slightly scorched.
Going around meeting the crew on my first day there, I discovered that one of the carpenters, Andrew Gordon, lived in Petaluma also. It turned out he lived about a half a block from me, and he had been to more Grateful Dead shows than anybody I ever knew. After that we commuted together for the rest of that project and many others.
The foreman on the Grizzly Peak house was Gary Crawford. He was the only person in my construction life that would get to the job earlier than I did. Andrew and I would pull up there a good half hour or more early every day and he'd already be there, standing by his truck in the dark smoking a cigarette. He had a good layer of blubber that insulated him so well that he wore short pants in any weather, but his weight never slowed him down. He was a relentless hard worker and equally relentlessly careful about what we were building. And of course he was funny. He kept running gags going and had a nickname for everybody. At first I was "Gun Turret John," for the weeks I had been doing the finish work from planks on a scaffold up in the turret-like skylights at the Grizzly Peak house. By the time we reached Coarsegold, a couple of projects later, I was "Johnny Be Good."
In 1994, Gary invited Andrew and me to build a house with him from the ground up. The site was near Coarsegold in the Sierra foothills about 25 miles south of Yosemite. Gary's approach to building was as if he had arrived at Deming's methods independently. It was an eye-opener to work with him closely building an entire house. Up until then we had done remodels and additions together. We began with a bare plot of ground in the woods adjoining a 6-acre lake, with no other houses within sight. There wasn't any electricity to the job yet and a couple of poles would need to be added to bring wires in from the road, so we built most of it on generator power.
The care with which Gary proceeded began with a solid set of batter boards and precisely located screws to tie the strings for our layout. A backhoe and operator were hired to make the foundation trenches, and to grade the site. Now, working with most builders, the concrete foundation walls that are below grade are of little concern. Gary had us keep the forms for the stem walls perfectly plumb all the way to the bottom of the trench. Most builders will say you don't need to do that as long as the level where the sill plates are bolted down is in the right place. Gary's idea was that when your start compromising and saying "good enough" at the bottom of the trench, by the time you get to the peak of the roof, it's going to add up to "not good enough." That's pure Deming thinking.
At first it was just Gary and I until we got to the point of building the forms and tying the steel. After that, Andrew drove up to join us. One day shortly after Andrew arrived Gary said he was going to buy us lunch, so we all got into one truck and drove down to a local burger joint. Gary went in first and stepped up to the counter to order, "Three cheeseburgers and two chocolate shakes." Andrew and I looked at each other wondering why only two drinks when Gary turned to us and asked, "And...what are you guys having?" His capacity to put away food was a continual amazement.
Once the concrete was poured and the forms were stripped, we set to work on the sill plates and floor joists, taking great care to keep the plates as straight as a string while bolting them down. The floor joists we cut to the precise measurement, even though most builders would be fine with plus or minus 1/16", Gary's concept was; "Why not cut it exactly right?" To start with, he introduced us to the idea of calibrating our tape measures.
We all have these steel tape measures and they have a way of becoming inaccurate through normal use. It's that hook on the end that can make trouble. All the inches on the tape stay exactly the same, but that first inch can lead you to an inaccuracy of 1/16" by itself, and since all measurements have that first inch, they are all off. Which wouldn't be too much of a problem if we were all using the same tape for the whole project. But that isn't the case; we all had our own tapes. That hook on the end of the tape slides in and out the thickness of the hook, so that if you butt the end of the tape into something it will slide in that tiny amount, and if you were going measure the length of something you could hook on the end, it would slide out that same amount to compensate. So what you do to calibrate the tape is to dial in that hook. It can get its little slot the rivets go through clogged with dirt, glue, caulking or plaster, and it won't slide right making your butted measurements not agree with the hooked ones. That seldom happens. What does happen often is that hook gets bent.
So what you do is get a board or workbench surface with a sharp edge and measure a perfect inch from the edge using any other part of the tape but that first inch. You use the sharpest pencil to make the thinnest line. Then you hook your tape on the edge and see what it reads. Most often if it's off, it's because the hook has bent inward, so the reading will be a little short. You take your hammer and tap that hook until it reads just right. When you've got three or four carpenters working on the same structure, just the fact that their tape measures read slightly differently adds up to a certain amount of slop. In most builders' minds, it's good enough, but not for Deming, Gary Crawford, and now me.
Likewise most framers will use a thick flat carpenter's pencil to mark their cuts, and as it loses its sharpness, the line it's making gets fatter and fatter until it's more than 1/16". Then you've got to decide which side of the line your saw cut is going to be on. Again, with multiple individuals the collective result is slop, so we kept sharpening our pencils by stropping them on a flat piece of sandpaper.
On the little lake there were about seven mallards and one large white duck. We took to feeding them pieces of our donuts one morning break. After that whenever we shut down the generator for break or lunch the ducks would turn and swim toward our shore, the big white duck first with the mallards following. He was definitely their leader. He was the first to waddle up to where the donut pieces were landing, with the mallards cautiously coming up behind. If you threw a whitish piece of donut or bread they would rush to be the first to get it, but if you tossed a piece of a brownie they would turn and run away figuring that it was a rock. After we got power company electricity and stopped using the generator, Gary would make quacking sounds at break and lunch and Howard, one of the white duck's nicknames, would turn and swim to join us with his mallard entourage paddling along behind.
You could build an entire house measuring and marking each piece with a square and pencil, and cutting it with a skilsaw on a pair of sawhorses. There would be a lot of variation in the squareness of the end cuts and the lengths of nominally identical members. Factor in fatigue, discomfort, and tedium and the sloppiness compounds itself.
In a Deming analysis, it's not the carpenters' fault, though some would perform better than others. It's nobody's fault; it's a flag that the situation needs (1) better procedures and training; like calibrating tapes, sharpening pencils and adding a half second of sharpened awareness before pulling the trigger and making the cut. And/or the situation would benefit from (2) better tools: like when there are a lot of members that need to be the same length, like studs, or cripples and trimmers for windows that are the same size, you set up a chop saw table so that the cut ends square every time, and you can set up stop blocks to cut repeated lengths and a built-in tape on the fence so that no measuring and marking even needs to be done. And/or less variation might be achieved with (3) better materials.
So you get a big load of lumber delivered from a local supplier and you may have blown it already if you only considered the price. Framing lumber has irregularities. You do the best you can and still it isn't perfect. You take the stack of materials that are going to be the wall studs. In California, most suppliers will sell you precisely precut 92 1/4" studs, so that with the addition of a bottom plate and two top plates, the stud wall is 96"+ tall. You are going to need a lot of shorter cuts of that same material for cripples and trimmers, so you get more studs than you need so you can "upgrade your stock," as we call it. We set aside the best for the studs. The others; the ones with curves in them or big knots that will weaken them, or some "wayne" which means a place where one edge is bark; those we put aside to use as shorter pieces and braces. Sometimes you'll get a load of 2" x 10"s, for instance, of the different lengths your plan calls for, and for whatever reason, maybe they came from different mills, some of them were 9 1/4" wide and the others more like 9 3/8". That's going to be a problem if they are end to end in the same floor, for instance. We always have a planer on the job to make the framing members the same dimension...another way of upgrading your stock.
It made sense to rent a vacation house at nearby Bass Lake, rather than rent motel rooms. In the fall off-season the lake was drained, so rents went way down. We each had our own bedroom and we had a kitchen to make our meals and lunches. We began calling Gary "Housemother." When we started, it was the middle of the summer and we were working 12 hours a day in the sweltering heat. As time went on the days got shorter and the nights got cooler so by the end it was winter and it was so cold we were putting our hands on the blazing halogen work lights until our gloves started to smolder, then we could go back to work. A hard worker day and night, Gary spent most evenings going over the plans, doing material take-offs for the next lumber drop, and going over the details.
Framing a house is not rocket science. It's pretty simple, and before long we had built the subfloor and had all the stud walls standing on the top plates. At that point, most builders would just start putting the plywood sheathing on. Gary added a whole day's work to fine tune the skeletal building in a process he called "plumbing and lining." He'd start by having us use string to check the straightness of the top plates, adjusting them and holding them in place with temporary braces, and making sure the walls were exactly as long they were supposed to be. He had invented some Plexiglas jigs that would clamp over the double top plate and hold the string of a plumb bob exactly 1" away, and with 1" x 1" blocks we could quickly check the stud walls for plumb, adjust them and use long diagonal braces to keep them from racking back.
With all our care in using the straightest lumber for the studs, some of them would still bow in or bow out. You can take a long string attached to the screw on your batter board and pull it up across the outer plane of the stud wall to check which studs need correction. Gary had a few methods to straighten the studs in place by notching and pinching, or by notching and shimming. Every stud corrected was held in position by a brace to the floor. Rejected studs are fine for temporary braces.
Then we plumbed and lined the interior walls. With all the framed walls braced every which way, the inside of the building was a web of lumber, and the outside all in an uncluttered plane. When we put the 1/2" plywood sheathing on, it became obvious that all the care we had put into the fine tuning made this step come together with no adjusting, fudging, shimming or trimming. Edges of the 4' x 10' sheets land exactly on the center of the studs, and the surface becomes a perfect plane, with no bulges or dips. Once that is nailed off every 6" on the edges and 8" in the field, those walls are not going to rack, but you leave all the braces in place until they need to be removed. There's no hurry to take them down and for a few weeks, all that green lumber can dry while held in position. It wants to stay the way it has dried.
Well, I'm an ace when it comes to building hip roofs, and it was my pleasure to build the pyramidal roofs on the two wings of this house. With the perimeter of the top plate a dependable perfect square, everything fit together like the parts of a Mazda transmission.
The windows and doors went in with minimal adjustments of the rough openings, and then it was time for Andrew and I to go home. The rest of the project was done by Gary and the local subs he had rounded up: electricians, plumbers, sheetrockers, painters, cabinetmakers, roofers, and for the exterior, a stucco contractor. It was years before I returned there and took pictures of the finished project.
We had nicknames for Gary: "Mr. Large," or "Oh Large One," when addressing him directly, and "Housemother," of course, but in the following years we spoke of him respectfully as "The Great White Duck."
© 2017, 2022 John Oliver
All Rights Reserved