Not every project I build comes out perfectly. As a matter of fact, I can’t recall too many that didn’t have at least one minor mistake. Of course, I mean something that no one else would notice, though some of you might. Without a doubt, I’ve never built a perfect pivot bookcase, but I’m getting a lot closer!

Even the bookcase in this article isn’t perfect. Each time I build one, I learn something new. After all, hidden bookcase doors are a lot more complicated than an ordinary door—there are a lot of variables, both in design and construction, especially on openings that have to swing out.

In this article, I’ll point out a few of the mistakes I made so hopefully you won’t make them—and maybe I won’t make them again. If you notice any others, please let me know. Hidden door bookcases aren’t easy to design or build, but they’re intriguing. Maybe one day we’ll all be able to build one that’s perfect in every way.

Hinges and Wheels

I’ve seen and installed a lot of bookcase doors, many that swing on regular butt hinges. I’ve always used 4 1/2 or 5-in. heavy-duty ball bearing hinges, and they work alright, though the hinges tend to sag a little when the case is really loaded down with books. And they always need some adjustment down the road. Plus, they require a lot of jamb clearance, which has never seemed right to me. Besides, butt hinges only work on swing-in bookcases—there’s no way to hide them completely on a swing-out design.

I’ve also seen cabinet shops build these types of doors, using euro hinges. Trust me, those never work, no matter how many of those little hinges you use, they always sag. I’ve seen carpenters use piano hinges, too, but then it’s tough to take the case off or adjust the hinge. Besides, even a piano hinge is hard to hide in the trim on a swing-out case.

Swinging bookcases always sag a little, too. I’ve tried installing wheels and rollers on the bottoms of swinging bookcases, and they work okay, as long as the floor is a smooth, hard surface, and if there are no throw rugs, though sometimes the roller leaves a tell-tale track on the floor, especially over carpet.

When you use a roller, at the very least you have to leave a gap at the bottom of the case for floor clearance, and that’s a dead giveaway, too. Plus it’s almost impossible to really hide the joints in the baseboard, no matter how cleverly you disguise them. From what I’ve learned, the best way to design and build a durable swing-out bookcase door, one that can be adjusted easily, and one that’s truly invisible, is to design the door to swing above the baseboard, and hang it on a center-hung pivot hinge.

Start With a Drawing

There are few projects I work on today without doing a scale drawing first. When in comes to bookcases, especially swinging ones, SketchUp has saved my life. I started this project with a two-dimensional drawing, one that allowed me to pivot the door in the drawing. That’s how I found the correct location for the pivot point, which took some experimenting. The two most important issues are: 1: The case has to swing clear of the hinge jamb; 2: The case has to open 90 degrees. If you don’t know how to animate Sketchup drawings, watch this tutorial that Todd Murdock has put together:

I wanted the case to have a minimal amount of clearance between the jambs, so it would just clear the trim on the hinge side, and wouldn’t require wide trim on the strike side. That clearance is determined by the setback of the pivot perpendicular to the face of the wall.

(Note: Click any image to enlarge)

When wide open, the door butts against the trim on the hinge side. That clearance is determined by the depth of the bookcase and the location of the pivot, measured from the hinge jamb toward the strike jamb–parallel with the wall.

Bottom Clearance

The real improvement in this design is swinging the bookcase above the baseboard, so it won’t drag on a throw rug and can be trimmed out without any visible gaps. I wanted to end up with the case about 2 3/4 in. above the floor, to clear 2 1/2-in. baseboard. For a taller base, the bottom of the case would be even farther from the floor. If you’re not familiar with Rixson pivot hinges, scroll down to that section below right now.

Another drawing, this one three-dimensional and detailing the hinge parts and clearance requirements, confirmed that mounting the pivot base on two layers of 3/4 plywood would get me close to 2 1/2 in. above the floor. Because I could install the toe kick after swinging the case, the exact dimension didn’t matter, which made execution a lot easier.

Bookcase Construction

(Note: Click any image to enlarge.)

To prevent the case from sagging, I dadoed the sides to accept the shelves (see photo, right), something I don’t always do for built in cases. For cutting dados, I normally use a templates guide on my router, which makes it easier to build a compact template, and provides a cleaner tighter dado, but I was lazy. I didn’t have a Porter-Cable-style 3/4-in. template guide for this new router, and rather than running to the tool store, I made the router template exactly the width of the router base. I installed the cross pieces allowing enough space for both bookshelf sides plus an extra 3/16 in.—so I could slide the template up and down without hanging up—and used a long shim and spring clamps to lock the template in place.

An even easier tool for cutting dados is a Festool MFT table and router guide rail. This system is designed perfectly for the task and requires no template and no special clamping setup. Simply layout the book shelf sides with clear pencil lines for each dado (I used a Sharpie so the lines would be more visible in the photographs). Rather than running my router bit dangerously close to the guide rail, I adjust the router so that it cuts almost 1/4 in. away from the rubber edge.

To make it easier to align the boards for each cut, I attached a sacrificial fence to the table. The first pass cut a neat dado in the fence, and I aligned all the cuts with that dado. To make sure the boards didn’t slip as I moved them through the cutting station, I screwed a 3/4 in. cleat on top of the layout marks for one of the shelves. Once that cleat came up near the guide rail, I removed it and pressed it into the dado, where it locked the two boards together.

Here’s a trick I learned at Festool School: the dust collection system will collect almost all the saw dust if you don’t dado right through the first piece. Instead, plunge the router into the workpiece about 1/2 in. from the edge, cut the dado, then clean up the front when you’re finished. That little dam is all that’s needed to stop the dust from shooting out the dado, leaving it at the mercy of the dust collector.

Edgebanding Plywood Shelving

I’ve done a lot of edgebanding and always hated the hair-line crack that develops between the plywood and the solid stock. That gap is caused by the inner plywood endgrain swelling from the glue, which puts a little belly in the edge and forces the banding away from top and bottom of the plywood. To prevent edge swelling problems, I used a Collins Ply-Prep bit ($20.00) and ‘routed’ a slightly concave nose on each shelf.

In order to work properly, the Ply-Prep bit requires a router fence with infeed and outfeed surfaces slightly offset to accommodate the very slight amount of material removed from each shelf. I made a shallow pass, less than 1/16 in. deep, half-way across a temporary fence. A line etched into the bit helps center the bit vertically on the stock, which is vital—otherwise the edge won’t be cut square.

After fastening the solid mahogany banding on with glue and 23ga pins…

…I ran a laminate trimmer on each side to cut the surfaces flush.

The last piece I milled was the strike side of the case, which required a bevel. I made the first cut on my table saw, but the blade height wouldn’t cut to daylight, so I cleaned up the bevel with a power plane.

Assembly

Before assembling the pieces, I pre-finished everything, a lesson learned the hard way after making dozens of bookcases—it’s just too hard to finish all those inside corners and edges without getting runs, drips, and finish all over my wrists. I used a water-based polyurethane and a roller, brushing out each piece to remove air bubbles. If I were smarter, I’d own an HVLP (high-volume, low-pressure) system, and spray the three coats on, but I’m not, and so I don’t.

To ensure a tight box that wouldn’t sag, I glued and fastened the shelves with screws, too, brushing the glue into each dado.

Finished sides, added after the case is swinging, cover the screws. I also cut the finished sides 1/2 in. wider, so that they cover the 1/2-in. plywood back. That way, the sides don’t require rabbets.

I glued and screwed the back flush with the sides, so that the case would never rack.

Hardware Preparation

Pivot hinges are the only way to fly when it comes to supporting a heavy bookcase and achieving an invisible door. I used a Rixson Model 370 bottom pivot, which can accommodate up to 500 lbs. and doors up to 3-ft. 8-in. x 8-ft. 6-in. The bottom pivot includes two pieces: the bottom pivot spindle which mounts directly to the floor (upper right, in photo to the right), and the bottom bearing (lower right), which must be mortised into the bottom of the door. The top pivot is a standard model 340, consisting of a retractable jamb-mounted pivot spindle and finished cover plate (middle and upper left), which are mortised into the jamb head, and a top guide (lower left), which is mortised into the top of the door.

I learned a long time ago to always make templates for door hardware, especially hinges—first, because it’s easier to position and cut the mortises perfectly, which means mortise depth, too; and second, because once you’ve used any special type of hardware, you’re bound to use it again and soon—it’s just a law of the jungle, like thermodynamics. In this case, the bearing guides and the top jamb pivot are the same width and thickness, but because their centers vary, along with their lengths, each piece of hardware requires a custom template.

I started by ripping stock for the center spreaders. A standard door-hanging template guide and router bit (1/2-in. bit and 9/16-in. template guide) will cut 1/16 in. short of the template bushing, so I made the template openings 1/8 in. wider and longer than the hardware. I ripped the spreader stock to 1 3/8 in. for the 1 1/4-in. plates. I centered the spreaders between two outer rails, spacing the spreaders apart the length of each plate plus 1/8 in., then fastened the templates together with pocket screws.

The centers vary on each piece of hardware, so make individual templates, one for the top guide and one for the bottom bearing (on left).

Laying out the template stops was critical because that’s what positions the pivots perfectly. For each template, I marked a center line on both axes (parallel to the wall, and perpendicular to the wall), then measured from those center lines to locate the stops. For the bookcase templates, I measured 2 1/4 in. from the pivot center to the back of the first side, knowing the second finished side would add an additional 3/4 in., resulting in a 3 in. backset. For the front backset, I measured 1 3/4 in. from the pivot to the front of the template, and I attached stops on that line.

Setting the router depth was simply a matter of adjusting the depth stop above the turret to exactly the thickness of the hardware.

I clamped both templates to the case and mortised the brackets without a second thought.

I fastened the bottom bearing immediately (below, left), pre-drilling the double-thick bottom shelf for the #10 screws. The top guide (below, right) mounts flush with the top of the case-the bushing must be mortised into the case. I traced the location of the center of the bushing…

…drilled out the hole with a paddle bit, then mounted the bracket. The top shelf is only 3/4 in. thick, but a false shelf, installed after the case is swinging, hides the bushing.

I designed the case 3/4 in. short to allow for this second jamb head, which I mortised in my shop, before installing the case.

The top jamb bracket includes a linkage arm that draws the pivot spindle out of the top bushing in the case, so it’s easy to install and remove the case or a door. I drilled a 1 in. hole at each end of the mortise for the linage arm…

…then I connected the holes with a jig saw.

At the closet door jamb, I traced the mortise for the linkage arm onto the existing head jamb.

Then I drilled out and cleaned up the mortise, and installed the top jamb pivot. I can’t stress how important it is to check the laser plumb dots by also measuring to the jamb—regardless of what type of door you’re hanging, whether it’s new construction or a remodel. Remember, the jamb might not be plumb and you have to hang the case to ‘fit’ the jamb! It’s vital to have a complete understanding of the whole picture, otherwise you have to move hardware after everything is installed (one guess how I know this).

Sometimes, dead plumb and perfectly square aren’t the only concerns when hanging a door, bookshelf or otherwise. I wanted the ‘door’ to fit the jamb, with even gaps. The opening was a little cross-legged, too, and I wanted the casing to fit flat against the case—the case had to be almost perfectly flush with the jamb. The measurement mark was off by only 1/8 in., so I followed that rather than the laser plumb marks.

A laser works great for transferring the plumb line. Just place the red dot on the center of the top pivot and mark the location of the bottom pivot.

Rixson also offers an accessory plumb bob that mounts directly to the top pivot—a slick way of finding the bottom pivot location.

Notice that the bottom support base is 1/2 in. back from the face of the jamb. That 1/2 in. allowed me to recess the bottom toekick so the case would project over the kick, thereby hiding the 1/8 in. gap between the top of the kick and the bottom of the case.

Hanging the case isn’t difficult. Like with most doors, I retracted the top pivot spindle by backing out the set screw. When I’m hanging a door, I usually set the door perpendicular to the jamb, place it on the bottom pivot, then lean it back against the top pivot. That way, I have comfortable control over the door while backing out the set screw and retracting the top spindle. It’s easy to position the door directly under the spindle, then run the set screw back in, pinning the door into place. But with a bookcase it’s not so simple.

Fortunately this was one problem I anticipated, which made me feel pretty good. I made the case 1/4 in. short of the opening, providing just the right gap between the top of the case and the head jamb. I backed out the set screw half way, then placed the case on the bottom pivot and straightened it up in the opening. The top of the case barely scraped across the bottom of the set screw, while the top jamb pivot spindle dragged over the top of the case and then dropped like magic right into the pivot guide. Amazing!

I installed the false sides on both sides of the case, driving fasteners from inside the case, so they wouldn’t be visible as the ‘door’ opened. Of course, no one would ever see the finished side near the hinge, unless they stood inside the closet.

Before starting the trim, I installed a shim made from UHMW (ultra-high molecular weight) plastic, which is pretty slippery stuff ($18.00 from www.smallparts.com). I ripped a 1 1/4-in. length of the material from a 3/4 in. x 12 x 12 blank ($17.00), then I cut a long shim using a Festool guide and saw. I sized the shim to just touch the bottom of the case when the door is closed, which prevents any minor settling. That way, moving joints in the trim at the top of the case stay tight.

Trimming the top of the case is tricky. The joint between the architrave molding (parting bead) and the top of the case must be invisibly tight, yet still provide 1/16 in. clearance for the case to swing. And that’s where I made another mistake. I should have ripped the new top jamb down—to make it at least 1/2 in. back from the face of the jamb—so that the architrave molding would run back inside the jamb, past the bookcase, which would help to hide the joint.

Realizing I couldn’t hide the joint any other way, I swallowed hard, then removed everything from the opening. After ripping down and replacing the head jamb, I hung the case back in the opening and started installing the trim again. Another good reason not to use a piano hinge.

The horns on the architrave molding must be scribed to fit the wall and butt against the head jamb inside the opening.

I next installed a frieze board, and finished the entablature with a two-step cap rabbeted in several passes on my table saw.

The base details went on next. With the case closed, I milled a piece of mahogany toe kick and scribed it to the floor, leaving 1/8 in. clearance to the bottom of the case.

I attached the plinth blocks with trim head screws, and the casing, too, especially the strike side piece that remains on the cabinet and acts as stop when the cabinet swings closed. Notice that the toe kick is recessed inside the jamb–it’s not flush with the jamb. That way, the bottom shelf projects over the toe kick making it impossible to see the clearance gap between the top of the toe kick and the bottom of the bookcase.

It was at that moment I realized I couldn’t reach the set screw with a screw driver: I couldn’t run the screw in to secure the case completely, and I couldn’t back the screw out to remove the case. I didn’t feel so smart anymore, and it got worse.

On my first attempt at drilling a simple 3/8 in. access hole through which I could reach the set screw with a narrow screw driver, I couldn’t seem to find a drill bit sharp enough to drill through the plywood. I dried a paddle bit first, then a twist drill. On the third attempt, I realized I was drilling right into the top guide hardware.

Determined to overcome my own stupidity, I thought through the problem carefully and found a second access hole located on a radius layout, so I could swing the case clear of the top guide and reach the top pivot set screw. Fortunately, the new hole lined up perfectly. I turned the screw and drove the pivot spindle all the way into the top guide. Notice that the first hole is aligned perfectly with the hardware mounted in the top of the case.

With the case tight against the wall and under pressure from a slight amount of cross leg, I drilled a 3/4-in. hole through the side and into the jamb. A 3/4-in. x 5-in. long dowel, with a mahogany grip, locks the case in the opening. I hide the grip with a stack of books so no one will know how to open it.

There’s no door knob, and the case rubs just a hair on the UHMW plastic shim, but a slight tug on the shelves slips the case free from the shim, and the door swings open with a swoosh of air. Sure, one day I might even tape and mud the joint between the jamb and the wall…but no one but me and my dog should ever see that anyway.

True to my original drawings, the case pivots back from the hinge-side trim and just clears the strike jamb as it swings open to exactly 90 degrees. Don’t try this in a small closet. In fact, a 3/0 closet would work best, though this 2/8 opening, with a 7 in. deep case, allows enough access for a skinny guy like me.

Back then, I thought about building the new wooden gate myself, but I didn’t have the time. I thought it might be more sensible to ‘hire a professional.’ I know this magazine is read mostly by professionals, and I don’t mean this as criticism of the entire industry, but believe me, not everyone who says they’re a professional is professional.

The first, professionally installed gate. (Note: Click any image to enlarge.)

For example, when my contractor’s crew started installing the flat 5 1/2-in. boards across the gate frame, they started on the left side and ended with a small 2-in. strip of wood on the right side, near the latch. I asked them to redo the boards, so they’d be centered—that didn’t earn me any friends. Of course, by the time I realized the workmanship was questionable, they were setting the finish. I later learned that that wasn’t the only area where quality was sacrificed for speed.

After three years, the gate started falling apart, and the warranty was long expired.

The concrete footings were crumbling, and could be broken with my bare hands.

The posts were out of level, and the gate was dragging on the ground, despite my continuous adjustments.

Rather than contract with the same company again, I decided to re-build the gate my way.

Design

I started by drawing plans for the upgrade in Google SketchUp. The design was much more structural, with a stucco wall extending off of the garage and a stucco pillar on the hinge side of the gate.

The main challenge was dealing with the varying angles and grades. After many weeks of measuring, planning, thinking, and drawing, I came up with my final measured drawings of the foundation and wall structure.

Some of you might be thinking: “Wow, if I spent that much time designing a simple gate, I’d never make a living.” You’re probably right! Fortunately, I wasn’t trying to make a living building this gate—I just wanted it to last.

I chose to build the wall out of traditional wood framing instead of concrete block, because I’m not that experienced with block, which turned out to be a good decision. When I went to Building and Safety to get a permit, and showed them my plans, they told me that I didn’t need a permit for a wood structure fence—but I would need a permit if it were concrete block!

Construction

The first step was excavation. This required a bit of irrigation work, jack hammering to get out some of the old concrete, and lots of digging. The total depth of the footing and stem wall was 25 in., but I dug a bit deeper to get to undisturbed soil. I then back-filled with 3/4-in. crushed gravel.

I set up shop in my garage and got to work on the forms for the foundation. These were extremely challenging due to the multiple angles. Plus, I wanted to do a monolithic pour of the footing and stem wall, which meant that a set of forms for the stem wall had to sit on top of the footing forms.

First, I set the forms for the footing.

Next, I set the forms for the stem walls. I used #5, 5/8-in. rebar, drilled and epoxied into the garage foundation.

My calculations were for just under a 1/2-yard of concrete. I rented a ready-mix trailer from a local equipment rental yard, which made things a lot easier.

The forms made it easy to screed the concrete level and smooth. I set 1/2-in. j-bolt anchors, and let the foundation cure for 48 hours before starting framing.

Most TiC readers probably already know that a wall that’s simply bolted to a foundation will never be rigid enough to hold a gate, but, being new to construction, I had to learn the hard way. I should have set a steel post at both sides of the gate, right into the foundation, and then framed around the posts. I didn’t know that. So, I re-worked the design.

The new design required an addition to the foundation, which changed the freestanding wall into an “L” shape. More digging…more rebar…more forms…and more concrete.

The next step was two layers of Grade-D building paper and 20-guage, self-furred stucco netting, which was installed with 1 1/2-in. staples with a 1-in. crown. None of the staples were installed on the flat horizontal surfaces of the wall tops, and they were kept 2 in. down the wall from the top edge.

Also note that I didn’t install a weep screed. I specifically chose not to install this detail because it simply wouldn’t have looked good with our house, which was a 1927 Spanish Revival. I did, however, continue the lath onto the foundation, which was attached with Ramset nails.

With building paper and lath installed, the project was ready for stucco.

The scratch coat came next.

Then came the brown coat, which brought the thickness out to the same level as the existing garage wall.

Finally, the topcoat. Our house had a skip trowel texture, so blending the new section to the exiting stucco was relatively easy.

I kept the wall moist for 72 hours while the stucco cured. Additionally, the stucco had to cure for at least 28 days before painting. I chose not to use a colored topcoat, because our house was painted.

The next step was to start working on the brick pathway and threshold leading up to the gate. I started with excavation, then 4 in. of 3/4-in. crushed gravel, and a 4-in. slab of concrete with 6 in. wire mesh. The photo to the right was taken just after I bull-floated the concrete. After the water bled out, I floated and lightly scratched it to give the mortar something to grab onto.

Brick was set in a 3/8-in. bed of mortar. A masonry saw was used for the handful of angles.

Grout was completed the next day, and the brick was washed with muriatic acid.

The Gate

Finally! It was time to build the gate! I have to admit that during this whole process, my wife and I hadn’t decided on the gate design. We couldn’t even decide between wood or wrought iron, which meant that I couldn’t install the proper jambs. Of course, I didn’t want that decision to hold me up, so I stuccoed the entire wall.

Ultimately we chose a wooden gate—with mortise and tenon joinery and floating panels. For a wooden gate, I would have preferred to install the jambs before the stucco, and then key the stucco into a rabbet at the back of the jamb, but that’s not how things worked out.

Even though I have a pretty good shop at home, I felt that I needed more space, and a few tools I didn’t own, like a bandsaw, so I used a local professional wood shop that a friend of mine uses to build furniture. The shop had three table saws, one of which was an Altendorf sliding table saw. This was an absolutely awesome machine to use! They also had a hollow chisel mortiser, oscillating spindle sander, oscillating edge sander, bandsaw, and many more fine tools that aided in the gate construction.

I chose Vertical Grain Douglas Fir, because it was readily available without special order. Honestly, I wanted to build the gate out of Cedar, but it was special order, and I couldn’t wait.

The gate design was to be strictly mortise and tenon joinery without any fasteners or pins, and the center panels would float in a dado. There would be a total of four rails, which I designated from the top down as: Top Rail, Top-Middle Rail, Bottom-Middle Rail, and Bottom Rail. The two Middle Rails would have standard tenons. The Top and Bottom Rails required haunched tenons, because of the dado that ran the entire length of the stiles.

I started by milling the 2×6 stock for the stiles and rails. The first step in the process was smoothing one edge on a jointer. The next step was smoothing a face on the jointer so that these two surfaces were square to each other. After this, I ran the opposite face in a thickness planer to achieve my final thickness of 1 3/8 in. Finally, the last edge was ripped on the table saw to a width of 5 1/4 in.

The next step involved setting up a 1/2-in. dado blade on the table saw to cut the dados in the stiles and rails. These were cut to a depth of 3/4 in. The inside edges of the stiles, both edges of the two Middle Rails, and the top edge of the Bottom Rail had a through dado cut in them. The dado in the bottom edge of the Top Rail was not cut at this time because of the curve—this will be discussed later in the article.

The setup was simple: install a 1/2-in. wide dado blade, set the blade height to 3/4 in., and set the fence to 7/16 in. away from the blade. Each piece was cut once, turned end-for-end, and then cut again. This ensured that the dado was perfectly centered in the workpiece.

Curved Template

After cutting the dados, it was time to start drawing and creating the curved template for the Top Rail. I had to create the template before cutting the mortises in the stiles, because it would indicate exactly where the Top Rail and Top-Middle Rail would be.

I laid out the radius-to-rail the Egyptian way: I drew it full-scale on a piece of 1/4-in. plywood, so I could get the right curve. I didn’t have a specific radius in mind for the curve, so I set up trammel points on a long piece of scrap stock and played around. The final radius I chose was 36 in. I drew the bottom curve, and then moved the pivot point vertically up the centerline 5 1/4 in. to draw the top curve. From there I drew horizontal lines perpendicular to the sides, one at the apex of the top curve and another at the bottom points of the bottom curve—these are labeled as Top Line and Bottom Line in the following diagram. I also drew vertical lines 5 1/4 in. in from each side to represent the width of the stiles.

(Click to enlarge)

Once the template reached this stage, the template was cut exactly on the two outside parallel lines, which represented the gate width. I cut the template a few inches above and below the Top Line and Bottom Line. The exact distance away from the Top and Bottom Lines didn’t really matter, because the template would eventually be cut later at the curved lines.

The top rail template with haunched tenon layouts on both ends.

The gate was six feet tall at the top of the curve, so I measured six feet up from the bottom of each stile. I then placed the template on the stile with the Top Line aligned with the six-foot mark. Next, I marked tangent points at stiles A & B.

I also measured and marked the layout of the tenons on the template. There wasn’t any exact science here. The tenons on all of the other rails were 3 in. long, but the top rail was special because of the curve. I settled on a length of 2 in.—the deepest I could go into the stile and still leave a fair amount of material near the top edge.

I cut the plywood template with a jigsaw and smoothed the curves on an 8-in. oscillating edge sander.

Using the template, I marked the tops of the stiles, then measured and drew the exact location of the Top Rail mortises based on one of the lines drawn on my template.

The next step was to mark the location of the Top-Middle Rail, but this was where I made a serious mistake. I referenced off of the lines at Point B on each stile to find the location of the Top-Middle Rail instead of at Point D. Unfortunately, I didn’t discover this until too late. In fact, it was at the time of assembly, during glue-up, that I became aware of my mistake. (I’ll talk about this again later in the article.)

These reference points and lines allowed me to measure down the stile to find the location of the Top-Middle Rail, which was 5 1/4 in. below the Top Rail. The Bottom-Middle Rail was located and marked by measuring 5 1/4 in. up from the top of the Bottom Rail.

With this information, I was able to lay out the exact locations of all the mortises with lines drawn on the faces of both stiles. Afterwards, I transferred the lines to the edges, which were ultimately referenced when cutting the mortises.

Mortise layout for the Bottom Rail and the Bottom-Middle Rail.

Mortise and Tenons

I used an Oliver #91-D Vertical Hollow Chisel Mortiser to cut the mortises. (Unfortunately, I didn’t take pictures of that process.) With the mortises completed, I went to work on the rails. I cut them all on a chop saw to a length of 35 3/4 in. (29 3/4 in. for the length of the rail between the stiles, plus 3 in. for each tenon). I cut the Top Rail to the same length, even though its final length would be 2 in. shorter because of the 2 in. tenons. Having equal-sized rails made it easier to maintain perfect layout tolerance on the center sections.

I cut the tenons on a table saw using a dado blade, and I set a rip fence as a stop for the tenon length. After making the initial shoulder cut, I moved the material away from the fence and removed the remaining waste. I used a band saw to cut the haunched tenons (see photo, right).

For the curved top rail, I just traced my template onto a glue-up of 2×6 stock, which was wide enough to get the full radius. I cut the top rail with a band saw and sanded close to the line with an oscillating spindle sander. I smoothed the stock using a Stanley #113 circular plane.

All of the rails are complete, with the exception of the dado on the bottom, concave side of the top rail.

To cut the dado in the radius top rail, I used a 1/2-in. slot cutting router bit with a ball bearing guide mounted in a router table.

Tongue and Groove Panels

A 1/4-in. dado was cut down the center of one edge on each board.

The panels were milled from 1×6 lumber to a final thickness of 1/2 in., and a width of 5 1/4 in. The final width of the panels would be 30 3/4 in., and would be made up of seven individual boards. This would allow for 1/2 in. of expansion. Each board was tongue and grooved and glued together. The tongues and grooves were made on the table saw, but I started by milling the grooves first.

I cut the tongues next, keeping the board oriented vertically with the face up against the fence. An alternative method would have been to cut with the face against the table. Either way would have worked, but a zero-clearance insert was mandatory when cutting it in a vertical fashion. Also note that a featherboard was used when cutting both the tongues and grooves.

Using a dado blade and a zero-clearance insert, both sides of each board were cut with the same setup, resulting in a tongue perfectly centered.

Bevels on both sides of the groove edge were cut on the sable saw with the blade tilted to 45 degrees.

Bevels on both sides of the tongue edge were cut with a shoulder plane.

When assembled, a simple V-notch groove was formed between each board.

The individual boards were glued and clamped together using Tightbond III waterproof glue.

After the glue set, the panels were cut to final dimension and pre-stained. I used Cabot Solid Color Acrylic Decking Stain. I chose this finish because it didn’t need priming, and soaked into the wood instead of remaining on the surface as a film, like paint or polyurethane. I didn’t want to use spar polyurethane, because of the way it yellows and flakes with age. I think the long-term maintenance of the decking stain will be easier—just a fresh coat every three to five years. I also pre-stained all of the dados in the stiles and rails before assembly.

Gate Assembly

The next step was assembling the gate.

The gate was glued and clamped, using Tightbond III waterproof glue. The center panels floated within the frame to allow for expansion and contraction throughout the seasons. However, I used a small bead of “Big Stretch” acrylic latex caulking on the bottom edge of each panel, on both sides of the gate, where it fit in the dados. This will, hopefully, inhibit any water from getting down into the dados.

The stiles were left long during assembly, but after the glue set-up, I trimmed the stiles and the Top Rail close to the line with a jigsaw. I then clamped the curved template onto the gate directly on my cut line, and used a flush trim router bit with a ball bearing guide.

Because of a through-dado on the stiles, a haunched tenon was necessary on both the top and bottom rails.

Gate Installation

Gate installation was relatively easy. I installed the jambs on the stucco walls using 1/2-in. lag bolts. I used two 4-in. ball bearing hinges and an exterior door lockset from Emtek. I didn’t want a traditional gate handle. The jambs were installed to the wall using four lag bolts on each jamb that were recessed into the wood and plugged. The recess was cut with a forstner bit, and the plugs were cut with a plug cutter, then trimmed flush.

After two coats of stain, the gate was finished (below, left). Once I put a bit of paint on the stucco (below, right), the project was finally complete.

Once completed, we received many compliments from neighbors and friends who all stated the new gate and wall looked as if they had always been a part of the house. We agreed.

LOOKING BACK

Along the way, I learned a lot, and would have done several things differently:

• First off, I wish I had built the gate 1 3/4-in. thick instead of 1 3/8-in. thick. This is for two reasons: 1. The gate is warping. 2. A thicker gate would have allowed me to use better hardware.

• I wish I’d fixed the design flaw of the free-standing wall earlier…while in the design stage, not the concrete stage. That would have meant a lot less stress.

• And speaking of stress, when it came time to pour the concrete, I ordered a half-yard. My calculations were just under this number, but I should have ordered 3/4-in. of a yard. I was freaking out during the pour, worried that I’d run out of concrete. But scraping the drum gave me just what I needed.

• Finally, I should have done a full mock-up of the gate, or at least drawn the full-size gate on plywood. As I mentioned earlier, I made a glaring error: My drawing and initial idea was for the top and bottom panels to be of the same height where they meet the stiles. Unfortunately, I laid out the mortises incorrectly and didn’t notice it until after the glue up when I stood back and said “Ugh!” Of course, my wife said they looked fantastic, and no one but me knows the truth.

I made sure to disclose at the beginning of this article that I’m not a professional contractor, nor do I work in the industry. But I am a very serious do-it-yourselfer. My day job is in the film industry. I’m a sound mixer and I work on a swing shift, which starts at 3:30 in the afternoon. That allowed me to work on my project for several hours a day before I went “to work,” as well as over a few long weekends. In all, the entire project took about eight weeks—which, I know, will sound like a long time to a lot of TiC readers. But I really enjoy doing this type of work myself—not only is the final product fulfilling, but every step along the way was rewarding—even the hard-learned lessons. And I definitely welcome any and all feedback—I relish the idea of learning “the easy way.”

I would like to thank Peter Vogel for his patience and guidance with building the gate. Peter is an exceptional woodworker and artist. Additionally, I would like to thank Kirk Giordano, of Kirk Giordano Plastering, Inc., for his informative videos. Kirk’s videos show his level of expertise and professionalism, which aided me considerably in completing my project.

• • •

AUTHOR BIO

Mike Boden is a re-recording mixer at 20th Century Fox, where he has worked since 2005. With over twenty years of experience in the film industry, Mike has also held positions at Universal Pictures, Sony Entertainment, and several other smaller studios.

After college, Mike noticed that his mother’s home was in need of some serious repairs. Mike decided to tackle them himself, which served as a great entry point into the craft of woodworking and construction. From there he bought his first home in 2001 and embraced the opportunity to build many upgrades himself, which included a laundry room remodel, French doors, skylight, cedar closets, interior doors, landscaping, pergola, and much more.

Woodworking and construction offer Mike a gratifying counterpoint to sitting in a dark studio, mixing audio. Mike dreams of someday having his own dedicated woodshop instead of a shared garage.

When not working at the studio or on the house, Mike enjoys traveling with his wife, cooking, playing with his three dogs, and photography. His photography portfolio can be viewed at www.mikeboden.com.

My first thought was to rebuild the barn at my house. We don’t keep horses anymore, and the old barn is full of stuff I’ve saved that I  could just never bring myself to throw away—from ten years of doing JLC Live! shows, and remodels all over the county in north eastern Pennsylvania, where I live.

Emptying that barn would take longer than re-building it, and besides, a friend of mine that has a shop at his house said it is a pain because people—friends you never knew you had, if you catch my drift—are always stopping in looking for a favor, like: “I need six feet of this,” or “Can you help me fix this?” or “I don’t know how to make this?” My friend also said that it was hard to go out there to work sometimes, being so close to the house.

On top of that, I wasn’t sure what my shop would turn out to be. At one time, before I started making a living by the mile and was swinging a hammer full time, I thought about starting a business making dovetail drawers, or custom moldings, or fireplace mantles…you know what I mean.  And I figured if I ever did start a business and wanted to sell it, I wouldn’t have much to sell if the business and shop were in my barn.

So I started looking for a small piece of land—just the right spot. Not too far from home, but not too close, either. It took a lot of years to find. A piece came up a mile down the road. A real nice and well-known spot called “Milk Can Corners.” High-visibility for being out in the middle of nowhere, and on a paved road, too, if you can imagine—that’s big for my neck of the woods. The property also had 3-phase power, which I figured might be nice in the future—I had big plans for machinery.

I figured that, at the very least, the land was a good investment. Who knows, maybe in a hundred years they’ll put a strip mall on it. But I knew I’d always be able to sell it down the road, if I ever had to. And it was just the right distance from my home: far enough to make going there like going to work, but not so far that it wouldn’t be fun on a Sunday.

(Note: Click any image to enlarge)

Milk Can Corners is a pretty spot, with a large pond to the southeast, surrounded rolling hills that are blanketed with snow in the winter, and covered with hardwood trees that turn every color in the rainbow throughout the year—especially in the fall. Since the spot is so visible, I wanted the building to look nice for the neighbors (and for me, too) every time we drove by it—I didn’t want to build another steel pole barn.

So I started looking at barns for ideas—and I looked everywhere. I’m lucky to work with the Katz Roadshow. We travel all over the country. I found this place in the Pacific Northwest, while we were driving from Seattle up through the San Juan Islands. As soon as I saw this barn, I knew it was the one (see photo, left). I liked the peaked gable roofs, with the hay-pulleys. I don’t even know what you call that style, but that was it.

On the next few road trips, we worked on a SketchUp drawing. I wanted to use that Greek Revival trim design, too, the one Gary demonstrates at the roadshows. I’ve seen that same style of window and door trim on barns and buildings all around me (see photo, right), and figured it would look great on my shop—I wanted the building to fit right into the area where I live, as if it had always been there.

Then I started to over-think the building: could I build it so someone else could turn it in to a house? Or maybe a retail store—it was right at the intersection of two main county roads. I was going to put in a foundation and stick build. But after figuring the cost, I realized I couldn’t afford it: I live way out in the country and I’m in construction—there’s not a lot of money to be made out here.

A good friend of mine, Don Hohn, owns a construction company that does a lot of pole barns. He built them for farms, and commercial use, and for retail, too—some huge buildings. He told me that for basically the price of a foundation I could have half the materials of a pole barn. And he told me that if I was careful while building the barn, I could tighten up the cost even more by paying attention to all the small details. Plus, I could tighten up the building, too—seal it up and insulate it really well—which where I live is pretty important. It gets cold here!

So I built a pole barn with the idea that it could be a house some day. Don’t get me wrong: once you take this path, it does get costly—any building does. But I think it will all pay off down the road.

Don Hohn’s crew came and set all the poles first, and attached purlins on the walls. Then they straightened everything up with the same string-and-line and bracing we use for stick-built homes.

To carry the trusses, Lvls are set on top of the poles. The gable end poles run high to support the gable trusses at each end—trusses can rack and fall in high winds, especially tall ones, like my 8/12 pitch barn.

My trusses were also too tall to transport on the truck, so they shipped them in two pieces. Notice that the peak is missing on each truss.

The first truss is set on the outside of the poles and will line up with wall purlins in the same plane for siding. Notice the crew set a 2×6 every 4 ft. on center, sandwiched between the two lvl’s, with the roof pitch cut on top, so the 2×6 blocks wouldn’t stick up past the trusses. Setting the trusses was easy. Each truss was pulled tight against the blocks then nailed off.

Once all the main trusses were set, the tops were added, then the 24-in. ladder-type bracing for the overhangs. And, finally, the flying peaks were added—where farmers always attached a pulley to load hay into their barn.

I framed the dormers and cupola on the ground, and we set all of them with a crane, which made it easy to frame—a lot easier than working on the steep roof. The dormers were set right on top of the roof purlins. I framed in the window shafts later.

Before installing any metal on the main roof, we finished the dormers and cupola. It’s much easier to work on roof purlins than on slick metal roofing. But we made one mistake…we didn’t put the wrb (housewrap) over the side wall flashing. I regret that, a lot. There are several small leaks in my roof!

We set roof jacks and planks beneath the dormers, and laid a sheet of osb with cleats on it so we’d be more comfortable and have someplace to put stuff. If you ever worked on 24-in on-center purlins you know how sore your legs and hips get.

Roof almost done. Finishing up the ridge vent and fascia.

In the photo to the left, you can see us running the copper ground wire for the lighting rods along the ridge.

The roof is almost finished, and you can see the lightening rods and the weather vane, plus the pointed peaks on every gable, which give the whole building the look of an old barn. Yes, that’s snow falling! When you’re working on a metal roof, snow or rain can be very scary.

Carl Hagstrom (standing to my right) came out to help me set the windows. You can tell that I’m pretty happy to be a barn-builder and owner, but Carl’s wondering what he’s doing outside wearing a tool belt when the temperature is in the teens!

All the windows were framed in flush to the purlins, and were now ready for housewrap and board-and-bat siding.

We wrapped WRB on the outside of the purlins to help block wind and rain. Plus, having the WRB outside the purlins allowed me to spray insulation foam behind the posts. Rough-cut green lumber tends to split a lot as it dries, so we only nailed one side of the siding and batts, then let everything dry and shrink six to eight months before nailing off the second side. You’ll also notice that we set a temporary 2×4 with a laser right at the bottom of the siding, which made it easy to install the 16-ft. tall boards.

We installed flexible flashing for each sill pan, then applied a good bead of sealant. You can also see in this picture the 2×6 framing for each window opening—on the right side, the 2×6 framing box is secured to the horizontal purlins.

The WRB was cut flush with the window opening, then lifted up high enough to clear the window flashing. We ran sealant up the sides and across the top of each window, but left the bottoms open, so they’d drain.

Carl’s an old hand at window installs—I think he’s installed a few hundred windows just at building shows alone!  We checked for plumb, level, and sash function before securing each window.

Yes, it’s too cold to be installing adhesive flashing! We tried to keep it warm in the truck, but the weather wasn’t going to hold up my barn!

Here’s the finished shop! Well…almost. I was in a hurry, so I sided right over the front door and side window—I cut both of those in later.

Both the top hay door and hay dolly at the peak are in—they’re both dummies, just for looks. The siding boards are 16-ft. so we packed out the purlins with 1-in. batts, and ran the gable siding right over the top of the wall siding.

I hung a pair of sliding barn doors, so it would really look like an old barn, but I also installed a 2-in. insulated garage door behind it them. I got the rolling-door design from a building I saw in the neighborhood. I really like that raked top rail.

I really wanted the building to look like it had always been there, for a few hundred years. Once the fresh-cut lumber darkens, I think it will. The real crime is…I’m thinking about an addition.

Even before I insulated the walls, the WRB made for a great wind block—and a nice place to work while we poured the slab. Notice that the 16-ft. walls and scissor trusses leave plenty of room for a second floor or loft down the road.

On the inside, we poured a monolithic slab over 8 in. of tamped stone. I set a radiant-heat system in the slab, too, so the slab is insulated around the perimeter with 2-in. blue board 16 in. tall on the wall and 16 in. wide at the bottom of the pour.

We poured the perimeter 12-in. x 16-in. and the interior floor 6 in. thick with 3/8-in. insulation rolled under the whole slab.

We put expansion board around the poles, and hung the rebar wired to 20d, so it would stay in place while pouring.

After laying down all the tubes, I put 3/8 insulation over the tubes where we were going to cut the expansion joints. I don’t know why I did that…it just felt good.

Conduit protected the tubes where they came up out of the concrete—that way we wouldn’t slice through them while finishing the floor. We installed 8 loops, so none of the loops would be too long.

To protect the tubes during the pour, we wheeled the mud in on sheets of OSB. The heat tubes were pressurized, so if we had a leak we would know right away and could repair it right then and there. I had plenty friends to help.

I was especially lucky to have Harry Aldrich on the job that day. He’s one of the best flat-work guys in my area. He’s old school, with plenty of patience, and he does beautiful work.

My plumber and friend Keith Birchard stayed the whole day—just in case we hurt one of the tubes, he was there ready to repair it.

After the pour, I divided the slab in to 6 sections and cut the expansion joint about 1 1/2 in. deep, praying I wouldn’t hit any tubes that might have floated up in the concrete.

I dreamed for years of having a building and floor like this one.

• • •

AUTHOR BIO

Tom (right) letting loose at a machine gun rally in Kentucky.

Tom Brewer lives and tries to work in Northeastern Pennsylvania, but, unfortunately, he’s not home much, and has yet to set up his new shop! Tom travels about seven months out of every year as Road Manager for the Katz Roadshow.

Still, all that traveling has a few rewards. Steady work; touring historic homes and locations; and, occasionally, some real fun.

A client on a tight budget sent me a picture of an arch he wanted built in his family room. It looked a little complicated; it wasn’t a simple radius. That was my first encounter with an ellipse. He had found a millwork shop that would make the arch for a competitive price. As usual for that time, I was happy doing just the rough framing and installing the owner-supplied trim…at least until the piece showed up on the jobsite.

That old saying—if the price is too good, there is something wrong—proved true. What was supposed to be an elliptical arch looked more like someone traced a large garbage can lid and two coffee cans—an unsuccessful attempt at a three-centered arch.

I wasn’t pleased, and neither was my client. While it’s usually against my nature to criticize another craftsman’s work, I couldn’t tolerate that trim. It had to go. That is the moment I decided to learn more about the ellipse—to learn not only how to draw one, but how to make one.

I turned to George Collings’ Circular Work in Carpentry and Joinery and fell deeply into the mystery of circular work and its use in millwork. Along the way, I discovered that the ellipse can be the perfect form for arches on homes. It can be used in places where a segmental arch won’t look right, or when a semicircular arch won’t fit, like in flanking arched openings with different spans, or an arch with an extremely low rise. No matter what the height or width of an opening, the shape of an elliptical arch is always pleasing and consistent.

The Ellipse Defined

Even with Collings’ great book in my hands, understanding how to layout an ellipse wasn’t easy. Just look at this online definition: A curved line forming a closed loop, where the sum of the distances from two points (foci) to every point on the line is constant (source).

And here’s another description, with a formula, that I found online: A closed conic section shaped like a flattened circle and formed by an inclined plane that does not cut the base of the cone. Standard equation x2/a2 + y2/b2 = 1, where 2a and 2b are the lengths of the major and minor axes. Area: πab

Well, regardless of what Gary Katz says about my abilities with math, I’m not very good at understanding advanced equations. Like most carpenters, I need to get a handle on things—I need to get my hands on something tangible, something physical, in order to understand it.

The Basics

(Note: Click any image to enlarge)

The easiest way for me to explain what an ellipse looks like is to share a simple illustration of something that any carpenter can visualize. If you take a 4″ PVC pipe and cut it on your miter saw at a 90˚ angle, (zero on most miter saws!), the cut end of the pipe forms a circle with a 2 in. radius, and a diameter of 4 in.

If you cut that same pipe at an angle, by swinging the saw to 22 ½˚ or 45˚, the cut end of the pipe will form an ellipse. And the size and shape of the ellipse is mathematically predictable.

Drawing an Elliptical Arch (the string method)

A circle only has one axis—its diameter, but an ellipse has two: a large axis called the Major Axis, and a smaller one called the Minor Axis.

If we look at that piece of pipe we cut on the miter saw, the minor axis would be the diameter of the pipe.

If we are going to use this shape to create an arch, there are a few important features we need to identify in order to really understand an ellipse. Those features are:

•    The Rise and the Run of the arch (the Rise is one half of the minor axis, and the Run is equal to the major axis; since we are only using half of the ellipse)
•    The focal points

In the illustration of the ellipse we can see a few elements that define the shape. For a carpenter Rise and Run are more familiar, so I’ll use those terms instead of major axis and ½ the minor axis. The ellipse we will draw for this article will have a Run of 40 in. and a Rise of 14 in.

Starting with a horizontal baseline (the spring line of the arch), mark off the 40 in. in addition to the midpoint at 20 in. (see below).

Using a square, draw a perpendicular line from the midpoint of the arch’s Run to define the Rise of the arch. In this example the rise is 14 in. (see below).

Next, find the ellipse’s focal points by using a measurement of ½ the Run length (20 in. in this example) to strike a mark on the Run line measuring from the top of the rise line. I usually make a small story pole for this to make it easier.

Set two screws at each end of the Run, and then connect a non-stretch line or cable between the two points. This gives us a string with a measurement equal to the Run, the major axis of the ellipse.

NOW, move the cable connections to the focal points without changing the length of the string.

All that’s left to do is to stretch out the string and draw your ellipse. The shape that is drawn can be cut with a jig saw with a reasonable degree of accuracy for rough framing. I use this technique for drywall arches and for framing barrel ceilings and porches.

Note: An alternate method for setting the string length is to set a third screw at the top of the Rise line. Tightly stretch the string from a screw set at one focal point, over the height line screw, and secure it to a screw set at the opposite focal point. Next, remove the Rise line screw and use the string as a guide to trace the arch’s shape.

…

…

…

Cutting an Elliptical Arch with a Router

The string method is normally not accurate enough for molding and case work, at least not in my hands. I suppose there are carpenters with the skills and patience to make it work—but I prefer to use a router to cut my trim. The technique and layout may be different, but now that we understand the shape it’s really not hard at all.

First determine the layout of the arch—the Rise and the Run. These dimensions will determine how to set up the router’s trammel arm. The trammel that I use consists of a piece of 1/8 in. thick aluminum stock, and two sliding shower door rollers to act as pivots.

To set up the trammel, mount your router at one end of the trammel arm and drill a hole to allow the cutting bit to drop through. Measuring from the appropriate cutting side of the bit, mount the rollers along the trammel arm as shown below. To make life easy, I run a score line down the center of my trammel to help in layout. This allows me to locate the pivots on the center of the arm very quickly. Both rollers must be placed accurately in order to create a predetermined shape.

Now that the trammel is set up, we need to create a T-slot for the trammel’s pivot rollers to ride in. The T-slot is set along the Run line (the spring line of the arch) with its perpendicular slot centered on the Rise line. The width of the slots corresponds to the width of the pivot rollers being used. I create this T-slot by cutting two rectangles out of whatever scrap I happen to have onsite, and use the roller wheels, or spacers of the same width, to set the slot width. When everything is aligned and positioned correctly, I screw the pieces down to a backer board, including the piece I’m going to cut, which forms the top of the ‘Run’ track.

With the trammel rollers dropped into the slots, this jig cuts an almost perfect elliptical arch. There is no need to locate the focal points, the Rise and Run dimensions are constrained by the T-slot, and the geometry is automatically created.

…

You’ll find that once you make of a few of these, there are a lot of things you can do with the ellipse. Exterior ornaments, arched trim heads for bookcases, arched passage ways, and a host of other cool projects ….

The real trick is never to let your client see just how easy it is!

(SketchUp drawings by Wm. Todd Murdock)

(Note: Click any image to enlarge)

Without further chatter let’s build some doors.

First, to build doors from stock, you have to laminate your stiles and rails. I try to select vertical grain materials, and Douglas Fir is a perfect choice. The species is known for limited movement—great stability, and distinct hardness for a ‘softwood’. Maybe this is why a lot of wood doors are made from fir. When laminating, it’s best to orient your lumber so the grain is opposed.

I know that this photo doesn’t show that opposed grain as well as it might (see photo, right)—we got pretty lucky with vertical grain being…well…vertical. In fact, some of the grain actually turns at the end and runs the same direction as the piece it’s laminated to. I also inspect the lumber and will sometimes compromise the grain direction to bury a defect in the glue face.

Next we cut a dado into the stile and rail stock to accept the panels. I do this before I cut the rails to exact length because it saves some time. This is best done with a shaper, but I wanted to keep this job limited to tools most carpenters have access to, so I used the table saw.

Just a few passes and we have the dado. A few minutes with a sharp chisel and you have a nice clean bottom dado.

Next I cut the rails to length. Be sure to leave extra length for the tenons on each end.

There’s high dollar machinery for this step too, but I found the MFT table to be a fantastic substitute, with my OF1000 router riding on the rail with a ¾” dado bit making a single pass per tenon side.

Cutting the Tenons

This is an extremely simple setup. First, set your router up on the guide rail. Pick your shoulder cut location and set up a stop on the MFT fence. Dial in the cutter depth and then have at it. The MFT table and the router is almost too sweet for this use; the precision and repeat-ability insure each rail is exactly the same, which is critical for building doors square. This was one of those moments where the cost of a Festool product was immediately justified by the time it saved me to perform the task at hand. Tenoning each rail took less than two minutes each.

Excuse me while I take another sip of the green koolaid here…umm my, that is delicious.

Here’s a tip, don’t fuss too much trying to make your stile and rail mortise and tenon joints super snug—you don’t want it sloppy, but you shouldn’t have to fight it. Having a little wiggle space here makes the glue up a lot easier. Once the domino is added into the mix, those joints will tighten right up anyway.

I cut the mortise pockets in the rails first before cutting the tenons—so the rails would have square ends and better support for the Domino. But I’ll admit I also made those cuts first because I was over eager to fondle the XL. I cut the tenons later.

Once you have all your stiles and rails run, cut the panels. I had ¼” VG Fir ply in stock but I wanted a ½” panel so I laminated two sheets together with spray adhesive.

The secret weapon here is 3M Spray 90. I love this stuff.

Chamfer the edges of the panels and the tenons on the rails a little to help them slide into the stiles. I used the RO90 for this.

Honestly, this wasn’t intended to be a fix for Festool junkies, it just feels that way.

If you don’t round over the tenons a little, it’s nearly impossible to get everything together when you’re gluing up. Test your assembly before glue up—including your clamps. More than likely you will have to ‘tune up’ a rail or two to get all the joints tight. I always do. So, keep that MFT table setup until you finish the glue-up.

I like to keep my stiles long for extra clamping space, and so I don’t have to fuss around keeping them perfectly flush. It’s easy to cut the tails off with a tracksaw after the glue-up dries.

So How About that XL?

My first impression of the XL was that I was surprised at how small it was. I was expecting a behemoth given the sizes of the new dominoes, but it really isn’t a whole lot bigger than the 500. While similar in size, the units are completely different.

I prefer the ergonomics of the new XL much more than those on the 500. With the 500 I’ve found myself actually holding the cord end where it meets the machine for ‘keep it flat control’. But the XL seems to lock onto the material much better with a more positive forward placement of the front hand. From my perspective, the improved ergonomics provides a superior clamping force compared to the 500, which frees the back hand to focus on controlling the plunge cut.

Actually, both hand placements are an improvement, providing more consistent plunge cuts and improved control. This is really important given the length of these new dominoes. If you are off in plane between holes, these long behemoths won’t allow the joinery to close. I’ve had the same problem with the 500 but only when I wasn’t paying close enough attention and allowed the machine to sag off the work piece slightly. That ‘sagging’ is usually the only cause of misalignment.

The new hand positions resolve the issue, and now as long as I’m applying significant pressure to the front hand, the machine feels ‘locked’ down.

I also like that I can see the bit doing its thing. While you can’t actually see it plunge into the material, just being able to see it gives me a bit more confidence. You also don’t have to break the machine apart to change the bit. It makes it easier if you do, but I was able to make a change without taking off the front handle/table assembly.

The XL maxes out at 70mm deep, half the exact length of the 140mm dominoes. I had to use the 100mm dominoes because my mortise-and-tenon joint ate away a half inch (12-13mm) of depth.

Adding a little more plunge depth would be my first suggestion for an improvement on this new machine as it does shine in the construction of doors, and doors are traditionally going to have a mortise-and-tenon joint even if it is dowelled (or domino-ed).

The new XL offers two mortise widths versus the three widths available on the 500. Given the specialty nature of the XL, I expect this will be adequate; I found the 3MM oversize a nice amount of play when assembling the doors. I chose to bore a tight hole in the rails (horizontal members) and use the oversize hole for the stiles. This gave me a little wiggle adjustment between the three rails to accommodate any slight discrepancies in my markings, as well as allowed for adjustments for keep things square.

To adjust the width of the mortise you switch a lever (see photos, below).

Another improvement over the 500 is a easy-to-view display of the mortise width adjustment, right on top of the tool.

Wiggle room is definitely helpful during glue up: having the ability to tap the rails around a little may compromise strength a tad, but if you can’t get the door together perfectly square and flat, what’s the point of strength? Maximum strength can only be insured with tight holes on both sides.

If I had a bunch of doors to do, I might set up a story pole to be precise about placement, especially for heavier exterior doors where strength is more of an issue.

I discovered a few other tidbits during my introductory evaluation. First, let’s look at the positives.

The indexing pins on the front of the machine called ‘stop pins’ can now be locked up and out of the way.

There are now six stop pins, which provides a lot more options for spacing mortises, which is handy for quickly setting up repeatable cuts on a variety of projects.

And when you don’t need them, just push them up and they’ll click out of the way.

You’ll find the depth of mortise adjustment is different too—on the XL there’s a lot more settings available so you can dial in the exact depth of cut.

As I said before, it’s easy to remove the motor from the fence. In fact, it’s a little bit like the Domino 500. You use the same technique, with the wrench, but in this case you lift a lever that’s on the base of the fence (see photos, below).

Of course no review is good without a gripe or two. The doors I built were 1-3/8″ thick, standard thickness for interior doors—which is just under 35mm. Naturally, I wanted my mortises in the center of that dimension, at about 17.5mm. Unfortunately, you can’t dial in a custom mortise location based on the thickness of your stock. Like it’s smaller brother, hard stop settings are 10, 15, 20, 25, 30, and 40mm.

No, this isn't a picture of the Domino 500. It's the XL. They're almost identical.

I’m not a big fan of pre-set depth adjustment settings. Either what you want isn’t available, or—as happened in my biscuit-jointing past—vibration or an accidental drop knocks off the adjustment setting just enough to ruin the job. The good thing about the XL is that the stops are ‘positive,’ they won’t move accidentally.

To dial in the exact depth I wanted, I used a 2mm spacer placed under the faceplate of the tool set at 20mm. This put me .5mm off but it was close enough as long as I marked all the tops of the parts.

I admit I did a little rough handling test of the XLs friction lock for the height adjustment. Like I said before, the lock held solid. But I’d still like to see a micro adjustment device here, similar to a router, given the precise nature of the tool and how accuracy affects the success of the joint. Especially when—like in this example—you have a domino going through a secondary tenon and alignment is critical within that tenon.

And while I’m on a rant, given that this machine excels at door building, why not have centerline stops for 1-3/8″ (17.5mm) and 1-3/4″ doors (22mm)? Just sayin’.

But all gripes aside, watch the video below and marvel at an awesome machine, one that effortlessly cuts the most precise mortise pocket known to man. Visibility and seeing your layout marks is superb—same as the 500, which makes placement a snap.

Once I had all the holes cut and dominoes in place, I test fit the whole assembly again. Of course, this is when I discovered the depth loss in the mortise hole due to the height of the stile-and-rail tenon (you don’t think I figured that out before I made the cuts, do you?).

Once satisfied with the fit of all the joinery, I disassembled the whole door again and started the glue up.

I don’t use a whole lot of glue. In fact, I don’t apply any to the panels, so they’ll float and move. I apply a small bead of glue along the bottom of the tenon on each side.

The reason for this is squeeze out. These are stain grade doors, and glue wreaks havoc on stain grade. In this application, less is more. You should have a little consistent squeeze out on each seam. Nothing more. Let that squeeze out set up for an hour or two then carefully chisel or scrape it away while it is still a little pliable but not runny. When it comes time to finish sand the doors, you’ll be glad you waited.

And when you clamp up, be sure to have clamps on both sides of the door to create opposing clamping force. This keeps the clamp from pulling the door into a bow or a belly. Also, use a straight edge to insure your assembly stays flat. Adjust your clamps as needed.

Additionally, I lay full-length strips of 1/8-in. plywood beneath the clamps, to protect the door—which also makes sanding a little easier. It takes quite a bit of force to get everything tight with all that joinery going on. You don’t want to dent up your work with the clamps.

Allow setup to dry overnight, and enjoy a frosty beer while you watch the glue dry. If you’re inclined, fondle the XL a little more and marvel at its magnificence. It just saved you a ton of time. Once the glue has dried, unclamp and cut off your stile horns with the track saw, sand out imperfections and glue joints, then set up a router for hinge mortising.

While sipping a brew, I got to thinking of other ways the XL could benefit my operation. We’re a custom door shop, first and foremost, so anytime I’m building a door from scratch there’s no doubt the XL will have a part in it—for building doors, this unit is a phenomenal time saver. In fact, you could even build doors right on a jobsite, like our grandfathers did. Wait a minute…how did they do that without a Domino XL?

• • •

I don’t know about you, but I have a lot of drawers in my shop that are crammed with tools. It’s difficult to find stuff when I need it, and every time I open a drawer, I’m always worried that my sharp tools are banging around, getting dull or chipped. Especially my new lathe tools.

Jesse Wright sent me a text message recently with a photo of Kaizen Foam. He was writing a review for Tools of the Trade. The minute I saw the stuff, I knew it was the answer.

(Note: Click any image to enlarge)

I called Paul Akers at FastCap and ordered a sample package of each type along with the tools Paul recommended for marking and cutting. Kaizen Foam is not expensive—the thick material I used for my large lathe tools costs about $20.00 for a 2 ft. x 4 ft. sheet; the thinner foam is about half that much.

Paul also sent me a TriBlade utility knife—the type with the long replaceable blades, and a marking tool—one with a long-nose tip that makes tracing tools a snap, even when you’re drawing on black foam.

The marker comes with a long cap that protects the entire length of the tip. I was surprised at how dark a line it traced.

Kaizen Foam comes in several different thicknesses and styles. But they all share the same characteristic: each piece is made up from multiple laminations, so it’s easy to tear layers out and maintain a consistent depth—that’s the real secret to this innovative product.

I don’t have a lot of free time on my hands, so it was late at night, about 10:30, when I finally had a chance to use the foam in my shop. But working with this stuff isn’t brain surgery—it’s not something you have to do first thing in the morning. In only a few minutes, I had all my lathe tools organized perfectly. Now I’ve got my eye on a few other drawers.

Today, few homes lend themselves to such extravagant design, so the four-centered arch has largely been abandoned, except for high-end Tudor or Tudor Revival homes, which makes sense: the four-centered arch is often called a ‘Tudor Arch’ because of it’s origin in Jacobean architecture.

These bookcases are framed with four-centered arches. Though the entablature looks a little busy, this design might easily fit in a ‘library’ today. (Note: Click any image to enlarge.)

Four-centered arches were once found only in Gothic or Gothic revival homes, like Lyndhurst, in the Hudson River Valley.

The mirror in the Lyndhurst dining room over-mantle (see photo, above) is framed with a four-centered arch featuring finial-like tracery. This Gothic theme is continued in the flanking two-centered arches. A closer look also reveals a depressed four-centered arched doorway on the left.

The four-centered arch is not seen very often in modern homes, but when the style dictates, it can make a very dramatic statement. The compound curves of this type of arch can offer a regal feel to the space.

When defining an arch using four different arc centers, the possible configurations are almost endless. The shape of the arch can vary, even with the same span and rise.

Four-centered openings are often framed with a square surround. This creates a triangular-shaped space above the opening, and is called a 'spandrel.' It is often used as an ornament, featuring a decorative panel or carving.

The Gothic style is often described as the ‘Gothic order,’ as opposed to the classical orders—which explains why this entablature is included in an 18th century pattern book Gothic Architecture, by Batty Langley. Notice the pointed 4-centered ‘ogee’ arches decorating the frieze.

Pattern books often included proportional drawings and instructions on layout, like this four-centered doorway. Following the instructions isn’t always a simple task.

The Pseudo Four-Center

A variation of the four-centered arch is the ‘pseudo four-centered arch.’ This type of arch is often used on openings with a short rise. In this variation, the larger arcs that create the pointed top are replaced with straight lines that are tangent to the outer circular arcs.

A pseudo four-centered arched door decorates this walk-in cabinet. If you’ve been noticing recent trends in kitchen designs, then you’ll recognize the influence that Gothic architecture currently has on woodwork and appliance surrounds—especially stove hoods.

The pseudo four-centered arch framing the fire box of the mantelpiece below is subtle, but its Gothic influence makes a definite statement.

This inglenook at the Frederick Holland Day home in Norwood, MA is an extreme example of a pseudo four-centered arch.

Another home along the Hudson River Valley, Olana, built by the painter Frederic Church, is also decorated in the Gothic style, though the masonry and tile work—and colors—aren’t what you’d expect! These are examples of pseudo four-centered arches with a much greater rise.

As you can see from the examples above, the four-centered arch can be used in a variety of ways—not only to decorate a passageway or doorway. Certainly, elliptical and three-centered arches are more common than four-centered arches—most homes in America are based, in one way or another, on classical designs, not Gothic designs, which explains why four-centered arches are rarely used today—but they should be. And one of the reasons they aren’t used is because few carpenters know how to lay them out—especially when the arch proportions must be adjusted to fit an existing opening. Here are some Quick Reference Guides to help you.

The Classic Four-Centered Arch

This example uses only the width of the arch to determine proportion. The relationships between the four centers in this example are not the only ones possible, but are the most commonly used.

Four-Centered Arches with a Known Height & Width

The following steps will help you lay out a four-centered arch when you know the required height and width of the opening.

Pseudo Four-Centered Arches

The following procedure can be used for drawing out a pseudo four-centered arch. This variation is often used on openings with a very short rise. Trying to fit a traditional four-centered arch within these constraints can require radii that are very large and difficult to work with. Replacing the larger arcs with straight lines is much easier and creates a different feel.

I’ll preface this article by saying that my first job within this industry was at a small custom cabinet shop, so it’s fair to say that I’ve made a couple cabinets before. Why in the world would I go to cabinet making class when I’m totally satisfied with my process now?

There are always new ways to accomplish the tasks we’ve done for years. Some ideas we discard, some we embrace, and some we just glean a little tidbit from and add it to our bag of tricks. Plus, I have always been interested in how to make cabinets without a table saw. I was hoping this class would shed a little light on that subject, so I could make an informed decision on whether I should change or tweak my approach. Other than that, I really didn’t know what to expect when I walked into the class.

There were eight people in my group—a little more than normal—but, just by happenstance, we had not one but two great instructors, each of whom specialize in two very different aspects of the Festool system. Steve Bace, the head trainer at Festool Las Vegas, came from a deep background in solid surface fabrication before he joined the Festool family; and Brian Sedgeley, the head of the Indianapolis location, specializes in cabinet and furniture construction.

After a brief introduction from our trainers, and a five-minute roundtable with all the students, we walked into the ‘classroom’. If you are even vaguely familiar with the Festool line, let me tell you—walking through their door is the equivalent of being a kid and having Willy-Wonka open the gate to the chocolate factory.

(Note: Click any image to enlarge)

Everyone’s eyes got really wide and started to dilate; I think one guy even wept…no, it wasn’t me (okay, yes it was). Hey, tens of thousands of square feet of green and black might do it to you, too.

An MFT-3 was your desk, the Kapex stations and multiple guide rails were your assistants, and the one hundred systainers filled with every tool or accessory you could think of was the choir. We walked around for a few minutes, just taking it all in. Finally we gathered around and got to work.

They split us up into three groups and gave us our instructions. Each group was to create an FF lower cabinet and euro upper—both complete with 5mm adjustable shelf pin holes. The nice thing about this process is that you can set up this operation at your shop or on-site. Space is your only limitation; and, as you’ll see, you don’t always have to have a ton of space to do quality work. We were also encouraged to ask questions and work at our own pace. “We are not here to build two quality cabinets,” they insisted. “We are here so that when you leave you will have the knowledge to do this again one hundred times over, with confidence.” I really appreciated that approach.

We started by breaking down sheet goods for our carcasses.

They walked us through the proper way to set up a MFT…

…and attach the parallel guides for our long rips.

They really did walk us through a good deal of the Festool family—both old and new. Speaking for myself, it allowed me to learn the idiosyncrasies of these tools, some of which I had owned for a couple of years. For instance, I had no idea that if you measure from the bottom plate of the Domino to the center of the cutter, it is exactly 10mm. Why is that important? Because if you want to put a fixed shelf between two side panels that is not more than an inch or two from either end, that darn flip down plate is gonna be in the way. Knowing this little tidbit about the 10mm offset enables you to draw a line where your shelf (or shelves) will go, clamp a straight edge to your work piece, and your Domino will line up almost dead center of your 19mm shelf (did I mention this class dealt heavily in metric?). I never knew this trick, and the Domino was the first Festool tool I owned!

We also got a demonstration on the Kapex for sizing our dimensional stock to finished length, which was nice, because I was still trying to figure out if it was really worth the extra coin. Of course, the action was sweet, and the footprint was small, and dust collection was…you get the idea.

But I already knew these things. What really stood out to me was the laser and angle finder that comes with the unit. I had read about how accurate each of them was, but seeing was truly believing.

When you take a reading in (or around) a corner, you simply bring the jig back to the saw, turn on the laser…

…and align the center line on the jig with the dashed laser line. You have now established at what degree your miter will be cut.

My least-skilled employee could master this within minutes. I like anything that takes the guesswork or calculation out of the equation. If there’s an easy way for me to do something, while getting a quality result, without having to think about it too much, I’m in.

For routing dados, and for trimming edge banding, we turned to the MFK 700—Festool’s ‘little router’. I say that with a little bit of a chuckle, because with its nearly one horsepower motor, micro-adjustability, and wide offset bases, it’s really the router I reach for most often. It’s so comfortable in the hand, and is the only router that comes to mind with which you can place more than 50% of the base on your work. Most router bases are, of course, round, with the bit being dead center. If you are edge routing you will have nearly half your baseplate on the work and the other half, well, that’s up to you. With the offset base on the 700, you can keep the vast majority of the plate on your work—and the handle gives you added stability.

I haven’t even mentioned the other base that lets you do edge trimming. By that I mean flush trimming materials applied to edges. For this project we used iron-on birch edge banding, and, as we all know, you have to trim it. Where I would normally use an edge banding trimmer consisting of a couple plastic components with flush cutting blades and some springs in the middle, we used a router that was tilted almost 90 degrees. This stuff was really thin, so of course it trimmed easily; but the nice thing about this setup is that if we had glued on a piece of solid stock that was, say 1/2 in. thick, we could still have used this exact same setup to trim it.

Okay, back to the class.

For boring the 5mm adjustable shelf holes we used the LR 32. I’ve been eyeing this part of the system for a while, since I already had the OF 1010 router (the 1400 will work too, but the setup is slightly different), and my jig/drill approach did not always give me the best results. I also didn’t want to have a dedicated machine for drilling holes or for cup hinges. Blum makes a great press that will do both, but I don’t have space to lug one around with me. Tools that will do more than one job win out nearly every time in my book.

I will say that I was a little bit lost during the initial setup, but the nice thing about this class is that they don’t just describe how things are setup—they tell you how, then they let you do it. You can’t beat hands-on experience. Any questions are answered along the way, and, in the end, you feel very confident in what you are doing.

We discussed layout, the standard backset of shelf holes, distance from top or bottom of panels for hinge plates, and whether or not a stop- or through-hole is needed. Once you’ve figured out the setup, the actual boring is a breeze—and really fast, too. You can almost race through the process and still have an absolutely clean hole every single time. (Just as a side note, after this class I went home and bought this system, and I’ve used it a bunch of times already. Melamine, plywood (both pre- and unfinished), and MDF all respond the same: Flawless every time.)

On the second day, we finished our projects about ninety minutes early, and were given two options: leave early, or stay and putter around the shop. Duh, which one do you think we chose?

This is where individual questions and opinions started coming out. One student kept mentioning how much he loved his RAS and the instructor echoed how much he liked it as well. After a couple more mentions of this tool I finally blurted out, “I’m sorry, what’s a RAS?” The room grew quiet, and all eyes turned to me. Imagine crickets chirping. Finally, the instructor said, “Let me show you.” I felt a little bit better when over half the class told me that they had no idea what a RAS was, either.

They brought out the Rotary Sander/Grinder/’Animal’ they called the RAS, drew an arbitrary ‘scribe line’ on a piece of plywood—sort of like what you might encounter on a filler strip butting to an irregular wall—and started in on it. Holy cow, this thing found the line as quickly as a bloodhound, and followed the trail all the way to the end in a minute. And we removed a lot of material. I couldn’t believe the dust collection either. All that stock was sanded off and without any distinguishable dust. Wow.

As I mentioned earlier, Steve Bace comes from a solid surface background. He had mentioned that the OF 2200 was what he used to rout complex edges on some of his material. When someone spotted a chunk of solid surface that was nearly 2 in. thick, the question was posed, “Can you show us how to rout that?” He immediately pulled out a complex detail router bit that was nearly the size of a baseball. “This router can put a clean edge on this stock in one pass.” I think he saw the look of disbelief in my eyes, which is why he looked at me and said, “and Matt’s gonna show us.” What?!? I’ve never used this router before, and all my instincts were telling me that this was not a good idea. But I had to step up, right?

The first thing I noticed when I grabbed this tool is that it’s HEAVY. Actually, 17.2 lbs. to be exact. My first thought was, “talk about user fatigue….” But then, with the enormous bit raised above the baseplate, I set the router on the table, turned it on, and let go (don’t try that at home; the instructor was right next to me). It just sat there humming and didn’t move a millimeter. The weight is actually comforting, because you know there is not only a lot of power, but a lot of beef, too. When I finally began to rout I didn’t even realize that the bit had contacted the edge until the bearing came into contact with it. I moved forward cautiously, but soon realized that the only thing slowing me down was me. This machine carved right through like a champ, and with the dust collection on, the air was as clean as a whistle.

Every question we could think to ask was not only answered, but also demonstrated, when possible. I walked out of those two days with new ideas on how to use my tools (and also with a few more on my wish list).

In the end, did I totally revamp the way I do things? Not really. I have a slightly different approach when it comes to cabinet backs, but I still like my way of joining face frames together. I did walk away with a bunch of new ideas on how to do things that I might not have ever thought of. Some apply to cabinets, some to woodworking and trim carpentry, others I may never use, but I now have the knowledge in my skill set, should the need ever arise for me to use it.

This class also re-instilled the ‘systems approach’: Get a system and use it—whether it is tools, techniques, software, whatever. Figure out the best way that meets your needs, and then repeat, repeat, repeat. I was a believer long before I took this class, but I have an even deeper appreciation for it after those two days.

If you are thinking of attending one of these classes—do it (and if the class is full, send Festool an email and tell them you want to go. My guess is that if they are inundated with requests they may be forced to add a class or two). You may hear some things that you already know, but odds are you will also hear some that you don’t. You may even find that spark of genius that inspires you to try something new and innovative.

http://festooltraining.com/

——————-

Be sure to read Part 1 of this series on arches: Circular-Based Arches

——————-

This depressed type of arch, like the Segmental and Drop arch, can be used when the design requires the rise—or height—of the arch to be reduced. While segmental and ‘elliptical shaped’ arches both share a rounded top, the elliptical variation provides the benefit of a clean vertical transition, and respects traditional design principles.

A three-centered arch is an elliptical approximation using three tangent arcs. (Click any image to enlarge.)

A true ellipse is the shape created by making a diagonal section-cut through a cone or cylinder. The ellipse has two focal points and a constantly changing arc radius.

It can be difficult to determine if an arch is a true ellipse, or just one composed of simple tangent arcs, swung from three centers. Either way, elliptically shaped arches are more commonly found in traditional homes based on colonial styles—though their use depends more upon the skill of the architects, millwrights, and finish carpenters.

I’ve heard some carpenters say (and I won’t mention any names!) that the popularity of segmental arches—sometimes one of the most boring and ugly forms of architecture—results more from a lack of knowledge and technique than from an understanding of classical forms—both Gothic and Colonial.

These carpenters believe that elliptical arches—or, at the very least, three-centered arches—are far more attractive, but that the technique is beyond the skill of most contemporary carpenters. I don’t necessarily agree. I don’t think the segmental arch should be completely avoided.

A segmented arch forms a pleasing and handsome frame, as long as the arches (the rise, the radius, the span) are nearly identical in size.

In the first part of this series, I shared some images of segmented arches gone wrong. But, when designed and executed properly, a segmented arch forms a pleasing and handsome frame, as long as the arches (the rise, the radius, the span) are nearly identical in size. But, if the openings have variable spans, a three-centered arch is a better answer!

At this point, I can’t help but mention Gary Striegler’s article in JLC about building an arched passage door. I’m including a PDF of that article here. It’s a critical part of this study, both because it will help readers form a better understanding of complex arches (arches with more than two centers, and elliptical arches), and because Gary’s article provides techniques for constructing a three-centered arch, which is much easier than milling elliptical molding! In fact, mill shops often use a similar technique to create their elliptical moldings, sometimes using five or more centers to create a more accurate elliptical shape.

This coffered three-centered arch passageway features raised panels and uses the Ionic capitals of the pilasters as imposts to provide visual strength and support.

Another example of where a three-centered arch is easier on the carpenter, as opposed to a true ellipse, is in a coffered passageway. The curved panels of the head only require two different radii. In the photo to the right, you can see that the panels across the top share the same curvature, and panels with a tighter radius are used as the arch terminates on each side.

Getting back to the purpose of this article—how do we layout this pseudo ellipse? Well…it all depends on what you are given to work with. Although being involved at the planning stages is ideal, most of the time it’s not a reality.

Hopefully, the following Quick Reference Guides will help you deal with any ‘curve’ you’re thrown.

The Classic Three-Centered Arch

This layout is for the classic three-centered arch. You only need to know the required width or span of the arch. The rise of the arch will be determined by proportion only.

.

Three-Centered Arches with a Known Height & Width

This layout is used when you must fit an arch within a predetermined height and width.

.

Three-Centered Arches with Known Radii

This layout is used for creating a three-centered arch when the two radii to be used are predetermined. This is the situation used in Gary Striegler’s article.

.

Keep an eye out for the last part in this series, on Four-Centered Arches!

One of the main culprits in the attack against space is the pesky table saw. While it is an essential tool, the portable table saw takes up the largest chunk of real estate, whether in use or packed away. Most carpenters I know are always trying to find a smaller table saw—but we’re also loath to sacrifice quality. After all, a table saw isn’t worth a nickel if it won’t cut well or operate safely.

(Note: Click any image to enlarge)

This article will focus on two of the smallest table saws out there: the Bosch GTS1031 (52 lbs.) and the DeWalt DW745 (45 lbs.). I wanted to see if these saws were up to a real-world challenge on a jobsite, or if they were simply designed for the occasional DIY project. Ironically, a lot of the carpenters I’ve been working with, and we have a good-size crew, have been interested in the results of my head-to-head study; in fact, many of them participated in this review.

Most portable table saws these days are pretty much a standard size, and many manufacturers offer some sort of collapsible-wheeled stand as an accessory. Wheeled stands are great if you have a step-van or a trailer—and an endless amount of available space. But if you’re working out of a regular van or pickup truck, you’ll have to start making serious sacrifices with the tools you carry when you decide to load your table saw. And if you do load your table saw, you’d better have help!

Table saws mounted to wheeled stands weigh over 100 lbs., more weight then I like to lift twice a day, alone. Some carpenters still swear by these stands, and I suppose I might, too, if I worked on small jobs where my tool setup was always close to my vehicle. But I work on large jobsites, on high-end custom homes, and some days I see my truck only twice a day. My on-the-job shop varies from a basement wine cellar to a third-floor master suite. And the grounds are always torn up with trenches, concrete work, and landscapers. Wheeling a saw stand around is not an option.

At the same time, portable table saws are too small to really work on, even if you’re just ripping trim and shelving. And for cutting cabinet parts, they’re nearly worthless. That’s why, for this review, I tested both ‘compact’ portable saws using a Rousseau 2745 table-saw stand with an out-feed table.

A little about the Rousseau stand: right out of the box I had issues. First, of the eight screws that secure the table top, two fell out when I turned it right side up, and two more were stripped! Those aren’t very good odds. There was also welding slag left on the main crossbar that impeded the fence from sliding smoothly and functioning properly. In order to get the fence to work, I had to sand off the little metal beads under the powder coating, which, of course, removed the finish. I was not impressed to say the least, especially since the stand costs just as much as one of these saws.

To Rousseau’s credit, when I brought this to their attention, they sent out a replacement stand. In fact, my note to them sparked a full-on company meeting and review of quality control issues. They thanked me for criticizing their stand! I’d like to see more companies step up and take responsibility for their products the same way. Believe me, if you ever have an issue with a Rousseau product, you can expect to get good service.

Now, back to how this affects the saws and their levels of performance. Rousseau stands provide portable saws—especially these new compact saws—a much larger work surface, greater stability, and improved safety. The stands come with a shop-saw-style rip fence, and there are a multitude of add-ons and modifications you can also purchase to suit your needs. I used the Rousseau 2720 out-feed table to go along with my stand.

Blades

Now, on to the saws.

My first suggestion when it comes to these portable saws is to remove the factory-supplied blade and go buy a good blade! You can keep the original blade around for those times when you need a sacrificial blade—when you know there are nails or something that might ruin a good blade. And while I’m on the subject, never buy a thin-kerf blade. I know that saw manufacturers recommend thin-kerf blades for these saws because the motors aren’t nearly as powerful as a shop saw, but, honestly, most of the work we do with a small portable saw is ripping trim material—not a lot of 8/4 hardwood.

Both of these saws have more than enough power to run a full-width saw blade. If you’re running a thin-kerf blade to save material…well, I honestly don’t think you’ll ever save enough material to make it worth your while. The biggest headache of a thin-kerf blade is deflection. When I’m cutting hardwoods—sometimes even when I’m ripping softwood—and I want to rip off anything under 1/8 in., deflection really pisses me off. And I’m often trying to rip off less than 1/16 in.!

So for this review, I ended up using three different blades: I tried a Ridgid Titanium 50 tooth blade and a Forrest Woodworker blade on both saws, in addition to the factory supplied blades. I actually liked the less-expensive Ridgid blade in the Dewalt more than the thin-kerf Forrest. But, in the Bosch both alternate blades seemed to wobble more than the original blade so we used the factory-supplied blade in the 1031. 

Multi-purpose tools?

While I am on this rant, I’m also not a believer in making your out-feed table a multi-purpose Swiss-army knife. I see a lot of carpenters installing everything from router inserts to accessory clamps in their out-feed tables (sorry, Gary!). I may be the only one—and I apologize if I’m insulting all the other out-feed table fanatics—but maybe I’m the only unfortunate soul that runs into that open router hole, or that slightly proud lip or screw, while I’m making a delicate and expensive rip.

Hang-ups like that also create a dangerous situation when you have a spinning blade, binding material, and irreplaceable fingers. I know it’s tempting—after all, just look at all that free space! But unless you are extremely diligent about making everything absolutely flush, unless you use solid router inserts every time you rip, you could be putting yourself in a dangerous situation. Okay, that’s enough lecturing for today.

The Bosch GTS 1031

TiC has already examined the DeWalt 745, so let’s look closely at its rival. The Bosch compact saw has many of the same features—after all, manufacturers are beginning to recognize the importance of these details.

A large paddle-switch makes it easy to turn the saw on, and especially easy and fast to turn the saw off!

Like the DeWalt saw, the Bosch table extends to the right. Lift the lever to slide the table out, then lock the lever by pressing it back down.

On the DeWalt 745 the fence slides out on a cool rack-and-pinon gear, but relies on a Rube-Goldberg flip-over arm to support the stock. But on the Bosch extension system a small section of the table actually slides out. There’s no rack-and-pinion control, but there is good support for the workpiece. Of course, on our jobsite, we rarely used these fences because the saws were mounted in Rousseau stands.

Riving Knives

Riving knives are now required accessories on all saws—the days of having to remodel a saw guard and make your own riving knife are fortunately over. Like most carpenters, I’ve grown to like riving knives so much—and have learned to rely on how well they prevent kickback—that I’m reluctant to use a saw without a riving knife. You should be, too.

Both the DeWalt and the Bosch come with similar guard systems. The lever that releases the riving knife on the Bosch saw is slightly larger than the DeWalt’s, but it’s not painted yellow, so it’s harder to see in this photo.

The riving knife is really nothing more than the splitter, stripped of the guard and anti-kick back pawl, with the height adjusted to about 1/8 in. to 1/4 in. below the teeth on the blade.

Once the riving knife is lifted to its highest position, the guard slips onto the front…

…and the anti-kickback pawl snaps on to the back.

I’ll be the first to raise my hand and admit the truth: Like most carpenters, our crew rarely used the saw with the full guard in place. We like to see the blade—there’s no other way to make precise measurements. In fact, most of the time we put on the guard only when we heard the jobsite safety inspector was around the corner—and at those moments, it was nice that the guard installs so easily and so quickly. (Yes, on some of our jobs, there’s a safety inspector! For insurance and liability purposes, many large contractors have an OSHA-style inspector that will fine companies for frayed cords, not having guards on saws, pinned back safeties, etc.)

The guard stores beneath the saw. A flick of the finger releases it.

The anti-kickback pawl stores beneath the saw, too, and snaps in securely.

• • •

Accessory Storage & Handles

While we’re looking at the bottom of the saw, notice that the whole base of the saw is protected by a roll-bar cage. That may be the reason the Bosch weighs 7 lbs. more than the DeWalt, but it is good protection.

The cage provides a secure handle for lifting and carrying the saw.

Comfortable handles are also installed at the top of the saw, on both sides of the table.

Handy cord storage can be found beneath the back of the saw, inside the cage.

I really appreciate how saw manufacturers are thinking more about the problems we face with tool accessories. Even the miter gauge—which I never use—stores beneath the table, at the back of the saw.

And if you use the rip fence, it can also be stored upside down beneath the table. Unfortunately, in that position, the rip fence interferes with dropping the saw into a Rousseau stand, but hey, you can’t expect to win every time!

A stout push stick also stores on the side of the saw, within easy reach.

Head-to-Head on the Jobsite

Remember, we used these saws one at a time. And they were often the only table saw on the jobsite. So we used each saw a lot—sometimes asking a little too much of it. But that’s reality, right? My overall impression of the Bosch saw is that it’s okay. I’m not the type to bash anyone or anything, but I tried two different blades on the saw and they both had a serious wobble—and one of them was the blade that came with the saw. The wobble was especially noticeable on startup, and although it straightened out—or seemed to—I wasn’t happy with how it left the edges of the stock: rough, and often with saw marks, which meant extra work cleaning up edges that wouldn’t normally need that kind of effort. It was pretty disappointing.

The Bosch saw bevels past 0 and 45 degrees, which is very handy!

But the lock/unlock lever is a knuckle-buster at the 45 degree angle. You can’t release the lock without bashing your knuckles into the table saw extension release lever. I guess that’s another price we pay for compact portable saws.

I like the cool riving knife system, but more dust seems to fly in your face than out the back port, especially if you don’t have a vacuum hooked up—which is another thing to carry, and another reason why a wheeled saw stand doesn’t work for me. Overall, I don’t have a lot of great things to say about the Bosch. It’s a mediocre tool, a judgment reflected by the voices of my other crew members: They all asked if I could bring the DeWalt back. That about sums it up.

Top Pick – DeWalt

We started working with the DeWalt, and in the end we went back to it. I didn’t play easy with this saw just because of its size, and neither did the other guys on our crew. Like I said, it was often the only table saw on the job site, so it was used for everything from making custom plinth blocks out of 8/4 hardwood to ripping sheet goods down to size.

Overall, both saws are loud. Hearing protection is a must when using either of these tools. Gary tested the decibels and found that the Bosch was slightly louder.

Both saws have 15 amp motors and work just fine for what I would call “standard” ripping, but both struggled somewhat with thicker hardwoods. Their ripping capacities without the Rousseau stand are limited. You could, of course, supplement this with a track saw, but that means finding space for it. And that is what this article is about: finding tools that work within our confined spaces—both on the jobsite and in our vehicles. Do we have to sacrifice space for function? And what exactly is the sacrifice?

The truth is, I made both of these saws work for me on cramped jobsites for over a month each, and our work is demanding. In the end, the DeWalt won the war. For those of us with limited space, this saw is a viable option. But that doesn’t mean the DeWalt won with acclaim. The saw has some deficiencies that might really bother a fanatic. Two black metal tabs at the rear of the blade insert are not flush with the insert. I had to tape over those pieces to stop wood from catching on the proud lip.

The DeWalt has some serious plusses, too, like the rack-and-pinion fence, though unfortunately, because I used the Rousseau stand, I didn’t get to use the best feature on the saw—it’s very easy to make accurate adjustments in small increments with that fence!

But the DeWalt is a good little saw. It handled everything we threw at it. Sure, there were a few hiccups, like burn marks and chatter, and scant power at times—when we really pushed the little guy. And really, given the price, size, and weight of the saw, all of these complaints are minor; they should be expected. Call me a pessimist, if you will, but I don’t expect cabinet-saw performance from a portable unit. In my opinion, for the money (and for the size!), this little DeWalt saw performs just fine, even on the very demanding jobs where I work—where installations are often unacceptable if they’re off by 1/32 in.

And I have to say this in support of both Bosch and DeWalt: I think manufacturers are starting to catch on that people like us make our living out of the back of a pickup, a van, or half of a garage, and we need all the help we can get!

• • •

AUTHOR BIO

Michael Inskeep is a foreman at Millworks By Design in southern California. As a young man he realized he had a talent for creating things, which grew into a love for building furniture, painting, drawing, and making music. As a professional carpenter, he cut his teeth building stairs. From there he made the transition to other aspects of finish carpentry. Along the way Michael had the fortune to work with some exceptional carpenters who taught him a few “tricks of the trade.” He also enjoys passing those “tricks” on to others who are willing to learn. His attention to detail, and ability to learn quickly, have led him to work on some of the largest and best projects in southern California. But, at the end of the day, his true passions are his two baby boys. The smiles on their faces make all the stress of deadlines and dust worthwhile!