Courtyard Gate – #4 Gluing up Stiles & Lower Panels, Making Latilla Sample

Gluing up the three pieces to make the stile blanks are next.  The process is the same as with the rails except these assemblies are about two and a half times longer and I can only do one at a time so with curing time it takes me four days to glue all of them up.

Once glued up all the stiles get the same cleanup and trimming as the rails so no need to go through that again.  Here’s the set of the four stiles plus the six rails needed with their edges trimmed down close to the final width.  Just in case a problem shows up I am going to let these blanks set for a few days before moving on to the next step.

Next is to edge glue two pieces together for each of the three boards needed for the lower door panels shown highlighted in the drawing.  After cutting the parts oversize they get glued up.  Shown in the right photo are three panels glued along with the remaining three cut and ready to be glued.

The upper section upper section of the gate is mostly open but has three latillas in each gate section shown highlighted in the drawing.  I wanted to use cedar rather than pine for rot resistance but have been unable to locate anything other than pine ones.  It would be easy to glue up three or four layers of the cedar (top right photo) and turn them down on the lathe to a straight cylinder (bottom right photo) or a decorative turning like a stair baluster.  However, neither of those fit in with the desired southwestern design. 

As an alternate I will do some testing to see if I can create the hand peeled/hewn look starting with the cylinder using different methods.  Below are different tools to try.  Some may work and some or all may not work.  Starting at the top left corner working clockwise is my power plane, angle grinder with carbide teeth and a coarse grit sanding disk that can also mount on the grinder, a pneumatic die grinder with carbide burr, three wood rasps a hand plane and pneumatic impact chisel.

To replicate the hand peeled/hewn look on a turned cylinder I needed a faceted layout to work from and chose six faces.  To do the layout a compass is set to the cylinder’s radius and with the point positioned on the edge an arc is drawn through the center.  The point where that arc intersects the cylinder’s perimeter is my center point for the next arc.  After that the intersecting points along the edge are connected with a straight line.  Result is the layout in the top photo.  The bottom photo shows the cylinder set back in the lathe and has straight lines drawn on the face to act as my guide.

My first tool was the pneumatic impact chisel and it didn’t work at all because the blade is free to rotate and I couldn’t control the angle of the cut.  The power plane and hand plane gave me a flat smooth surface that looked too flat, precise and not organic enough.  The pneumatic die grinder with carbide burr gave me the ability to create a more organic surface but the result was too rough as in the top photo below.  Same result with the hand rasps.  The angle grinder with carbide teeth gave me the needed organic look but left a surface with its teeth marks and the surface was not smooth enough.  Changing the carbide cutter to a coarse grit sanding disk helped but still left unacceptable sanding arcs (bottom photo) in the finished surface.

I didn’t have a finer grit sanding disk for the angle grinder so took a 120-grit disk made for a random orbital sander and cut it to fit in the angle grinder.   The good news is it gave me just what I wanted for a finish shown in the top photo but the 120-grit disk is not nearly sturdy enough for the purpose.  Knowing the grit needed I got a 120-grit disk (right photo) purpose made for use with the grinder.

With a successful test the leftovers from making the stiles and rails get cut down to make the six four-layer latilla blanks needed.  Here are three blanks glued up along with the three waiting for the first group’s glue to cure before they get glued together. 

Once the second set of blanks get glued the glue runs are cleaned off, trimmed like the stiles and rails then set aside until needed later.  At this point I think that all of the laminating needed for the project is done.  There has been a lot of time spent to go from the rough sawn cedar fence boards to smooth, straight, flat and with parallel edges then gluing all the pieces together.  As a matter of fact, it has taken three plus quarts of waterproof Type III glue to get to this point.

Because I stopped surfacing the rough sawn planks when a smooth surface was achieved the thickness varied from .55” to .48” thick.  Mixing and matching them gave me stile and rail glue-ups all about four hundredths of an inch over the nominal target of 1 ½” thick.    My next step is to run these pieces through the thickness sander to get them all to a consistent thickness.   The thickness sander is the tool to do that but I noticed while making the File Handles & Case project the sander did not give parallel surfaces.  The problem was the edge nearest the motor was slightly thinner than the outboard edge.  This isn’t a problem as my old thickness sander had to be adjusted too.  Process is to take a couple of spacers the same thickness (I used two 3/8” drill bits) and adjust things until the bed and drum were parallel.  The photos show the bits set in place for adjustment.  When they both just touch the drum the bed and drum are parallel.  

When making the adjustment a problem showed up in that there was not enough adjustment range to bring the drum and bed parallel.  The solution was to add a shim (red arrow) to get the drum and bed almost parallel then fine tune with the machine’s built-in adjustment.  In the end I got the difference down to two hundredths of an inch.

Next Up – Cutting Pieces to Length, Sanding to Thickness & Making Mortises

Courtyard Gate – #3 Edge Trimming, a Twisted off Screw & Gluing up Rails

The boards are left to rest for a few days flat with spacers for air movement to see if any of them moved and all are smooth, flat and most have one true edge.  They then get sorted into stacks by my estimation of how wide they will be after the one rough edge gets cut parallel to the routed edge.  The photo shows them broken down in 1/16” steps from 5 3/8” wide on the left and 5” wide on the right. 

To rip the second long edge parallel to the routed edge the table saw is used.  As I said above most of the boards were fine but a few of the edges had bowed a tiny bit, less than 1/16” while they sat.  To help mitigate those inconsistencies in the routed edge the 8’ long level gets clamped to the fence and the routed edge is run against that.

With all the boards now having their two long edges straight and parallel I can finally start cutting blanks that will be face glued together to make the required 1 ½” thick pieces.  The first group are the top, bottom and middle rails shown highlighted in the top left drawing.  The top right photo shows a typical set of boards cut slightly oversize to allow for trimming.  The bottom photo shows the 12 pieces needed to make the top and bottom rails for both doors.  The center rail pieces are not shown.

To keep the three layers aligned while they get glued up a screw gets added at each end about ½” in from the edge.  A stop block and fence locates the hole so I don’t have to measure each one.  The top left photo show shows that setup and the lower left photo is a closer view.  The right photo is of the screw installed after countersinking the hole.

Because the gate will be outside a waterproof glue is used to prevent glue failure.  Here are the three pieces that make up one of the top or bottom rails.  The black thing next to the glue bottle is a toothed glue spreader that spreads the glue to a consistent thickness much faster than can be done otherwise.  Spreading the glue isn’t shown as it’s always a mad rush to get the glue on, pieces aligned and clamped within the glue’s working time.  It’s going to be more fun getting this done for the stiles when each of the boards are almost three times longer.

This is what two of the rails look like glued up.  First a 6-mil piece of plastic the glue does not stick to goes down so the assembly doesn’t glue itself to the work bench.  The glued three-layer assembly with the screws run in for alignment is next.  That’s followed by a thick flat plank to spread the clamp’s pressure evenly across the assembly.  Last four heavy duty clamps are added and tightened down until the boards cry uncle.

The first set of rails went fine however in the second batch one of the screws used for holding the boards in alignment twisted off when I went to remove it.  This leaves me with a real problem in that my planned saw cut for trimming to length goes right through the center of the hole now occupied by a broken off screw.  Removing the broken part of the screw was going to be a problem no matter how it got done.  After some thought I decided to work my way from the top down using a wood chisel until I could see the end of the screw then carefully remove the wood until the screw was exposed.  The top photo shows one of the screws that cooperated when removed and the other is the broken one.  The bottom photo is a closer view of the broken screw that now can be removed.  I believe the problem was when the screw was run in after the glue was applied the threads got coated and the screw got glued in place.  In the future the screw threads will get coated with wax and removed an hour after being clamped while the glue has not yet fully cured.

After an overnight cure the clamps are removed and the glue mostly cleaned off with a paint scraper.  The bottom photo shows the edge cleaned up (red arrow) so none of the remaining glue sticks out beyond the clean top board’s edge.  That edge will act as my true straight edge when it’s run through the table saw to trim the opposite edge.  The piece will then be flipped and the first reference edge trimmed so all the pieces are flush with the assembly and a little oversize.  There is one more set of rails to do and then it’s on to the stiles or vertical frame pieces.

First, step for the stiles is to cut them to rough length on the chop saw but before doing that I screwed a fence extension to the top of my workbench to the left of the saw to make sure that long pieces were square with the saw blade.  This had been on my to-do list for a long time and now seemed like a good time to add it.  The top photo shows the finished MDF piece (red arrow) screwed down.  To make sure it’s in-line with the fence attached to the saw I clamped my 8’ level to the saw’s fence then clamped the MDF extension to the level and lastly screwed the extension to the workbench’s top.

The new fence to the left of the saw’s blade is used when the right end is cut square. 

That’s followed by sliding the board to the right where a stop block (red arrow) is set so left end can be cut.  This two-step process gives me square cuts on both ends and makes all pieces the same length.

The table saw is then used to cut the boards to their oversize width trimming off knots, or other bad bits off the edges.  The red arrows point to the knot that gets cut away and the insert is a closer view.  Sometimes only one edge needs to be cut other times there are flaws on both sides that need to be removed.  When done almost all the pieces are clear wood with no defects at all.

Next Up – Gluing up Stiles & Lower Panels, Making Latilla Sample

Courtyard Gate – #2 Surfacing & Truing One Edge

This is a view of the shop setup to be used for running the boards through the planer.  The red arrow points to the planer.  To its right is a red large shop vac that will be connected to the planer for chip collection.  The blue roller stand in the foreground will support the cedar on the infeed side and the yellow roller stand in the background will provide support as the cedar exits the planer.

Preparation before surfacing is to check the boards for a few things.  First, is to look for loose knots (just one found) and remove.  Second, is to check for any staples or metal.  Each of the boards had a plastic tag stapled on so all those get removed.  There was also one staple on the face of one board and it got removed.  Reason I am so careful is that planer knives spinning at 8,000 RPM’s hitting a nail or metal staple nicks the knife chipping the knife’s edge leaving a ridge on the surfaced board the size of the chip which is not desirable.  There are a few fixes but none quick or cheap so it pays to carefully check before running the boards through the planer.  Third, is to check and see if the board has a cup.  If it does the crown goes up as shown in the drawing below.  That way the outer edges will ride on the planer’s base keeping the board from rocking.  The bottom photo shows a board after the first pass through the planer with the original rough surface on the left and the fresh cut surface at the right.  Another pass or two through the planer will flatten this face getting rid of the rough surface.

Because you can always plane a board thinner but you can’t add material back on if too much it taken off very thin passes of about 1/64” get removed.  As I was working through the stack unknown to me there were four boards that had split as they had dried.  On the left are two of them and the others are similar.  Because most of the board is in good shape I just cut off the split part then ran them through the planer with no problems.  After making several passes getting the first face about 95% flat I flipped the boards over, made a couple of passes then called it a day.  All in all, it took about four hours of preparation and planer work to get to this point.  While doing some cleanup I checked the to see how full the 32-gallon shop-vac was since before starting it had been emptied and to my surprise it was completely full as shown in the right photo. 

Once all the boards have one rough surface flattened, smoothed and the second face nearly smoothed establishing one straight edge is next.  I could have continued surfacing but wanted a break from it.  All of the boards have slightly uneven rough sawn edges and most of them are slightly curved.  Usually, a straight edge is created on the jointer or by attaching my 8’ level to the table saw fence and running the board against that.  There is a common problem with either method in that the reference surface on the jointer or the 8’ level is too short to fully engage the 6’ board’s length both before and after the cutting point.  Ideally a 12’ long truly straight edge is needed.   My fix is to use a router and an 8’ long straight aluminum guide clamped to the board.  This works since the board is not moving and I’m matching the edge to the guide   I have two routers that would work but one has a plunge base with a flat side that will work very well for following the straight edge.  First, the clear base currently on the router has a large opening required for large diameter bits and here a small ½” diameter bit will be used.  Changing the base to one with a smaller opening gives me more contact surface between the router and the board making the router more stable.  When changing bases, I like to make sure the hole in the base is centered on the router’s drive shaft.  It’s not essential here but if a guide were mounted to the base, it would be.  Centering is done using a centering cone (red arrow) shown in the photo below.  Once the cone centers the base the screws holding it on are tightened.

Here are two views of the setup used to give me a good straight edge using the router.

Before clamping the board in place, the edge is checked and the edge’s crown or high point of the curve gets selected to be the cut face.  The red line in the drawing shows the desired straight edge on a very bowed board.  Here the photo shows a combination square (red arrow) set for the offset needed for the router to trim just a tiny bit off this edge.  The same offset is set at the other end.  Now any bow extending beyond the line in the drawing gets trimmed off resulting is a nice smooth straight edge.

Here everything is set to be routed with the router in position ready to go.  After a pass down the edge, it’s straight and ready to have the rest of the surfacing done so both faces are flat and smooth.  One down and 36 to go.

More surfacing is next where all the boards are surfaced down 1/64” per pass until all the defects are gone and the faces are smooth.  The two photos below show all the cedar pieces with both faces surfaced smooth and with one edge routed straight.  The material is sorted in stacks by thickness because some boards were slightly thicker, flatter or smoother that others and I quit surfacing when both sides were smooth.  On the far-left side the pieces are .55” thick and each stack is .01” thinner until you get to the far-right stack which is jumps by .02” and is .48” thick.  Having different thickness means I can mix and match them for gluing up the main pieces to hit my minimum 1½” thickness plus a little cushion in case a problem shows up when the three pieces get laminated together.

Next Up – Edge Trimming, a Twisted off Screw & Gluing up Rails

Courtyard Gate – #1 Design and Material Selection

In late summer I got an inquiry asking if I was interested in taking a look at repairing or making a replacement for an old wooden courtyard gate that had seen better days.  Since I had nothing planned, we set up a time to meet and do an evaluation of the existing gate shown below.  At first glance it looked like there was just a missing piece along with some rotted infill panels both of which could be repaired.  However, upon closer inspection the rot was not limited to the infill panels but had gotten into the structural elements.  There were also some cracked framing members that when all considered made the gate a candidate for replacement rather than repair.  Fortunately, the jambs attached to the courtyard wall were in good shape and could be reused.

After taking measurements I did some research sending them about 20 different images of potential styles and designs.  That list quickly got narrowed down to a handful and the decision was made to keep the same general style of the gate but add some ornamentation to the infill panels.  More discussion ensued with a decision to add the Zia sun symbol.  In pure form it is a circle with groups of rays pointing in four directions, representing the four seasons, four cardinal directions, or the four stages of life.  It’s from the Zia Pueblo people of New Mexico.  The symbol is featured on the New Mexico state flag and gets applied to all sorts of things.  Everything from clothes, to lights, jewelry and stained glass.

With the general parameters set I made a drawing or should I say several drawings exploring a bunch of different options for using the Zia.  As I say, “The good thing about a custom design is you can have anything you want and the bad thing is you can have anything you want.”  You just have to work through all the options and decide what that is.  Below are some of the options looked at.

The final design or mostly final design settled on is shown below on the left.  The X-ray drawing on the right shows an exploded view and in it the tenons and mortices can be seen.    My guess is that the drawing is about 95% done.  Now there may be some changes regarding the infill panel’s tenons and mortises.  I guess I should say constructions details will change as the build process evolves as they always do.  I also need to double check the gate height and more importantly its width before the cutting gets started.

Because we wanted the gates to be rot resistant that limited us to a few material choices.

• Pressure treated construction lumber is really wet.  For 1½” thick material to dry I would like to allow at least 9 months and that’s so short only because we live in an arid desert climate.  Additionally, when drying it can display all sorts of problems like warp, twist, cup split and kink plus it generally has flaws like knots that are really unacceptable in this application.
• White oak is a possibility but it costs quite a bit and it’s pretty heavy with a specific gravity of about .75.
• Cypress or Redwood are good choices with a specific gravity of about .4.  They are naturally rot resistant; light somewhat expensive and decent material is not available around here.  I could order some but who knows what it would look like which is why I insist on sorting through materials stacks and picking my own pieces out.
• Cedar is a good rot resistant choice with a specific gravity of about .35 but is not available locally in the thickness needed.  However, rough surface cedar fencing boards 5/8” thick by 5 ½” wide and 6’ long are locally available allowing us to sort through and pick good boards.  The down side is they are not dry and will have to set to dry out before using.  We went to the local big box store and spent a few hours sorting through their stock pulling out flat, straight grained material with minimal small defects.  We ended up with 37 pieces after sorting through about 450 boards.  Here they are stacked up to dry with spacers between each board to allow airflow around them to minimize uneven drying.

Right after stacking the boards their moisture content got checked and it registered 45.7% which is really wet with water making up much of the weight of the board.  Here are the moisture meter readings at initial testing on left, two weeks in the center and five weeks on the right with it down to less than 6% which is great.  One advantage to living in the desert where the humidity is frequently in the single digits.  The boards being relatively thin also helped speed the drying process.

After about a month the boards got moved from the wood storage area into the shop.  They then got restacked and checked for any problem children that were no longer flat, straight and true.  Surprisingly, there was only one board that had any real problems.  In the top photo the red arrows indicate how much its warped.  The inset measures it at a little over ¾”.  The bottom photo illustrates how I am trying to remove the warp.  It may work since the board is not completely dry and I am not real confident of success but nothing ventured nothing gained.

One drawback in using cedar fence boards is that they are not surfaced smooth.  The photo shows the roughness of the boards in their current state.  They aren’t as rough as regular rough sawn hardwood planks but not as smooth as needed for gluing or for a finished surface.  The first step will be to run them through a planer taking off the minimum needed to get a surface suitable for gluing.

Next Up – Surfacing & Truing One Edge

File Handles & Case – #9 Finishing & Hardware Installation

Drilling the holes for the hinge pins is next.  The top photo shows where the shims have been reinstalled between the base and the lid along with a pair of .01” thick shims between the beveled edge on the top and base.  With the shims all in the lid and base set in their exact finished location tape is used to hold the top and base in place while the holes get drilled.  Last, shown in the bottom photo is to mark the hinge pin locations and center punch that point for reference when drilling.

This is the setup for drilling.  Everything is clamped square and tight in place so the tip of the drill bit is lined up with the punched center point.  The drill has a blue tape depth marker (red arrow) so when it touches the oak wood face the hole is drilled to the proper depth.

Once the hinge pin holes are drilled the tape holding the lid and base together along with the shims get removed.  Any final light sanding is done and the two pieces are blown off with compressed air in preparation for finishing.  However, before the finish gets applied I need to plug the hinge pin holes to keep the finish out because an unfinished surface is needed for the glue to bond to.

Since the hinge pin holes are 3/16” in diameter I went to cut a piece of 3/16” dowel and found I didn’t have any.  Rather than go and try to buy a dowel that size I decided to just turn some since only two pieces about an inch each are needed.  After cutting a ¼” square piece of soft wood it’s turned to just a bit over 3/16” then 120 grit sand paper is used to put a slight taper on the end until it fits in the hole.  It’s then parted down so there is only a little holding it on then it’s snapped it off.  Making them took way less time than driving to one of the big box places to see if they even had a dowel that size.

Just like the file handles I am going to use Danish Oil Finish (DOF).  This is my typical finish for tool handles, toys or anything that could be exposed to hard use.  DOF is a lot easier to touch up if it’s damaged as compared to the lacquer finish, I use on other items.  Like the file handles three coats of DOF are applied.  This photo shows what the lid looks like when the first coat is applied.  The red arrow points to one of the plugged hinge pin holes.

When getting ready to apply the second coat of DOF I noticed that I had missed making the finger notch in the base to make opening the lid easier.  Fortunately, with this type of finish it’s not a problem.  The notch is added using a small drum in the oscillating drum sander and fine sandpaper.  The photo shows the top of the tool tilted to 30 degrees and the small drum in the process of making the notch.

Here you can easily see the light area where the notch was sanded to the desired depth.  Only other work is to lightly soften the sharp edges with 220 grit sandpaper until they feel right.  When the second and third coats of DOF get applied the color of the notch will match perfectly.

Drilling holes for the feet is next.  A brad point bit is used for drilling since the center point helps me align the bit’s point with the punched hole to accurately locate the hole just where I want it.  The bit also cuts its perimeter with a spur on its outer edge so that helps minimize tear out there.  The bottom left photo shows the brad point drill used on the left and a regular twist drill on the right.  The bottom right photo is a closer view of the drill and hole.

Gluing in the feet is done by using a toothpick to spread a little glue inside the hole then pressing the brass foot in.  To get the feet set with the same amount exposed a spacer with a small hole is set over the foot and a clamp is used to press the foot flush with the spacer.  Installing them this way takes care of any variance in the foot length or the hole depth.

The last work on the case is to install the two brass pivot pins, one of which is shown in the lower left.  Process is to take two drill bits the same size as the pins then use them to align the holes in the lid and base with the stainless-steel spacer in-between.  Cut playing cards are used as spacers to hold the lid in place during the pin insertion.

Next the assembled case is stood on edge and a drill removed.  A drift punch (red arrow) is used to make sure the wood pieces and steel spacer are still aligned.  Using a toothpick glue is carefully put in the case hole but not the lid since the lid needs to freely rotate on the pin.  The pin is then inserted and pushed in partway by hand.  The process is repeated for the other pin.  Last is to take a clamp and press in both pins until they are flush with the lid’s sides.

This is the finished case ready to be used with the just installed brass pivot pin visible at the right.  The photo also shows the continuous flow of grain on the top between the lid and the fixed part of the base.

Here the case is shown with the lid up and with the lid rotated to act as a stand holding the tools upright.

I know that this case may be a bit much for something to hold six detail files along with their handle but I enjoy using good tools and associated items.  I like to think that there is a mental component when using them that helps me do good work.  Besides the wood is from a scrap piece that was split, cupped, bowed along its length and twisted.  The brass rod and stainless-steel washers were left over from other projects and the Danish Oil Finish was on hand.  Last but not least I enjoyed designing and building the piece.  That’s my flawed reasoning and I’m sticking to it. 😊